perceptions of risk factors and mitigating strategies for
TRANSCRIPT
University of Calgary
PRISM: University of Calgary's Digital Repository
Graduate Studies The Vault: Electronic Theses and Dissertations
2016
Perceptions of Risk Factors and Mitigating Strategies
for Water Related Zoonotic Diseases on Small-Scale
Integrated Farms in Vietnam
Le, Quynh Ba
Le, Q. B. (2016). Perceptions of Risk Factors and Mitigating Strategies for Water Related Zoonotic
Diseases on Small-Scale Integrated Farms in Vietnam (Unpublished doctoral thesis). University of
Calgary, Calgary, AB. doi:10.11575/PRISM/27287
http://hdl.handle.net/11023/2862
doctoral thesis
University of Calgary graduate students retain copyright ownership and moral rights for their
thesis. You may use this material in any way that is permitted by the Copyright Act or through
licensing that has been assigned to the document. For uses that are not allowable under
copyright legislation or licensing, you are required to seek permission.
Downloaded from PRISM: https://prism.ucalgary.ca
UNIVERSITY OF CALGARY
Perceptions of Risk Factors and Mitigating Strategies for Water Related Zoonotic Diseases on
Small-Scale Integrated Farms in Vietnam
by
Quynh Ba Le
A THESIS
SUBMITTED TO THE FACULTY OF GRADUATE STUDIES
IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE
DEGREE OF DOCTOR OF PHILOSOPHY
GRADUATE PROGRAM IN VETERINARY MEDICAL SCIENCES
CALGARY, ALBERTA
MARCH, 2016
© Quynh Ba Le 2016
ii
Abstract
Small-scale integrated (SSI) farming is a common agricultural model in Vietnam. However, SSI
farmers use contaminated water, have no access to hygienic latrines, and have very limited
awareness about water quality or environmental sanitation. There are potential interactions in the
context of this model that may result in transmission of water-related zoonotic diseases (WRZD).
The government of Vietnam plans to restructure the agriculture system by 2020. Study of the
model in the context of water and public health can contribute to shaping the future of Vietnam’s
agriculture system. I used a cross-sectional study to develop a socioeconomic profile of SSI
farmers, examine their perceptions of risk factors for transmitting WRZD, assess basic microbial
and related quality of water used on their farms, and explore factors that are associated with on-
farm water quality as well as SSI farmers’ engagement in strategies to reduce transmission of
WRZD. Data were collected from 600 SSI farms in two provinces in North and South Vietnam.
The typical profile of the participating SSI farmers was a 45 year old married individual with two
children, seven years of formal education, low income (c. $1200 p.a.), and nine years farming
experience. Most SSI farmers had basic awareness of avian influenza prevention, but very
limited awareness of WRZDs such as Escherichia coli (E. coli). Water on the majority of SSI
farms had unacceptable levels of E. coli and was significantly associated with SSI farmers’
characteristics and perceptions (e.g., years of farming, number of poultry on a farm, and
perceived self-efficacy in managing livestock). SSI farmers’ characteristics and perceptions (e.g.,
education, income, and perceived self-efficacy in managing livestock) influenced their
engagement in strategies to reduce transmission of WRZD. Future research and policies in
Vietnamese agriculture need to consider a transdisciplinary approach (e.g., EcoHealth) to
increase SSI farmers’ awareness of water public health and their engagement in mitigating
iii
strategies for WRZD transmission. Consideration for the health of farmers, animals, and the
environment should be interwoven with livestock production and be an equally important part of
a holistic integrated SSI farming model.
iv
Acknowledgements
There are many people - from rural communes in Vietnam to various research
organizations in both Vietnam and Canada – to whom I am grateful for their support throughout
my doctoral study program. Their support has helped me this far in my journey/adventure of
further education and career development moving beyond my human medicine and public health
background and into a fascinating, challenging, and holistic research area that involves not only
humans but also animals and their environment – i.e., EcoHealth/One Health/Global Health.
Ultimately, I am grateful for the tremendous guidance and mentorship from my primary
supervisor, Dr. David Hall, not only throughout my study in Canada and in the field study in
Vietnam but also my life in Canada. His experiences, patience, support, and faith in me has
empowered me to strive to become a transdisciplinary development researcher. I would not have
completed my study without my co-supervisor, Dr. Susan Cork, who provided advice,
leadership, and vision that helped me position my study and career. I am grateful for the advice
and support from my committee members Drs. Carl Ribble, Margaret Russell, and Sylvia
Checkley. Their expertise, knowledge, and extremely patient feedback have shaped my research
thinking and skills.
I am also grateful for the support from Dr. Jeff Davidson, who procured a seed grant that
enabled me to connect with local researchers and pilot my study tools. I highly appreciate the
help of staff members at UCVM, Dr. James Cross, Dr. John Matyas, Dr. Tara Christie, Lorraine
Toews, Katie Douglas, Ingrid Middleton, Kasia Judycki, Lisa Clough, Jess Nakaska, and my
fellow graduate students who helped me back on track. I would like to thank Chris Morley for
supporting my research and connecting to his network researcher in Vietnam.
v
Especially, I would like to extend my sincere thanks to the 600 farmers in Thai Binh and
An Giang who participated in my field data collection and enthusiastically shared their thoughts,
ideas, and concerns, which were the foundation for my research. A huge benefit for me was the
unique opportunity to work with and learn from the stakeholders and partners across disciplines
and sectors in Vietnam and Canada. The list of stakeholders, partners, and staff who helped me
during the field data collection is long and my apologies to those who I am not able to name in
the following: Drs. Vu Hoang Hoa, Tran Van On (Hanoi Water Resources University), Dr.
Nguyen Viet Hung, Dr. Pham Duc Phuc, Luu Quoc Tuan, Lam Thi Binh, Dang Van Chinh,
Nguyen Duy Tien, and Le Thanh Huyen (Hanoi School of Public Health), Drs. Nguyen Dinh
Minh, Nguyen Viet Khong, Nguyen Lan Anh, Pham Thi Ngoc (National Institute of Veterinary
Research), Dr. Phan Thi Van (Research Institute of Aqua-agriculture No 1), Dr. Nguyen Van
Duc and his team in Thai Binh, Drs. Huynh Van Nen and his team in An Giang, Dr. Pham
Quang Khai (NUSA Vietnam), and Kimberly Schiefer (Bluewater BioSciences Inc, Canada).
Their support and collaboration were extremely helpful for me to complete my data collection in
Vietnam. This dissertation was professionally copy edited by Kathleen McWilliams, with
permission from my supervisor and following the thesis editing regulations of the Faculty of
Graduate Studies, University of Calgary.
My work would not have been possible without enabling support from different funders
across Canada. I highly appreciate funding support for my research from IDRC Doctoral
Research Awards/International Development Research Centre (IDRC), Stars in Global
Health/Grand Challenges Canada (GCC), Seed Grant/Canada Excellence Research Chair
(CERC), and the Building Ecohealth Capacity in Asia (BECA) project that was implemented by
the University of Calgary and administered by Veterinarians Without Borders Canada. The
vi
UCVM Graduate Student Entrance Awards and the DEPH Studentship and Fellowship Funds
from the Department of Ecosystem and Public Health, Faculty of Veterinary Medicine were
crucial for me to start and maintain my study.
I thank my relatives and friends in Vietnam and Canada for their love, support,
understanding, and patience during my extended long commitment to PhD study. I am proud of
my grandparents, parents, brothers, and sisters for believing in me and supporting me
unconditionally. I am extremely lucky for the unconditional love and support from my wife and
my kids – Fish, Chip, and Bob – who have been always with me through my difficult times, kept
me optimistic and on my feet when I was about to fall apart.
Thank you to all of you without which this work would not have been possible!
vii
Dedication
“May be tomorrow I will settle down, until tomorrow I will just keep moving on!”
From my favorite TV series “Littlest Hobo”
viii
Table of Contents
Abstract ............................................................................................................................... ii Acknowledgements ............................................................................................................ iv
Dedication ......................................................................................................................... vii Table of Contents ............................................................................................................. viii List of Tables ................................................................................................................... xiv List of Figures and Illustrations .........................................................................................xx List of Symbols, Abbreviations and Nomenclature ......................................................... xxi
Glossary .......................................................................................................................... xxii
CHAPTER ONE: INTRODUCTION ................................................................................24
1.1. Overview of the Research ......................................................................................24 1.2. Background ............................................................................................................26
1.2.1. Small-Scale Integrated farming as an important agricultural model in
Vietnam ............................................................................................................26
1.2.2. Rural water and public health issues may challenge the SSI farming model.
28
1.2.3. SSI farmers’ perceptions about risk factors for transmitting WRZD may be
associated with their engagement in mitigating strategies to reduce transmitting
WRZD ..............................................................................................................29
1.3. Key Research Questions ........................................................................................30 1.4. Research Methods ..................................................................................................31
1.5. A Description of Chapters in the Dissertation .......................................................32
1.6. Connection Between Chapters ...............................................................................37
References ......................................................................................................................38
CHAPTER TWO: DEVELOPING A SOCIOECONOMIC PROFILE OF SMALL-SCALE
INTEGRATED (SSI) FARMS IN THE PROVINCES OF THAI BINH AND AN
GIANG IN VIETNAM .............................................................................................48 2.1. Introduction ............................................................................................................48
2.1.1. Overview of Small-Scale Integrated (SSI) farming in Vietnam .............48 2.1.2. Rationale for the research .......................................................................50
2.1.3. SSI farming and risk factors for transmission of HPAI ..........................52 2.1.4. Research questions and objectives ..........................................................53
2.2. Methods..................................................................................................................57
2.2.1. Research design.......................................................................................57
2.2.1.1. Methods and phases of the research ...................................................57 2.2.1.2. Research areas ....................................................................................58 2.1.2.3. Selection criteria ................................................................................59 2.2.1.4. Sampling design: ................................................................................61 2.2.1.5. Variables ............................................................................................62
2.2.2. Pilot study ...............................................................................................63 2.2.3. Data collection for the full research study ..............................................63 2.2.4. Analysis of the full field survey’s data ...................................................66
2.3. Results ....................................................................................................................68 2.3.1. Demographics .........................................................................................68
ix
2.3.1.1. Age, years of farming, years of attending school, and household size68 2.3.1.2. Gender and gender role ......................................................................69
2.3.2. On-farm production.................................................................................72 2.3.2.1. Fish production ..................................................................................72
2.3.2.2. Livestock production .........................................................................73 2.3.2.3. On-farm production other than fish and livestock production ...........73 2.3.2.4. Occupation other than farming among participating farmers ............73
2.3.3. Farm economics ......................................................................................75 2.3.3.1. Income of SSI farmers .......................................................................75
2.3.3.2. Contribution to total farm income from various kinds of agriculture
production ...............................................................................................76 2.4. Discussion ..............................................................................................................78
2.4.1. SSI farming practice and its profile in the provinces of Thai Binh and An
Giang ................................................................................................................78 2.4.2. SSI farming and EcoHealth.....................................................................79
2.5. Conclusions ......................................................................................................83 References ......................................................................................................................85
CHAPTER THREE: SMALL-SCALE INTEGRATED FARMERS’ PERCEPTIONS
ABOUT RISK FACTORS FOR TRANSMITTING WATER-RELATED ZOONOTIC
DISEASE (WRZD) IN THE PROVINCES OF THAI BINH AND AN GIANG IN
VIETNAM ..............................................................................................................100 3.1. Introduction ..........................................................................................................100
3.3.1. Rationale for studying farmers’ perceptions of risk factors for transmission
of WRZD .......................................................................................................101
3.1.2. Research question and objective ...........................................................102 3.2. Methods................................................................................................................103
3.3. Results ..................................................................................................................109 3.3.1. Perceptions of water quality in the villages ..........................................109 3.3.2. Perceptions of threat to health and/or wellbeing ...................................112
3.3.2.1. Perceived susceptibility to diseases .................................................113 3.3.2.2. Perceived severity of diseases ..........................................................114
3.3.3. Expectations ..........................................................................................117 3.3.3.1. Perceptions of barriers to taking mitigating actions ........................117
3.3.3.2. Perceptions of benefits of taking or not taking mitigating actions ..117 3.3.3.3. Perceptions about self-efficacy in performing mitigating actions ...121
3.3.4. Cues to actions ......................................................................................124 3.4. Discussion ............................................................................................................125 3.5. Conclusions ..........................................................................................................130 References ....................................................................................................................132
CHAPTER FOUR: WATER QUALITY AND PUBLIC HEALTH AMONG SMALL-
SCALE INTEGRATED (SSI) FARMS IN THE PROVINCES OF THAI BINH AND
AN GIANG IN VIETNAM ....................................................................................138 4.1. Introduction ..........................................................................................................138
4.1.1. Rationale ...............................................................................................138 4.1.2. Research questions and hypotheses ......................................................141
x
4.1.2.1. Research questions: ..........................................................................141 4.1.2.2. Hypotheses: ......................................................................................141
4.2. Methods................................................................................................................142 4.3. Results: .................................................................................................................147
4.3.1. Sources of water and frequencies of use in participating farms ...........147 4.3.2. Results of on-farm water quality assessed using the Vietnamese national
laboratory protocol .........................................................................................148 4.3.3. Quality of on-farm water assessed by community health/veterinary workers
(CHW/CVWs)................................................................................................149
4.3.4. Factors associated with the auality of on-farm water ...........................149 4.3.4.1. Description of variables used for modelling ....................................150 4.3.4.2. MLR Model - E. coli cfu in sources of water used for drinking (DV) vs. a
set of IVs ...............................................................................................150 4.3.4.3. MLR Model - E. coli cfu in water sources for domestic use (DV) vs. a set
of IVs.....................................................................................................157
4.3.4.4. SUR Model - E. coli cfu in sources of on-farm water used for domestic
purposes ................................................................................................158
4.4. Discussion ............................................................................................................164 4.4.1. Sources and frequency of use of on-farm water for drinking and domestic
purposes .........................................................................................................164
4.4.2. Low quality of on-farm water for both drinking and domestic purposes167 4.4.3. Associated factors of on-farm water quality .........................................171
4.4.4. Rural water and its regulatory framework on SSI farms in Vietnam....174 4.5. Conclusions ..........................................................................................................177
References ....................................................................................................................179
CHAPTER FIVE: FACTORS ASSOCIATED WITH SMALL-SCALE INTEGRATED (SSI)
FARMERS' ENGAGEMENT IN MITIGATING STRATEGIES TO REDUCE
TRANSMISSION OF WATER-RELATED ZOONOTIC DISEASES (WRZD) IN THE
PROVINCES OF THAI BINH AND AN GIANG IN VIETNAM ........................189
5.1. Introduction ..........................................................................................................189 5.1.1. Rationale ...............................................................................................189
5.1.2. Research questions and hypothesis .......................................................193 5.1.2.1. Research questions ...........................................................................193
5.1.2.2 Research hypothesis .........................................................................194 5.2. Methods................................................................................................................194
5.2.1. Methods used to identify SSI farmers’ engagement in mitigating strategies
to reduce the transmission of WRZD.............................................................194 5.2.2. Methods used to explore factors that are associated with SSI farmers’
engagement in mitigating strategies...............................................................196 5.3. Results ..................................................................................................................198
5.3.1. SSI farmers’ actions and engagement in mitigating strategies to reduce
WRZD transmission.......................................................................................199 5.3.1.1 Livestock management (LM) actions ..............................................199 5.3.1.2. Source water protection (SWP) actions ...........................................199 5.3.1.3. Water storage treatment and distribution (WSTD) actions ..............200 5.3.1.4. Point of use/household (POU) actions .............................................202
xi
5.3.1.5. SSI farmers’ engagement in mitigating strategies to reduce WRZD
transmission ..........................................................................................203 5.3.2. Factors associated with SSI farmers’ engagement in the mitigating strategies
to reduce transmission of WRZD ..................................................................204
5.3.2.1. Results of probit analyses of SSI farmers’ engagement in mitigating
strategies to reduce WRZD transmission ..............................................204 5.3.2.2. Marginal probability effects of the factors that are associated with SSI
farmers’ engagement in mitigating strategies to reduce WRZD transmission
205
5.4. Discussion ............................................................................................................212 5.4.1. SSI farmers’ choices of actions and their engagement in mitigating strategies
to reduce the transmission of WRZD.............................................................212
5.4.1.1. Wide differences in choices of specific actions to mitigate WRZD
transmission ..........................................................................................212 5.4.1.2. SSI farmers’ broad engagement in mitigating strategies to reduce WRZD
transmission ..........................................................................................214 5.4.2. Factors associated with SSI farmers’ engagement in mitigating strategies to
reduce transmission of WRZD .......................................................................214 5.4.2.1. Factors associated with SSI farmers’ engagement in LM as a mitigating
strategy to reduce WRZD transmission ................................................216
5.4.2.2. Factors associated with SSI farmers’ engagement in SWP as a mitigating
strategy to reduce WRZD transmission ................................................218
5.4.2.3. Factors associated with SSI farmers’ engagement in POU as a mitigating
strategy to reduce WRZD transmission ................................................219
5.4.2.4. Factors associated with SSI farmers’ engagement in WSTD as a
mitigating strategy to reduce WRZD transmission ...............................220
5.4.2.5. Common factors associated with SSI farmers’ engagement across four
mitigating strategies to reduce WRZD transmission ............................221 5.5. Conclusions ..........................................................................................................222
References ....................................................................................................................225
CHAPTER SIX: CONCLUSION ....................................................................................231
6.1. Overview ..............................................................................................................231 6.2. Summary of the Literature Review ......................................................................232
6.3. Research Questions, Hypotheses, and Methods ..................................................233 6.4. Contribution to Knowledge..................................................................................235
6.4.1. An understanding of SSI farming model in the context of rural water and
public health ...................................................................................................235 6.4.1.1. A socioeconomic profile of the SSI farming practice in Thai Binh and An
Giang .....................................................................................................236 6.4.1.2. SSI farmers’ limited perceptions of risk factors for transmission of
WRZD ...................................................................................................238 6.4.1.3. SSI farmers’ engagement in mitigating strategies to reduce transmission
of WRZD...............................................................................................239 6.4.1.4. Low quality of water used on SSI farms for both drinking and domestic
purposes ................................................................................................239
xii
6.4.2. An understanding of factors associated with on-farm water quality and SSI
farmers’ engagement in mitigating strategies to reduce the transmission of WRZD
240 6.4.2.1. Factors that are associated with on-farm water quality ...................240
6.4.2.2. Factors that are associated with SSI farmers’ engagement in mitigating
strategies to reduce the transmission of WRZD ....................................242 6.5. Limitations ...........................................................................................................242 6.6. Overall Recommendations ...................................................................................244 6.7. Conclusions/Final Thoughts ................................................................................247
References ....................................................................................................................248
APPENDIX A: THEORETICAL MODEL .....................................................................251
APPENDIX B: QUESTIONNAIRES USED DURING THE FIELD STUDY ..............252 B.1. Questionnaires in Vietnamsese: ...........................................................................252 B.2. Questionnaires in English: ...................................................................................269
APPENDIX C: CONSENT SCRIPTS .............................................................................285
C.1. Consent Scripts in Vietnamese .............................................................................285 C.2. Consent Scripts in English ...................................................................................290
APPENDIX D: ETHICS APPROVALS .........................................................................294 D.1. University of Calgary Ethics ID: REB13-0470 ...................................................294 D.2. Hanoi School of Public Health Ethics Approval: No. 151/2013/YTCC_HD3 ....294
D.3. Hanoi School of Public Health Ethics Approval: No. 008/2014/YTCC_HD3 ....294
APPENDIX E: CHAPTER TWO’S APPENDIX............................................................295
APPENDIX F: CHAPTER FOUR’S APPENDICES ......................................................300 F.1. Brief Ecology of Escherichia coli and WHO’s Recommendations for Assessing Rural
Water Quality .......................................................................................................300 F.2. Protocols Used for Assessing On-Farm Water Quality ........................................303
F.2.1. Protocol for Laboratories to Assess Basic Microbial and Related Water Quality
........................................................................................................................303
F.2.2. Protocol for Community Health/Veterinary Workers to Quantify E. coli ...305 F.3. Detailed Model Building Strategies for MLR and SUR .......................................307
F.3.1. Modeling building strategies ........................................................................307
F.3.2. Multiple linear regression (MLR) model .....................................................307 F.3.3. Seemingly Unrelated Regression (SUR) Model ..........................................309 F.3.4. Detailed Results of Sources of Water Used and Corresponding Frequencies of
Use among SSI Farmers in the Provinces of Thai Binh and An Giang .........315
F.4. Detailed Results of Basic Microbial and Related Indicators of On-Farm Water .316 F.4.1. Quality of On-Farm Water Used for Drinking .............................................316 F.4.2. Quality of On-Farm Water Used for Domestic Purposes ............................317 F.4.3. Detailed Results of Drinking Water Quality Assessed by CHW/CVW using
ColiplateTM .....................................................................................................318 F.5. Detailed Results of MLR Models for E. coli in Sources of Water On Farms ......319
xiii
F.6. Detailed Results of SUR Model for E. coli in Sources of Water On Farms ........331
APPENDIX G: CHAPTER FIVE’S APPENDICES .......................................................334 G.1. Summary Statistics of Variables Used in the Probit Models ...............................334 G.2. Summary Statistics of SSI Farmers’ Perceptions Related Variables Used in the Probit
Models..................................................................................................................335 Table G.2. Summary Statistics of SSI Farmers’ Perceptions Related Variables Used in Probit
Models to Explore Factors Associated with SSI Farmers’ Engagement in Mitigating Strategies to
Reduce WRZD Transmission ..........................................................................................335 References ....................................................................................................................336
xiv
List of Tables
Table 2.1. Risk Factors for Transmission of Highly Pathogenic Avian Influenza (HPAI) and
Their Possible Connection with Small-Scale Integrated (SSI) Farming .............................. 55
Table 2.2. Demographics of Small-Scale Integrated (SSI) Farm Study Participants in the
Provinces of Thai Binh and An Giang, Vietnam .................................................................. 68
Table 2.3. Types of School Attended by Small-Scale Integrated (SSI) Farmers in the
Provinces of Thai Binh and An Giang, Vietnam .................................................................. 69
Table 2.4. Gender of the Participating Farmers in the Provinces of Thai Binh and An Giang
in Vietnam and Their Relationship with Other Household Members .................................. 71
Table 2.6. Mean Quantity of Young Fish Produced Per Year on Small-Scale Integrated (SSI)
Farms in the Provinces of Thai Binh and An Giang, Vietnam, Stratified for Education
and Income ............................................................................................................................ 72
Table 2.7. Annual Livestock Production on Small-Scale Integrated Farms (SSI) in the Thai
Binh and An Giang Provinces in Vietnam ............................................................................ 73
Table 2.8. On-farm Production Other Than Fish and Livestock on Small-Scale Integrated
(SSI) Farms in the Thai Binh and An Giang Provinces in Vietnam ..................................... 74
Table 2.9. Occupations Other Than Farming Among Small-Scale Integrated (SSI) Farmers in
the Provinces of Thai Binh and An Giang, Vietnam ............................................................ 75
Table 2.10. Income of Small-Scale Integrated (SSI) Farmers in the Provinces of Thai Binh
and An Giang, Vietnam ........................................................................................................ 76
Table 2.11a. Farmers’ Perceptions of the Importance of Contribution to Total Farm Income
from Various Types of Agricultural Production on Small-Scale Integrated (SSI) Farms
in the Province of Thai Binh, Vietnam ................................................................................. 77
Table 2.11b. Farmers’ Perceptions of the Importance of Contribution to Total Farm Income
from Various Types of Agricultural Production on Small-Scale Integrated (SSI) Farms
in the Province of An Giang, Vietnam .................................................................................. 77
Table 3.1. Perceived General Condition of the Water Environment in the Villages of Small-
Scale Integrated (SSI) Farmers in the Provinces of Thai Binh and An Giang, Vietnam .... 110
Table 3.2. Perceived Harm to Health When Using Different Sources of Water among Small-
Scale (SSI) Farmers in the Provinces of Thai Binh and An Giang, Vietnam ..................... 111
Table 3.3. Factors Influencing Small-Scale Integrated (SSI) Farmers’ Thoughts about Water
Quality in the Provinces of Thai Binh and An Giang, Vietnam ......................................... 112
xv
Table 3.4. Perceived Susceptibility to Highly Pathogenic Avian Influenza (HPAI), Diarrhoea,
Coliform Bacteria, and Parasites from Untreated Water Sources among Small-Scale
Integrated (SSI) Farmers in the Provinces of Thai Binh and An Giang, Vietnam ............. 115
Table 3.5. Perceived Severity of Diseases from Highly Pathogenic Avian Influenza (HPAI),
Diarrhoea, Coliform Bacteria, and Parasites among Small-Scale Integrated (SSI)
Farmers in the Provinces of Thai Binh and An Giang, Vietnam ........................................ 116
Table 3.6. Perceived Barriers to Taking Actions to Mitigate Transmission of WRZD among
Small-Scale Integrated (SSI) Farmers in the Provinces of Thai Binh and An Giang,
Vietnam ............................................................................................................................... 118
Table 3.7. Perceived Benefits that Encourage Small-Scale (SSI) Farmers in the Provinces of
Thai Binh and An Giang, Vietnam, to Either Take or Not Take Actions to Mitigate
Transmission of WRZD ...................................................................................................... 119
Table 3.8. Perceptions of Water and Wastewater Value for Livestock Production among
Small-Scale Integrated (SSI) Farmers in the Provinces of Thai Binh and An Giang,
Vietnam ............................................................................................................................... 120
Table 3.9. Levels of Importance for Water Use Issues in Reducing Transmission of WRZD
among Small-Scale Integrated (SSI) Farmers in the Provinces of Thai Binh and An
Giang, Vietnam ................................................................................................................... 122
Table 3.10. Perceived Self-efficacy to Perform Mitigating Actions among Small-Scale
Integrated Farmers in Thai Binh and An Giang, Vietnam .................................................. 123
Table 4.1. Mean Frequency of Use for On-Farm Water Used for Drinking and Domestic
Purposes in Both Provinces of Thai Binh and An Giang, Vietnam .................................... 151
Table 4.2. Comparison of Mean pH, Turbidity, and E. coli cfu of on-Farm Water Used for
Drinking, Assessed Using Samples Tested by National Laboratories, in the Provinces of
Thai Binh and An Giang, Vietnam ..................................................................................... 152
Table 4.3. Comparison of Mean pH, Turbidity, E. coli cfu of On-Farm Water Used for
Domestic Purposes, Assessed Using Samples Tested by National Laboratories, in the
Provinces of Thai Binh and An Giang, Vietnam ................................................................ 153
Table 4.4. Prevalence of E. coli in On-Farm Water Tests (using ColiplateTM) Used for
Drinking by Source in Thai Binh and An Giang, Vietnam ................................................. 154
Table 4.5. Summary Descriptive Statistics for the Dependent E. coli Variables Used in the
Regression Models .............................................................................................................. 154
Table 4.6a. Summary Statistics for the Independent Variables (IVs) Considered for Use in
Regression Models .............................................................................................................. 155
xvi
Table 4.6b. Summary Statistics for the Independent Variables (IVs) Considered for Use in
Regression Models (cont’d) ................................................................................................ 156
Table 4.7a. Summary of the Best MLR Model Fittings for E. coli cfu in Each of the Four
Sources of Drinking Water in the Provinces of Thai Binh and An Giang, Vietnam .......... 159
Table 4.7b. Summary of the Best MLR Model Fittings for E. coli cfu in Each of the Four
Sources of Drinking Water in the Provinces of Thai Binh and An Giang, Vietnam
(cont’d) ................................................................................................................................ 160
Table 4.8. Summary of the MLR Best Model Fittings for E. coli in Pond Water and River
Water for Domestic Purposes in the Provinces of Thai Binh and An Giang, Vietnam ...... 161
Table 4.9a. SUR Model Estimation of the Two Equations of E. coli cfu in Water Used for
Drinking with Cross-Equation Constraints ......................................................................... 162
Table 4.9b. SUR Model Estimation of the Two Equations of E. coli cfu in Water Used for
Drinking with Cross-Equation Constraints (cont’d) ........................................................... 162
Table 4.9c. SUR Model Estimation of the Two Equations of E. coli cfu in Water Used for
Drinking with Cross-Equation Constraints (cont’d) ........................................................... 163
Table 5.1. Small-Scale Integrated (SSI) Farmers’ Mitigating Strategies and Their
Corresponding Actions in Reducing Risk Factors for Water-Related Zoonotic Disease
(WRZD) Transmission (*) .................................................................................................. 196
Table 5.2. Summary Statistics of Variables Indicating SSI Farmers’ Engagement in
Mitigating Strategies for WRZD Transmission .................................................................. 204
Table 5.3a. Maximum Likelihood Estimates of Factors that are Associated with Small-Scale
Integrated (SSI) Farmers’ Engagement in Mitigating Strategies of Water Related
Zoonotic Diseases (WRZD) Transmission Using Probit Analyses .................................... 206
Table 5.3b. Maximum Likelihood Estimates of Factors that are Associated with Small-Scale
Integrated (SSI) Farmers’ Engagement in Mitigating Strategies of Water Related
Zoonotic Diseases (WRZD) Transmission Using Probit Analyses (cont’d.) ..................... 207
Table 5.3c. Maximum Likelihood Estimates of Factors that are Associated with SSI Farmers’
Engagement in Mitigating Strategies of Water Related Zoonotic Diseases (WRZD)
Transmission using Probit Analyses (cont’d.) .................................................................... 208
Table 5.4. Predicted Frequency of Mitigating Strategies ........................................................... 209
Table 5.5a. Marginal Effects (at Means) of Factors Associated with SSI Farmers’
Engagement in Mitigating Strategies for WRZD Transmission Using Probit Analyses .... 210
xvii
Table 5.5b. Marginal Effects (at Means) of Factors that are Associated with Small-Scale
Integrated (SSI) Farmers’ Engagement in Mitigating Strategies of WRZD Transmission
Using Probit Analyses ......................................................................................................... 211
Table E.1. Annual Fish Production on Small-Scale (SSI) Farms in the Provinces of Thai Binh
and An Giang in Vietnam ................................................................................................... 296
Table E.2. Quantity and Kilograms of Fish Produced per Year on Small-Scale Integrated
(SSI) Farms in the Provinces of Thai Binh and An Giang, Vietnam (without outliers) ..... 297
Table E.3. Quantity of Young Fish Produced per Year on Small-Scale Integrated (SSI) Farms
in the Provinces of Thai Binh and An Giang in Vietnam Stratified over Education and
Income (with outliers) ......................................................................................................... 299
Table F.1a. Pairwise Correlation between Independent Variables used for Regression Models 311
Table F.1b. Pairwise Correlation between Independent Variables used for Regression Model
(cont’d) ................................................................................................................................ 312
Table F.1c. Pairwise Correlation between Independent Variables used for Regression Models
(cont’d) ................................................................................................................................ 313
Table F.1d. Pairwise Correlation between Independent Variables used for Regression Models
(cont’d) ................................................................................................................................ 314
Table F.2. Mean Frequency of Use of On-Farm Water for Drinking in the Provinces of Thai
Binh and An Giang, Vietnam .............................................................................................. 315
Table F.3. Mean Frequency of Use of On-Farm Water for Domestic Purposes in the
Provinces of Thai Binh and An Giang, Vietnam ................................................................ 316
Table F.4. Mean pH, Turbidity, and E. coli cfu of On-Farm Water Used for Drinking in the
Provinces of Thai Binh and An Giang, Vietnam ................................................................ 317
Table F.5. Mean pH, Turbidity, and E. coli cfu of On-Farm Water Used for Domestic
Purposes in the Provinces of Thai Binh and An Giang, Vietnam ....................................... 318
Table F.6. Mean MPN a of E. coli cfu Using ColiplateTM Tests in On-Farm Water for
Drinking by Sources and by Number of Incubation Day in the Provinces of Thai Binh
and An Giang, Vietnam ...................................................................................................... 318
Table F.7a. Summary of the Best MLR Model Fittings for E. coli cfu in Stored Rainwater for
Drinking by Selection Procedures in the Provinces of Thai Binh and An Giang, Vietnam 319
Table F.7b. Summary of the Best MLR Model Fittings for E. coli cfu in Stored Rainwater for
Drinking by Selection Procedures in the Provinces of Thai Binh and An Giang, Vietnam
(cont’d) ................................................................................................................................ 320
xviii
Table F.8a. Summary of the Best MLR Model Fittings for E. coli cfu in Pipe Water for
Drinking by Selection Procedures in the Provinces of Thai Binh and An Giang
Provinces, Vietnam ............................................................................................................. 321
Table F.8b. Summary of the Best MLR Model Fittings for E. coli cfu in Pipe Water for
Drinking by Selection Procedures in the Provinces of Thai Binh and An Giang, Vietnam
(cont’d) ................................................................................................................................ 322
Table F.9. Summary of the Best MLR Model Fittings for E. coli cfu in Well Water for
Drinking by Selection Procedures in the Provinces of Thai Binh and An Giang, Vietnam 323
Table F.10. Summary of the Best MLR Model Fittings for E. coli cfu in Well Water for
Drinking by Selection Procedures in the Provinces of Thai Binh and An Giang, Vietnam 324
Table F.11a. Summary of the Best MLR Model Fittings for E. coli cfu in River Water for
Drinking by Selection Procedures in the Provinces of Thai Binh and An Giang
Provinces, Vietnam ............................................................................................................. 325
Table F.11b. Summary of the Best MLR Model Fittings for E. coli cfu in River Water for
Drinking by Selection Procedures in the Provinces of Thai Binh and An Giang, Vietnam
(cont’d) ................................................................................................................................ 326
Table F.12a. Summary of the Best MLR Model Fittings for E. coli cfu in River Water for
Domestic Purposes by Selection Procedures in the Provinces of Thai Binh and An
Giang, Vietnam ................................................................................................................... 327
Table F.12b. Summary of the Best MLR Model Fittings for E. coli cfu in River Water Used
for Domestic Purposes by Selection Procedures in the Provinces of Thai Binh and An
Giang, Vietnam (cont’d) ..................................................................................................... 328
Table F.13a. Summary of the Best MLR Model Fittings for E. coli in Pond Water for
Domestic Purposes by Selection Procedures in the Provinces of Thai Binh and An
Giang, Vietnam ................................................................................................................... 329
Table F.13b. Summary of the Best MLR Model Fittings for E. coli in Pond Water for
Domestic Purposes by Selection Procedures in the Provinces of Thai Binh and An
Giang, Vietnam (cont’d) ..................................................................................................... 330
Table F.14. SUR Model Estimation of the Two Equations of E. coli cfu in Water Used for
Drinking .............................................................................................................................. 331
Table F.15. SUR Model Estimation of the Two Equations of E. coli cfu in Water Used for
Drinking (cont'd) ................................................................................................................. 331
Table F.16. SUR Model Estimation of the Two Equations of E. coli cfu in On-Farm Water .... 332
xix
Table G.1. Summary Statistics of Demographic Variables Used in Probit Models of Factors
Associated with Small-Scale Integrated (SSI) Farmers’ Engagement in Mitigating
Strategies to Reduce WRZD Transmission ........................................................................ 334
xx
List of Figures and Illustrations
Figure 1.1. Water-Related Zoonotic Disease Interactions in the Water Environment ................. 25
Figure 1.2. Dissertation’s Outline ................................................................................................. 33
Figure 2.1. Chapter Two’s Outline ............................................................................................... 56
Figure 2.2. Research Areas – The Provinces of Thai Binh and An Giang in Vietnam ................ 60
Figure 3.1. Chapter Three’s Outline ........................................................................................... 104
Figure 3.2. Triggers/Cues to Taking Mitigating Actions against Transmission of Water
Related Zoonotic Diseases (WRZD)................................................................................... 124
Figure 4.1. Chapter Four’s Outline ............................................................................................. 143
Figure 5.1. Outline of Chapter Five ............................................................................................ 191
Figure 5.2. Livestock Management Actions ............................................................................... 201
Figure 5.3. Source Water Protection Actions ............................................................................. 201
Figure 5.4. Water Storage, Treatment, and Distribution Actions ............................................... 202
Figure 5.5. Point of Use/Household Actions .............................................................................. 203
Figure F.1. The Coliform Group ................................................................................................. 302
xxi
List of Symbols, Abbreviations and Nomenclature
Abbreviations Definition
AI Avian Influenza
AIV Avian Influenza Viruses
CHWs Community Health Workers
CVWs Community Veterinary Workers
CHWs/CVWs Community Health Workers and Community
Veterinary Workers
DVs Dependent Variables
E. coli Escherichia Coli
EPA United States Environmental Protection Agency
HPAI Highly Pathogenic Avian Influenza
IVs Independent Variables
MDG Millennium Development Goals
MoC Ministry of Construction
MoH Ministry of Health
MARD Ministry of Agriculture and Rural Development
MF Membrane Filtration
MPN Most Probable Numbers
MLR Multiple Linear Regression
NTU Nephelometric Turbidity Units
ISO International Organization for Standardization
OECD Organizations for Economic Cooperation and
Development
pH A numeric scale used to specify the acidity or
alkalinity of on-farm water
SSI Small-Scale Integrated
SUR Seemingly Unrelated Regression
TSA Trypone Soy Agar
UN United Nations
UNICEF United Nations Children’s Funds
WB World Bank
WHO World Health Organization
WRZD Water Related Zoonotic Diseases
xxii
Glossary
Association Association is a statistical dependence between 2 variables, and
indicates the degree to which an outcome is higher or lower in
those with or without exposure. An association can also be
referred to as a link, relationship or correlation (Oxford
Handbook of Epidemiology for Clinicians, 2012)
Cues/Triggers to Action Strategies to activate “readiness” condition (Health Belief
Model) or stimuli needed to trigger the decision-making
process to accept a recommended health action:
http://www.utwente.nl/cw/theorieenoverzicht/Theory%20Clust
ers/Health%20Communication/Health_Belief_Model/
Domestic water Domestic water is defined as “water used for domestic use but
not for direct drinking or processing food” (QCVN 02:
2009/BYT - National Technical Regulation on the Domestic
Water Quality, Department of Preventive Medicine and
Environment, Ministry of Health – Vietnam)
Drinking water Drinking water is defined as “water used for direct drinking or
processing food” (QCVN 01: 2009/BYT - National Technical
Regulation on the Drinking Water Quality, Department of
Preventive Medicine and Environment, Ministry of Health –
Vietnam)
Mitigation A mitigating strategy of water related zoonotic disease
(WRZD) transmission is a plan (e.g., a plan for managing
livestock) that includes one or more activities a small-scale
integrated (SSI) farmer could choose to reduce the transmission
of WRZD. An example of a mitigating strategy is good
livestock management which may include actions or tactics
(e.g., vaccinate domestic animals and/or treat animal waste at
source) that SSI farmers could choose to reduce WRZD
transmission (adapted from World Health Organization)
Perceived susceptibility One’s opinion of chances of getting a condition (Health Belief
Model):
http://www.utwente.nl/cw/theorieenoverzicht/Theory%20Clust
ers/Health%20Communication/Health_Belief_Model/
Perceived severity One’s opinion on how serious a condition and its consequences
are condition (Health Belief Model):
http://www.utwente.nl/cw/theorieenoverzicht/Theory%20Clust
ers/Health%20Communication/Health_Belief_Model/
Perceived benefits One’s belief in the efficacy of the advised action to reduce risk
or seriousness of an impact condition (Health Belief Model):
http://www.utwente.nl/cw/theorieenoverzicht/Theory%20Clust
ers/Health%20Communication/Health_Belief_Model/
Perceived barriers One’s opinion of the tangible and psychological costs of the
advised action condition (Health Belief Model):
xxiii
http://www.utwente.nl/cw/theorieenoverzicht/Theory%20Clust
ers/Health%20Communication/Health_Belief_Model/
Risk factors Something that increases a person’ chance of getting a disease.
This definition assumes that there is a causal relationship rather
than simply a statistical association. Risk factors are often
described as modifiable (e.g., environmental exposures and
behaviours) or fixed (e.g. age or ethnicity) (Ward, Toledano,
Shadick, Davies, and Elliott, 2012)
Self-efficacy Confidence in one’s ability to take action condition (Health
Belief Model):
http://www.utwente.nl/cw/theorieenoverzicht/Theory%20Clust
ers/Health%20Communication/Health_Belief_Model/
Small-scale integrated
model/farming
Small-scale integrated (SSI) farming is defined as farming
practices in which farmers combine some types of livestock and
either directly or indirectly depend on agriculture.
VAC model A traditional small-scale integrated farming model in which:
V= Vuon (i.e., Vegetable garden); A= Ao (i.e., Fish pond), C=
Chuong (i.e., Livestock housing)
Water related zoonotic
diseases (WRZD)
Diseases in humans that come from animals and related to
water such as Highly Pathogenic Avian Influenza, some
parasites, or diseases caused by a range of bacteria (e.g.,
pathogenic strains of Escherichia coli). Criteria for determining
WRZD include: 1) the pathogen must spend part of its life
cycle within one or more animal species; 2) it is probable or
conceivable that some life stage of the pathogen will enter
water; and 3) transmission of the pathogen from animals to
humans must be through a water related route (World Health
Organization)
Page 24 of 340
Chapter One: Introduction
1.1.Overview of the Research
In this dissertation, I present my research developed from a preliminary study in the
provinces1 of Thai Binh (in the Red River Delta) and An Giang (in the Mekong River Delta) in
Vietnam where I completed a field evaluation for controlling and preventing Highly Pathogenic
Avian Influenza (HPAI). In the preliminary study, I observed small-scale integrated (SSI)2
farmers interacting among themselves, their livestock, and the water environment on farms, and I
witnessed how these interactions could potentially transmit water-related zoonotic diseases
(WRZD). In my research, I used the World Health Organization’s (WHO) definition of WRZD
as diseases in humans that come from animals and are related to water such as HPAI viruses,
some parasites, or diseases caused by a range of bacteria (e.g., pathogenic strains of Escherichia
coli such as enterohaemorrhagic E. coli) (WHO, 2004). The criteria for determining WRZD
include: 1) the pathogen must spend part of its life cycle within one or more animal species; 2) it
is probable or conceivable that some life stage of the pathogen will enter water; and 3)
transmission of the pathogen from animals to humans must be through a water-related route
(Moe, 2004).
In Figure 1.1 below, I illustrate potential interactions in the context of SSI farms that may
result in transmission of WRZD.
1 Vietnam is divided into 64 provinces; each province is sub-divided into districts, then communes, and villages. 2 Small-scale integrated (SSI) farming is defined as farming practices in which farmers combine some forms of
livestock and either directly or indirectly depend on agriculture (Hall, Thao, Minh, and Lien, 2006; Irini and
Rapsomanikis, 2005; Pica-Ciarmarra, Tasciotti, L., Otte, and Zezza, 2011).
Page 25 of 340
Figure 1.1. Water-Related Zoonotic Disease Interactions in the Water Environment
The conceptual model for my research was based on the observation that farmers make
decisions, including management choices related to animal and human health, that are influenced
by their impressions about the water quality on farms. The theoretical foundation for my research
was adapted from the Health Belief Model and the Theory of Planned Behaviour (Becker, 2012;
Janz and Becker, 1984; Rosenstock, Strecher, and Becker, 1988). Both the Health Belief Model
and the Theory of Planned Behaviour explain how individuals perceive a personal threat to
health and/or wellbeing, together with their belief in the benefits and barriers of an action, will
likely predict their health-related behaviours (see Appendix A). Based on the theoretical
foundation, the overall aim of my research was the following: to understand the association
between SSI farmers’ perceptions about the risk factors for transmitting WRZD and their
engagement in mitigating strategies to reduce transmitting WRZD in the provinces of Thai Binh
and An Giang in Vietnam. In my research, I defined a mitigating strategy to reduce transmitting
Adapted from WHO, 2004
On-farm water environment
Microbial pathogens
FarmersLivestock
Water related
zoonotic diseases
Page 26 of 340
WRZD as a plan (e.g., a plan for managing livestock) that includes one or more activities an SSI
farmer could choose to reduce the transmission of WRZD. An example of a mitigating strategy is
good livestock management, which may include actions, or tactics (e.g., treat animal waste at
source) that SSI farmers could choose to reduce transmitting WRZD (Carr and Bartram, 2004;
WHO, 2004). Moreover, since the Vietnamese Ministry of Agriculture and Rural Development
is proposing to restructure the agriculture system in Vietnam by 2020 (MARD, 2012), studying
the SSI farming model in the context of rural water and public health in Vietnam can contribute
to shaping its agriculture system in the near future.
1.2. Background
1.2.1. Small-Scale Integrated farming as an important agricultural model in Vietnam
Integrated aquaculture-agriculture farming is a common agricultural model in rural areas
in Vietnam (Dang, Milstein, Verdegem, and Verreth, 2006; Vu, Tran, and Dang, 2007). This
model is often referred to by various names such as smallholder, the VAC model3, small-scale
producers, and/or small-scale integrated farms (Bui, Hien, and Dang, 2010). In this model,
farmers combine some types of livestock and directly or indirectly depend on agriculture (Hall et
al., 2006; Irini and Rapsomanikis, 2005; Pica-Ciarmarra et al., 2011). In my research, farming
practices with this combination are referred to as small-scale integrated (SSI) farming and rural
farmers who practice SSI farming are called SSI farmers. The following paragraph provides a
brief review of SSI farmers and SSI farming in Vietnam.
It is reported that 10 million SSI farmers in Vietnam live without adequate sanitation and
9.3 million of them lack safe drinking water (Rechenburg and Herbst, 2006; WHO, 2006).
3 The VAC model: V= Vuon (i.e., Vegetable garden), A= Ao (i.e. Fish pond), C= Chuong (i.e., Livestock housing)
Page 27 of 340
However, SSI farmers benefit the least from national social economic policies and national
economic growth (VPMO, 2010). More than 80% of SSI farmers in Vietnam own some forms of
livestock (Hoang, 2011; Irini and Rapsomanikis, 2005; Pica-Ciarmarra et al., 2011; WB, 2002;
WB, 2006; WB, 2006), with 87% of them keeping primarily poultry (Irini and Rapsomanikis,
2005).
More than 50% of SSI farmers in Vietnam depend on livestock for food and income
(Hall, Benigno, and Wantanee, 2006; Irini and Rapsomanikis, 2005) and most of their cash
revenue comes from selling live animals (e.g., at home and/or in live poultry markets rather than
selling livestock products) (Irini and Rapsomanikis, 2005). Moreover, although SSI farmers have
been applying disease control methods such as vaccination and biosecurity (Morley, 2010), the
most common way they deal with infected animals is to salvage them (i.e., sell or eat the sick
animals). However, limited support and access to experienced animal health services further
promotes the spread of disease by magnifying the level of panic (Gleeson and Dung, 2010).
Backyard livestock rearing also plays an integral role in recycling nutrients in farming
systems and in generating valuable cash income, however, its development has been hampered
by infectious diseases such as classical swine fever and reproductive failure for sows
(Kamakawa, Thu, and Yamada, 2006). Scavenging ducks are also common and it is believed
they play an important role in maintaining and transmitting the H5N1 (HPAI) virus throughout
the food chain, yet published information is lacking on H5N1 infection among scavenging ducks
(Henning et al., 2011; Morley, 2010). Chapter Two has more detailed information on SSI
farming.
Page 28 of 340
1.2.2. Rural water and public health issues may challenge the SSI farming model.
Water is the vehicle for transmitting about 80% of all water related-diseases in people
worldwide (Ford and Colwell, 1996). Traditionally, Vietnamese farmers have considered water
as the most important factor for sustainable agriculture (Bach, 2007). However, in the SSI
farming model, a significant amount of untreated excreta from animals and humans is used as
fertilizer, food for fish, irrigation, or is dumped, which contaminates water sources with excreted
pathogens such as E. coli (Dalsgaard, 2001; Dao, 2006; Pham et al., 2011; Porphyre and Nguyen,
2007; Rechenburg and Herbst, 2006; Vu et al., 2007). Thus, although over 70% of SSI farmers
use unsanitary water and have no latrines, their awareness of the importance of a clean water
supply and environmental sanitation is limited (MoC and MARD, 2000).
The transmission of WRZD is considered a key issue in the context of an inadequate
clean water supply and sanitation in the rural areas of Vietnam (Hanington, Hall, and Anyan,
2016; Thang, Tuan, and Huton, 2008). For example, some of the reported important risk factors
for transmission of HPAI viruses were the proximity and increased accessibility of poultry flocks
to water sources, the increased density of ponds and streams, and temporal flooding (Gilbert et
al., 2006; Gilbert et al., 2008; Henning, Pfeiffer, and Vu, 2009; Nguyen, Vuong, Pham, and Vu,
2011; Pfeiffer, Minh, Martin, Epprecht, and Otte, 2007). Water bodies are thought to be
indicators for transmission of HPAI viruses and could provide a temporary reservoir for them at
the village and farm level (Morris and Jackson, 2005). This is important because AI viruses can
persist for extended periods in water, sewage, excreta and animal wastes depending on different
factors such as water temperature (Boere, Galbraith, and Stroud, 2006; Morris and Jackson,
2005; Webster, Peiris, Chen, and Guan, 2006; WHO, 2007). More specifically, the HPAI
(H5N1) virus can persist in water for over a 60-day period, and constitutes a potential source of
Page 29 of 340
waterborne AI transmission for wild birds and domestic ducks (Brown, Swayne, Cooper, Burns,
and Stallknecht, 2007; Domanska, Minta, Smietanka, Sylvie, and Berg, 2010; Nazir et al., 2010;
Roche et al., 2009). The increasing interaction between domestic and wild birds in a common
water source (Boere et al., 2006; Morris and Jackson, 2005; Zheng et al., 2010) also creates
favourable conditions for co-infection, re-assortment, and evolution of the viruses among host
species (Stallknecht, Shane, Kearney, and Zwank, 1990; Ward, Maftei, Apostu, and Suru, 2009;
Webster, Peiris, Chen, and Guan, 2006). Increasing livestock populations additionally pollute
both soil and surface waters (Bui, 2010; Trinh, 2001), with concerns for the presence of a high
concentration of pathogenic microorganisms in livestock wastes (Vu, 2001; Yajima and
Kurokura, 2008). The direct use of this contaminated water can result in an actual risk to public
health and appropriate risk mitigation measures such as source water protection and/or livestock
management are crucial (Guan and Holley, 2003; Toews, 2009; Trang, Bui, Molbak, Phung, and
Dalsgaard, 2007; Yajima and Kurokura, 2008).
1.2.3. SSI farmers’ perceptions about risk factors for transmitting WRZD may be associated
with their engagement in mitigating strategies to reduce transmitting WRZD
I have not found other studies that explore SSI farmers’ perceptions of risk factors or
their mitigating strategies for reducing WRZD transmission in Vietnam. Some studies have
shown generally that SSI farmers are not aware of, or do not rank highly, the health risks
associated with their use of water contaminated with waste from humans and from livestock
(Keraita et al., 2010). Rather, they view these as unavoidable risk factors related to livestock
production (Knudsen et al., 2008). Moreover, for farmers, health risks from using contaminated
water are not as serious as water that remains on the skin and causes skin problems; however,
health risks from non-composted smelly feces are serious because they enter the body through
Page 30 of 340
polluted air (Knudsen et al., 2008). Although farmers know proper hygiene, such hygiene
measures are ineffectively practiced (e.g., hand washing without soap). Most likely, this is due to
misconceiving risk factors and/or lacking knowledge about cause-effect relationships, habits,
and/or not ranking the risks high enough while using water (Herbst et al., 2009; Keraita et al.,
2010; Rechenburg and Herbst, 2006).
In some cases, farmers are aware of health risk factors, however, they assess the value of
potential mitigation measures by the overall economic impact on work efficiency rather than by
taking into account the potential health benefits to be gained (Keraita et al., 2010). A number of
studies have shown that some farmers have a basic awareness of health risks relating to
contaminated water use (Herbst et al., 2009; Keraita, Jim'nez, and Drechsel, 2008; Owusu,
Bakang, Abaidoo, and Kinane, 2012; Owusu, Bakang, Abaidoo, and Kinane, 2011;
Weldesilassie, Boelee, Drechel, and Dabbert, 2011). However, these studies only targeted small
samples of peri-urban farmers with reference to crops instead of livestock among rural SSI
farmers. Thus, it is important to understand SSI farmers’ perceptions about the risk factors for
transmitting WRZD, their mitigating strategies to reduce this threat, and the factors associated
with their strategies. This understanding could be useful for the development of mutually
acceptable risk-management strategies and/or interventions (Keraita et al., 2010), especially in a
Vietnamese context.
1.3. Key Research Questions
My research thus addresses the following research questions:
1. What is the socioeconomic profile of SSI farmers in the provinces of Thai Binh and An
Giang in Vietnam?
Page 31 of 340
2. Is the socioeconomic profile of SSI farmers in the provinces of Thai Binh and An Giang
consistent with an EcoHealth approach in the context of SSI farming?
3. What are the perceptions of SSI farmers concerning the risk factors for transmitting
WRZD in the provinces of Thai Binh and An Giang?
4. What sources of water do SSI farmers use for drinking and domestic purposes4 and what
are the corresponding basic microbial and related indicators5 of water quality for those
sources in the provinces of Thai Binh and An Giang?
5. How does the quality of on-farm water that SSI farmers in the provinces of Thai Binh and
An Giang used for drinking and domestic purposes relate to their perceptions about both
water quality and risk factors for transmitting WRZD?
6. Do SSI farmers in the provinces of Thai Binh and An Giang engage in mitigating
strategies to reduce transmitting WRZD?
7. What factors are associated with SSI farmers’ engagement in mitigating strategies to
reduce the transmission of WRZD in the provinces of Thai Binh and An Giang?
1.4. Research Methods
In my research, I combined qualitative and quantitative methods in a cross-sectional
study with three separate phases: 1) establishing, testing, piloting, and revising study tools (e.g.,
questionnaires and water testing protocols); 2) gathering data from target groups of farmers
through a full field survey that included questionnaires and water quality testing; and 3)
analyzing data and publishing results. Initially, I conducted a pilot study to test and revise the
4 In my research, drinking water is defined as “water used for direct drinking or food processing”. Domestic water is
defined as “water use for domestic purposes but nor for direct drinking or processing food” (MoH, 2009; MoH,
2009) 5 The WHO recommends Escherichia coli (E. coli) colony forming unit (cfu), turbidity, and pH as basic microbial
and related indicators for assessing rural water quality (WHO, 2012)
Page 32 of 340
study tools before using them in the full field survey where I randomly selected a sample size of
600 eligible households representing SSI farming to participate in my research. Key variables
included demographics, socioeconomic and livestock production characteristics, SSI farmers’
perceptions of risk factors for transmitting WRZD, and the actions SSI farmers take to reduce
transmitting WRZD. Variables surrounding SSI farmers’ engagement in mitigating strategies to
reduce transmitting WRZD were generated using WHO’s recommended mitigating strategies to
cluster the data on farmers’ actions to reduce transmitting WRZD.
Most waves of the HPAI (H5N1) epidemic during the 2003 – 2006 period occurred in the
river delta areas of my two target research areas in Thai Binh and An Giang provinces: these are
two typical provinces of Vietnam in the Red River Delta and Mekong River Delta respectively.
Detailed information of the overall methods used in my research is presented in Chapter
Two. Chapters Three, Four, and Five provide additional details of methods that specifically
relate to each of the chapters.
1.5. A Description of Chapters in the Dissertation
This dissertation consists of six individual chapters that are structured in a manuscript
style format. Beside the introduction and conclusion chapters, I present the research results as a
collection of four manuscript-based chapters that were prepared for submitting to peer-reviewed
journals preferably in the field of agricultural economics, water quality, and public health.
Although each manuscript-based chapter is a standalone manuscript, I arranged the chapters of
this dissertation in a logical and continuous order to present the results according to the order of
my research questions.
In the next section, I present a flow chart outlining how the research questions relate to
each manuscript-based chapter (See Figure 1.2) and a description of chapters in the dissertation.
Page 33 of 340
Figure 1.2. Dissertation’s Outline
Chapter 2:
Develops a socioeconomic
profile of SSI farmers in
Thai Binh and An Giang
Provinces of Vietnam
Chapter 3:
Presents SSI farmers'
perceptions of risk factors
for WRZD transmission
in Thai Binh and An
Giang provinces of
Vietnam
Chapter 4:
Presents the water quality
among SSI farms in Thai
Binh and An Giang
provinces of Vietnam
Chapter 5:
Presents factors
associated with SSI
farmers' engagement in
mitigating strategies to
reduce transmission of
WRZD in Thai Binh and
An Giang provinces of
Vietnam
Research question 1:
What is the socioeconomic profile of SSI farmers in Thai
Binh and An Giang provinces of Vietnam?
Research question 2:
Is the socioeconomic profile of SSI farmers in Thai Binh and
An Giang consistent with an EcoHealth approach in the
context of SSI farming?
Research question 3:
What are SSI farmers’ perceptions of risk factors for water
related zoonotic disease (WRZD) transmission in Thai
Binh and An Giang provinces of Vietnam?
Research question 4:
What are sources and the basic quality of water used by
SSI farmers for drinking and domestic purposes in Thai
Binh and An Giang provinces of Vietnam?
Research question 5:
How does the quality of water used by SSI farmers relate
to their perceptions of water quality and risk factors for
transmission of WRZD?
Research question 6:
Do SSI farmers engage in mitigating strategies to reduce
transmission of WRZD?
Research question 7:
What are factors associated with SSI farmers’ engagement
in mitigating strategies to reduce transmission of WRZD?
Page 34 of 340
Chapter One provides an overview of the research, a background of my research, a
summary of each individual chapter in the dissertation, and connections between chapters in the
dissertation. I also acknowledge the contributions of other authors to manuscripts that were
prepared based on this dissertation.
Chapter Two explores the first two research questions by presenting a profile of SSI
farming in the regions of research within the context of the increased documented risk factors
associated with the transmission of WRZD. The first objective of this chapter is to describe the
characteristics of SSI farming within the geographical research regions while considering the risk
factors for transmitting WRZD. The second objective is to use an EcoHealth approach6 to
discuss the potential ecological and health implications of SSI farming in the context of risk
factors for transmitting WRZD. Before conducting the research, I anticipated that the SSI
farming practices used in Thai Binh and An Giang would expose farmers to risk factors for
transmitting WRZD. Furthermore, I expected that not all the characteristics of SSI farming
practices in Thai Binh and An Giang were consistent with an EcoHealth approach. Finally, I
used descriptive and analytical statistics to assist in the analysis of data collected for this chapter.
Based on this chapter, two manuscripts were prepared and submitted for journal publication. My
committee members are acknowledged in the manuscripts for their comments and advice on this
chapter.
6 In my research, EcoHealth (or ecosystem approaches to health) is referred to as an approach that explores the
interface between human health, animal health, and environmental health by using six key pillars/principles:
complexity, trans-disciplinarity, community participation, gender and social equity, sustainability, and knowledge to
action (Hall, 2010; IDRC, 2012). The International Development Research Centre (IDRC) initiated EcoHealth and
developed it based on many years of health and environment research(IDRC, 2012).
Page 35 of 340
Chapter Three addresses the third research question by presenting a study of SSI farmers’
perceptions about the risk factors for transmitting WRZD in the study areas. Here, I use both the
conceptual and theoretical models for my research (See Appendix A). I also discuss the
importance of these perceptions in researching how SSI farmers engage in mitigating strategies
to reduce transmitting WRZD. My specific objective was to describe four constructs surrounding
SSI famers’ perceptions of risk factors for transmitting WRZD. The first construct was SSI
farmers’ impressions about the quality of general water environment in their villages. The second
construct was SSI farmers’ anticipation of threat to health and/or wellbeing of transmitting
WRZD including the projected susceptibility to and severity of transmission of WRZD. The third
construct was SSI farmers’ perceived expectations including sensed barriers, supposed benefits,
and believed self-efficacy in mitigating the transmission of WRZD. The fourth construct was
beliefs about triggers/cues7 to taking actions that would mitigate transmitting WRZD (See
Appendix A). Descriptive statistics were used to assist in the data analysis for this chapter.
Chapter Four addresses the fourth and fifth research questions by presenting a study of
the types and qualities of water sources that SSI farms in the study area typically use. I also
explore the association between on-farm water quality and the perceptions identified in Chapter
Two and discuss the public health implications of this association. The objectives of my research
in Chapter Four were to 1) describe the sources and assess the basic microbial and related water
quality SSI farms use and 2) to explore the connection between farm water quality and farmers’
perceptions of risk factors for transmitting WRZD. I expected SSI farmers to use multiple water
sources for drinking and domestic purposes. I hypothesized that the frequency of water source
7 A trigger/cue to action is “the stimulus needed to trigger the decision making process to accept a recommended
health action”. Examples of possible cues to actions are chest pains, mass media campaigns, advice from others, a
reminder postcard from a physician, illness of family member or friend, etc.) (BUMC, 2013)
Page 36 of 340
use for drinking and domestic purposes among SSI farmers in Thai Binh was different from
those in An Giang, and that these water sources in both provinces are of low quality compared to
water quality standards in Vietnam. My other hypothesis in this chapter was that the
socioeconomic status of SSI farmers and their perceptions about risk factors for transmitting
WRZD are associated with microbial content of water on farms. Different methods of data
collection, specifically questionnaires, in-depth interviews, farm visits, and focus group
discussions were used to gather data on the water sources SSI farmers have access to and their
corresponding frequency of use for drinking and domestic purposes.
In Chapter Four, I use two protocols to assess objectively the basic microbial and related
quality indicators (i.e., pH, turbidity, and E. coli colony forming unit count) of on-farm water for
drinking and domestic purposes. These protocols are: 1) a protocol for laboratories to assess on-
farm water quality and 2) a protocol for community health/veterinary workers (CHWS/CVWs) to
quantify E. coli in on-farm water. These protocols follow WHO’s recommendations for assessing
basic indicators of rural water quality (WHO, 2011; WHO, 2011; WHO and OECD, 2003) and
conform with Vietnam’s national technical regulations on drinking and domestic water quality
(MoH, 2009; MoH, 2009). This chapter additionally discusses implications for the research
results for rural water, public health, and for the water regulatory framework for SSI farms in
Vietnam. Descriptive statistics were used to analyze and present the results related to water
sources, frequency of use, and on-farm water quality. Regression analysis was used to model and
test the association between on-farm water quality and SSI farmers’ perceptions of water quality
and risk factors for transmitting WRZD.
Chapter Five addresses the sixth and seventh research questions by presenting SSI
farmers’ engagement in, and the factors associated with mitigating strategies to reduce
Page 37 of 340
transmitting WRZD (e.g., livestock management). I hypothesized that the socioeconomic status
of SSI farmers and their perceptions of risk factors for transmitting WRZD are associated with
their mitigating strategies to reduce transmitting WRZD. This chapter also discusses implications
for associations between formulating interventions and policies to improve the health of farmers
and animals, and their environment. Descriptive statistics and probit regression were used to
assist in the data analysis, to test the hypothesis, and to present the results.
In Chapter Six, I summarize the results presented in Chapters Two to Five and discuss the
limitations of my research. Then, I provide recommendations for future research and formulating
and revising programs and policies related to WRZD, livestock production, rural water, and
public health in Vietnam. Finally, I conclude my research study.
1.6. Connection Between Chapters
Overall, the results of the research are connected through the chapters in the dissertation.
The results from Chapter Two and Three were used to assist the studies in Chapter Four and
Five. In particular, the results from the research in chapters Two and Three were used as a
foundation for Chapter Four to model factors that are associated with on-farm water quality.
Similarly, the results from chapters Two, Three, Four were used in Chapter Five to explore
factors that are associated with SSI farmers’ engagement in mitigating strategies to reduce
transmitting WRZD. In order to ensure that each manuscript-based chapter is a stand-alone
chapter, there will be some repetition of information in the text of the chapters.
Page 38 of 340
References
Bach, A. H. (2007). Equivalence in the translation of Vietnamses cultural words in the book
"Wandering through Vietnamese culture" by Huu Ngoc. Vietnam College of Foreign
Languages (VHUN), Hanoi.
Becker, M. H. (2012). The Health Belief Model and Personal Health Behavior. (4 ed.) (vols. 2).
Boere, G. C., Galbraith, C. A., and Stroud, D. A. (2006). Waterbirds Around the World The
Stationery Office, UK.
Brown, J. D., Swayne, D. E., Cooper, R. J., Burns, R. E., and Stallknecht, D. E. (2007).
Persistence of H5 and H7 Avian Influenza viruses in water. Avian Diseases Digest, 2,
e22.
Bui, T. D., Hien, N. T., and Dang, T. T. (2010). Distribution of freshwater snails in family-based
VAC ponds and associated waterbodies with special reference to intermediate hosts of
fish-borne zoonotic trematodes in Nam Dinh Province, Vietnam. Acta Tropica, 116, 15-
23.
Bui, X. M. (2010). Duck farming systems and avian influenza in the Mekong Delta of Vietnam.
In FAO Rome.
BUMC (2013). The Health Belief Model. BU School of Public Health [On-line]. Available:
http://sphweb.bumc.bu.edu/otlt/MPH-Modules/SB/SB721-Models/SB721-Models2.html
Carr, R. and Bartram, J. (2004). The control envelope and risk management. In (pp. 67-92). IWA
Publishing.
Dalsgaard, A. (2001). Health Aspects of the Reuse of Wastewater in Agriculture and
Aquaculture in Vietnam. In Wastewater Reuse in Agriculture in Vietnam: Water
Page 39 of 340
Management, Environment and Human Health Aspects International Water Management
Institute.
Dang, N. K., Milstein, A., Verdegem, M. C. J., and Verreth, J. A. V. (2006). Food inputs, water
quality and nutrient accumulation in integrated pond systems: A multivariate approach.
Aquaculture, 261, 160-173.
Dao, L. H. (2006). Situation and future direction of environmental protection in livestock
production, Vietnam Department of Animal Husbandary, Ministry of Agriculture and
Rural Development, Vietnam.
Domanska, B. K., Minta, Z., Smietanka, K., Sylvie, M., and Berg, A. T. (2010). H5N1 High
Pathogenicity Avian Influenza Virus Survival in Different Types of Water. Avian
Diseases Digest, 5, e180-e181.
Ford, T. E. and Colwell, R. (1996). A Global Decline in Microbiological Safety of Water: A Call
for Action. American Academy of Microbiolog.
Gilbert, M., Chaitaweesub, P., Parakamawongsa, T., Premashthira, S., Tiensin, T., Kalpravidh,
W. et al. (2006). Free-grazing ducks and highly pathogenic avian influenza, Thailand.
Emerging Infectious Diseases, 12, 227-234.
Gilbert, M., Xiao, X., Pfeiffer, D. U., Epprecht, M., Boles, S., Czarnecki, C. et al. (2008).
Mapping H5N1 highly pathogenic avian influenza risk in Southeast Asia. Proceedings of
the National Academy of Sciences, 105, 4769-4774.
Gleeson, L. J. and Dung, D. H. (2010). Mid-Term Review: Vietnam Integrated National
Operational Program for Avian and Human Influenza (OPI) 2006-2010: HPAI Control
and Eradication in the Agriculture Sector.
Page 40 of 340
Guan, T. Y. and Holley, R. A. (2003). Pathogen survival in swine manure environments and
transmission of human enteric illness – a review. Journal of Environmental Quality, 32
(2), 392.
Hall, D. C. (2010). The Ecohealth Approach: An introduction to the key concepts. Building
Ecohealth Capacity in Asia, Veterinarians Without Border Canada [On-line]. Available:
https://www.vetswithoutborders.ca/asia/ecohealth
Hall, D. C., Benigno, C., and Wantanee, K. (2006). The impact of avian influenza on small and
medium scale poultry producers in South East Asia (preliminary findings) Food and
Agriculture Organization (FAO).
Hall, D. C., Thao, T. D., Minh, D. V., and Lien, L. V. (2006). Competitiveness of the livestock
sector in Vietnam World Bank paper. WB-EASRD and FAO-TCIP.
Hanington, P., Hall, D. C., and Anyan, W. (2016). The role of environmental management in
snail-borne zoonotic diseases. In S. C. Cork, D. C. Hall, and K. Liljebjelke (Eds.), One
Health Case Studies: Addressing Complex Problems in a Changing World (Expected
2016). 5M Publishing, Sheffield, UK.
Henning, J., Henning, K. A., Morton, J. M., Long, N. T., Ha, N. T., Vu, L. T. et al. (2011).
Highly pathogenic avian influenza (H5N1) in ducks and in-contact chickens in backyard
and smallholder commercial duck farms in Viet Nam. Preventive Veterinary Medicine,
101, 229-240.
Henning, J., Pfeiffer, D., and Vu, L. (2009). Risk factors and characteristics of H5N1 Highly
Pathogenic Avian Influenza (HPAI) post-vaccination outbreaks. Vet.Res., 40.
Page 41 of 340
Herbst, S., Benedikter, S., Koester, U., Phan, N., Berger, C., Rechenburg, A. et al. (2009).
Perceptions of water, sanitation and health: a case study from the Mekong Delta,
Vietnam. Water Science and Technology, 60, 699-707.
Hoang, B. T. (2011). Rural employment and life: Challenges to gender roles in Vietnam's
agriculture at present Research Centre for Gender, Family and Environment in
Development, Viet Nam.
IDRC (2012). Ecohealth Research in Practice: Innovative Applications of an Ecosystem
Approaches to Health. International Development Centre (IDRC); Springer.
Irini, M. and Rapsomanikis, G. (2005). Contribution of Livestock to Household income in
Vietnam: A household typology based analysis (Rep. No. PPLPI Working Paper N.21). A
Living from Livestock, FAO.
Janz, N. K. and Becker, M. H. (1984). The Health Belief Model: A Decade Later. Health
Education and Behavior, 11, 1-47.
Kamakawa, A., Thu, H. T. V., and Yamada, S. (2006). Epidemiological survey of viral diseases
of pigs in the Mekong delta of Vietnam between 1999 and 2003. Veterinary
Microbiology, 118, 47-56.
Keraita, B., Drechsel, P., Seidu, R., Amerasinghe, P., Cofie, O., and and Konradsen, F. (2010).
Harnessing Farmers' Knowledge and Perceptions for Health-Risk Reduction in
Wastewater-Irrigated Agriculture. In P.Drechsel, C.A.Scott, L.Raschid-Sally,
M.Redwood, and and A.Bahri (Eds.), Wastewater Irrigation and Health: Assessing and
Mitigating Risk in Low-Income Countries (pp. 337-353). London: Earthscan Publications
Ltd.
Page 42 of 340
Keraita, B., Drechsel, P., Seidu, R., Amerasinghe, P., Cofie, O., and Konradsen, F. (2010).
Harnessing Farmers' Knowledge and Perceptions for Health-Risk Reduction in
Wastewater-Irrigated Agriculture. In P.Drechsel, C. A. Scott, L. Raschid-Sally, M.
Redwood, and A. Bahri (Eds.), Wastewater irrigation and health : assessing and
mitigating risk in low-income countries (pp. 337-353). London: Earthscan Publications
Ltd.
Keraita, B., Jim'nez, B., and Drechsel, P. (2008). Extent and implications of agricultural reuse of
untreated, partly treated and diluted wastewater in developing countries. Agr Vet Sci Nutr
Nat Resour, 3, 15.
Knudsen, L. G., Pham, D. P., Nguyen, H. T., Helle, S., Jensen, P. K., Dalsgaard, A. et al. (2008).
The fear of awful smell: risk perceptions among farmers in Vietnam using wastewater
and human excreta in agriculture. The Southeast Asian journal of tropical medicine and
public health, 39, 341-352.
MARD (2012). Proposal on restructuring the agricultural sector towards greater added value
and sustainable develoment Hanoi: Vietnam Ministry of Agriculture and Rural
Development.
MoC and MARD (2000). National Rural Clean Water Supply and Sanitation Strategy up to the
Year 2020 Hanoi, Vietnam: Ministry of Construction and Ministry of Agriculture and
Rural Development of Vietnam.
Moe, C. L. (2004). What are the criteria for determining whether a disease is zoonotic and water
related. In World Health Organization (WHO). Waterborne Zoonoses: Identification,
Causes and Controls (pp. 27-45). London: IWA Publishing.
Page 43 of 340
MoH (2009). National technical regulation on domestic water quality (QCVN 02: 2009/BYT).
Hanoi: Department of Preventive Medicine and Environment, Ministry of Health,
Socialist Republic of Vietnam.
MoH (2009). National technical regulation on drinking water quality (QCVN 01: 2009/BYT).
Hanoi: Department of Preventive Medicine and Environment, Ministry of Health,
Socialist Republic of Vietnam.
Morley, C. (2010). Bio-Security Stakeholder Analysis, Vietnam (Rep. No. OSRO/RAS/604/USA
B6). The Emergency Centre for Transboundary Animal Diseases (ECTAD), Food and
Agriculture Organization of the United Nations (FAO).
Morris, R. S. and Jackson, R. (2005). Epidemiology of H5N1 Avian Influenza in Asia and
Implications for Regional Control Rome, Italy: Food and Agriculture Organisation of the
UN.
Nazir, J., Haumacher, R., Ike, A., Stumpf, P., Bohm, R., and Marschang, R. E. (2010). Long-
term study on tenacity of avian influenza viruses in water (distilled water, normal saline,
and surface water) at different temperatures. Avian Dis, 54, 720-724.
Nguyen, H. V., Vuong, T. A., Pham, D. P., and Vu, V. T. (2011). Safe use of wastewater in
agriculture and aquaculture (Rep. No. 2). The National Centre of Competence in
Research (NCCR) North-South.
Owusu, V., Bakang, J. E. A., Abaidoo, R. C., and Kinane, M. L. (2012). Perception on untreated
wastewater irrigation for vegetable production in Ghana. Environ Dev Sustain, 14, 135-
150.
Page 44 of 340
Owusu, V., Bakang, J. E., Abaidoo, R., and Kinane, M. (2011). Perception on untreated
wastewater irrigation for vegetable production in Ghana. Environment, Development and
Sustainability, 1-16.
Pfeiffer, D. U., Minh, P. Q., Martin, V., Epprecht, M., and Otte, M. J. (2007). An analysis of the
spatial and temporal patterns of highly pathogenic avian influenza occurrence in Vietnam
using national surveillance data. Vet J, 174, 302-309.
Pham, D. P., Nguyen, V. H., Hattendorf, J., Zinsstag, J., Phung, D. C., and Odermatt, P. (2011).
Risk factors for Entamoeba histolytica infection in an agricultural community in Hanam
province, Vietnam. Parasites and Vectors, 4, 102.
Pica-Ciarmarra, U., Tasciotti, L., Otte, J., and Zezza, A. (2011). Livestock Assets, Livestock
Income and Rural Households: Cross-Country Evidence from Household Surveys (Rep.
No. ESA Working Paper No. 11-17). Rome: World Bank, FAO, AU-IBAR, ILRI.
Porphyre, V. and Nguyen, Q. C. (2007). Pig production development animal-waste management
and environment protection: a case study in Thai Binh Province, Northern Vietnam.
PRISE Publications, France.
Rechenburg, A. and Herbst, S. (2006). Hygienic Aspects of Sanitation and Water in the Rural
Areas of the Mekong Delta, Vietnam (Rep. No. "Prosperity and Poverty in a Globalised
World Challenges for Agricultural Research"). University of Bonn, Institute for Hygiene
and Public Health, Germany.
Roche, B., Lebarbenchon, C., Gauthier-Clerc, M., Chang, C. M., Thomas, F., Renaud, F. et al.
(2009). Water-borne transmission drives avian influenza dynamics in wild birds: the case
of the 2005-2006 epidemics in the Camargue area. Infect Genet Evol, 9, 800-805.
Page 45 of 340
Rosenstock, I. M., Strecher, V. J., and Becker, M. H. (1988). Social Learning Theory and the
Health Belief Model. Health Education and Behavior, 15, 175-183.
Stallknecht, D. E., Shane, S. M., Kearney, M. T., and Zwank, P. J. (1990). Persistence of Avian
Influenza Viruses in Water. Avian diseases, 34, 406-411.
Thang, P., Tuan, H., and Huton, G. (2008). Economic impacts of sanitation in Vietnam. World
Bank.
Toews, D. W. (2009). Food, Global Environmental Change and Health: EcoHealth to the
Rescue? Mcgill Journal of Medicine, 12, 85-89.
Trang, D. T., Bui, T. T., Molbak, K., Phung, D. C., and Dalsgaard, A. (2007). Epidemiology and
aetiology of diarrhoeal diseases in adults engaged in wastewater-fed agriculture and
aquaculture in Hanoi, Vietnam. Trop Med Int Health, 12, 23-33.
Trinh, T. L. (2001). Wastewater and its Treatment System in the Southern Part of Vietnam. In
International Water Management Institute: Proceedings of a Workshop held in Hanoi,
Vietnam 14 March 2001 Southern Institute of Water Resources Research.
VPMO (2010). About the organisation of the national poverty survey to prepare for the
implementation of social secuity program and policy in the period of 2011- 2015 (Rep.
No. 1752/ CT-TTg). Vice Prime Minister Office (VPMO), the Government of Vietnam.
Vu, T. K. V., Tran, M. T., and Dang, T. T. S. (2007). A survey of manure management on pig
farms in Northern Vietnam. Livestock Science, 112, 288-297.
Vu, T. T. H. (2001). Treatment of Domestic Wastewater and its Reuse in Farm Irrigation in the
Red River Delta. In Wastewater Reuse in Agriculture in Vietnam: Water Management,
Environment and Human Health Aspects Center for Irrigation and Water Supply,
Vietnam Institute for Water Resources Research.
Page 46 of 340
Ward, M. P., Maftei, D. N., Apostu, C. L., and Suru, A. R. (2009). Association Between
Outbreaks of Highly Pathogenic Avian Influenza Subtype H5N1 and Migratory
Waterfowl (Family Anatidae) Populations. Zoonoses and Public Health, 56, 1-9.
WB (2002). Memorandum of the President of the International Development Association and the
International Finance Corporation to the Executive Directors on a Country Assistance
Strategy of the World Bank Group for the Socialist Republic of Vietnam (Rep. No. Report
No: 24621-VN). Vietnam Country Management Unit East Asia and the Pacific Region.
The World bank.
WB (2006). Vietnam Food Safety and Agricultural Health Action Plan (Rep. No. Report No.
35231 VN).
WB (2006). Vietnam Plan Action for Food Security and Animals, Plant Protection Hanoi: World
Bank.
Webster, R. G., Peiris, M., Chen, H., and Guan, Y. (2006). H5N1 outbreaks and enzootic
influenza. Emerg Infect Dis [serial on the Internet].
Webster, R. G., Peiris, M., Chen, H., and Guan, Y. (2006). H5N1 outbreaks and enzootic
influenza. Emerg Infect Dis [serial on the Internet]. Emerging Infectious Diseases.
Weldesilassie, A. B., Boelee, L. I. N., Drechel, P. A. Y., and Dabbert, S. T. E. P. (2011).
Wastewater use in crop production in peri-urban areas of Addis Ababa: impacts on health
in farm households. Environment and Development Economics, 16, 25-49.
WHO (2004). Waterborne Zoonoses: Identification, Causes, and Control. IWA Publishing.
WHO (2006). Volume II: Wastewater use in agriculture.
WHO (2007). Review of latest available evidence on potential transmission of avian influenza
(H5N1) through water, sewage and ways to reduce the risks to human health (Rep. No.
Page 47 of 340
WHO/SDE/WSH/06.1). Geneva: World Health Organisation: Water, Sanitation and
Health Public Health and Environment.
WHO (2011). Guidelines for drinking-water quality. (Fourth ed.) World Health Organisation.
WHO (2011). Microbial aspects. In Guidelines for drinking-water quality (pp. 117-153). World
Health Organization.
WHO (2012). Rapid assessment of drinking water quality - A handbook for implementation.
World Health Organization.
WHO and OECD (2003). Assessing Microbial Safety of Drinking Water: Improving Approaches
and Methods. IWA Publishing.
Yajima, A. and Kurokura, H. (2008). Microbial risk assessment of livestock-integrated
aquaculture and fish handling in Vietnam. Fisheries Science, 74, 1062-1068.
Zheng, T., Adlam, B., Rawdon, T. G., Stanislawek, W. L., Cork, S. C., Hope, V. et al. (2010). A
cross-sectional survey of influenza A infection, and management practices in small rural
backyard poultry flocks in two regions of New Zealand. New Zealand Veterinary
Journal, 58, 74-80.
Page 48 of 340
Chapter Two: Developing a Socioeconomic Profile of Small-Scale Integrated (SSI) Farms
in the Provinces of Thai Binh and An Giang in Vietnam
2.1. Introduction
2.1.1. Overview of Small-Scale Integrated (SSI) farming in Vietnam
In Vietnam, as in other developing countries, economic growth relies primarily on
agricultural production and industrialization associated with rapid urban development. In this
environment in Vietnam, integrated aquaculture agriculture farming is a common model in rural
areas (Dang et al., 2006; Vu et al., 2007). In my research, SSI farming is defined as farming
practices in which farmers combine some types of livestock rearing and directly or indirectly
depend on agriculture (Hall et al., 2006; Irini and Rapsomanikis, 2005; Pica-Ciarmarra et al.,
2011). SSI farming and/or farmers are often referred to by various terms such as smallholder
farmers, the VAC model8, small-scale producers, and/or SSI farms (Bui et al., 2010). However,
most farmers practicing SSI farming benefit the least from national social economic policies and
economic growth (Morley, 2010). For instance, in Vietnam, it is estimated that 10 million rural
farmers still live without adequate sanitation and 9.3 million lack safe drinking water (VPMO,
2010).
The VAC model is an example of SSI farming that encompasses a predominantly small-
scale operation of poultry, pigs, fish, and crops (Bui et al., 2010). The VAC model is a common
integrated aquaculture, agricultural farming model (Bosma et al., 2006; Bui et al., 2010; Dang et
al., 2007; Le et al., 2007). Some farmers also rely on raising free-grazing ducks (Henning et al.,
2011) and dairy cattle (Geurden et al., 2008). Backyard livestock rearing is another example of
SSI farming model, which plays an integral role in recycling nutrients in farming systems and
8 The VAC model: V= Vuon (i.e., Vegetable garden), A= Ao (i.e., Fish pond), C= Chuong (i.e., Livestock housing)
Page 49 of 340
generating valuable cash income. This model can work effectively but the development of
backyard livestock rearing has been hampered by infectious diseases such as fatal epizootics of
piglets and reproductive failure of sows caused by classical swine fever (CSF) virus (Kamakawa
et al., 2006). Moreover, in this model, a significant amount of untreated excreta of animals and
humans is used as fertilizer, fish food, irrigation, or is dumped, which contaminates water
sources with excreted pathogens (Vu et al., 2007; Pham et al., 2011; Dalsgaard, 2001).
More than 80% of SSI farmers own some forms of livestock (Pica-Ciarmarra et al., 2011)
with 87% of them specifically rearing poultry (Hall et al., 2006; Irini and Rapsomanikis, 2005).
More than 50% of farmers also depend on livestock for nutrition and income (Irini and
Rapsomanikis, 2005) and most of their cash revenue comes from selling live animals rather than
livestock products such as broiler meat (Kamakawa et al., 2006).
In SSI farming, economic factors are one of the most important considerations in driving
access to health care for people in the rural areas. For instance, SSI farmers often consider the
quality of health services and sociocultural and economic factors (e.g., living with extended
family or cost of services), rather than geographical access in utilizing health services (Duong et
al., 2004). Some studies have also reported the health behaviour of SSI farmers as a concern. For
instance, although SSI farmers have been engaged in disease control methods such as
vaccination and bio-security (Gleeson and Dung, 2010), they most commonly deal with infected
animals by salvaging them (i.e., sell or eat sick animals). Gleeson and Dung (2010) further
argued that when SSI farmers confronted a problem (e.g., a zoonotic disease) that no one,
including animal health services, had experience with, they panicked, which further promoted
the spread of the disease. Such situations emphasize the importance of understanding the
farmers’ sense of self-efficacy in dealing with public health or animal health problems.
Page 50 of 340
2.1.2. Rationale for the research
Water has a profound influence on human health and water-related zoonotic diseases
(WRZD)9 place a heavy burden on the population health services of many countries, especially
developing countries(WHO, 2011). Farmers appear to consider water use an important factor
mainly because of its implications for agriculture and economic development, rather than for
their own health or that of their animals (Bui et al., 2010). Specifically, in Vietnam, there are
increasing reports of WRZDs especially among SSI farmers and their animals in the Red River
and Mekong River Deltas (ProMED, 2014; MoH and MARD, 2011; MARD and MoH, 2006).
However, their awareness level about the health risk factors associated with transmitting WRZD
remains low (Knudsen et al., 2008; Keraita et al., 2010). Thus, it appears that although SSI
farming represents one of the largest agricultural sectors in Vietnam (MARD, 2012), it will be
challenging to reduce transmission of WRZD (see Table 2.1).
In Canada, the International Development Research Centre (IDRC) has initiated
ecosystem (or EcoHealth) approaches to health that have their foundation in many years of
health and environment research (Charron, 2012). Ecosystem approaches to health explore the
interface between humans, animals, and the environment through using six key pillars/principles
consisting of complexity, trans-disciplinarity, community participation, gender and social equity,
sustainability, and knowledge to action (Hall, 2010; IDRC, 2012).
9 In my research, I used WHO’ definition of WRZD as diseases in humans that are spread from animals and are
related to water such as Highly Pathogenic Avian Influenza, some parasites, or diseases caused by a range of
bacteria (e.g., pathogenic strains of Escherichia coli) (WHO, 2004). Criteria for determining WRZD include: 1) the
pathogen must spend part of its life cycle within one or more animal species; 2) it is probable or conceivable that
some life stage of the pathogen will enter water; and 3) transmission of the pathogen from animals to humans must
be through a water related route (Moe, 2004).
Page 51 of 340
Internationally, case studies and reports have been presented that bring forward the issues
of SSI livestock management, improvement in famers’ incomes, and reduction of WRZD
(Waltner-Toews, 2011; IDRC, 2012). However, to date in Vietnam, I have not found any studies
that provide an EcoHealth insight into SSI farming and that explore farmers’ perceptions of risk
factors for transmitting WRZD and link their mitigation strategies10 for reducing the transmission
of WRZD. Therefore, my research is significant because I use an EcoHealth approach to
examine and describe the socioeconomic profile of SSI farming in the two main river deltas in
Vietnam, while focusing on general socioeconomic status, demographics, and characteristics of
livestock production.
My research highlighted avian influenza (AI) including Highly Pathogenic Avian
Influenza H5N1 (HPAI) because it is one of the WRZD global threats, and it has had a
significant impact on the health and livelihoods of SSI farmers in Vietnam. The major proportion
of the HPAI epidemics happened during 2003-2005 in the rural areas of the two main river deltas
in Vietnam where the most vulnerable SSI farmers reside (Hoang, 2006; Rushton et al., 2006).
Moreover, a number of reported risk factors for transmitting WRZD relate to SSI farming
practices. Water bodies are also believed to be modes of transmission and temporary reservoirs
for AI viruses at the village and farm level (Morris and Jackson, 2005). As well, AI viruses can
persist for extended periods in water, sewage, excreta, and animal wastes varying on different
factors (Boere et al., 2006; Morris and Jackson, 2005; Webster et al., 2006; WHO, 2007).
10 In my research, I defined a mitigating strategy to reduce the transmission of water-related zoonotic diseases
(WRZD) as a plan (e.g., a plan for managing livestock) that includes one or more activities an SSI farmer could
choose to reduce transmitting WRZD. An example of a mitigating strategy is good livestock management, which
may include actions, or tactics (e.g., treat animal waste at source) that SSI farmers could choose to reduce
transmitting WRZD (Carr and Bartram, 2004; WHO, 2004).
Page 52 of 340
The values of incorporating the principles of an EcoHealth approach are considered in
my research. Together with the research outlined in other chapters of this dissertation, the
research results in this chapter will assist human and animal health authorities, as well as
research and development agencies, to understand better how to influence current SSI farming
management practices. Particularly, the research results will be useful for SSI farmers, health
workers, veterinarians, researchers, and policy makers who have the potential to help revise,
formulate, and implement government policies in SSI farming and in water and health. The
research results will be useful for considering an EcoHealth approach in the context of SSI
farming to reduce risk factors for transmitting WRZD.
2.1.3. SSI farming and risk factors for transmission of HPAI
Highly Pathogenic Avian Influenza (HPAI) during the 2003-10 period in Vietnam caused
119 human cases of HPAI disease - the second highest human case incidence in the world and
case-fatality of 50% (Soares et al., 2010; WHO, 2010). HPAI also had significant impact on
Vietnamese socioeconomic condition with an estimate of 50 million poultry culled (Figuie,
2007; Otte, Hinrichs, Rushton, Roland-Holst, and Zilberman, 2008; Soares et al., 2010) while
68% of poor rural households rely on poultry as a primary source of income (FAO, 2010). The
reported risk factors for transmitting the HPAI virus are related to poultry production and trade,
behaviour and culture, interactions among wild and domestic birds, the water environment,
farming practices, and spatial and temporal variation (See Table 2.1.). The risk factors for
transmitting HPAI are also more likely driven by characteristics operating at the household level
rather than at the village level (Phan, 2010). Ducks pose a significant risk factor in the spread of
low pathogenic avian influenza (LPAI) as the virus can infect ducks without producing clinical
signs. In fact, several studies have concluded that ducks are asymptomatic reservoirs and
Page 53 of 340
transmitters of HPAI (Bui, 2010; Hulse-Post et al., 2005; Munster et al., 2005; Tracey et al.,
2004). Scavenging ducks are common and are thought to play an important role in maintaining
and transmitting the HPAI virus, but there is a lack of published information about the HPAI
virus infection status among these animals (Henning et al., 2009). Consuming some products
(e.g., eating fresh duck blood) is also considered a high risk factor in the transmission of HPAI to
humans (Care, 2011; Zuckerman, 2011). With respect to seasonality, HPAI outbreaks are now
occurring in the warmer months of the year with ducks being the prominent affected species in
the HPAI (H5N1) outbreaks (Minh et al., 2009).
SSI farming practices are also thought to play an important role in maintaining and
transmitting some infectious diseases (e.g., H5N1) due to the increased potential for interspecies
mixing among livestock, humans, and wild animals (Geurden et al., 2008; Henning et al., 2011).
Table 2.1 provides more information about the main groups of risk factors for transmitting HPAI
as reported by other researchers. Most of these risk factors can be linked with SSI farming
practices.
In summary, SSI farming practices play an important economic role for Vietnam’s rural
population. Nevertheless, there are differences between SSI farming practices in the Red River
Delta and those in the Mekong River Delta. Exploring whether characteristics of these SSI
farming practices are consistent with an EcoHealth approach can help determine potential health
risk factors, and reduce undesirable health consequences for SSI farmers, their animals, and their
environment.
2.1.4. Research questions and objectives
This chapter addresses the first two questions of my research. These questions are 1)
What is the socioeconomic profile of SSI farmers in the provinces of Thai Binh and An Giang in
Page 54 of 340
Vietnam? and 2) Is the socioeconomic profile of SSI farmers in the provinces of Thai Binh and
An Giang consistent with an EcoHealth approach in the context of SSI farming? More
specifically, my research in this chapter has two objectives. The first objective is to describe
characteristics of SSI farming in the research area in the context of the risk factors for
transmitting WRZD. The second objective is to discuss any potential ecological and health
implications of SSI farming in the context of risk factors for transmitting WRZD through using
an EcoHealth approach (See Table 2.1). I expect that the SSI farming practices used in the
provinces of Thai Binh and An Giang expose farmers to risk factors for transmitting WRZD.
Furthermore, I expect that not all characteristics of SSI farming practice in Thai Binh and An
Giang are consistent with an EcoHealth approach in the context of SSI farming.
Page 55 of 340
Table 2.1. Risk Factors for Transmission of Highly Pathogenic Avian Influenza (HPAI)
and Their Possible Connection with Small-Scale Integrated (SSI) Farming
Reported risk factors for transmission of HPAI
(order based on subjective priority11)
References
1. Poultry production (e.g., number of chickens,
unevenly distributed poultry supply-and-demand
system, contact with sick or dead poultry)
(Biswas P.K. et al., 2009a; Dinh et al.,
2006; Gilbert et al., 2004; Sheta et al.,
2014; Tiensin et al., 2009; Zhao et al.,
2013; Zhou et al., 2009)
2. Poultry trade (e.g., live poultry trade, poultry
slaughter house, and trade of poultry between farms)
(Desvaux et al., 2014; Fasina et al., 2011;
Henning et al., 2009; Paul et al., 2011;
Sims, 2007; Soares et al., 2010)
3. Duck production (e.g., number of ducks, and
scavenging ducks in rice field)
(Gilbert et al., 2006; Tiensin et al., 2009;
Paul et al., 2014)
4. Behaviours and culture (e.g., eating fresh duck
blood, visiting a stall or market selling live poultry
during the week before the illness)
(Care, 2011; Zuckerman, 2011)
5. Interaction between wild and domestic birds (e.g.,
abundance of domestic ducks and geese, and kept
birds/animals other than poultry)
(Boere et al., 2006; Sheta et al., 2014;
Tiensin et al., 2009; Zheng et al., 2010)
6. Relating to water and environment (e.g., water
bodies at the village and farm level and
environmental characteristics)
(Morris and Jackson, 2005; Paul et al.,
2014)
7. On-farm management practices (e.g., farm
visitors, visiting other farms, and farm biosecurity
management/practices)
(Fasina et al., 2011)
8. Spatial and temporal variation (e.g., river length
within a commune, poultry in close proximity and
increased accessibility to a water body/sources,
temporal flooding, seasonality, and distance to
water sources)
(Biswas P.K. et al., 2009b; Dang et al.,
2006; Fang et al., 2008; Vu et al., 2007;
Ward et al., 2008)
9. Other on-farm production (e.g., vegetation indices
and rice production)
(Gilbert et al., 2004; Pfeiffer et al., 2007)
11 In this table, based on my literature reading and my own judgment of priority, I subjectively grouped the risk
factors of transmission of HPAI (reported by other studies) by categories in order to show their possible connection
to small scale integrated farming systems.
Page 56 of 340
Figure 2.1. Chapter Two’s Outline
Chapter 2:
Presents a developed socioeconomic profile of SSI farmers in Thai Binh and An Giang Provinces of Vietnam,
including:
Used to assist modeling factors associated with on-farm water quality (Chapter 4) and factors associated
with SSI farmers’ engagement in mitigating strategies to reduce transmission of WRZDs (Chapter 5) in
the provinces of Thai Binh and An Giang in Vietnam
Research question 1:
What is the socioeconomic profile of SSI farmers in Thai Binh
and An Giang provinces of Vietnam?
Discussion of SSI
farming and EcoHealth
Farm economics (i.e., the
SSI farmers’ income and
contribution to farm
income of agricultural
production)
Livestock production
and fish production on
the SSI farms
Demographic profile of
the SSI farmers
Research question 2:
Is the socioeconomic profile of SSI farmers in Thai Binh and An
Giang consistent with an EcoHealth approach in the context of
SSI farming?
Page 57 of 340
2.2. Methods
2.2.1. Research design
2.2.1.1.Methods and phases of the research
In my research, quantitative and qualitative methods were used in a cross-sectional
research design. The following three phases were used to conduct the research: 1) establishing,
testing, piloting, and revising the questionnaires and survey protocol; 2) gathering data from the
target groups of farmers through the full field survey; and 3) analyzing the collected data.
Quantitative and qualitative data were collected at the same time. The quantitative data
were collected using questionnaires and objective assessments of microbial and related indicators
of water quality. Quantitative methods were used to: confirm my choice of hypothesis for testing
in my research; clean and check data; and summarize and analyse data12. Key qualitative method
steps included: identifying the target population (i.e., the eligible farmers in the research areas);
designing and testing the survey instruments; and defining and identifying variables that cross-
reference the research hypothesis. The following qualitative method tools were used to explore
culturally and meaningfully farmers’ perceptions and characteristics: in-depth interviews, focus
group discussions (FGD), meetings, field visit notes, observations, desk review, and audio-visual
materials. Qualitative data were used to assist the discussion of research results.
In the first phase of my research, I designed tools for the study based on my knowledge
of the research issues, the conceptual and theoretical models of the research, feedback from local
partners, and a review of the literature. These study tools included questionnaires and qualitative
data collection tools (i.e., guidelines for in-depth interviews, observations, and FGD). The study
tools were tested and used in a pilot study before being revised for use in a full field survey. In
12 STATA version 13 was employed for quantitative analysis.
Page 58 of 340
the pilot study, I tested the questionnaires and qualitative data collection tools by asking the
enumerators of the field survey to read all the questions and to share their understanding of the
meaning and clarity of the questions. Enumerators were also asked to role-play in pairs
specifically to ask and answer questions contained in the questionnaires, and to review the
guidelines in the qualitative data collection tools and then provide me feedback on the
appropriateness of the questions and procedures for the field survey. Following this, I revised the
questionnaire and the qualitative data collection tools considering the feedback I received from
the enumerators. The enumerators then administered the questionnaire to 30 farmers in Thai
Thuy district of Thai Binh province. The questionnaire was revised once again before I used it in
the second phase (i.e., the full field survey). Details of the pilot study is in section 2.2.2.
In the second phase of my research, I conducted the full survey in the two provinces to
gather data from 600 eligible rural households (i.e., farms), which were randomly selected to
participate in my research. The selection of these rural households is described in the sampling
design section below. Key variables included demographic and socioeconomic characteristics,
livestock production focusing on poultry production, and other key types of on-farm production.
Local partners including commune and village leaders, as well as commune and village
health/veterinary workers played an important role in the field data collection. In the third phase
of my research, the collected data were cleaned, coded, and organized before analyzing.
2.2.1.2.Research areas
The research team’s previous evaluation visit to the Red River and Mekong River Deltas
provided them with a background understanding of the related problems and the situation of SSI
farmers in the two river deltas including the provinces of Thai Binh and An Giang. This
background understanding led to selecting the provinces of Thai Binh and An Giang as the two
Page 59 of 340
target research areas for my research (See Figures 2.2). More importantly, these two provinces
are located in the centre of the Mekong River Delta and the Red River Delta where the HPAI
epidemic waves were focused (Hoang, 2006; Rushton et al., 2006).
There are 1.6 million and 1.5 million people respectively living in the provinces of Thai
Binh and An Giang. In 2010, the total number of farms/livestock farms were 3,376/2,388 (i.e., in
Thai Binh) and 17,273/218 (i.e., in An Giang) (GSO, 2012), which produced (as of October 1st,
2010) 8,899 (i.e., in Thai Binh) and 4,067 (i.e., in An Giang) heads of poultry (GSO, 2010). The
total area of surface water for aquaculture is 13.4 (i.e., in Thai Binh) and 1.9 (i.e., in An Giang)
thousand hectares (MARD, 2012). However, only 27.7% and 47.5% of the total numbers of
communes in Thai Binh and An Giang respectively have a common sewage system (GSO, 2010;
Nguyen et al., 2010b)
2.1.2.3.Selection criteria
I used four criteria to select participants/rural household residents from the research
areas. The first criterion was rural household residents who produced backyard poultry and
expected to do so for at least the next six months. The second criterion was rural household
residents who had a poultry flock size of about 50 birds (Burgos et al., 2007). The third criterion
was rural household residents who had access to at least one of the main water sources for rural
farmers (e.g., pond, river, well, or canal). The last criterion was rural household residents who
were interested in and willing to consent to participate in the research study. There were no
exclusion criteria.
Page 60 of 340
Figure 2.2. Research Areas – The Provinces of Thai Binh and An Giang in Vietnam13
13 This figure uses maps produced by FAO and D. Pfeiffer including: (1) Map of poultry density in Vietnam and (2) Map of irrigated and drained agricultural
land (FAO, 2015); and (3) Map of HPAI epidemic waves 2004 (Pfeiffer, 2015). Greater numbers are indicated by darker colours.
Page 61 of 340
2.2.1.4. Sampling design:
The choice of using both qualitative and quantitative methods together with insufficient
available information relating to effect size and variability challenged my ability to estimate the
sample size for my research. The sample size for the full field survey followed a
recommendation in which a minimum sample size pertaining to the most common qualitative
and quantitative and correlation designs would be 82 participants per group for a two tailed
hypothesis with 0.80 power at the 5% level of significance (Denise, 2010). Moreover, within a
5% to 10% level of significance assuming the large and medium conventional estimated
population correlations of 0.10 and 0.30, a sample size of 785 and 85 participants respectively
would be sufficient to keep the power of 0.80 and risks of statistical errors to standard levels
(Browner et al., 2013). Given this recommendation and the available resources for my research, a
sample size of 600 participants was feasible and acceptable.
I first conducted a pilot study among 30 farmers in which the questionnaires, the
qualitative data collection tools, and the survey procedures were tested, used in the field, and
then revised. The research team was already familiar with the target provinces through being
previously involved in health and agricultural research in these provinces. One district per
province14 was selected. This was based on a deliberate process in which the two districts chosen
had the closest matching population and number of communes to the mean population and
number of communes of all the districts in their respective province. The communes in these
districts reflected: 1) farming villages in Vietnam with a diverse ecosystem (e.g., ponds and
canals) that farmers were using for their poultry production; 2) farming systems engaging in
14 Vietnam is divided into 64 provinces. Each province is sub-divided into districts, then communes, and villages.
Page 62 of 340
various types of SSI farming and other agricultural activities; and 3) typical risk factors for
WRZD related to livestock production. Within the provinces of Thai Binh and An Giang, the
districts selected were Quynh Phu and Chau Thanh respectively. The desired sample size was
300 households per district.
A two-stage cluster sampling method was used to select target communes as the clusters,
and households in each cluster as the research participants. In the first stage, the existing number
of communes in each district was identified as the existing sampling frame. The number of
clusters selected in each district was defined by dividing the sample size by the estimated mean
number of population elements in each cluster. The clusters were selected based on the sampling
interval (S) obtained by dividing the total population by the number of clusters, and on a random
number (N) using a random number table generated by Microsoft Excel (Microsoft, 2014). In the
second stage, the technique of selecting a random sample from a table of random numbers was
used to select households to participate in the research. Commune leaders and the communal
farmers’ associations actively assisted the research team by providing the list of farmers for the
random selection of the sample.
2.2.1.5.Variables
For the research in this chapter, variables were collected in the following areas of
interest: 1) personal characteristics (e.g., age, education, and other demographic characteristics);
2) household socioeconomic characteristics (e.g., income, poverty status, gender, and family
size); and 3) on-farm and livestock production characteristics (e.g., type of livestock, size of
flock, cropping pattern, and water sources).
Data collection methods included the use of questionnaires, in-depth interviews, focus
group discussion, observations, and local health and veterinary reports. In particular, structured
Page 63 of 340
questionnaires were designed for interviewing the randomly selected farmers whereas semi-
structured questions and guidelines were designed for conducting the in-depth interviews, the
focus group discussions, and the observations of the participating farmers. Questionnaires
contained both open-ended and closed-ended questions. For the close-ended questions, farmers
had the option either of selecting a scaled response or of refusing to answer the questions.
2.2.2. Pilot study
From July 2012 to June 2013, a pilot study titled “Animals, water, and public health in
Vietnam” was conducted on 30 small-scale farms in the Thai Thuy district of Thai Binh province
in Vietnam. The Canada Excellence Research Chairs (CERC) program funded the pilot study
and the University of Prince Edward Island and the University of Calgary jointly led the one-
year study. The objectives of the pilot study were to examine the types of fish and livestock
raised on SSI farms in Vietnam and to test the study tools (i.e., the questionnaires, the qualitative
data collection tools, and the survey’s procedures). Testing and revising the study tools ensured
that the farmers easily understood the questions, and that the questions were free of ambiguity
while ensuring the validity and reproducibility of the responses. Farmers who participated in the
pilot study were not selected again for the full research study.
2.2.3. Data collection for the full research study
For both the pilot study and the full study, data collection followed the standard
ethical/consent procedures of the University of Calgary and the local research ethics board at the
Hanoi School of Public Health (see Appendix D). The field investigator recruited and trained the
enumerators on how to conduct the survey before they collected data in the field. Commune
collaborators/local guides assisted the field investigator and the enumerators throughout the field
survey. The research team worked with their local partners to identify commune collaborators
Page 64 of 340
who then informed the selected households of the nature of the survey and requested members of
each household to discuss and nominate a representative to participate in the survey.
Based upon their knowledge, the household representatives were asked to provide their
birthday to determine the ages of farmers. Those who did not remember their birthdays were
asked to check their birth certificate and/or check it with other household members. The lists of
farmers provided by the commune collaborators also noted farmers’ birthdays for cross-
reference. Prior to the interview, consent forms were explained carefully to the selected farmers,
and only farmers who understood and agreed with the consent scripts were interviewed.
Informed consent was also attained for the focus group discussions, in-depth interviews, and
observations. At the end of each interview with a farmer, a bar of soap and a hand towel were
provided to the farmer as tokens of appreciation. No farmers knew about the tokens of
appreciation until after the completion of the interviews. The questionnaires were pretested and
integrated into five mini iPads using licensed Survey Pocket15 and QuestionsPro16 software. The
enumerators administered the questionnaires during in-person interviews with farmers. In each
province, a senior local researcher and I supported the enumerators throughout the data
collection (e.g., accompanying enumerators, observing enumerators in actions, answering
questions if any, having a briefing session with all enumerators before each day of data
collection and a reviewing session at the end of each day of data collection).
In the procedures for collecting qualitative information, the research team purposefully
selected farmers and other key stakeholders (informants) for the focus group discussions and in-
15 SurveyPocket is a survey application to conduct field surveys on iPad, iPhone or Android tablet and smart phone
devices (http://www.surveypocket.com/ ). 16 QuestionPro is a web-based software for creating and distributing surveys
(http://www.questionpro.com/home/howItWorks.html ).
Page 65 of 340
depth interviews. Local authorities were asked to share relevant documents for review, if they
were available. Each focus group discussion included from five to ten farmers and took
approximately one hour to complete. The enumerators also facilitated the focus group
discussions by using the predesigned guidelines and guided questions. For the in-depth
interviews, each enumerator asked open-ended questions face-to-face with each farmer who was
representing a household. Heads of households and/or the person with responsibility for
livestock production, water, and related issues in the households were the primary respondents.
These procedures provided opportunities for participants to express qualitative information such
as questions, issues, comments, and inputs, which may not been sufficiently addressed in the
questionnaires.
Both the field investigator and the enumerators were experienced in conducting
interviews and administering questionnaires and they were familiar with the local agriculture
context. The enumerators also not only helped the field investigator in administering the
questionnaires, but in recording the interviews, and probing for answers, comments, and
expressions of the responding farmers. Audio records were then analyzed and summarized.
Farmers were also informed in advance that codes would be used in place of names in the
research to ensure confidentiality and to prevent bias.
Instructions and checklists were used to guide the interviews and the focus group
discussions to ensure consistency and to avoid missing important information. With respect to
the secondary data relating to livestock production, environment, and farmers’ health, the field
investigator asked local authorities to share information. However, the obtained secondary
information was limited as it was not always available, nor comprehensive, and sometimes it was
too general. Information obtained from the in-depth interviews, the focus group discussions, and
Page 66 of 340
the literature review was used in discussing the results. This information primarily relates to
livestock production, water use, and farmers’ health in the studied areas.
2.2.4. Analysis of the full field survey’s data
The raw data collected from all the farmers who met the selection criteria and fully
participated in the field research were entered into a dataset, cleaned, coded, and edited.
Descriptive statistics were used to inspect frequency distributions (Long and Freese, 2014) and
central tendency (continuous variables) relating to the socioeconomic characteristics of the
participating farmers and their livestock production. Statistical tests were used to test significant
differences of research results between two provinces (e.g., Pearson Chi-squared test, t-test, and
Fisher's exact test).
As the primary researcher, I visually inspected the dataset using graphs and looked for
outliers. Fifteen farmers had young fish production numbers that appeared to deviate markedly
from those of other farmers and might reflect some abnormality or errors in the measurement or
recording. Therefore, I examined the variable for young fish production further and found that 10
farmers in the province of An Giang (i.e., 10 observations in the dataset) had young fish
production (i.e., 1.3 billion young fish per year) that lay outside the normal range of production
compared to the other participating farmers. In addition, the mean years of attending school and
the mean income/person/year were much greater than the rest of the SSI farmers (i.e., 9.4 years
of schooling, and > 20 million Vietnamese Dong (VND)17 respectively compared to 5,769 young
fish per year, 6.9 years of schooling, and < 20 million VND respectively). I also calculated the
interquartile range (IQR) as well as the outer fences (i.e., the lower outer limit and the upper
17 1 CAD = 17,206.42 VND as of Oct. 12, 2015.
Page 67 of 340
outer limit) of the data on young fish production. The numbers of young fish produced in these
10 farms were more than three times the IQR above the 75th percentile (Q3) of the number of
young fish produced in the farms that produced fish. Furthermore, the numbers of young fish
produced on the 10 farms were much greater than the national mean number of fingerlings
produced per farm, per year, which was 16,472,868 (Ngoc, 2013).
Given the above calculations, I considered the numbers of young fish produced in these
10 farms as extreme outliers, and that including these numbers in analyzing young fish
production may bias the dataset. Therefore, I dropped these young fish production numbers (but
not the 10 farms) in describing young fish production in the participating farms. Furthermore, I
examined the profile of the 10 farms with extreme numbers of young fish production further by
stratifying the data over various income and education levels. This stratification was done to
improve the accuracy (i.e., valid representation) of the strata for the number of young fish
produced among the stratified participating farmers. I also used frameworks for SSI farming
(Dixon et al., 2004; Nguyen, 2014) in further examining the profile of these 10 farms concerning
the numbers of adult fish produced, volume of fish production in kg, level of education, on-farm
income, and number of livestock. I found that two of the 10 farms with extreme outliers in young
fish production were definitely not SSI farms. Therefore, I decided to remove these two farms
entirely from the dataset. Qualitative information gathered from the focus group discussions, in-
depth interviews, and secondary data collections were used to inform with the discussion and
interpretation of the results of the quantitative analysis. The focus of this dissertation was not on
reporting qualitative results analyzed using Nvivo.
Page 68 of 340
2.3. Results
2.3.1. Demographics
2.3.1.1.Age, years of farming, years of attending school, and household size
The mean ages of participating farmers in Thai Binh and An Giang were 47.5 years and
44.4 years respectively (range 17-85) (Table 2.2). Participating farmers had a wide range of
years of integrated farming experience (i.e., a range of 1-40 years and a range 1-50 years in An
Giang and Thai Binh respectively). At the time of the study, farmers in An Giang appeared to
spend less years working as SSI farmers than those in Thai Binh did (i.e., 8 years and 12 years
respectively). Participating farmers spent nearly six years and eight years in school in An Giang
and Thai Binh respectively. The mean family size was four people in both An Giang and Thai
Binh. The mean number of children (under 18 years old) was about one.
Table 2.2. Demographics of Small-Scale Integrated (SSI) Farm Study Participants in the
Provinces of Thai Binh and An Giang, Vietnam
Variable Obs Mean Std. Dev. Min Max
Thai Binh
Age in years 299 47.5*** 11.9 17 85
Years of schooling 300 8.2*** 2.7 2 18
Household size 300 4.2*** 1.3 2 9
No. of people <18 yrs
old 300 1.0*** 0.9 0 3
Years of farming 300 11.7*** 8.7 1 50
An Giang
Age in years 298 44.4 10.5 19 76
Years of schooling 298 5.6 3.2 0 16
Household size 298 4.7 1.3 2 10
No. of people <18 yrs
old 298 1.4 1.0 0 5
Years of farming 298 7.6 7.2 1 40
(***) indicates a statistical significant difference between Thai Binh and An Giang p < 0.01
Page 69 of 340
Table 2.3 shows that most farmers in both provinces attended some years in school up to
high school grades18. However, only 54/300 and 37/298 of farmers in Thai Binh and An Giang
respectively attended 12 years of school (high school level). Most of the participating farmers in
Thai Binh spent some years attending secondary school while most farmers in An Giang
attended primary school only from grade one to grade six. Participating farmers in An Giang had
mainly attended primary school while those in Thai Binh had mostly attended up to the
secondary school level. Sixteen farmers in Thai Binh (5% of the participating farmers in Thai
Binh) had also attended post-secondary school. This number was much greater than the number
of farmers who attended post-secondary school in An Giang (i.e., one or 0.3% of the
participating farmers in An Giang).
Table 2.3. Types of School Attended by Small-Scale Integrated (SSI) Farmers in the
Provinces of Thai Binh and An Giang, Vietnam
Types of school
attended*
Provinces
Thai Binh An Giang Total
Number Percent Number Percent Number Percent
No school 0 0.0 7 2.3 7 1.2
Primary school 56 18.7 200 67.1 256 42.8
Secondary
school 174 58.0 53 17.8 227 38.0
High school 54 18.0 37 12.4 91 15.2
Post-high school 16 5.3 1 0.3 17 2.8
Total 300 100.0 298 100.0 598 100.0
(*) significant difference between Thai Binh and An Giang with p = 0.000, two-tailed Fisher's exact test
2.3.1.2.Gender and gender role
Most farmers who were nominated by their family members to participate in the survey
were men (206/300 or 68.7% and 222/298 or 74.5% in Thai Binh and An Giang respectively)
18 In Vietnam, a primary school includes grade one to grade five classes, a secondary school includes grade six to
grade nine classes, and a high school includes grade ten to grade twelve classes
Page 70 of 340
and most of them were heads of households (195/300 or 65.0% and 208/298 or 69.1% in Thai
Binh and An Giang respectively (Table 2.4). However, it was common for both men and women
to share responsibility for family health (159/299 or 53.2% and 197/292 or 67.5% of households
in Thai Binh and An Giang respectively) (Table 2.5). In Thai Binh, male farmers were mainly
responsible for livestock production, fish production, and family health in 144/295 (38.6%),
134/287 (46.7%), and 67/299 (22.4%) of households respectively. Meanwhile, female farmers
were mainly responsible for livestock production, fish production, and family health in 54/295
(18.3%), 29/287 (10.1%), and 73/299 (24.4%) respectively. In An Giang, male farmers were
mainly responsible for livestock production, fish production, and family health in 130/294
(44.2%), 53/186 (28.5%), and 64/292 (21.9%) of households respectively. Meanwhile, female
farmers were mainly responsible for livestock production, fish production, and family health in
37/294 (12.6%), 12/186 (6.5%), and 31/292 (10.6%) respectively (Table 2.5).
Page 71 of 340
Table 2.4. Gender of the Participating Farmers in the Provinces of Thai Binh and An
Giang in Vietnam and Their Relationship with Other Household Members
Province/
Gender Relationship with other household members
Head of household a Other roles b Total
Thai Binh Number Percent Number Percent Number Percent
Male 195 65.0 11 3.7 206 68.7
Female 24 8.0 70 23.3 94 31.3
An Giang
Male 206 69.1 16 5.4 222 74.5
Female 16 5.4 60 20.1 76 25.5
Total
Male 401 67.0 27 5.0 428 72.0
Female 40 6.0 130 22.0 170 28.0 (a) no significant difference between Thai Binh and An Giang with p = 0.677, Pearson Chi-squared test
(b) no significant difference between Thai Binh and An Giang with p = 0.690, Pearson Chi-squared test
Table 2.5. Gender Roles in Small-Scale Integrated (SSI) Farm Households in the Provinces
of Thai Binh and An Giang in Vietnam
Gender roles
Provinces
Thai Binh An Giang Total
Number Percent Number Percent Number Percent
Mainly responsible for livestock production a
Male 114 38.6 130 44.2 244 41.4
Female 54 18.3 37 12.6 91 15.4
Both 127 43.1 127 43.2 254 43.1
Total 295 100.0 294 100.0 589 100.0
Mainly responsible for fish production
Male 134** 46.7 53 28.5 187 39.5
Female 29** 10.1 12 6.5 41 8.7
Both 124** 43.2 74 39.8 198 41.9
Not sure 0** 0.0 47 25.3 47 9.9
Total 287 100.0 186 100.0 473 100.0
Mainly responsible for family health
Male 67** 22.4 64 21.9 131 22.2
Female 73** 24.4 31 10.6 104 17.6
Both 159** 53.2 197 67.5 356 60.2
Total 299 100.0 292 100.0 591 100.0 (a) no significant difference between Thai Binh and An Giang with p = 0.121, Pearson Chi-squared test
(**) significant difference between Thai Binh and An Giang with p = 0.000, Pearson Chi-squared test
Page 72 of 340
2.3.2. On-farm production
2.3.2.1.Fish production
In Thai Binh, SSI farmers with either secondary school or post-high school or incomes of
10 to 40 million VND per person per year produced the greatest mean number of young fish per
year. In An Giang, SSI farmers with either primary school or secondary school and incomes of
10 to 40 million VND per person per year produced the greatest mean number of young fish per
year (Table 2.6).
Table 2.6. Mean Quantity of Young Fish Produced Per Year on Small-Scale Integrated
(SSI) Farms in the Provinces of Thai Binh and An Giang, Vietnam, Stratified for
Education and Income
Provinces/
Income b
Highest level of educational attainment a
1 2 3 4 5
Thai Binh
< 4.8 - 3,800.0 2,500.0 - -
- (5,377.7) (4,203.6) - -
4.8-10 - 2,469.2 6,103.6 933.6 520.0
- (2,593.1) (17,732.2) (896.5) (678.8)
10-20 - 9,555.6 5,146.1 3,472.7 14,190.0
(12,875.5) (17,010.5) (3,733.4) (21,759.3)
20-40 - 1,000.0 10,000.0 1,000.0 4,000.0
- n.ac (11,742.0) n.a n.a
>40 - 10,000.0 1,508.3 800.0 200.0
- n.a (985.1) (244.9) n.a
An Giang
< 4.8 3 1,307.4 75.0 129.5 -
- (2,001.0) (35.4) (247.0) -
4.8-10 - 1,357.6 3,100.0 1,850.0 -
- (1,544.8) (3,489.3) (2,333.4)
10-20 - 18,006.1 6,000.0 7,006.7 -
- (27,239.7) n.a (11,263.2) -
20-40 - 14,000.0 100,000.0 100,000.0 -
- (8,485.3) n.a n.a -
>40 - - - - - (.) Numbers in brackets are standard deviations; a 1 = no school; 2 = primary school; 3 = secondary school; 4 = high
school; and 5 = post-high school; b Income was reported in millions of Vietnamese dong (VND) c No s.d. is reported where only one observation exists
Page 73 of 340
2.3.2.2.Livestock production
Table 2.7 shows that SSI farms in Thai Binh raised more chickens, ducks, and pigs than
SSI farms in An Giang. However, the mean number of cattle raised on a SSI farm in An Giang
was about three times greater than in Thai Binh.
2.3.2.3.On-farm production other than fish and livestock production
In addition to livestock production and fish production, participating farmers in An Giang
and Thai Binh also had other on-farm production of which rice and vegetable production were
the two most common types (Table 2.8). However, very few farmers were producing cassava,
bamboo, or peanuts.
Table 2.7. Annual Livestock Production on Small-Scale Integrated Farms (SSI) in the Thai
Binh and An Giang Provinces in Vietnam
Province/Livestock Annual on-farm livestock production by province
No. of farms Percent of farms Mean sd Max Min
Thai Binh
Number of pigs 193 65 24.8*** 18.9 70 1
Number of ducks 126 42 57.1*** 56.9 200 1
Number of chickens 272 91 33.5*** 22.4 100 1
Number of cattle 65 22 1.4*** 1.0 5 1
An Giang
Number of pigs 101 34 7.7 9.7 50 1
Number of ducks 119 40 14.1 18.8 100 1
Number of chickens 201 67 11.9 10.7 52 1
Number of cattle 128 43 3.9 2.1 10 1 (***) indicates a statistical significant difference between Thai Binh and An Giang p < 0.01, t-test
2.3.2.4.Occupation other than farming among participating farmers
Participating farmers were asked to indicate any other work or employment they had in
addition to working as integrated small-scale livestock farmers. Table 2.9 shows that in both
provinces, almost half of the participating farmers had no additional work. Among those who
had additional jobs, the most common in Thai Binh and An Giang were owning a small
Page 74 of 340
business/being a vendor, working communal committees19, or doing seasonal work respectively.
The most commonly reported seasonal work included temporary work relating to construction,
brick production, mechanics, motorbike taxi, rice processing, and on-call locally hired workers.
Table 2.8. On-farm Production Other Than Fish and Livestock on Small-Scale Integrated
(SSI) Farms in the Thai Binh and An Giang Provinces in Vietnam
Province Numbers and percent of farms by commodity and province
Rice Vegetables Fruits Corn Peanuts Cassava Bamboo
Thai Binh a
Number 234.0 216.0 195.0 26.0 22.0 5.0 7.0
Percent 78.0 72.0 65.0 8.7 7.3 1.7 2.3
An Giang b
Number 207.0 77.0 46.0 29.0 6.0 23.0 15.0
Percent 69.5 26.8 15.4 9.7 2.0 7.7 5.0
Both provinces c
Number 441.0 293.0 241.0 55.0 28.0 28.0 22.0
Percent 73.7 49.0 40.3 9.2 4.7 4.7 3.7 a n = 300 (in the province of Thai Binh) b n = 298 (in the province of An Giang) c n = 598 (in the provinces of Thai Binh and An Giang)
19 This may include a position at a commune health centre, commune veterinary station, commune farmers’ union,
or commune women’s union
Page 75 of 340
Table 2.9. Occupations Other Than Farming Among Small-Scale Integrated (SSI) Farmers
in the Provinces of Thai Binh and An Giang, Vietnam
Occupation other
than farming
Numbers of farmers by occupation and province
Thai Binh An Giang Total
Number Percent Number Percent Number Percent
No additional work 163 55.1 137 46.1 300 50.6
Communal staff 20 6.8 7 2.4 27 4.6
Small
business/vendor 60 20.3 39 13.1 99
16.7
Small production 29 9.8 7 2.4 36 6.1
Seasonal work 19 6.4 107 36.0 126 21.2
Retired 5 1.7 0 0.0 5 0.8
Total 296 100.0 297 100.0 593 100.0
2.3.3. Farm economics
2.3.3.1.Income of SSI farmers
Most SSI farmers reported their income from farming (expressed in VND per person per
year) as being between 4.8 million VND (equivalent to $244 CAD)20 to 20 million VND
(equivalent to $1,015 CAD) (Table 2.10). A limited number of farmers had income per year that
was higher than 40 million VND (equivalent to $2,030 CAD) – specifically 18 farmers (6.1%)
and six farmers (2.0%) in Thai Binh and An Giang respectively. Eleven (4%) and 118 (36%) of
participating farmers in Thai Binh and An Giang respectively had annual incomes that were
lower than the cut-off income level21 for poverty set by the Vietnamese government.
20 At the exchange rate of $1CAD = 19,700 VND on 5th December 2013 (Source: Vietinbank) 21 Less than 4.8 million VND per person per year – equivalent to $244 CAD per person per year
Page 76 of 340
Table 2.10. Income of Small-Scale Integrated (SSI) Farmers in the Provinces of Thai Binh
and An Giang, Vietnam
On-farm income
per person per
year (VND
million)
Number and% of farmers by income and province
Thai Binh An Giang Total
Number Percent Number Percent Number Percent
< 4.8 11*** 3.7 107 36.1 118 20.0
4.8-10 106*** 36.1 104 35.1 210 35.6
10-20 127*** 43.2 61 20.6 188 31.9
20-40 32*** 10.9 18 6.1 50 8.5
> 40 18*** 6.1 6 2.0 24 4.1
Total 294 100.0 296 100.0 590 100.0 (***) significant difference between Thai Binh and An Giang with p = 0.000, Pearson Chi-squared test
2.3.3.2.Contribution to total farm income from various kinds of agriculture production
Participating farmers were also asked to rate the level of income that they obtain from
their various on-farm production activities. Table 2.11a-b shows that although poultry, fish,
duck, pig, and rice production brought more income to SSI farmers than other vegetable
production, about 28% to 53% of farmers in An Giang and Thai Binh respectively indicated that
they receive either moderate or moderate to high farm income from poultry production. Yet,
approximately 5% of farmers in the province of Thai Binh considered fish, duck, and pig
production as the top three farm activities for providing “extremely high” contributions to on-
farm income compared to other forms of agricultural production. In the province of An Giang,
less than 2% of farmers considered fruit, fish, duck, and pig production as farm activities for
providing “extremely high” contributions to on-farm income compared to other forms of
agricultural production.
Page 77 of 340
Table 2.11a. Farmers’ Perceptions of the Importance of Contribution to Total Farm
Income from Various Types of Agricultural Production on Small-Scale Integrated (SSI)
Farms in the Province of Thai Binh, Vietnam
Levela Contribution of on-farm production to farm income (%) by agricultural activity
Poultr
y
Fish Duck Pig Rice Cassav
a
Bamboo Peanut Cor
n
Vegetab
le
Fruit
1 2.2 c 5.4 4.6 4.2 - - 10.0 - - - -
2 22.3 57.9 26.2 60.6 6.7 - 10.0 - 21.4 9.1 5.8
3 53.1 23.0 57.9 28.6 53.1 - 20.0 40.0 50.0 50.0 54.7
4 15.0 10.8 6.7 6.1 34.3 - 50.0 50.0 28.6 16.8 21.6
5 7.3 2.9 4.6 0.5 5.9 - 10.0 10.0 - 24.0 17.9
Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0
Nb 273 278 195 213 239 0 10 20 28 208 190
a Levels of contribution to income: 1 = extremely high contribution; 2 = high contribution; 3 = moderately high
contribution; 4 = low contribution; and 5 = no contribution; b N = number of respondents; c For example, of 273
respondents who raised poultry, 2.2% indicated that poultry production made an “extremely high” contribution to
farm income
Table 2.11b. Farmers’ Perceptions of the Importance of Contribution to Total Farm
Income from Various Types of Agricultural Production on Small-Scale Integrated (SSI)
Farms in the Province of An Giang, Vietnam
Levela
Contribution of on-farm production to farm income (%) by agricultural activity
Poultr
y
Fish Duck Pig Rice Cassav
a
Bamboo Peanut Cor
n
Vegetabl
e
Fruit
1 - - 0.4 0.4 1.9 - - - - - 0.5
2 3.9 c 12.5 2.4 10.3 14.1 6.6 - 2.0 8.1 4.1 -
3 28.5 18.1 17.9 17.3 38.9 5.2 0.5 3.9 4.3 4.6 1.4
4 28.5 16.0 16.3 10.7 15.9 - 6.2 1.0 1.4 16.8 16.0
5 39.1 53.4 63.1 61.3 29.3 88.2 93.3 93.3 86.1 74.6 82.2
Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0
Nb 256 232 246 243 270 211 209 208 209 220 219
a Levels of contribution to income: 1 = extremely high contribution; 2 = high contribution; 3 = moderately high
contribution; 4 = low contribution; and 5 = no contribution; b N = number of respondents; c For example, of 256
respondents who raised poultry, 3.9% indicated that poultry production made an “high” contribution to farm income
Page 78 of 340
2.4. Discussion
2.4.1. SSI farming practice and its profile in the provinces of Thai Binh and An Giang
Overall, SSI farming practice including the VAC was promoted as a model in rural areas
in Vietnam for a couple of decades after the introduction of the “Doi moi” reform policy in
198622. This is an important policy milestone with a shift from a centrally planned and
subsidized system to a market oriented system. This policy shift has led to the allocation of
production materials such as land to individual farmers instead of to collective ownership. The
VAC model was introduced as an integrated farm management system focusing on three core
components – the garden, the fishpond, and animal husbandry – with the aim of improving
farming practices and achieving high economic efficiency through optimizing the use of land and
water with low capital investment and inputs (Le, 2001; Pham, 2010). In some other Asian
countries, several integrated farming practices that are similar to the VAC model also exist such
as integrated fish-horticulture farming (India), a culture of short-cycle species in seasonal ponds
and ditches (Bangladesh), fodder-fish integration practices (Malaysia), and embankment fish
culture and integrated grass-fish farming (China) (Pullin, 2001).
My research in this chapter shows that there were differences between ages of the farmers
from the provinces of Thai Binh and An Giang. The participating farmers from An Giang were
younger and had fewer years of schooling compared to the participating farmers in Thai Binh.
Rural farmers are still practicing the small-scale integration of various on-farm production
activities with livestock, fish, crop, and so on, which is similar to the reported structure of SSI
farming in Vietnam (Devendra and Thomas, 2002). However, the participating farmers in this
22 In 1896, the Vietnamese government introduced “Doi moi” policy, which indicates a shift the Vietnamese
economy from a centrally planned economy to a market oriented economy (Dang, 2015; Le, 2013).
Page 79 of 340
study preferred to refer to their farming practice as a normal/traditional way that most farmers
practice farming instead of as an SSI model. This may be explained by the fact that SSI farming
has been introduced in rural areas of Vietnam over the last couple of decades (Hop, 2003).
Farmers with secondary school education and greater income appeared to produce the greatest
mean number of young fish per year. Concerning fish production in An Giang, a number of
farmers had recently stopped producing fish on farms and the qualitative study revealed three
main reasons for this. First, producing fish required greater investment yet yielded low profits.
Second, the restructuring of the irrigation system limited the volume and quality of water input
for ponds. Third, farmers felt it was more convenient to catch fish from canals or rivers or to buy
fish for food than to produce fish on the farm. The insight from this is that programs and policies
targeting livestock production among SSI farms should address different kinds of livestock and
place an emphasis on poultry production.
A majority of reported risk factors for HPAI are related to SSI farming (See Table 2.1)
yet future food availability is projected to depend on improving the productivity for livestock
production in SSI farming systems (CGIAR, 2011; Thomas et al., 2002), with Vietnam being
among the top five countries in Asia to have a large number of small farms (Oksana, 2005).
Moreover, small-scale livestock production is the fastest growing sub-sector of agriculture in
Vietnam and will likely be a significant sub-sector for the next 20 years. Therefore, it is
important to review SSI farming using an EcoHealth approach to determine how to improve SSI
farming in the area of managing risk factors for emerging infectious diseases.
2.4.2. SSI farming and EcoHealth
With respect to complexity and trans-disciplinarity, SSI farming practice is truly a
complex environment where there is an interface between humans, animals, and the
Page 80 of 340
environment. The integration of gardens, fishponds, and animal husbandry is an example of
complexity where animal manure is used as fertilizer and food for plants and fish. Livestock
production can also bring income and food for farmers, food for animals (e.g., using throwaway
parts of fish as food for animals), and wastewater for gardens. Products from gardens can
likewise be used as food for fish and animals, and bring income for farmers. Such examples of
integration can help optimize on-farm resources. However, this integration correspondingly has
potential linkages to increasing risk factors for transmitting infectious diseases from livestock
and fish to humans (Dang and Dalsgaard, 2012; Huong et al., 2014; Markwell and Shortridge,
1982; Nguyen et al., 2010a; Nguyen et al., 2010b). Moreover, complexity is also reflected in
geographical variation in SSI farming. For instance, the mean volume of fish production for
participating farmers in An Giang is significantly greater than in Thai Binh, and the mean
number of cattle raised in a household in An Giang was significantly greater than in Thai Binh.
Therefore, this complexity requires the knowledge and involvement of various disciplines such
as veterinary health, public health, water security, the private sector, and so on to not only
improve farming practices and generate income but also to enhance the health of farmers and
their animals and to help them avoid polluting their environment.
Regarding community participation, although farmers are the owners and implementers
of this SSI farming practice, their involvement in designing, implementing, and monitoring
policies related to livestock production, rural water supplies, and sanitation is limited (Le et al.,
2014). Yet, to optimize the limited water resources available for farmers and reduce the negative
socioeconomic and health impacts on the rural population, it is important to more effectively
enhance community participation in the policy making process not only with policy makers but
Page 81 of 340
also with farmers themselves. Such actions could more explicitly help address the issues related
to SSI farming such as water, health, and small-scale livestock production.
In connection with gender equity, my research results show that male farmers played a
greater role in the SSI farms than female farmers did. More specifically, male farmers headed
most of the SSI farms. The percentages of farms with male farmers mainly responsible for
livestock production, fish production, and family health were greater than those percentages for
female farmers except for family health in Thai Binh. I think these results are interesting. I have
not found any other studies that explore the same gender equity issues among SSI farms (i.e.,
gender roles in livestock production, fish production, and family health) to compare with the
results of my research. However, some case studies of roles of women in agriculture reported
that women heavily participate in livestock production and the aquaculture sector (FAO, 2011;
Ranney et al., 2011). My research results focused on gender roles and main responsibilities for
livestock production, fish production, and family health. However, cautions may be needed in
generalizing the results as gender roles may change depending on types/scales of livestock
production on farms. Male farmers may have reported differently as they may feel uncomfortable
with female farmers being mainly responsible for livestock production, fish production, and
family health. It may also be cultural issues where men claim responsibility for women’s roles.
Furthermore, there may be differences between, for example, being responsible for livestock
production (e.g., making decisions) and doing tasks relating to livestock production (e.g., feeding
livestock). Therefore, future research may need to consider these differences/possible changes in
order to understand more fully gender roles among small-scale farming.
Concerning social equity among SSI farms, the application of the VAC model was
reported to generate a mean income from about 50 million VND to 70 million VND per year, per
Page 82 of 340
household – equivalent to $2,530 CAD to $3,335 CAD per household, per year (Pham, 2010).
Among the study’s participating farmers, the mean number of people in a household was 4.4.
Most of the participating farmers earned 4.8 million VND to 10 million VND per person, per
year, which is equivalent to 21.12 million VND to 44 million VND per household, per year. This
level of combined household income – equivalent to $1,072 CAD to $2,233 CAD respectively –
is lower than the income level Pham (2010) reported from the VAC model. Further, about one in
five of the SSI farmers had income that fell below the existing poverty line set by the
government of Vietnam. Approximately, ten times more SSI farms were below the poverty
income level in An Giang than in Thai Binh. The mean family sizes of the participating
households in Thai Binh and An Giang were slightly larger than the 2009 mean family size in
Vietnam which was 3.8 (GSO, 2011).
Given my research results shows a decrease in mean income with a slightly larger mean
family size, it is understandable that farmers are looking for off-farm seasonal work to earn
additional income. However, with limited education and training, it is a challenge for farmers to
look for off-farm seasonal work. Moreover, the percentage of farmers who attended university in
the provinces of Thai Binh and An Giang (i.e., 5.3% and 0.3% respectively) was lower than the
percentage of those in the general population in the Red River Delta and the Mekong River Delta
who completed university – 6.3% and 2% respectively (GSO, 2011). In-depth interviews with
farmers showed that farmers who took on off-farm seasonal work were from the younger
generation and had to work far from home in poor conditions with limited social benefits.
Moreover, although their income from off-farm seasonal work was greater than that from
livestock production, their away-from-home living expenses were also high.
Page 83 of 340
Pertaining to the idea of introducing the SSI farming model was the goal of enhancing the
sustainability of farmers’ livelihoods. Sustainability was introduced at the beginning of the
model in which the nature of integrating various on-farm production activities – from fish and
livestock to crop production – were for optimizing on-farm resources to produce greater
sustainability for farmers’ livelihoods (Hop, 2003).
However, from the standpoint of an EcoHealth approach in the context of SSI farming,
several key issues may influence the sustainability of both the SSI farming model itself and the
health of farmers and their animals. These issues include the following: the level of income from
integrated farming and the trend among farmers of doing off-farm work; the attention paid to the
health of farmers and their livestock; guidelines for managing health risk in the model; private
sector connections and markets; and protection of the environment, particularly the quality of on-
farm water while practicing integrated farming. In addition, SSI farmers are not equipped to face
challenges and competitiveness when Vietnam joins the World Trade Organization (Nguyen,
2015; Vu, 2016; Pham, 2008; Vo, 2005). Meanwhile, existing public policies have not been
updated to support farmers, which may worsen the undesirable consequences of SSI farming,
instead of improving them (Hall and Le, 2009; Le et al., 2014). Among the above issues, the
income of farmers was explored in the SSI model, as well as for the participating farms in my
research. However, other issues, such as those mentioned above, need to be studied further to
better understand potential implications for an EcoHealth approach in the context of SSI farming
2.5. Conclusions
The SSI farming model is a common agricultural practice in Vietnam and will continue to
be an important part of agriculture in Vietnam for the future. This model has proven to be
successful in helping farmers optimize their on-farm resources and generate food and income.
Page 84 of 340
Reported risk factors for transmission of WRZD with respect to SSI farmers could lead to
unintended health consequences for not only farmers and their animals, but also for their
environment. The characteristics of SSI farmers described in this study can also be used as a
basis for future studies of the SSI farming model. The EcoHealth approach has been considered a
good model for community development (Finkelman et al., 2008). In Vietnam, there are some
valuable examples of applying the EcoHealth approach in academic research, policy advocacy,
and community development (Hall et al., 2012; Nguyen-Viet et al., 2013; Nguyen-Viet et al.,
2015). Nevertheless, when using an EcoHealth perspective and taking into consideration the
increase of WRZD, this model must be reviewed to ensure that it will not be a simple agricultural
economic model in which food and income are outputs. Rather, it should be a more holistic
EcoHealth-based model in which the health of farmers and the health of animals are interwoven
with on-farm production and the protection of the environment, and these are considered an
equally important output of the model.
This chapter presented results relating to the profile of farmers who practice the SSI
farming model in the provinces of Thai Binh and An Giang in Vietnam. However, further studies
are needed to understand the SSI model more comprehensively with respect to an EcoHealth
approach. Researchable issues relating to an SSI farming model may include risk factors for
transmission of WRZD among humans and animals; farmers’ perceptions of health-related
factors; the contributing factor of farmers’ mitigating strategies for WRZD transmission; on-farm
environmental health, especially water public health; and the contribution of improved SSI
farming methods to reduce the exposure to risk factors in WRZD transmission.
Page 85 of 340
References
Biswas P.K., Christensen J.P., Ahmed S.S.U., Das A., Rahman M.H., Barua H. et al. (2009a).
Risk for infection with highly pathogenic avian influenza virus (H5N1) in backyard
chickens, Bangladesh. Emerg Infect Dis [serial on the Internet].
Biswas P.K., Christensen J.P., Ahmed S.S.U., Das A., Rahman M.H., Barua H. et al. (2009b).
Risk for infection with highly pathogenic avian influenza virus (H5N1) in backyard
chickens, Bangladesh. Emerg Infect Dis [serial on the Internet].
Boere, G. C., Galbraith, C. A., and Stroud, D. A. (2006). Waterbirds Around the World The
Stationery Office, UK.
Bosma, R. H., Phong, L. T., Kaymak, U., Van Den Berg, J., Udo, H. M. J., van Mensvoort, M. E.
F. et al. (2006). Assessing and modelling farmers' decision-making on integrating
aquaculture into agriculture in the Mekong Delta. NJAS - Wageningen Journal of Life
Sciences, 53, 281-300.
Browner, W. S., Newman, T. B., and Hulley, S. B. (2013). Estimating Sample Size and Power:
Applications and Examples. In S.B.Hulley, S. R. Cumming, W. S. Browner, D. G. Grady,
and T. B. Newman (Eds.), Designing Clinical Research (Fourth Edition ed.,
Philadelphia: Wolters Kluwer Health and Lippincott Williams and Wilkins.
Bui, T. D., Hien, N. T., and Dang, T. T. (2010). Distribution of freshwater snails in family-based
VAC ponds and associated waterbodies with special reference to intermediate hosts of
fish-borne zoonotic trematodes in Nam Dinh Province, Vietnam. Acta Tropica, 116, 15-
23.
Bui, X. M. (2010). Duck farming systems and avian influenza in the Mekong Delta of Vietnam.
In Rome: FAO.
Page 86 of 340
Burgos, S., Hong Hanh, P. T., Roland-Holst, D., and Burgos, S. A. (2007). Characterization of
Poultry Production Systems in Vietnam. International Journal of Poultry Science, 6, 709-
712.
Care (2011). Community - based Surveillance for Avian Influenza in Vietnam: Experiences,
Results, Discussions, and Recommendations Care International.
Carr, R. and Bartram, J. (2004). The control envelope and risk management. In (pp. 67-92). IWA
Publishing.
CGIAR (2011). Pigmeat value chain in Vietnam: Background proposals for the CGIAR
Research Program on Livestock and Fish International Livestock Research Institute
(ILRI).
Charron, D. F. (2012). Ecohealth Research in Practice: Innovative Applications of an Ecosystem
Approaches to Health. International Development Centre (IDRC); Springer.
Dalsgaard, A. (2001). Health Aspects of the Reuse of Wastewater in Agriculture and
Aquaculture in Vietnam. In Wastewater Reuse in Agriculture in Vietnam: Water
Management, Environment and Human Health Aspects International Water Management
Institute.
Dang, H. V. (2015). A new approach to explain policy reforms in Vietnam during doi moi by
developing and validating a major policy change model for Vietnam (Order No.
3554150).
Dang, N. K., Milstein, A., Verdegem, M. C. J., and Verreth, J. A. V. (2006). Food inputs, water
quality and nutrient accumulation in integrated pond systems: A multivariate approach.
Aquaculture, 261, 160-173.
Page 87 of 340
Dang, N. K., Phong, L. T., Verdegem, M. J. C., Duong, L. T., Bosma, R. H., and Little, D. C.
(2007). Integrated freshwater aquaculture, crop and livestock production in the Mekong
delta, Vietnam: Determinants and the role of the pond. Agricultural Systems, 94, 445-
458.
Dang, S. T. T. and Dalsgaard, A. (2012). Escherichia coli Contamination of Fish Raised in
Integrated Pig-Fish Aquaculture Systems in Vietnam. Journal of Food Protection, 75,
1317-1319.
Denise, F. P. (2010). Correlation and Simple Regression. In Maura Connor, Pamela Fuller, and
Patrick Walsh (Eds.), Statistics and Data Analysis for Nursing Research (Second Edition
ed., pp. 197-222). Julie Levin Alexander.
Desvaux, S., Nguyen, C. O., Vu, D. T., Henriquez, C., Ky, V. D., Roger, F. et al. (2014). Risk of
Introduction in Northern Vietnam of HPAI Viruses from China: Description, Patterns and
Drivers of Illegal Poultry Trade. Transboundary and Emerging Diseases.
Devendra, C. and Thomas, D. (2002). Cropanimal systems in Asia: importance of livestock and
characterisation of agro-ecological zones. Agricultural Systems, 71, 5-15.
Dinh, P. N., Long, H. T., Tien, N. T. K., Hien, N. T., Mai, L. T. Q., Phong, L. H. et al. (2006).
Risk factors for human infection with avian influenza A H5N1, Vietnam, 2004. Emerg
Infect Dis.
Dixon, J., Tanyeri-Abur, A., and Wattenbach, H. (2004). Framework for analyzing impacts of
globalization on smallholders. Food and Agriculture Organization of the United Nations
(FAO) [On-line]. Available: http://www.fao.org/docrep/007/y5784e/y5784e02.htm
Duong, D. V., Binns, C. W., and Lee, A. H. (2004). Utilization of delivery services at the
primary health care level in rural Vietnam. Social Science and Medicine, 59, 2585-2595.
Page 88 of 340
Fang, L. Q., de Vlas, S. J., Gong, P., Gong P, and et al. (2008). Environmental Factors
Contributing to the Spread of H5N1 Avian Influenza in Mainland China. PLoS ONE, 3.
FAO (2011). The State of Food and Agriculture: Women in Agriculture - closing the gender gap
for development. Food and Agriculture Organization (FAO) [On-line]. Available:
http://www.fao.org/docrep/013/i2050e/i2050e.pdf
FAO (2010). The fight against Highly Pathogenic Avian Influenza and other Emerging
Infectious Diseases. FAO Vietnam [On-line]. Available:
http://www.fao.org/docrep/015/an164e/an164e00.pdf
FAO (2015). Maps: Poultry Production and Avian Influenza in Vietnam. Food and Agriculture
Organization of the United Nations (FAO) [On-line]. Available:
http://www.fao.org/ag/againfo/programmes/en/pplpi/map_hpaivietnam.html
Fasina, F. O., Rivas, A. L., Bisschop, S. P. R., Stegeman, A. J., and Hernandez, J. A. (2011).
Identification of risk factors associated with highly pathogenic avian influenza H5N1
virus infection in poultry farms, in Nigeria during the epidemic of 20062007. Preventive
Veterinary Medicine, 98, 204-208.
Figuie, M. (2007). Future of Poultry Farmers in Vietnam after HPAI. FAO and MARD Workhop
[On-line]. Available: https://agritrop.cirad.fr/539844/1/document_539844.pdf
Finkelman, J., MacPherson, N., Silbergeld, E., and Zinsstag, J. (2008). External Review of the
IDRC Ecohealth Program Initiative: Final Report. International Development Research
Centre (IDRC) [On-line]. Available: https://idl-bnc.idrc.ca/dspace/handle/10625/37572
Geurden, T., Somers, R., Thanh, N. T. G., Vien, L. V., Nga, V. T., Giang, H. H. et al. (2008).
Parasitic infections in dairy cattle around Hanoi, northern Vietnam. Veterinary
Parasitology, 153, 384-388.
Page 89 of 340
Gilbert, M., Chaitaweesub, P., Parakamawongsa, T., Premashthira, S., Tiensin, T., Kalpravidh,
W. et al. (2006). Free-grazing ducks and highly pathogenic avian influenza, Thailand.
Emerging Infectious Diseases, 12, 227-234.
Gilbert, M., Wint, W., and Slingenbergh, J. I. (2004). The ecology of Highly Pathogenic Avian
Influenza in East and Southeast Asia: outbreak distribution, risk factors and policy
implications. (Rep. No. 43). Food and Agriculture Organization.
Gleeson, L. J. and Dung, D. H. (2010). Mid-Term Review: Vietnam Integrated National
Operational Program for Avian and Human Influenza (OPI) 2006-2010: HPAI Control
and Eradication in the Agriculture Sector.
GSO (2010). Summary results of the Vietnam household living standard survey 2010 Vietnam
General Statistics Office (GSO).
GSO (2011). The 2009 Vietnam Population and Housing Census General Statistics Office,
Ministry of Planning and Investment Vietnam.
GSO (2012). Results of the 2011 Rural, Agricultrure and Fishery Census: Part 4 - Households
using aquacultural land by the size of land used and province Hanoi: Statistical
Publishing House - General Statistics Office.
Hall, C. D., Dinh, X. T., Vu, H. H., Tran, P., and Le, B. Q. (2012). Eco-health message on
dengue in Vietnam.
Hall, D. C. and Le, Q. B. (2009). The midterm evaluation of the Government of Vietnam-UN
Joint Programme to Fight Highly Pathogenic Avian Influenza 2007-2010: Strengthening
the Management of Public Health Emergencies in Vietnam – with a focus on the
Prevention and Control of Diseases of Epidemic Potential including Highly Pathogenic
Page 90 of 340
Avian Influenza (HPAI). The Government of Vietnam-UN Joint Programme to Fight
Highly Pathogenic Avian Influenza.
Hall, D. C. (2010). The Ecohealth Approach: An introduction to the key concepts Building
Ecohealth Capacity in Asia (BECA).
Hall, D. C., Thao, T. D., Minh, D. V., and Lien, L. V. (2006). Competitiveness of the livestock
sector in Vietnam World Bank paper. WB-EASRD and FAO-TCIP.
Henning, J., Henning, K. A., Morton, J. M., Long, N. T., Ha, N. T., Vu, L. T. et al. (2011).
Highly pathogenic avian influenza (H5N1) in ducks and in-contact chickens in backyard
and smallholder commercial duck farms in Viet Nam. Preventive Veterinary Medicine,
101, 229-240.
Henning, J., Pfeiffer, D., and Vu, L. (2009). Risk factors and characteristics of H5N1 Highly
Pathogenic Avian Influenza (HPAI) post-vaccination outbreaks. Vet.Res., 40.
Hoang, T. D. (2006). HPAI Epidemic and lessons learned in Viet Nam. In APEC.
Hop, L. T. (2003). Programs to Improve Production and Consumption of Animal Source Foods
and Malnutrition in Vietnam. The Journal of Nutrition, 133, 4006S-4009S.
Hulse-Post, D. J., Sturm-Ramirez, K. M., Humberd, J., Seiler, P., Govorkova, E. A., Krauss, S. et
al. (2005). Role of domestic ducks in the propagation and biological evolution of highly
pathogenic H5N1 influenza viruses in Asia. Proc Natl Acad Sci USA, 102, 10682-10687.
Huong, L. Q., Madsen, H., Anh, L. X., Ngoc, P. T., and Dalsgaard, A. (2014). Hygienic aspects
of livestock manure management and biogas systems operated by small-scale pig farmers
in Vietnam. Science of the Total Environment, 470471, 53-57.
IDRC (2012). Ecohealth Research in Practice: Innovative Applications of an Ecosystem
Approaches to Health. International Development Centre (IDRC); Springer.
Page 91 of 340
Irini, M. and Rapsomanikis, G. (2005). Contribution of Livestock to Household income in
Vietnam: A household typology based analysis (Rep. No. PPLPI Working Paper N.21). A
Living from Livestock, FAO.
Kamakawa, A., Thu, H. T. V., and Yamada, S. (2006). Epidemiological survey of viral diseases
of pigs in the Mekong delta of Vietnam between 1999 and 2003. Veterinary
Microbiology, 118, 47-56.
Keraita, B., Drechsel, P., Seidu, R., Amerasinghe, P., Cofie, O., and Konradsen, F. (2010).
Harnessing Farmers' Knowledge and Perceptions for Health-Risk Reduction in
Wastewater-Irrigated Agriculture. In P.Drechsel, C. A. Scott, L. Raschid-Sally, M.
Redwood, and A. Bahri (Eds.), Wastewater irrigation and health : assessing and
mitigating risk in low-income countries (pp. 337-353). London: Earthscan Publications
Ltd.
Knudsen, L. G., Phuc, P. D., Hiep, N. T., Samuelsen, H., Jensen, P. K., Dalsgaard, A. et al.
(2008). The fear of awful smell: Risk perceptions among farmers in Vietnam using
wastewate and human excreta in agriculture. S E Asia J Trop Med Pub Health, 39, 341-
352.
Le, B. Q., Hall, C. D., and Cork, S. C. (2014). Water related policies (WRPs) and implications
for reductions of risk factors of waterborne diseases among livestock farmers in Vietnam.
In Ecohealth 2014 Conference Montreal, Canada: International Association for Ecology
and Health.
Le, H. H. (2013). Vietnam's domestic-foreign policy nexus: Doi Moi, foreign policy reform, and
Sino-Vietnamese normalization. Asian Politics and Policy, 5, 387.
Page 92 of 340
Le, L. T. (2001). Integrated Farming Systems: The VAC system in Northern Vietnam. Food and
Agriculture Organization of the United Nations (FAO) [On-line]. Available:
http://www.fao.org/docrep/005/Y1187E/y1187e00.htm#TopOfPage
Le, T. P., Udo, H. M. J., Mensvoort, M. E. F. v., Bosma, R. H., Le, Q., Dang, K. N. et al. (2007).
Integrated agriculture-aquaculture systems in the Mekong Delta, Vietnam: an analysis of
recent trends. Asian Journal of Agriculture and Development, 4, 51-66.
Long, J. S. and Freese, J. (2014). Regression Models for Categorical Dependent Variables Using
Stata. (Third ed.) Texas: Stata Press College Station.
MARD (2012). Provincial rural and agriculture database Vietnam: Planning Department,
Ministry of Agriculture and Rural Development.
MARD and MoH (2006). Vietnam Integrated National Plan for Avian Influenza Control and
Human Pandemic Influenza Preparedness and Response 2006-2008 Hanoi: Vietnam
Ministry of Health and Ministry of Agriculture and Rural Development.
Markwell, D. D. and Shortridge, K. F. (1982). Possible waterborne transmission and
maintenance of influenza viruses in domestic ducks. Applied and Environmental
Microbiology, 43, 110-115.
Microsoft (2014). Random Number Generation. Microsoft [On-line]. Available:
http://support.microsoft.com/kb/86523
Minh, P. Q., Morris, R. S., Schauer, B., Stevenson, M., Benschop, J., Nam, H. V. et al. (2009).
Spatio-temporal epidemiology of highly pathogenic avian influenza outbreaks in the two
deltas of Vietnam during 2003-2007. Preventive Veterinary Medicine, 89, 16-24.
Page 93 of 340
Moe, C. L. (2004). What are the criteria for determining whether a disease is zoonotic and water
related. In World Health Organization (WHO). Waterborne Zoonoses: Identification,
Causes and Controls (pp. 27-45). London: IWA Publishing.
MoH and MARD (2011). The Vietnam Integrated National Operational Program on Avian
Inlfuenza, Pandemic Preparedness and Emerging Infectious Diseases (APEID) 2010-
2015: Strengthening responses and improving prevention through a One Health
approach. Hanoi: Vietnam Ministry of Agriculture and Rural Development and Ministry
of Health.
Morley, C. (2010). Bio-Security Stakeholder Analysis, Vietnam (Rep. No. OSRO/RAS/604/USA
B6). The Emergency Centre for Transboundary Animal Diseases (ECTAD), Food and
Agriculture Organization of the United Nations (FAO).
Morris, R. S. and Jackson, R. (2005). Epidemiology of H5N1 Avian Influenza in Asia and
Implications for Regional Control Rome, Italy: Food and Agriculture Organisation of the
UN.
Munster, V. J., Wallensten, A., Baas, C., Rimmelzwaan, G. F., Schutten, M., Olsen, B. r. et al.
(2005). Mallards and Highly Pathogenic Avian Influenza Ancestral Viruses, Northern
Europe. Emerging Infectious Diseases, 11, 1545-1551.
Ngoc, T. (2013). Results of the national fingerlings production during the first 6 months of 2013.
Vietnam Directorate of Fisheries [On-line]. Available: http://www.fistenet.gov.vn/e-nuoi-
trong-thuy-san/a-san-xuat-giong/ket-qua-san-xuat-giong-thuy-san-6-thang-111au-nam-
2013
Page 94 of 340
Nguyen, A. T. L., Madsen, H., Dalsgaard, A., Nguyen, P. T., Dao, T. T. H., and Murrell, K. D.
(2010a). Poultry as reservoir hosts for fishborne zoonotic trematodes in Vietnamese fish
farms. Veterinary Parasitology, 169, 391-394.
Nguyen, C. K., Pham, D. P., Tran, H. B., and Nguyen, H. V. (2010b). Health Risks from Excreta
and Wastewater to Vietnamese Farmers.
Nguyen, V. B. (2015). Agriculture restructuring: Towards higher global competitiveness and
food security. Vietnamese Academy of Agricultural Sciences [On-line]. Available:
http://ageconsearch.umn.edu/bitstream/225580/2/VanBo2014.pdf
Nguyen, V. D. (2014). Results of the Control and Prevention of Animal Diseases in early 2014
Thai Binh: Thai Binh Department of Veterinary.
Nguyen-Viet, H., Pham, G., and Tung, D. X. (2013). Field building leadership initiative:
advancing ecohealth in Southeast Asia - the first year journey. 10th Year Anniversary of
Veterinary Public Health Centre for Asia Pacific. Veterinary Public Health Centre for
Asia Pacific (VPHCAP), Chiang Mai University: 192-197. Conferecne proceedings.
Nguyen-Viet, H., Doria, S., Tung, D., Mallee, H., Wilcox, B., and Grace, D. (2015). Ecohealth
research in Southeast Asia: past, present and the way forward. Infectious Diseases of
Poverty, 4, 5.
Oksana, N. (2005). Small farms: current status and key trends. Future of Small Farms Research
Workshop, Wye College [On-line]. Available:
http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.146.4632andrep=rep1andtype=
Page 95 of 340
Otte, J., Hinrichs, J., Rushton, J., Roland-Holst, D., and Zilberman, D. (2008). Impacts of avian
influenza virus on animal production in developing countries. CAB Reviews: Perspectives
in Agriculture, Veterinary Science, Nutrition and Natural Resources, 3.
Paul, M., Wongnarkpet, S., Gasqui, P., Poolkhet, C., Thongratsakul, S., Ducrot, C. et al. (2011).
Risk factors for highly pathogenic avian influenza (HPAI) H5N1 infection in backyard
chicken farms, Thailand. Acta Tropica, 118, 209-216.
Paul, M. C., Gilbert, M., Desvaux, S. p., Rasamoelina Andriamanivo, H., Peyre, M., Khong, N.
V. et al. (2014). Agro-Environmental Determinants of Avian Influenza Circulation: A
Multisite Study in Thailand, Vietnam and Madagascar. PLoS ONE, 9, e101958.
Pfeiffer, D. (2015). Epidemic Waves 2004-2005. Food and Agriculture Organization of the
United Nations (FAO) [On-line]. Available:
http://www.fao.org/ag/againfo/programmes/en/pplpi/docarc/hpai_05epidemicwaves.html
Pfeiffer, D. U., Minh, P. Q., Martin, V., Epprecht, M., and Otte, M. J. (2007). An analysis of the
spatial and temporal patterns of highly pathogenic avian influenza occurrence in Vietnam
using national surveillance data. Vet J, 174, 302-309.
Pham, D. P., Nguyen, V. H., Hattendorf, J., Zinsstag, J., Phung, D. C., and Odermatt, P. (2011).
Risk factors for Entamoeba histolytica infection in an agricultural community in Hanam
province, Vietnam. Parasites and Vectors, 4, 102.
Pham, V. K. (2008). Vietnamese agriculture: One year entering the WTO. Research Group
Discussion, Paper No.15, June 2008, School of Economics, The University
ofQueensland, Brisbane, Australia [On-line]. Available:
http://www.uq.edu.au/economics/eaerg/dp/EAERG_DP15.pdf
Page 96 of 340
Pham, V. T. (2010). VAC integrated system with entire energy chain in Vietnam. Food and
Agriculture Organization of the United Nations (FAO) [On-line]. Available:
http://www.fao.org/bioenergy/26371-0b904ea1d132260451c3231064e3f440c.pdf
Phan, Q. M. (2010). Epidemiological Studies of Highly Pathogenic Avian Influenza in Vietnam.
Massey University.
Pica-Ciarmarra, U., Tasciotti, L., Otte, J., and Zezza, A. (2011). Livestock Assets, Livestock
Income and Rural Households: Cross-Country Evidence from Household Surveys (Rep.
No. ESA Working Paper No. 11-17). Rome: World Bank, FAO, AU-IBAR, ILRI.
ProMED (2014). Number of alerts in Vietnam in the past 12 months for influenza. ProMED [On-
line]. Available: http://healthmap.org/promed/?p=152
Pullin, R. (2001). Integrated agriculture-aquaculture and the environment. Food and Agriculture
Organization of the United Nations (FAO) [On-line]. Available:
http://www.fao.org/docrep/005/Y1187E/y1187e07.htm#TopOfPage
Ranney, T., Enriquez, G., Cropenstedt, A., Gerosa, S., Lowder, S., Matuscke, I. et al. (2011). The
Role of Women in Agriculture. Food and Agriculture Organization (FAO) [On-line].
Available: http://www.fao.org/docrep/013/am307e/am307e00.pdf
Rushton, J., McLeod, A., and Lubroth, J. (2006). Managing transboundary animal disease.
Agriculture and Consumer Protection, FAO [On-line]. Available:
http://www.fao.org/docrep/009/a0255e/a0255e06.htm
Sheta, B. M., Fuller, T. L., Larison, B., Njabo, K. Y., Ahmed, A. S., Harrigan, R. et al. (2014).
Putative human and avian risk factors for avian influenza virus infections in backyard
poultry in Egypt. Veterinary Microbiology, 168, 208-213.
Page 97 of 340
Sims, L. D. (2007). Lessons learned from Asian H5N1 outbreak control. Avian Diseases Digest,
51, 174-181.
Soares, R. J., Ortiz-Pelaez, A., Lai, T. K. L., Dinh, Q., Otte, J., and Pfeiffer, D. (2010).
Associations between attributes of live poultry trade and HPAI H5N1 outbreaks: a
descriptive and network analysis study in northern Vietnam. BMC Veterinary Research,
6, 10.
Thomas, D., Zerbini, E., Parthasarathy Rao, P., and Vaidyanathan, A. (2002). Increasing animal
productivity on small mixed farms in South Asia: a systems perspective. Agricultural
Systems, 71, 41-57.
Tiensin, T., Ahmed, S. S. U., Rojanasthien, S., Songserm, T., Parntep Ratanakorn, P., and et al.
(2009). Ecologic Risk Factor Investigation of Clusters of Avian In?uenza A (H5N1)
Virus Infection in Thailand. Journal of Infectious Diseases.
Tracey, J. P., Woods, R., Roshier, D., West, P., and Glen, R. (2004). The role of wild birds in the
transmission of avian influenza for Australia: an ecological perspective. EMU Austral
Ornithology, 104, 109-124.
Trinh, T. L. and Nguyen, P. L. (2009). Assessment of surface water quality by water quality
index (WQI) at the Cai Sao canal, An Giang province, Vietnam. Livestock Research for
Rural Development [On-line]. Available: http://www.lrrd.org/lrrd23/7/lan23151.htm
Vo, T. X. (2005). Changing the Life of Vietnamese farmers. Angiang University, Vietnam [On-
line]. Available: http://www.crawfordfund.org/wp-content/uploads/2014/03/xuan-
Changing-the-Life-Vietnamese-Farmers.pdf
Page 98 of 340
VPMO (2010). About the organisation of the national poverty survey to prepare for the
implementation of social secuity program and policy in the period of 2011- 2015 (Rep.
No. 1752/ CT-TTg). Vice Prime Minister Office (VPMO), the Government of Vietnam.
Vu, H. Q. (2016). Vietnamese Farmers Face the WTO: Implications for Sociology. Vietnam
Institute of Public Opinion Research [On-line]. Available:
http://www.ios.sinica.edu.tw/cna/download/6b_Quang.pdf
Vu, T. K. V., Tran, M. T., and Dang, T. T. S. (2007). A survey of manure management on pig
farms in Northern Vietnam. Livestock Science, 112, 288-297.
Waltner-Toews, D. (2011). Ecohealth: A Primer. Veterinarians without Borders/ Vétérinaires
sans Frontières - Canada.
Ward, M., Maftei, D., Apostu, C., and Suru, A. (2008). Environmental and anthropogenic risk
factors for highly pathogenic avian influenza subtype H5N1 outbreaks in Romania,
20052006. Veterinary Research Communications, 32, 627-634.
Webster, R. G., Peiris, M., Chen, H., and Guan, Y. (2006). H5N1 outbreaks and enzootic
influenza. Emerg Infect Dis [serial on the Internet].
WHO (2004). Waterborne Zoonoses: Identification, Causes, and Control. IWA Publishing.
WHO (2007). Review of latest available evidence on potential transmission of avian influenza
(H5N1) through water, sewage and ways to reduce the risks to human health (Rep. No.
WHO/SDE/WSH/06.1). Geneva: World Health Organisation: Water, Sanitation and
Health Public Health and Environment.
WHO (2011). Water and Public Health. World Health Organization [On-line]. Available:
http://www.who.int/water_sanitation_health/dwq/S01.pdf
Page 99 of 340
WHO (2010). Cumulative number of confirmed human cases of avian influenza A(H5N1)
reported to WHO. World Health Organization [On-line]. Available:
http://www.who.int/influenza/human_animal_interface/H5N1_cumulative_table_archives
/en/
Yap, K. L., Yasmin, A. M., Wong, Y. H., Ooi, Y. E., Tan, S. C., Jegathesan, M. et al. (1992). A
one year community-based study on the incidence of diarrhoea and rotavirus infection in
urban and suburban Malaysian children. Med J Malaysia, 47, 303-308.
Zhao, K., Gu, M., Zhong, L., Duan, Z., Zhang, Y., Zhu, Y. et al. (2013). Characterization of
three H5N5 and one H5N8 highly pathogenic avian influenza viruses in China.
Veterinary Microbiology, 163, 351-357.
Zheng, T., Adlam, B., Rawdon, T. G., Stanislawek, W. L., Cork, S. C., Hope, V. et al. (2010). A
cross-sectional survey of influenza A infection, and management practices in small rural
backyard poultry flocks in two regions of New Zealand. New Zealand Veterinary
Journal, 58, 74-80.
Zhou, L., Liao, Q., Dong, L., Huai, Y., Bai, T., Xiang, N. et al. (2009). Risk Factors for Human
Illness with Avian Influenza A (H5N1) Virus Infection in China. Journal of Infectious
Diseases, 199, 1726-1734.
Zuckerman, J. N. (2011). Information on avian influenza ("bird flu") Academic Centre for Travel
Medicine and Vaccines and Royal Free Travel Health Centre, WHO Collaborating Centre
for Travel Medicine.
Page 100 of 340
Chapter Three: Small-Scale Integrated Farmers’ Perceptions about Risk Factors for
Transmitting Water-Related Zoonotic Disease (WRZD) in the Provinces of Thai Binh and
An Giang in Vietnam
3.1. Introduction
The foundations for my research in this chapter were both a conceptual model and a
theoretical model to determine the perceptions of small-scale integrated (SSI) farmers about risk
factors concerning the transmission of water related zoonotic disease (WRZD)23 in the provinces
of Thai Binh and An Giang in Vietnam. The conceptual model is that farmers make decisions,
including management decisions, related to animal and human health that are influenced by their
impressions about the quality of water used on farms. The theoretical model of my research is
adapted from both the Health Belief Model and the Theory of Planned Behaviour (Becker, 2012;
Janz and Becker, 1984b; Rosenstock et al., 1988). The theoretical model of my research indicates
that people’s perceptions of a personal threat to their health and or wellbeing, together with their
belief in the benefits and barriers of an action, will predict the likelihood of their health-related
behaviours (See section 3.2. and Appendix A). My research in this chapter explored the first part
of the theoretical model, which includes key constructs of perceptions. These constructs were: 1)
SSI farmers’ perceived quality of general water environment in their villages; 2) SSI farmers’
perceived threat of WRZD (e.g., perceived susceptibility to WRZD); 3) SSI farmers’
expectations (e.g., perceived barriers of taking actions to reduce the transmission of WRZD); and
23 In this research, I used the World Health Organization’s (WHO) definition of WRZD as diseases in humans that
are spread from animals and related to water such as Highly Pathogenic Avian Influenza, some parasites, or diseases
caused by a range of bacteria (e.g., pathogenic strains of Escherichia coli) (WHO, 2004). The criteria for
determining WRZD include: 1) the pathogen must spend part of its life cycle within one or more animal species; 2)
it is probable or conceivable that some life stage of the pathogen will enter water; and 3) transmission of the
pathogen from animals to humans must be through a water related route (Moe, 2004).
Page 101 of 340
4) SSI farmers’ perceived triggers/cues24 to taking actions that would mitigate25 the transmission
of WRZD.
3.3.1. Rationale for studying farmers’ perceptions of risk factors for transmission of WRZD
Perceptions in general are influenced by and reflected by individuals’ experiences and
actions, and can influence people’s health care seeking behaviors (Aburto et al., 2010; Ajzen,
1985; Levy and Myers, 2004; OECD, 2012). Farmers are not aware of, nor do they highly rank
health risk factors associated with their use of water contaminated by excreta from humans and
livestock. Rather, they view these risk factors related to production as unavoidable (Knudsen et
al., 2008b). Farmers did not consider health risks from water use to be significant because water
remains on the skin and can only cause skin problems; however, they thought that health risks
from non-composted smelly feces are serious because their odor enters the body through polluted
air (Knudsen et al., 2008b). Gender also plays a role because farmers mainly consider hygiene
and health to be women’s issues (Knudsen et al., 2008a).
Moreover, although farmers have a basic knowledge of proper hygiene, they carry out
hygienic practices (e.g., hand washing with soap) in an unsuitable manner (e.g., hand washing
without soap). Most likely, this comes from a misconception about the risk factors for the
transmission of WRZD, a lack of knowledge about cause-effect relationships, habits, and/or not
ranking the risk factors for transmission of WRZD highly while using water (Keraita et al.,
2010). In cases where farmers are aware of the health risk factors for transmitting WRZD, they
24 A trigger/cue to action is “the stimulus needed to trigger the decision making process to accept a recommended
health action”. Examples of possible cues to actions are chest pains, mass media campaigns, advice from others, a
reminder postcard from a physician, illness of family member or friend, etc.) (BUMC, 2013) 25 In this research, I defined “to mitigate transmission of WRZD” as to reduce the transmission of WRZD. An
example of an action to mitigate transmission of WRZD is to treat animal waste at source (Carr and Bartram, 2004;
WHO, 2004).
Page 102 of 340
assess measures to mitigate transmission of WRZD by considering its overall economic impact
on work efficiency rather than taking into account the potential health benefits of such efforts
(Herbst et al., 2009; Keraita et al., 2008a; Owusu et al., 2011; Weldesilassie et al., 2011).
A number of studies in Vietnam have shown that some farmers do have a basic
awareness of health risks related to water use (Herbst et al., 2009; Keraita et al., 2008b; Keraita
et al., 2010; MoC and MARD, 2000a; Owusu et al., 2011). However, these studies had small
sample sizes of peri-urban farmers, did not include SSI farmers, and did not consider the risk
factors for transmission of WRZD. In Canada, a research study investigated an in-depth
understanding of participants’ perceptions and reasons for using water sources other than
municipal water (e.g., bottled water) and using in-home water treatment devices (Quinlan, 2005).
However, participants in this research were selected from the public, and not from SSI farmers.
A more comprehensive understanding of SSI farmers’ perceptions of risk factors for
transmission of WRZD (i.e., this chapter) may assist future studies about on-farm water quality
and public health and factors that contribute to farmers’ mitigating strategies against transmitting
WRZD (i.e., Chapter Four and Chapter Five respectively). It may also inform future research
relating to livestock production, water quality, and WRZD. Furthermore, this understanding can
contribute to informing related policies, formulating WRZD risk-reduction measures, and
developing mutually acceptable risk-management strategies and/or interventions.
3.1.2. Research question and objective
In this chapter, I present an exploratory research study addressing the following research
question: What are SSI farmers’ perceptions about risk factors for transmission of WRZD in the
provinces of Thai Binh and An Giang in Vietnam? The objective of this chapter is to describe the
perceptions of SSI farmers in these provinces about the risk factors for transmitting WRZD.
Page 103 of 340
More specifically, in this chapter I describe the key constructs of SSI famers’ perceptions about
risk factors for the transmission of WRZD. I expect that SSI farmers have some impressions of
the general quality of water on farms and moderate threats of untreated water and WRZD to their
health and wellbeing. SSI farmers may indicate that economic related factors (e.g., values to
water for livestock production, and costs and economic gains of taking mitigating actions) as
their common barriers and benefits of concerns. SSI farmers may think that they have some
capacity in taking actions to reduce transmission of WRZD. I also hypothesize that there are
differences in perceptions of risk factors for transmission of WRZD between SSI farmers of the
two provinces.
This chapter’s results were also used in testing the hypotheses in Chapters Four and Five
of this manuscript. In the next section (i.e., Figure 3.1), I present a pictorial representation of the
structure of Chapter Three and its connection to Chapter Four and Five.
3.2. Methods
General information outlining the methods used in this research was presented in Chapter
Two (e.g., the research areas, the sample size, the pilot study, and the survey tools). In this
section, I will focus on aspects of the methods that I used to explore SSI farmers’ perceptions of
risk factors for the transmission of WRZD in the Vietnamese provinces of Thai Binh and An
Giang26. The questionnaires developed to collect information on farmers’ perceptions of risk
factors for the transmission of WRZD were designed based on a theoretical model adapted from
the Health Belief Model (Becker, 2012; Rosenstock et al., 1988) and the Theory of Planned
Behaviours (Payne et al., 2002) (See Appendix A).
26 Vietnam is divided into 64 provinces. Each province is sub-divided into districts, then communes, and villages.
Page 104 of 340
Figure 3.1. Chapter Three’s Outline
Used to assist modeling factors associated with on-farm water quality (Chapter 4) and factors associated with SSI farmers’ engagement in
mitigating strategies to reduce transmission of WRZDs (Chapter 5) in the provinces of Thai Binh and An Giang in Vietnam
Research question 3:
What are SSI farmers’ perceptions about risk factors for water related
zoonotic disease (WRZD) transmission in the provinces of Thai Binh
and An Giang in Vietnam?
Perception of trigger /cues
to taking mitigating
actions to reduce
transmission of WRZD
Perception of expectations
(e.g., perceived barriers of
and self-efficacy in taking
mitigating actions to reduce
transmission of WRZD
Perceptions of threat to
health and/or wellbeing
(i.e., perceived
susceptibility to and
severity of WRZD)
Perceptions of general
water quality in the
villages
Chapter 3:
Presents SSI farmers' perceptions about risk factors for WRZD transmission in the provinces of Thai Binh and An Giang in Vietnam
Page 105 of 340
More specifically, the theoretical model adapted from the Health Belief Model and the
Theory of Planned Behaviours states that people’s belief in a personal threat combined with their
belief in the benefits, barriers, and self-efficacy of an action will likely predict their behaviours.
Key constructs relating to people’s perceptions have been shown to have possible linkages to
their actions (See Appendix A). These constructs include the following: socioeconomic and
demographic factors; perceptions of threat (i.e., perceived susceptibility to and severity of
diseases); expectations (i.e., perceived benefits, perceived barriers, and self-efficacy); and cues to
action (e.g., media, personal influence, reminder) (Becker, 1974; Janz and Becker, 1984a;
Rosenstock et al., 1988). These constructs were used to guide the design of the questions used to
obtain the information relating to farmers’ perceptions of risk factors for the transmission of
WRZD, and to inform the hypothesis of the research. Various statistical tests (e.g., Pearson Chi-
squared test, t-test, and Fisher's exact test) were used to analyze the differences between farmers
of the two provinces regarding their perceptions of risk factors for WRZD transmission.
I used my knowledge of public health in Vietnam, especially of avian influenza
prevention and controls in Vietnam (Hall and Le, 2009; Le et al., 2014b; Le and Hall, 2011b; Le
and Hall, 2011a), to begin the design of my research study tools (i.e., the questionnaires, the
guidelines for in-depth interviews, and focus-group discussions). The study tools were then
revised based on my review of the literature, reports, and guidelines from the World Health
Organization (WHO, 2011; WHO, 2014a; WHO, 2014b), the Global Burden of Disease (GBD,
2014), and the Vietnamese government (MARD, 2012; MoC and MARD, 2000b; MoH, 2009b;
MoH, 2009a). A pilot study was conducted to test and further revise the study tools before using
them in the full field survey (see Chapter Two for more detailed information about the pilot
study).
Page 106 of 340
SSI farmers’ perceptions of risk factors for the transmission of WRZD were explored in
the study; these included their perceptions about the following: the general condition of water
sources in their villages; susceptibility to and severity of WRZD transmission through contact
with livestock and water; expectations with respect to WRZD transmission (i.e., perceived
barriers, perceived benefits, and perceived self-ability to take actions that would reduce WRZD
transmission); and triggers for taking actions to reduce WRZD transmission (see Appendix A).
Both qualitative survey tools (i.e., in-depth interviews, focus group discussions, and
observations) and quantitative survey tools (e.g., Likert scale questionnaires and a close-ended
question checklist) were used to collect the information about SSI farmers’ perceptions of risk
factors for transmitting WRZD, which forms the hypothesis of this research. The following
paragraphs briefly explain the questions asked during the survey (For more detailed information
about these questions, see Appendix A).
Concerning SSI farmers’ perceptions about the overall water environment in their
villages, I asked the SSI farmers who participated in the study to subjectively assess the general
condition of the water environment in their villages, and the levels of harm to human and animal
health from the different untreated sources of water on their farms. I also asked them to indicate
factors that may influence their thoughts on the quality of water on their farms. First, using a
Likert scale questions (1 = very poor to 5 = excellent), the SSI farmers were asked to rank their
thoughts about the general condition of the water environment in their villages. Second, farmers
were also asked to indicate their perceived levels of harm to their health if they used untreated
sources of water (i.e., not boiled or filtered) such as drinking untreated water27, handling
27 Drinking water is defined as “water used for direct drinking or processing food” (MoH, 2009b)
Page 107 of 340
domestic water28, handling pond water, and drinking bottled water29. Levels of harm were scaled
from 1 = no harm to 5 = extreme harm. Finally, SSI farmers were asked to indicate factors that
may influence their thoughts about water quality. These factors came from the literature as well
as the lessons learned from the pilot study. The SSI farmers had options to choose one or more of
these factors or to specify more details when answering this question.
Together, the perceived susceptibility towards diseases and the perceived severity of
diseases inform levels of perceptions of threat. Regarding perceptions of susceptibility to WRZD
as related to small-scale livestock production and water use, I asked the SSI farmers for their
thoughts about their levels of susceptibility to HPAI viruses, diarrhoea, coliform bacteria, and
parasites from using untreated water sources (i.e., not boiled or filtered). By using a scale from 1
to 5 (i.e., from 1 = no susceptibility to 5 = extremely high susceptibility), SSI farmers had the
option to indicate their thoughts about their levels of susceptibility to these diseases. This
question measured to what degree farmers perceived their susceptibility to a disease when using
untreated water; it was not about what causes them to be more or less susceptible to a disease. In
terms of farmers’ perceptions about the severity of diseases, I asked the SSI farmers to indicate
their thoughts about how severe diseases from HPAI viruses, diarrhoea, coliform bacteria, and
parasites are by using a scale of severity from 1 to 5 (i.e., 1= trivial to 5 = very severe).
Expectations are also a construct of perceptions. Expectations may contribute to
influencing farmers’ health behaviours. Expectations in this research have two sub-constructs
including 1) any perceived barriers that may prevent farmers from taking mitigating actions
28 Domestic water is defined as “water used for domestic use but not for direct drinking or processing food” (MoH,
2009a) 29 In my survey, bottled water was referred to as water packed in bottles by approved suppliers and available as a
commercial product. Farmers bought bottled water for drinking purpose without any further treatment by
themselves.
Page 108 of 340
against the transmission of WRZD and 2) perceived benefits that encourage farmers to take or
not take mitigating actions against transmitting WRZD. Regarding perceived barriers to actions,
I asked SSI farmers a question about what barriers might prevent them from taking mitigating
actions against the transmission of WRZD. This question had several options for the SSI farmers
to choose from, as well as an option to specify detailed answers. In terms of the perceived
benefits of taking or not taking actions, I asked the SSI farmers to specify what benefits
encouraged or did not encourage them to take mitigating actions against the transmission of
WRZD. Farmers were asked to choose one of the benefits listed in the questionnaire and/or to
indicate benefits that were not listed in the question.
Perceived self-efficacy is indicated in the theoretical model of this research as one
construct of perceptions that may influence people’s health behaviours. I asked the SSI farmers
to indicate their faith in their ability to perform four key clusters of mitigating actions against
transmitting WRZD. The first cluster of mitigating actions was livestock health management
(e.g., manage animal waste). The second cluster was taking mitigating actions at the water
storage, treatment, and distribution level (e.g., storing water in a confined tank). The third cluster
was taking mitigating actions relating to the protection of water sources (e.g., using of buffer
zones). The forth cluster was taking mitigating actions at points of use/household levels (e.g.,
hand washing with soap). These clusters were selected based on the WHO’s guidelines about the
control envelope and risk management for WRZDs (WHO, 2004).
Cues to actions such as media, personal influence, and reminders together with
perceptions of threats and expectations are indicated in the theoretical model of this research as
one of the key constructs that may influence people’s health behaviours. In the cue to actions
question, there were options for triggers or cues to actions as well as an option for specifying
Page 109 of 340
more detail. I asked the SSI farmers to select the triggers/cues (and provide specification if any)
that influenced their decisions to take mitigating actions against the transmission of WRZD, or to
specify any triggers or cues to their actions that were not listed in the questionnaire.
3.3. Results
3.3.1. Perceptions of water quality in the villages
Table 3.1 shows the results with respect to farmers’ perceptions about the general
conditions of the water environment in their villages. In both provinces, almost 70% of the SSI
farmers thought that the water environment conditions in their villages were either good or
adequate. A very small percentage thought their water environment was in either excellent or
very poor condition (1.2% and 2.9% respectively). None of the SSI farmers in Thai Binh
considered their water environment as excellent and 4.4% of them thought their water
environment was very poor. Meanwhile, 2.4% of the SSI farmers in An Giang considered their
water environmental condition to be excellent while 1.3% believed it to be very poor (Table 3.1).
In addition, the percentage of SSI farmers in An Giang who thought that their sources of
untreated water (i.e., not boiled or filtered) were safe for drinking were much greater than in Thai
Binh (17% and 1% respectively).
Page 110 of 340
Table 3.1. Perceived General Condition of the Water Environment in the Villages of Small-
Scale Integrated (SSI) Farmers in the Provinces of Thai Binh and An Giang, Vietnam
Scale Both provinces Thai Binh An Giang
Freq. Percent Freq. * Percent Freq. Percent
Excellent 7 1.2 - - 7 2.4
Good 104 17.4 18 6.0 86 28.9
Adequate 275 46.1 164 54.9 111 37.3
Poor 194 32.5 104 34.8 90 30.2
Very poor 17 2.9 13 4.4 4 1.3
Total 597 100.0 299 100.0 298 100.0
(*) indicates significant difference between Thai Binh and An Giang with p = 0.000, two-tailed Fisher's exact test
Most farmers thought there was at least some harm when using untreated sources of
water. In particular, farmers thought that if they consumed untreated drinking water or handled
untreated domestic water, there would be moderate to high levels of harm to their health (Table
3.2), however, that drinking bottled water imposed no or low harm to health. According to the
SSI farmers in both provinces, handling pond water was considered to result in high or extreme
levels of harm to health (Table 3.2).
Page 111 of 340
Table 3.2. Perceived Harm to Health When Using Different Sources of Water among
Small-Scale (SSI) Farmers in the Provinces of Thai Binh and An Giang, Vietnam
Province/Source of water Obs Mean a Std. Dev.
Both provinces
Untreated drinking water 574 3.8 0.7
Untreated domestic water 576 3.5 0.9
Pond water 511 4.1 0.8
Bottled water 475 1.7 0.8
Thai Binh
Untreated drinking water 280 3.7*** 0.8
Untreated domestic water 282 3.7*** 0.8
Pond water 275 4.3*** 0.6
Bottled water 202 1.8*** 0.7
An Giang
Untreated drinking water 294 3.9 0.7
Untreated domestic water 294 3.3 1.0
Pond water 236 3.9 0.9
Bottled water 273 1.6 0.8 a mean of level of harm with the scale from 1 = No harm, 2 = Low harm, 3 = Moderate harm, 4 = High harm,
to 5 = Extreme harm
(***) indicates a statistical significant difference between Thai Binh and An Giang p < 0.01
Table 3.3 shows that in Thai Binh the two most cited factors for farmers as influences on
their thinking about the quality of a water source were their impressions of water quality such as
the taste of the water and their previous experience with that water source (80% and 67% of the
SSI farmers in Thai Binh respectively). In contrast, in An Giang, previous experience with water
use was the factor that influenced most SSI farmers (73%) in their perception about water
quality; moreover, fewer of them indicated that impressions of water quality such as taste
influenced their perceptions about the quality of water. Media also influenced the perceptions of
SSI farmers about water in Thai Binh, and more so than in An Giang (i.e., 35% and 5%
respectively). Other listed factors in Table 3.3 appeared to influence the general perceptions of
Page 112 of 340
water quality among a smaller proportion of the SSI farmers in both provinces (i.e., ranging from
2% to 13% of the farmers).
Table 3.3. Factors Influencing Small-Scale Integrated (SSI) Farmers’ Thoughts about
Water Quality in the Provinces of Thai Binh and An Giang, Vietnam
Factors An Giang Thai Binh
Freq. Percent Freq. Percent
Impressions (e.g., taste) 72*** 24 239 80
Previous experiences 208 70 201 67
Risk perceptions 5*** 2 38 13
Attitudes towards water chemical
contaminants
7*** 2 34 11
Contextual cues relating to the
source of water supply
14** 5 33 11
Familiarity with specific water
properties
30 10 28 9
Trust in suppliers/sources 7*** 2 23 8
Media 14*** 5 104 35
(***) significant difference between Thai Binh and An Giang with p = 0.000, Pearson Chi-squared test
(**) significant difference between Thai Binh and An Giang with p = 0.01, Pearson Chi-squared test
3.3.2. Perceptions of threat to health and/or wellbeing
As indicated in the theoretical model of my research (see Appendix A), perceptions of
threat to health and/or wellbeing are one of the constructs of perceptions that may influence
people’s health behaviours. Together, the perceived susceptibility to diseases and perceived
severity of diseases inform the levels of perception of threat. The following sections present the
results of the level of perceived susceptibility as well as the level of perceived severity of key
diseases relating to livestock/poultry production (i.e., HPAI, diarrhea, coliform bacteria, and
parasites) when using various types of on-farm water (i.e., drinking water, handling domestic
water, handling pond water, and handling livestock wastewater) among SSI farmers.
Page 113 of 340
3.3.2.1.Perceived susceptibility to diseases
With respect to perceptions of susceptibility to diseases relating to SSI livestock
production and water use, Table 3.4 shows the perceived susceptibility among SSI farmers to
highly pathogenic avian influenza (HPAI), diarrhoea, coliform bacteria, and parasites from
untreated water sources. In Thai Binh, none of the SSI farmers thought that drinking and/or
handling untreated water has either a low or no level of exposure30 to HPAI, diarrhoea, coliform
bacteria, or parasites. On the contrary, SSI farmers thought that drinking untreated water and
handling untreated domestic water had a moderate to high level of susceptibility to HPAI,
diarrhoea, coliform bacteria, and parasites (i.e., mean level of susceptibility from 3.4 to 3.8).
Additionally, when it came to issues of handling untreated pond water and livestock wastewater
the SSI farmers in Thai Binh considered this to result in a high and extremely high level of
susceptibility to HPAI, diarrhoea, coliform bacteria, and parasites (i.e., mean level of
susceptibility from 4.2 to 4.6). Among the SSI farmers in An Giang, the perceived levels of
susceptibility to HPAI viruses, diarrhoea, coliform bacteria, and parasites from untreated water
sources were similar to the perceived levels of susceptibility among the Thai Binh SSI farmers.
However, there were a few exceptions. The first was that the An Giang SSI farmers
considered drinking untreated water as being a high susceptibility for and extremely high
susceptibility for diarrhea and parasites respectively (i.e., mean levels of susceptibility of 4.0 and
4.2 respectively). These mean levels of susceptibility were greater than those among the Thai
Binh SSI farmers (i.e., mean level of susceptibility 3.8 and 3.7 respectively). The second
30 This phrase emphasizes on farmers’ perceived susceptibility to HPAI, diarrhea, coliform bacteria, or parasites
from using untreated drinking water, not the act of drinking itself. I expected and observed during the data collection
that farmers’ perceptions of susceptibility to diseases were limited to as simple as exposure and risk of having a
disease, and did not include other aspects of susceptibility to a disease.
Page 114 of 340
exception was that the SSI farmers in An Giang believed they were less susceptible to HPAI,
diarrhoea, coliform bacteria, and parasites (i.e., mean levels of susceptibility from 3.7 to 3.9)
when handling untreated pond water than did their counterparts in Thai Binh (i.e., mean levels of
susceptibility from 4.2 to 4.5).
3.3.2.2. Perceived severity of diseases
With respect to SSI farmers’ perceived levels about the severity of diseases from HPAI
viruses, diarrhoea, coliform bacteria, and parasites, Table 3.5 shows that SSI farmers indicated
“severe” as the most common level of severity for HPAI, diarrhoea, coliform bacteria, and
parasites. The percentages of SSI farmers who considered bird flu very severe if contracted were
similar between the two provinces (i.e., 28%). When combining the responses in the perceived
severe and the perceived very severe levels, the results showed that about 80% and greater of SSI
farmers in both Thai Binh and An Giang considered HPAI, diarrhoea, and coliform diseases as
being severe and very severe if they had these diseases. Less than 12% of them considered it
mild or trivial if they contracted HPAI, diarrhoea, and coliform disease. None of the SSI farmers
in either province considered diarrhea or coliform diseases as being trivial.
Page 115 of 340
Table 3.4. Perceived Susceptibility to Highly Pathogenic Avian Influenza (HPAI),
Diarrhoea, Coliform Bacteria, and Parasites from Untreated Water Sources among Small-
Scale Integrated (SSI) Farmers in the Provinces of Thai Binh and An Giang, Vietnam
Perceived susceptibility to Thai Binh An Giang
Mean a Std. Dev. Mean Std. Dev.
HPAI from b
Drinking water 3.4* 0.9 3.9* 0.9
Handling domestic water 3.5* 0.9 3.6 0.9
Handling pond water 4.2* 0.8 3.9* 0.9
Handling livestock wastewater 4.5* 0.6 4.1* 1.0
Diarrhea from c
Drinking water 3.8*** 0.7 4.0*** 0.7
Handling domestic water 3.8*** 0.7 3.6*** 1.0
Handling pond water 4.4*** 0.6 3.9*** 0.9
Handling livestock wastewater 4.5*** 0.6 4.2*** 0.8
Coliform bacteria from d
Drinking water 3.7** 0.6 3.9** 0.6
Handling domestic water 3.7** 0.6 3.6** 0.8
Handling pond water 4.5*** 0.5 3.7*** 0.6
Handling livestock waste water 4.6*** 0.5 4.1*** 0.8
Parasites e
Drinking water 3.7*** 0.6 4.2*** 0.7
Handling domestic water 3.8 0.7 3.9 0.9
Handling pond water 4.5*** 0.5 3.9*** 0.7
Handling livestock wastewater 4.6*** 0.5 4.3*** 0.8 a scaled as 5 = extremely high susceptibility; 4 = high susceptibility; 3 = moderate susceptibility; 2 = low
susceptibility; and 1 = no susceptibility; b n = 456; c n = 498; d n = 207; e n = 382; (*) significant difference between
Thai Binh and An Giang p < 0.1; (**) significant difference between Thai Binh and An Giang p < 0.05; (***)
significant difference between Thai Binh and An Giang p < 0.01;
Page 116 of 340
Table 3.5. Perceived Severity of Diseases from Highly Pathogenic Avian Influenza (HPAI),
Diarrhoea, Coliform Bacteria, and Parasites among Small-Scale Integrated (SSI) Farmers
in the Provinces of Thai Binh and An Giang, Vietnam
Severity a Thai Binh An Giang
Freq. Percent Freq. Percent
HPAI
Very Severe 77 28 67 28
Severe 184 67 153 64
Moderate 13 5 16 7
Mild - - 2 1
Trivial 1 0 - -
Total 275 100 238 100
Diarrhea
Very Severe 8* 3 40* 14
Severe 205* 77 191* 65
Moderate 50* 19 59* 20
Mild 3* 1 5* 2
Trivial - - - -
Total 266 100 295 100
Coliform
Very Severe 8* 5 11* 26
Severe 121* 70 27* 63
Moderate 44* 25 3* 7
Mild - - 2* 5
Trivial - - - -
Total 173 100 43 100
Parasites
Very Severe 4* 2 18* 11
Severe 131* 54 72* 43
Moderate 101* 42 56* 34
Mild 5* 2 18* 11
Trivial - - 2 1
Total 241 100 166 100 a
scaled as 5 = very severe; 4 = severe; 3 = moderate; 2 = mild; and 1 = trivial
(*) significant difference between Thai Binh and An Giang with p = 0.000, two-tailed Fisher's exact test
Page 117 of 340
3.3.3. Expectations
The study results relating to farmers’ expectations included three sub-constructs: 1)
farmers’ perceived barriers that prevent them from taking mitigating actions against transmitting
WRZD; 2) farmers’ perceived benefits that encourage them to either take or not take mitigating
actions against the transmission of WRZD; and 3) farmers’ perceived self-efficacy to take
mitigating actions against the transmission of WRZD.
3.3.3.1. Perceptions of barriers to taking mitigating actions
The first sub-construct of perceptions of threat is perceptions of barriers to actions to
mitigate transmission of WRZD. Table 3.6 below shows the results of barriers to SSI farmers for
taking mitigating actions against the transmission of WRZD. The most common barriers to
actions were the responses of “do not know” and “cannot afford the cost” in Thai Binh and An
Giang respectively. In both provinces, “neighbours do not take actions” was indicated as a
barrier to taking mitigating actions by a small proportion of the SSI farmers in each province
(4% and 5% respectively). Beyond the above possible barriers to taking mitigating actions as
response options for SSI farmers, the only other barrier that SSI farmers specified was “lack of
land,” yet only three out of 598 SSI farmers indicated this.
3.3.3.2. Perceptions of benefits of taking or not taking mitigating actions
The second sub-construct of perceptions of threats is the perception of benefits from
either taking or not taking actions against WRZD transmission. Table 3.7 below shows five key
benefits of taking mitigating actions against WRZD transmission as indicated by SSI farmers in
both An Giang and Thai Binh. Farmers indicated the most common benefit in both Thai Binh
and An Giang as “economic gains” (i.e., 37% and 42% respectively). As for An Giang, only the
response “being a good model for others to follow” was indicated as a benefit for taking
Page 118 of 340
mitigating actions by less than 1% of the SSI farmers. On the other hand, 26% and 18% of the
SSI farmers in Thai Binh and An Giang respectively indicated that there were no benefits to
taking mitigating actions against WRZD transmission. Most SSI farmers in Thai Binh (98%)
indicated that there were no benefits from not taking any mitigating actions. However, in An
Giang, 49% and 42% of the SSI farmers respectively indicated that they received “no benefits”
but had “economic gains” from not taking mitigating actions against the transmission of WRZD.
Table 3.6. Perceived Barriers to Taking Actions to Mitigate Transmission of WRZD among
Small-Scale Integrated (SSI) Farmers in the Provinces of Thai Binh and An Giang,
Vietnam
Barriers to taking mitigating
actions Thai Binh An Giang
Freq. Percent Freq. Percent
Cannot afford the costs 95*** 32 144 48
Do not know how 114 38 118 40
Lack of understanding 90*** 30 31 10
Neighbors do not take actions 13 4 15 5
Too busy 106*** 35 47 16 (***) significant difference between Thai Binh and An Giang with p = 0.000, Pearson Chi-squared test
Page 119 of 340
Table 3.7. Perceived Benefits that Encourage Small-Scale (SSI) Farmers in the Provinces of
Thai Binh and An Giang, Vietnam, to Either Take or Not Take Actions to Mitigate
Transmission of WRZD
Benefits Thai Binh An Giang
Freq. Percent Freq. Percent
Benefits of taking mitigating actions a
Disease prevention for humans (a) 41* 14.0 54 18.9
Economic gains (b) 108* 36.7 121 42.3
Environment/water protections (c) 35* 11.9 25 8.7
(a)+(b)+(c) 28* 9.5 29 10.1
No benefit 77* 26.2 51 17.8
Disease prevention for animals 5* 1.7 4 1.4
Being good models for others to follow - - 2 0.7
Benefits not taking mitigating actions b
No benefit 295** 98.3 121 48.8
Time (d) 1** 0.3 19 7.7
Economic gains (e) 4** 1.3 103 41.5
Avoid being/feeling strange - - 1 0.4
(d) +(e) - - 3 1.2
Convenience - - 1 0.4 a Thai Binh (n = 294) and An Giang (n = 286),
b Thai Binh (n = 300) and An Giang (n = 248)
(*) significant difference between Thai Binh and An Giang with p = 0.1, two-tailed Fisher's exact test
(**) significant difference between Thai Binh and An Giang with p = 0.000, two-tailed Fisher's exact test
Further to exploring farmers’ perceptions about the benefits of taking mitigating actions
against the transmission of WRZD, Table 3.8 shows SSI farmers’ perceptions of the value of
water and wastewater for their livestock production. Concerning the value of their water, about
60% of the SSI farmers in both Thai Binh and An Giang indicated that water has moderate to
high value for their livestock production. Moreover, double the percentage of SSI farmers in
Thai Binh, compared to An Giang, considered water as being of extremely high value (13% and
6% respectively). The rest of the SSI farmers thought that water has low or no value to their
livestock production. Regarding wastewater, 30% and 77% of the SSI farmers in Thai Binh and
Page 120 of 340
An Giang respectively considered that the wastewater from livestock production has no value for
their livestock production (Table 3.8).
Table 3.8. Perceptions of Water and Wastewater Value for Livestock Production among
Small-Scale Integrated (SSI) Farmers in the Provinces of Thai Binh and An Giang,
Vietnam
Value levels a
Thai Binh An Giang
Freq. Percent Freq. Percent
Water b
Extremely high value 21* 13 17 6
High value 29* 17 107 39
Moderate value 74* 45 64 23
Low value 12* 7 25 9
No value 30* 18 63 23
Wastewater c
Extremely high value 2** 1 2 1
High value 40** 14 8 3
Moderate value 122** 41 31 12
Low value 43** 15 21 8
No value 89** 30 207 77 a scaled as 1 = no value, 2 = low value, 3 = moderate value, 4 = high value, and 5 = extremely high value
b Thai Binh (n = 166) and An Giang (n = 276),
c Thai Binh (n = 296) and An Giang (n = 269)
(*) significant difference between Thai Binh and An Giang with p = 0.000, Pearson Chi-squared test
(**) significant difference between Thai Binh and An Giang with p = 0.000, two-tailed Fisher's exact test
In addition, farmers’ perceived levels of importance about other issues relating to on-
farm water in reducing WRZD transmission are presented in Table 3.9. A total of 95% of SSI
farmers in both An Giang and Thai Binh indicated that on farm water-related issues31 were of at
least some importance in reducing risk factors for the transmission of WRZD (i.e., from low
important to extremely important). The response “highly important” was the most frequently
indicated level of importance of the on-farm water-related issues in reducing transmission of
WRZD (i.e., ranging from 44% to 71%). There was an exception to the issue of “rainfall”, which
31 Including water sources, sharing water between animals and humans, water sources-animals distance, lack of
water, rainfall, and water quality
Page 121 of 340
the SSI farmers in An Giang considered as being a “moderately important” issue instead of a
“highly important” issue in reducing risk factors for the transmission of WRZD.
3.3.3.3.Perceptions about self-efficacy in performing mitigating actions
Table 3.10 below shows farmers’ perceived actual capacity to take mitigating actions
against the transmission of WRZD (i.e., self-efficacy). Most SSI farmers in both provinces
considered they had a moderate to high ability to perform all of the key clusters of mitigating
actions. Moreover, about 10% of them thought they had a very high ability to do so, whereas
none of them considered themselves to have a weak ability to do so. No SSI farmers in either
province thought they had a weak or very weak capacity to manage livestock health. Less than
15% of the SSI farmers indicated that they were weak in undertaking mitigating actions at the
levels of protection of sources of water, water storage, treatment, and distribution, and point of
use/household level.
Page 122 of 340
Table 3.9. Levels of Importance for Water Use Issues in Reducing Transmission of WRZD
among Small-Scale Integrated (SSI) Farmers in the Provinces of Thai Binh and An Giang,
Vietnam
Water use issues/Importance Thai Binh An Giang
Freq. Percent Freq. Percent
Water sources a
Extremely important 33* 12 27 9
Highly important 164* 59 209 71
Moderately important 75* 27 47 16
Low important 4* 1 7 2
Not important 3* 1 3 1
Sharing water between animals and humans b
Extremely important 50* 18 46 16
Highly important 170* 62 156 54
Moderately important 49* 18 55 19
Low important 3* 1 20 7
Not important 1* 0 11 4
Water sources-animals distance c
Extremely important 52* 19 23 8
Highly important 157* 58 123 43
Moderately important 50* 19 82 29
Low important 10* 4 47 17
Not important 1* 0 8 3
Lack of water d
Extremely important 52* 20 10 4
Highly important 125* 48 120 47
Moderately important 75* 29 85 33
Low important 9* 3 31 12
Not important 10 4
Rainfall e
Extremely important 16* 6
Highly important 115* 44 19 8
Moderately important 102* 39 65 27
Low important 25* 10 125 51
Not important 2* 1 36 15
Water quality f
Extremely important 85* 33 19 8
Highly important 110* 43 140 57
Moderately important 51* 20 51 21
Low important 8* 3 27 11
Not important 1* 0 10 4 a Thai Binh (n = 279) and An Giang (n = 293); b Thai Binh (n = 273) and An Giang (n =288) c Thai Binh (n = 270) and An Giang (n = 283); d Thai Binh (n = 261) and An Giang (n =256) e Thai Binh (n = 260) and An Giang (n =245); f Thai Binh (n = 255) and An Giang (n =247)
(*) significant difference between Thai Binh and An Giang with p = 0.000, two-tailed Fisher's exact test
Page 123 of 340
Table 3.10. Perceived Self-efficacy to Perform Mitigating Actions among Small-Scale
Integrated Farmers in Thai Binh and An Giang, Vietnam
Perceived self-efficacy Thai Binh An Giang
Freq. Percent Freq. Percent
In livestock health management a
Very high 29 10 35 12
High 122 41 77 26
Moderate 125 42 149 50
Weak 19 6 37 12
Very weak - - - -
Protection of sources of water b
Very high 23 8 32 11
High 124 41 76 26
Moderate 115 38 157 53
Weak 36 12 32 11
Very weak 1 0 - -
In water storage, treatment and distribution c
Very high 19 6 29 10
High 121 41 70 24
Moderate 112 38 166 56
Weak 45 15 29 10
Very weak 1 0 - -
Taking mitigating actions at point of use d
Very high 44 15 32 12
High 94 32 80 31
Moderate 152 51 127 49
Weak 6 2 21 8
Very weak 1 0 - - a Thai Binh (n = 295), An Giang (n = 298), and significant difference between Thai Binh and An Giang with p =
0.000, Pearson Chi-squared test b Thai Binh (n = 299), An Giang (n =297); significant difference between Thai Binh and An Giang with p = 0.000,
two-tailed Fisher's exact test c Thai Binh (n = 298), An Giang (n = 294), and significant difference between Thai Binh and An Giang with p =
0.000, two-tailed Fisher's exact test d Thai Binh (n = 297) and An Giang (n =260), significant difference between Thai Binh and An Giang with p =
0.000, two-tailed Fisher's exact test
Page 124 of 340
3.3.4. Cues to actions
Figure 3.2 shows the common triggers/cues that SSI farmers in both provinces indicated
prompt them to take mitigating actions against the transmission of WRZD. In An Giang, the two
most frequently cited triggers/cues to action were “radios/newspapers” and “advice from
health/veterinary workers,” whereas in Thai Binh it was “worry about being infected or affected
with WRZD” and “radios/newspapers”. In contrast, the two least frequently cited triggers/cues to
action in Au Giang were “not busy with other tasks” and “lose income if do not take actions”
whereas in Thai Binh they were “not busy with other tasks” and “neighbors do the same
mitigating actions”. Worry about being affected by the transmission of WRZD was the most
common cue for taking mitigating actions in Thai Binh, while it was the third most common cue
in An Giang.
Figure 3.2. Triggers/Cues to Taking Mitigating Actions against Transmission of Water
Related Zoonotic Diseases (WRZD)
46
148
208
10
263
113
55
147
197
26
111
52
050
100
150
200
250
Thai Binh An Giang
Neighbors do the same mitigating actions (p = 0.308) Health workers' advice (p = 0.999)
Media - radios/newspapers (p = 0.399) Not busy with other tasks (p = 0.001)
Worried about being infected/affected by EID (p = 0.000) Lose income (p = 0.000)
Num
bers
of f
arm
ers
Graphs by Provinces
Trigger/cues to taking mitigation actions
Page 125 of 340
3.4. Discussion
The results of farmers’ perceptions of risk factors for the transmission of WRZD
supported my hypothesis formulated prior to conducting the survey that there are differences in
the perceptions of risk factors for transmission of WRZD between SSI farmers of the provinces
of Thai Binh and An Giang. They also appeared to have a good sense of their water environment
conditions. For example, SSI farmers in Thai Binh appeared to be more skeptical than those in
An Giang when assessing the conditions of their water environment. Meanwhile, other studies
have shown some differences in the density of livestock production and water resources between
Thai Binh and An Giang. For instance, the number of poultry produced in Thai Binh per year
(8,899,000) was almost double that in An Giang (4,067,000) (GSO, 2012a). The total number of
rural households using aquaculture land in Thai Binh was also about 10 times greater than that in
An Giang (GSO, 2012b). As well, the total area of surface water for aquaculture in Thai Binh
was much greater than that in An Giang (13.4 and 1.9 thousand hectares respectively) (GSO,
2010). Interestingly, the total number of communes in Thai Binh that have a common sewage
system was about half to those in An Giang (GSO, 2011). Therefore, it is likely that the general
water environment conditions in Thai Binh are not as safe as those in An Giang. I suggest that
SSI farmers with greater densities of livestock production and aquaculture land resources may be
more skeptical of their water environment conditions. However, further studies are needed to
assess the SSI water environment conditions and relate them to the corresponding perceptions of
SSI farmers.
The levels of harm to health from using various untreated sources of water as stated by
the SSI farmers in the survey are in line with my expectations before the survey. SSI farmers
were aware of the low quality as well as the potential harm caused by using untreated on-farm
Page 126 of 340
water sources. Nonetheless, SSI farmers shared in in-depth interviews and focus group
discussions that bottled water is expensive and not always affordable for them. Given the
relatively high price and high level of technology used to produce bottled water compared to
other sources of on-farm water, it is understandable why SSI farmers believed in the high quality
and low level of harm of bottled water.
In my previous experience in the studied area, SSI farmers judged their water quality
based on subjective factors such as water turbidity or colour, its taste in tea, or simply their
feeling after using the water for some time. Thus, it is not surprising that most SSI farmers in the
survey assessed their water quality according to their own impressions such as how it tastes,
rather than using any objective assessments by themselves or by local health authorities. I also
observed32 that SSI farmers had a great need for the water quality on their farms to be assessed.
However, they did not know how to do this, nor could they afford to have their water assessed.
Meanwhile, the local government agencies in Thai Binh and An Giang primarily monitor and
assess, to some extent, the water quality provided by industrial suppliers, but not the water
quality available at the farm/household level. As a result, there exists a dire need to introduce
simple and affordable on-farm water quality testing and to improve the capacity of local health
authorities to conduct rural water quality monitoring, assessment, and validation. Simple and
affordable on-farm water quality testing tools may help change the water-related behaviour of
farmers by enhancing visualization and ownership of water testing for farmers. Alternatively,
engaging the private sector may be a solution for delivering simple and affordable on-farm water
quality testing tools together with affordable options for on-farm water treatment and storage.
32 Field investigator/enumerators observed through questions asked in interviews and focused group discussion
Page 127 of 340
Nonetheless, more studies are needed to explore the concept of the objective assessment of on-
farm water using basic, no-frills, inexpensive on-farm water quality testing tools or other low-
priced options for on-farm water treatment and storage.
One outcome that I was not expecting in the results was the level of influence of media
such as radio and television, and print media such as newspapers on farmers’ perceptions in Thai
Binh, which was much greater than in An Giang. This may reflect regional differences in
lifestyles and cultures between the two river deltas. As a result, it may be necessary to consider
the differing levels of media influence on farmers’ perceptions in the two river deltas when
developing relating policies and programs.
Given the insights about farmers’ perceptions of the general conditions of their on-farm
water and the limited ways farmers engage to assess the quality and the level of harm of their on-
farm water, it is apparent that field-based objective assessments of on-farm water quality and/or
the risk of disease from on-farm water use are needed. Such field-based assessments may help to
quantify the quality of on-farm water and help better understand differences in the overall
perceptions and assessment of water quality among farmers in Thai Binh and An Giang. Further
studies in not only health but also communication may also be needed to explore the role of
media in influencing farmers’ perceptions in different regions of Vietnam.
Concerning the perceived threat of disease related to water and livestock production, the
results reflected my expectations that farmers were aware of and noticed the susceptibility to and
severity of diseases. Their levels of perceived susceptibility and perceived severity of diseases
were consistent with their concern for on-farm water quality. With understanding their perceived
level of susceptibility to disease, it may be useful to explore the association, if any, between
farmers’ perceptions and their mitigating strategies against transmitting WRZD. However, this
Page 128 of 340
research did not explore other indicators such as immunity and age that may influence the level
of susceptibility to diseases. Therefore, this research did not aim at assessing the correctness of
farmers’ perceived level of susceptibility to diseases. Moreover, given the various waves of the
H5N1 epidemic that have occurred in Vietnam since 2003, I expected my research to show a
high percentage of SSI farmers who perceived HPAI as a severe infectious disease. However,
their percentage was slightly lower than for the percentage of urban consumers (75%) who
considered HPAI a serious threat to them during the first peak of the outbreak in 2004 (Figuie
and Fournier, 2008). Therefore, key differences between rural and urban populations such as
demographic and socioeconomic characteristics, population density, and access to information
may be important contributing factors to people’s perception of the severity of various diseases
that require further study.
With respect to farmers’ expectations of conducting mitigating actions, it is
understandable and was expected that farmers considered economic factors important in taking
mitigating actions. Economic factors were perceived by SSI farmers not only as barriers for
taking mitigating actions (e.g., cannot afford the costs), but also as important benefits if they did
not take mitigating actions against the transmission of WRZD. This result is consistent with
other research which also reported the importance of overall economic impact on how farmers
assess mitigating actions against WRZD (Keraita et al., 2010). Farmers’ expectations of how
they perceive self-efficacy or their perceived actual capacity in taking mitigating actions may
also influence their health behaviours around control of water use and water quality (Carr et al.,
2011). The large proportion of SSI farmers who indicated having a moderate to high ability to
perform mitigating actions could possibly be explained by the fact that the self-efficacy
questions referred to the key aspects of traditional small-scale integrated farming practices.
Page 129 of 340
Self-efficacy is a person's belief in his/her ability to make a change or take an action
(Rosenstock et al., 1988). Perceived self-efficacy is indicated in the theoretical model of this
research as one of the constructs of perception that may influence people’s health behaviours. In
particular, perceived self-efficacy in this study was referred to as a farmer’s perceived actual
capacity in taking mitigating actions against the transmission of WRZD. Yet, the level of self-
efficacy may change when considering mitigating actions that go beyond or are out of the
context of traditional small-scale integrated farming practices. For example, farmers’ perceived
level of self-efficacy for private adaptive measures to climate changes (e.g., migration, changing
from farming to non-farming activities, and moving from crop to livestock) were reported as
being fairly low (Le et al., 2014a). Therefore, caution should be taken when generalizing the
levels of self-efficacy among farmers in the context of mitigating actions that meet certain
standards/recommendations rather than in the context of traditional small-scale integrated
farming.
In addition to perceptions of threats and expectations, triggers/cues to taking mitigating
actions against the transmission of WRZD were a construct in the theoretical model of this
research. Triggers/cues to action are considered as “a stimulus needed to trigger the decision-
making process to accept a recommended health action” (BUMC, 2013). Cues to actions (e.g.,
media) together with perceptions of threats and expectations are indicated in the theoretical
model of this research as one of the key constructs that may influence farmers’ health-related
behaviours. Given the huge impact of the H5N1 epidemics on the socioeconomic and health
conditions in Vietnam, the research results are reasonable in showing “worrying about being
affected by the diseases,” “radios/newspapers,” and “advice from CHWs/CVWs” as the top three
factors influencing SSI farmers’ reported mitigating actions. However, identifying triggers/cues
Page 130 of 340
to taking mitigating actions other than health-related benefits may be needed to influence
farmers’ adoption of safer practices (Keraita et al., 2010).
One interesting observation is that the SSI farmers indicated economic factors (e.g., loss
of income) as the fourth cue to taking mitigating actions. This is different from the results of
farmers’ expectations (i.e., farmers’ perceived barriers and perceived benefits of taking
mitigating actions to reduce transmitting WRZD) in which economic factors were indicated as
the top benefit as well as the top barrier to taking mitigating actions. This difference may relate
to the urgency caused by the rapid transmission of WRZD. Therefore, this difference may be an
important factor to consider when developing long-term or short-term programs and/or policies
that address mitigating actions against the transmission of WRZD. In addition, the striking
difference between the number of SSI farmers in An Giang and Thai Binh who considered
“worry about being affected by the transmission of WRZD” as a trigger/cue to taking mitigating
actions is not surprising when considering the differences in lifestyle and economic conditions
between Northern and Southern Vietnam. This may have implications for developing strategies
for intervention and for behavioural changes between the provinces of Thai Binh and An Giang.
3.5. Conclusions
This chapter presented and discussed the research results of SSI farmers’ perceptions
about the following: their overall on-farm water quality; the threat of transmitting WRZD
including the perceived susceptibility for and perceived severity of transmitting WRZD; the
barriers and benefits to taking mitigating actions and the belief in their ability to take an action;
and the triggers for their actions to mitigate the transmission of WRZD.
SSI farmers perceived the overall on-farm water condition in their villages to be
adequate. They were also aware of the potential harmful health consequences of using untreated
Page 131 of 340
water sources, and assessed water quality subjectively and primarily based on their own
impressions and previous experience. Most SSI farmers considered drinking untreated water and
handling untreated domestic water as being of moderate to high risk for the transmission of
HPAI, diarrhea, coliform bacteria, and parasites. If then infected, they believed that HPAI,
diarrhea, and coliform bacteria would be severe or very severe.
Most of the SSI farmers believed that they have moderate or high ability to conduct
mitigating actions against the transmission of WRZD. Moreover, economic factors were
perceived by most of the SSI farmers as both barriers and benefits to taking mitigating actions
against the transmission of WRZD. However, when it came to triggers/cues for taking mitigating
actions, economic factors were ranked as only the fourth factor that triggered action by most of
the SSI farmers. The top three factors were influence from the media, worry about being
infected, and advice from health workers.
The results presented in this chapter confirm my priori expectations that SSI farmers are
aware of the general quality of the water environment in their villages. They have some
impressions of risk factors for transmission of WRZD specifically moderate threats of WRZD to
their health and wellbeing, economic related factors as farmers’ common expected barriers and
benefits, and perceived moderate ability to taking actions to reduce transmission of WRZD.
Furthermore, this chapter also confirms my expected differences in perceptions of risk factors for
transmission of WRZD between SSI farmers of the two provinces. It will be useful to relate the
results of this chapter with objective assessments of on-farm water quality (see Chapter Four)
and to explore how farmers assess and make decisions about their mitigating strategies or how
these perceptions influence farmers’ choices about mitigating strategies (see Chapter Five).
Page 132 of 340
References
Aburto, N. J., Pevzner, E., Lopez-Ridaura, R., Rojas, R., Lopez-Gatell, H., Lazcano, E. et al.
(2010). Knowledge and Adoption of Community Mitigation Efforts in Mexico During the
2009 H1N1 Pandemic. American Journal of Preventive Medicine, 39, 395-402.
Ajzen, I. (1985). From intentions to actions: A theory of planned behavior. Prentice-Hall
Englewood Cliffs, NJ.
Becker, G. S. (1974). Rational choice theory: A Theory of Social Interactions. Journal of
Political Economy, 82, 1063-1093.
Becker, M. H. (2012). The Health Belief Model and Personal Health Behavior. (4 ed.) (vols. 2).
BUMC (2013). The Health Belief Model. BU School of Public Health [On-line]. Available:
http://sphweb.bumc.bu.edu/otlt/MPH-Modules/SB/SB721-Models/SB721-Models2.html
Carr, G., Potter, R. B., and Nortcliff, S. (2011). Water reuse for irrigation in Jordan: Perceptions
of water quality among farmers. Agricultural Water Management, 98, 847-854.
Carr, R. and Bartram, J. (2004). The control envelope and risk management. In (pp. 67-92). IWA
Publishing.
EPA (1993). Method 180.1: Determination of Turbidity by Nephelometry. EPA [On-line].
Available:
http://water.epa.gov/scitech/methods/cwa/bioindicators/upload/2007_07_10_methods_me
thod_180_1.pdf
Figuie, M. and Fournier, T. (2008). Avian Influenza in Vietnam: Chicken-Hearted Consumers?
Risk Analysis, 28, 441-451.
Page 133 of 340
GBD (2014). Global Burden of Disease Compare. Institute for Health Metrics and Evaluation
(IHME) [On-line]. Available: http://vizhub.healthdata.org/gbd-compare/
GSO (2010). Summary results of the Vietnam household living standard survey 2010 Vietnam
General Statistics Office (GSO).
GSO (2011). The 2009 Vietnam Population and Housing Cencus General Statistics Office,
Ministry of Planning and Invesment Vietnam.
GSO (2012a). Number of farms by province General Statistics Office.
GSO (2012b). Results of the 2011 Rural, Agricultrure and Fishery Census: Part 4 - Households
using aquacultural land by the size of land used and province Hanoi: Statistical
Publishing House - General Statistics Office.
Hall, D. C. and Le, Q. B. (2009). The midterm evaluation of the Government of Vietnam-UN
Joint Programme to Fight Highly Pathogenic Avian Influenza 2007-2010: Strengthening
the Management of Public Health Emergencies in Vietnam – with a focus on the
Prevention and Control of Diseases of Epidemic Potential including Highly Pathogenic
Avian Influenza (HPAI). The Government of Vietnam-UN Joint Programme to Fight
Highly Pathogenic Avian Influenza.
Herbst, S., Benedikter, S., Koester, U., Phan, N., Berger, C., Rechenburg, A. et al. (2009).
Perceptions of water, sanitation and health: a case study from the Mekong Delta,
Vietnam. Water Science and Technology, 60, 699-707.
Janz, N. K. and Becker, M. H. (1984a). The Health Belief Model: A Decade Later. Health
Education and Behavior, 11, 1-47.
Page 134 of 340
Janz, N. K. and Becker, M. H. (1984b). The Health Belief Model: A Decade Later. Health
Education and Behavior, 11, 1-47.
Keraita, B., Drechsel, P., and Konradsen, F. (2008a). Perceptions of farmers on health risks and
risk reduction measures in wastewaterirrigated urban vegetable farming in Ghana.
Journal of Risk Research, 11, 1047-1061.
Keraita, B., Drechsel, P., Seidu, R., Amerasinghe, P., Cofie, O., and Konradsen, F. (2010).
Harnessing Farmers' Knowledge and Perceptions for Health-Risk Reduction in
Wastewater-Irrigated Agriculture. In P.Drechsel, C. A. Scott, L. Raschid-Sally, M.
Redwood, and A. Bahri (Eds.), Wastewater irrigation and health : assessing and
mitigating risk in low-income countries (pp. 337-353). London: Earthscan Publications
Ltd.
Keraita, B., Jim'nez, B., and Drechsel, P. (2008b). Extent and implications of agricultural reuse
of untreated, partly treated and diluted wastewater in developing countries. Agr Vet Sci
Nutr Nat Resour, 3, 15.
Knudsen, L. G., Pham, D. P., Nguyen, H. T., Helle, S., Jensen, P. K., Dalsgaard, A. et al.
(2008a). The fear of awful smell: risk perceptions among farmers in Vietnam using
wastewater and human excreta in agriculture. The Southeast Asian journal of tropical
medicine and public health, 39, 341-352.
Knudsen, L. G., Pham, D. P., Nguyen, T. H., Samuelsen, H., Jensen, P. K., Dalsgaard, A. et al.
(2008b). The fear of awful smell: risk perceptions among farmers in Vietnam using
wastewater and human excreta in agriculture. Southeast Asian Journal of Tropical
Medicine and Public Health, 39(2), 341-352.
Page 135 of 340
Le, D. H., Li, E., Nuberg, I., and Bruwer, J. (2014a). Farmers's assessments of private adaptive
measures to climate change and influential factors: a study in the Mekong Delta,
Vietnam. Nat Hazards, 71, 385-401.
Le, Q. B., Hall, D. C., and Cork, S. C. (2014b). Vietnam National Rural Clean Water Supply and
Sanitation Strategy (2020): Implications for the reduction of waterborne diseases among
small scale farmers and their animals. In GRF One Health Summit 2112: One Health -
One Planet - One Future Risks and Opportunities Global Risk Forum Davos.
Le, Q. B. and Hall, D. C. (2011a). Stakeholder cooperation and ecohealth to control Highly
Pathogenic Avian Influenza (HPAI) in Vietnam. Canadian Public Health Association
Annual Conference 2011, Montreal Quebec, 2011 [On-line]. Available:
http://resources.cpha.ca/CPHA/Conf/Data/2011/A11-270ae.pdf
Le, Q. B. and Hall, D. C. (2011b). Using an ecohealth approach to livestock production:
community surveillance of newly emerging infectious diseases (nEIDs), International
Meeting on Emerging Diseases and Surveillance (IMED 2011.
Levy, B. R. and Myers, L. M. (2004). Preventive health behaviors influenced by self-perceptions
of aging. Preventive Medicine, 39, 625-629.
MARD (2012). Provincial rural and agriculture database Vietnam: Planning Department,
Ministry of Agriculture and Rural Development.
MoC and MARD (2000a). National Rural Clean Water Supply and Sanitation Strategy up to the
Year 2020 Hanoi, Vietnam: Ministry of Construction and Ministry of Agriculture and
Rural Development of Vietnam.
Page 136 of 340
MoC and MARD (2000b). National Rural Clean Water Supply and Sanitation Strategy up to the
year 2020 Hanoi, Vietnam: Ministry of Construction and Ministry of Agricultrue and
Rural Development of Vietnam.
Moe, C. L. (2004). What are the criteria for determining whether a disease is zoonotic and water
related. In World Health Organization (WHO). Waterborne Zoonoses: Identification,
Causes and Controls (pp. 27-45). London: IWA Publishing.
MoH (2009a). National technical regulation on domestic water quality (QCVN 02: 2009/BYT).
Hanoi: Department of Preventive Medicine and Environment, Ministry of Health,
Socialist Republic of Vietnam.
MoH (2009b). National technical regulation on drinking water quality (QCVN 01: 2009/BYT).
Hanoi: Department of Preventive Medicine and Environment, Ministry of Health,
Socialist Republic of Vietnam.
OECD (2012). "Determinants of farmer behavioural change", in Farmer Behaviour, Agricultural
Management and Climate Change. In ( OECD Publishing.
Owusu, V., Bakang, J. E., Abaidoo, R., and Kinane, M. (2011). Perception on untreated
wastewater irrigation for vegetable production in Ghana. Environment, Development and
Sustainability, 1-16.
Payne, N., Jones, F., and Harris, P. (2002). The impact of working life on health behavior: The
effect of job strain on the cognitive predictors of exercise. Journal of Occupational
Health Psychology, 7, 342-353.
Quinlan, J. A. (2005). The use and perception of drinking water in a Canadian community. The
Universitiy of Guelph.
Page 137 of 340
Rosenstock, I. M., Strecher, V. J., and Becker, M. H. (1988). Social Learning Theory and the
Health Belief Model. Health Education and Behavior, 15, 175-183.
Weldesilassie, A. B., Boelee, L. I. N., Drechel, P. A. Y., and Dabbert, S. T. E. P. (2011).
Wastewater use in crop production in peri-urban areas of Addis Ababa: impacts on health
in farm households. Environment and Development Economics, 16, 25-49.
WHO (2004). Waterborne Zoonoses: Identification, Causes, and Control. IWA Publishing.
WHO (2011). Guidelines for drinking-water quality. (Fourth ed.) World Health Organisation.
WHO (2014a). Avian influenza in humans World Health Organization.
WHO (2014b). Diseases and injury country estimates: Burden of disease. World Health
Organization (WHO) [On-line]. Available:
http://www.who.int/healthinfo/global_burden_disease/estimates_country/en/
Page 138 of 340
Chapter Four: Water Quality and Public Health among Small-Scale Integrated (SSI)
Farms in the Provinces of Thai Binh and An Giang in Vietnam
4.1. Introduction
4.1.1. Rationale
Water supply and water quality have a profound influence on public health(WHO, 2011b;
WHO, 2011d). It is also estimated that water is the vehicle for about 80% of all water-related
zoonotic diseases (WRZD) in people (Ford and Colwell, 1996). Specifically, in the Red River
and Mekong River Deltas in Vietnam, water is considered to be “the heart of development,” yet
this location was also the major focus for the highly pathogenic avian influenza (HPAI) epidemic
waves that took place between 2003 and 2005 (Gilbert et al., 2008a; Henning et al., 2009; Phan,
2010; Sims et al., 2005; Soares et al., 2010). The distribution of early HPAI outbreaks during the
2003-5 HPAI epidemic in Vietnam was concentrated at the farm level in the Red River Delta and
the Mekong River Delta. This distribution was thought to be associated with bodies of water on
farms (Morris and Jackson, 2005). Avian influenza viruses (AIV) can persist for extended
periods of time in water depending on water temperature and other environmental factors (Boere
et al., 2006; Morris and Jackson, 2005; Webster et al., 2006; WHO, 2007a). Moreover, highly
Pathogenic Avian Influenza (HPAI) viruses can persist in water over a 60-day period and
constitute a potential source of waterborne AIV transmission for wild birds and domestic ducks
(Brown et al., 2007; Domanska et al., 2010; Nazir et al., 2010; Roche et al., 2009).
Environmental transmission of viruses in general can occur due to consuming virus-
contaminated drinking water, having contact with virus-contaminated water bodies, and being
exposed to poorly managed excreta and wastewater (WHO, 2005; WHO, 2007b). For example,
wild birds carry and shed AIV (Costa et al., 2011; Ip et al., 2008). Factors that might facilitate
Page 139 of 340
AIV transmission between domestic and wild birds include: poultry operations located in wild
migratory bird flyways; watering and feeding areas open to access by wild birds and domestic
poultry; and waste runoff from domestic poultry operations that ends up in wetlands used by
wild birds (Rapport et al., 2006). These factors create favourable conditions for the co-infection,
re-assortment, and evolution of viruses among host species (Stallknecht et al., 1990; Ward et al.,
2009; Webster et al., 2006).
Infected wild birds, humans, and domestic animals shed AIV in their faeces and saliva
which could enter water environments (WHO, 2007a). High densities of livestock can also
pollute surface water such as lakes, ponds, streams, and rivers (Bui, 2010; Trinh, 2001).
Resultant water pollution levels can be 10 - 100 times greater than the acceptable level for use in
agriculture (Vu, 2001; Yajima and Kurokura, 2008). As a result, farmers need appropriate risk
mitigating measures because the direct use of untreated contaminated water can result in a
disease risk for humans residing in the areas (Dalsgaard, 2001; Guan and Holley, 2003; Toews,
2009; Trang et al., 2007a; Yajiam and Kurokura, 2008). Moreover, duck farms with a shared
water supply are often not fenced nor are they separated from streams and ponds and these farms
appear to be instrumental in AIV transmission in domestic ducks (Hubbard et al., 2004; Lin et
al., 2010; Markwell and Shortridge, 1982).
In Vietnam, farmers traditionally consider water to be the most important factor in
agriculture and this is commonly expressed in the proverb “nhất nước, nhì phân, tam cần, tứ
giống” meaning “water is the most important factor, and then comes fertilizer, industriousness,
and breeds” (Bach, 2007). However, over 70% of small-scale integrated (SSI) farmers use
contaminated water for drinking and other on-farm purposes and have no access to hygienic
Page 140 of 340
latrines. Meanwhile, they have very limited awareness about water quality or environmental
sanitation (MoC, 2000; MoC and MARD, 2000). SSI farming also presents a number of risk
factors for WRZD transmission. Some examples of these factors for HPAI transmission include
the proximity of poultry flocks to water sources and their increased accessibility to them, an
increased density of ponds and streams, and temporal flooding (Gilbert et al., 2008b; Nguyen et
al., 2011; Pfeiffer et al., 2009).
In contrast, appropriate densities for livestock production and rational water use33 can
help reduce environmental problems, control nutrient loss, and reduce the release of pathogens
into agricultural runoff (Hooda et al., 2000). Yet, the inappropriate use of water and its
consequences to the health of SSI farmers and their animals has not been well studied. Therefore,
prior to undertaking a public health intervention to reduce the incidence of WRZD in rural
communities, it is important to understand the status of water and its management on individual
farms. This includes developing an understanding of water quality on farms, uses of water
sources, farmers’ awareness and perceptions of water quality, and water use with respect to
microbial infection (Dinh et al., 2004; Huynh et al., 2006; Lye, 2002; Nguyen et al., 2008; Tran
et al., 2010; Trang et al., 2007b; Trinh and Nguyen, 2009). The basic recommended indicators to
assess rural water quality include pH, turbidity, and the presence of coliform bacteria.
Escherichia coli (E. coli) is also an indirect indicator organism of other faecal microbes in water.
Meanwhile, levels of pH and turbidity have positive correlations with the pathogen removal in
33 Rational use of water in the context of livestock production may include good livestock management activities
such as hygienic rearing condition for livestock, adopting water quality standards for livestock, and treating
livestock waste at source (Carr et al., 2004)
Page 141 of 340
water (WHO and OECD, 2003a; WHO, 2011a; WHO, 2011c; WHO, 2011b). Nevertheless,
methods for detecting viruses in rural water are limited (WHO, 2004).
Given the above, this research study had two objectives. First was to describe sources and
quality of water used on participating SSI farms. The second was to look for associations
between the quality of these water sources and farmers’ perceptions of risk factors for the
transmission of WRZD, which were presented in Chapter Three.
4.1.2. Research questions and hypotheses
4.1.2.1.Research questions:
This chapter addresses the fourth and fifth questions of my research, which are: 1) What
sources of water do SSI farms use for drinking and domestic purposes 34 and what are the
corresponding basic microbial and related indicators35 of water quality for those sources in the
provinces36 of Thai Binh and An Giang?, and 2) How does the quality of on-farm water SSI
farmers use relate to both their perceptions about water quality and risk factors for the
transmission of WRZD?
In Figure 4.1, I present a flow chart outlining the structure of Chapter Four.
4.1.2.2.Hypotheses:
1. The frequency of use of water sources for drinking and domestic purposes is different
among SSI farmers in the province of Thai Binh than in the province of An Giang.
34 In my research, drinking water is defined as “water used for direct drinking or food processing.” Domestic water
is defined as “water use for domestic purposes but nor for direct drinking or food processing” (MoH, 2009b; MoH,
2009a) 35 The WHO recommends Escherichia coli (E. coli) colony forming unit (cfu), turbidity, and pH as basic microbial
and related indicators for assessing rural water quality (WHO, 2012) 36 Vietnam is divided into 64 provinces. Each province is further sub-divided into districts, then communes and
villages.
Page 142 of 340
2. SSI farms in the provinces of Thai Binh and An Giang are faced with low quality
water sources for drinking and domestic purposes.37
3. SSI farmers’ socioeconomic status and perceptions of on-farm risk factors for WRZD
transmission are associated with the differing microbial content of water on farms.
4.2. Methods
In this section, I focus on presenting the methods that were used to: describe the water
sources used on farms; objectively assess on-farm water quality in the studied areas; and to
analyze the relationship between the objectively assessed on-farm water quality (i.e., pH,
turbidity, and E. coli) and farmers’ subjective assessments of their on-farm water quality
(presented in Chapter Three). A more detailed description of the research methods used in this
chapter is presented in Appendices 4.6.1-2. Other information concerning the research methods
such as sample size, studied areas, research questionnaires, and research ethics were presented in
Chapter Two.
I used questionnaires to ask participating farmers to indicate the water sources they have
access to and their corresponding frequency of use for drinking and domestic purposes. In-depth
interviews, farm visits, and focus group discussions were also employed to obtain more insights
37 These terms are defined in section 4.2 Methods.
Page 143 of 340
Figure 4.1. Chapter Four’s Outline
Used to assist modeling factors associated with SSI
farmers’ engagement in mitigating strategies to reduce
transmission of WRZDs (Chapter 5) in the provinces
of Thai Binh and An Giang in Vietnam
Research question 4:
What are sources and the basic microbial and related quality of
water used by SSI farmers for drinking and domestic purposes in
Thai Binh and An Giang provinces of Vietnam?
Factors associated with the microbial quality of
water used on SSI farms for drinking and domestic
purposes
Microbial
quality of water
used on SSI
farms for
drinking
(assessed using
Coliplate)
Microbial and
related quality of
water used on SSI
farms for drinking
and domestic
purposes (assessed
using laboratory
protocol)
Sources of water and
frequencies of use on
SSI farms
Chapter 4:
Presents the water quality among SSI farms in Thai Binh and An Giang provinces of Vietnam, including
Research question 5:
How does the quality of water used by SSI farmers relate to their
perceptions of water quality and risk factors for transmission of
WRZD?
Socioeconomic of
SSI farms (from
Chapter 2)
Perceptions of SSI farmers
toward risk factors for
WRZD transmission (from
Chapter 3)
Page 144 of 340
about on-farm water sources and frequencies of use. Drinking water was defined as “water used
for direct drinking or processing food,” and domestic water was defined as “water used for
domestic use but not for direct drinking or processing food” (MoH, 2009b; MoH, 2009a). I used
two protocols to assess the basic objective quality indicators for on-farm water for drinking and
domestic purposes (i.e., pH, turbidity, and E. coli). These protocols were 1) a protocol for
laboratories to assess on-farm water quality and 2) a protocol for community health/veterinary
workers (CHWS/CVWs) to quantify E. coli in on-farm water. These protocols followed WHO’s
recommendations for assessing the basic indicators of rural water quality (WHO, 2011b; WHO,
2011c; WHO and OECD, 2003b) and conformed with the Vietnamese national technical
regulations on drinking and domestic water quality to assess the basic indicators of the quality of
on-farm water (MoH, 2009b; MoH, 2009a). Detailed descriptions of these protocols are provided
in Appendix F2.
Descriptive and analytic statistics were used to analyze and present the results of the
water sources, frequencies of use, and objective assessments of on-farm water quality. Two sided
two-sample t-tests, one for drinking water and one for domestic water, using province
(DProvince) as a group variable were used to test the first hypothesis. In this hypothesis, the null
hypothesis indicated there were no differences between the mean frequencies of use of a water
source for drinking or domestic purposes between farmers in Thai Binh and An Giang (i.e., Ho:
diff = 0). The alternative hypothesis stated that the difference was not “0” (i.e., Ha: diff ≠ 0).
With respect to the second hypothesis, I used one-sided, one-sample t tests to compare the means
of pH, turbidity, and E. coli levels of water on farms with the standard levels of pH, turbidity,
Page 145 of 340
and E. coli in drinking and domestic water set by the Vietnamese Ministry of Health (MoH,
2009b; MoH, 2009a).
Regression analysis (Dooho et al., 2009) was used to model and test the relationship
between the objectively assessed on-farm water quality and farmers’ perceived on-farm water
quality and risk factors for WRZD transmission (i.e. the third hypothesis). Only E. coli test
results, determined using standard laboratory protocols, were used for testing this hypothesis.
Descriptions of variables used in the regression analysis are presented in section 4.3.4.1. An
overall strategy and four specific strategies were used to model factors that are associated with
on-farm water quality.
The overall strategy for modeling associated factors of on-farm water quality was to start
with a Multiple Linear Regression (MLR) model then use a Seemingly Unrelated Regression
(SUR) model. The MLR model included a number of separate single regression equations in
which E. coli cfu in various sources of water for drinking and domestic purposes were dependent
variables. The SUR model included two regression equations simultaneously for E. coli cfu in
both drinking and domestic water. The purpose of using both MLR and SUR models was to
provide further insights about the complexity of on-farm water quality.
The first specific strategy was specifying variables for a maximum model. E. coli cfu in
each source of on-farm drinking water samples were specified as the dependent variables (DVs)
for the MLR models. E. coli cfu in both drinking and domestic water were specified as the DVs
for the SUR model. A full set of potential independent variables (IVs) was specified based on the
theoretical model (See Section 3.2, Chapter Three), the hypothesis, the literature for this
Page 146 of 340
research, and my knowledge of on-farm water public health (See Section 3.4.1 for more detailed
description of IVs). The second specific strategy was to use two sets of considerations for
retaining IVs in the models. The first set of considerations was to include variables of my
primary interest in the research, variables known to be a potential confounder of primary factors
(e.g., farmers’ age), and evidence of change in the coefficients of primary factors when removing
a variable. The second set of considerations was to use p-values of the coefficients of individual
IVs and an adjusted R2 for the number of associated factors in the model to assess the model’s
goodness of fit. The third specific strategy was to fit models using various variable selecting
procedures (e.g., stepwise selection and forward selection). The final specific strategy was to use
statistical software (i.e., Stata 13) and my knowledge and experiences of water public health in
Vietnam to analyze data and interpret the results.
The MLR model and SUR model are written as follows and detailed descriptions of
model building strategies are presented in Appendix 4.6.3:
The MLR model:
Yi = β0+ βi Xi + ɛi; with i = 1, 2, 3... n
Yi is a continuous dependent variable representing on-farm water quality, considering
bundle i. Each Y represents the numbers of E .coli cfu in 100 ml of a source of on-farm
water.
Xi is a vector of IVs (continuous and/or dichotomous) believed to be relevant to the
consideration of ith equations (i = 1, 2, 3... n).
βo is a intercept or constant to be estimated,
Page 147 of 340
βi is a vector of unknown coefficients describing the direction and magnitude of the effect
of an Xi on Yi after controlling for the effects of the other Xi; βi to be estimated,
ɛi is the error term
The SUR model:
The SUR model includes two equations for 1) E. coli cfu in on-farm drinking water
versus a set of potential IVs; and 2) E. coli cfu in on-farm domestic water versus another set of
IVs. The two equations are written below with j = 1… m MLR equations:
yj = x’j βj + uj
These two equations are stacked into an SUR model:
[
𝑦1..
𝑦𝑚
] = [𝑥1 0 ⋯ 0⋮ ⋱ ⋮0 ⋯ 𝑥𝑚
] [
β1..
βm
] + [
u1..
um
]
The error terms are assumed to have a zero mean and to be independent across farmers and
homoscedastic. For a given farm, the errors are correlated across equations:
E(uij uij’|𝑋) = σjj’ and σjj’ ≠ 0 where j ≠ j’
Cross-equation restrictions were tested, and constraints were imposed for IVs that were
identified in the fitting of both equations and that were not significantly different from each
other. The SUR model was then re-estimated.
4.3. Results:
4.3.1. Sources of water and frequencies of use in participating farms
SSI farmers participating in the research study in both provinces depended on multiple
sources of water for both drinking and domestic purposes (Table 4.1). Most farmers in Thai Binh
Page 148 of 340
and An Giang used water from drilled well or river/canal respectively. Most farmers in An Giang
reported that they rarely or almost never used dug wells or drilled water wells as one of their
drinking water sources - 217/298 (73%) and 219/298 (73%) respectively. In Thai Binh, 143/300
(48%) of the farmers indicated that pond water or river/canal water were never or rarely used as
a source of drinking water. Farmers in both provinces also used combinations of different water
sources for domestic purposes. Among these, drilled wells and rivers/canals were frequently used
by 251/300 (84%) and 288/298 (97%) of the farmers in Thai Binh and An Giang respectively. In
both provinces, participating farmers rarely or never used water from bottles, pipes, or dug wells
for domestic purposes. The frequencies of use of water sources are statistically different between
two provinces. This supports one of the hypotheses tested in this chapter.
4.3.2. Results of on-farm water quality assessed using the Vietnamese national laboratory
protocol
Quality of on-farm water used for drinking purposes
Well water used for drinking on An Giang farms had a higher mean pH compared to
farms in Thai Binh (Table 4.2). The mean turbidity in rain and piped water on An Giang farms
was almost double that found in Thai Binh farms; however, mean numbers of E. coli cfu in An
Giang in those same two water sources were much lower than those in Thai Binh. Conversely,
mean E. coli cfu numbers in well water used for drinking in An Giang farms was about four
times higher than those on Thai Binh farms. For more results of the quality of drinking water, see
Appendix F 4.1.
Quality of on-farm water used for domestic purposes
Page 149 of 340
There were slight differences between the pH levels of domestic water between the two
provinces. The mean turbidity levels in these sources of water in An Giang were two to three
times greater than in Thai Binh (Table 4.3), yet the mean numbers of E. coli cfu in well water in
both Thai Binh and An Giang were similar. Pond water and river water used for domestic
purposes in both Thai Binh and An Giang had much greater mean numbers of E. coli cfu
compared to well water. For more results about the quality of drinking and domestic water,
please see Appendix F.4.2.
4.3.3. Quality of on-farm water assessed by community health/veterinary workers
(CHW/CVWs)
This section presents the results of the prevalence of E. coli in on-farm drinking water by
CVWs in the province of Thai Binh and CHWs in the province of An Giang using ColiplateTM
tests. Seventy-five percent (224/299) and 81% (242/298) of water samples used for drinking
were positive in E. coli tests in participating farms in Thai Binh and An Giang respectively
(Table 4.4). Farmers in Thai Binh used stored rainwater, piped water, well water, and bottle
water for drinking; those in An Giang also used flocculated river water and piped water stored in
reservoirs (For more detailed information of the quality of on-farm water assessed by
CHWs/CVWs, please see Appendix F.4.3).
4.3.4. Factors associated with the auality of on-farm water
This section includes a description of DVs representing E. coli levels in individual and in
all water sources and the potential IVs that are associated with the DVs (Subsection 4.3.4.1).
Pairwise correlation among the variables used for modelling is provided in Appendix F.3.2.
Page 150 of 340
Subsections 4.3.4.2 – 5 present the results of the best fittings of the regression model for each
DVs and the SUR model respectively.
4.3.4.1. Description of variables used for modelling
Table 4.5 shows the summary statistics of the dependent variables (DVs) used for
modelling. These DVs represent E. coli cfu in various on-farm water sources for drinking and
domestic purposes. Table 4.6 summarizes the possible IVs for modelling the DVs. On-farm
water quality may be associated with demographic factors (e.g., age), socioeconomic status (e.g.,
years of attending school), and subjective assessments of on-farm water quality (i.e., farmers’
perceptions of water quality). Most pairs of IVs were not significantly highly correlated (i.e.,
correlation coefficients greater than 0.5 with p-value< 0.05). These pairs of variables did not
essentially contain the same information (see Appendix F.3.2 for more details).
4.3.4.2.MLR Model - E. coli cfu in sources of water used for drinking (DV) vs. a set of IVs
Table 4.7 shows the results for the MLR regression analysis in which the number of E.
coli cfu in four sources of drinking water including rain, pipe, well, and river (i.e., the dependent
variables), each of which was regressed against a set of independent variables (IVs). The model
for E. coli cfu in stored rainwater for drinking was significant overall with p < 0.001 and
adjusted R2 = 0.17. Three out of ten variables were significant for the associated factors of E. coli
cfu in stored rainwater at p < 0.1. These significant associated factors included on-farm income
per person per year (Income; p < 0.09), fish production (Fish prod; p < 0.01), and satisfaction
with drinking water sources (DSatisfy-Drink, p < 0.01).
Page 151 of 340
Table 4.1. Mean Frequency of Use for On-Farm Water Used for Drinking and Domestic Purposes in Both Provinces of Thai
Binh and An Giang, Vietnam
Province/Sources of
water
Mean frequency of use for drinking by source of water (a)
Drinking use Domestic use
Obs Mean Std. Dev. Obs Mean Std. Dev.
Thai Binh
Rain 196 3.4*** 1.3 256 2.5*** 0.9
Drilled well 231 3.2*** 1.3 251 3.5*** 1.1
Bottled 169 2.4 1.1 108 1.2*** 0.5
Pipe 61 2.4 1.4 52 1.8*** 0.9
Dug Well 40 1.6*** 1.0 38 1.9*** 1.1
Pond 143 1.0*** 0.2 209 2.9*** 1.4
River/Canal 143 1.0*** 0.2 251 3.2*** 1.2
An Giang
River/Canal 297 3.0 1.3 288 3.8 1.0
Pipe 243 2.6 1.7 245 1.3 0.9
Bottled 280 2.5 1.2 254 1.5 0.8
Rain 277 2.4 1.0 267 2.1 1.0
Pond 271 1.3 0.6 260 1.4 0.7
Drilled well 219 1.0 0.2 227 1.1 0.6
Dug Well 217 1.0 0.1 226 1.0 0.1 (a) 1 = Never, 2 = Rarely, 3 = Sometimes, 4 = Often, to 5 = Very often
*** indicating values that are significant different between two provinces at p<0.01
Page 152 of 340
Table 4.2. Comparison of Mean pH, Turbidity, and E. coli cfu of on-Farm Water Used for Drinking, Assessed Using Samples
Tested by National Laboratories, in the Provinces of Thai Binh and An Giang, Vietnam
Variable/Province Rain water Pipe water Well water e Bottled water
River water
(fl.)
pHa
Thai Binh 6.4** 6.4** 6.5** - -
An Giang 7.9** 7.5** 6.7** 9.4 7.0
Turbidity (measured in Nephelometric Turbidity Units) b
Thai Binh 0.8** 0.9* 3.6 - -
An Giang 1.3** 2.0* 3.9 0.8 7.6
E. coli (measured in number of E. coli colony-forming units per 100 mL of sample water)c
Thai Binh 61.8** 19.5** 27.0** - -
An Giang 11.1** 7.1** 107.1** 8.0 12.8
Frequency of used
Thai Binh 3.4 2.4 2.4 2.4 1.0
An Giang 2.4 2.7 1.0 2.5 3.0 a = means pH of all sources of water were not smaller than “6.5” nor significantly greater than “8.5” at p < 0.1
b = means of turbidity of well and river water were significantly greater than “2” at p < 0.01
c = means of E. coli in all sources of water were significantly greater than “0” at p<0.01
d = Frequency of use of the source water for drinking using (1 = Never, 2 = Rarely, 3 = Sometimes, 4 = Often, and 5 = Very often).
e =both drilled well and dug well
** indicating values that are significantly different between the two provinces at p<0.05; * indicating values that are significantly different between the two
provinces at p<0.1
Page 153 of 340
Table 4.3. Comparison of Mean pH, Turbidity, E. coli cfu of On-Farm Water Used for Domestic Purposes, Assessed Using
Samples Tested by National Laboratories, in the Provinces of Thai Binh and An Giang, Vietnam
Variable/Province Well water e Pond water River water
pH a
Thai Binh 7.1* 6.7** 7.1
An Giang 6.1* 7.0** 7.3
Turbidity (measured in Nephelometric Turbidity Units) b
Thai Binh 10.2 34.4*** 25.3*
An Giang 31.5 59.2*** 62.1*
E. coli (measured in number of E. coli colony-forming units per 100 mL of sample water) c
Thai Binh 356.5 2,494.7 2,757.5***
An Giang 371.4 2,950.1 780.7***
Frequency of used
Thai Binh 2.2 2.9 3.2 An Giang 1.1 1.4 3.8
a = means pH of all water were not significantly smaller than “6.0” nor greater than “8.5” at p < 0.1 b = means of turbidity of well and river water were significantly greater than “5” at p < 0.01
c = means of E. coli in all sources of water were significantly greater than “20” at p < 0.01 d = Frequency of use of the source water for domestic purposes (1 = Never, 2 = Rarely, 3 = Sometimes, 4 = Often, to 5 = Very often). e = both drilled well and dug well
*** indicating values that are significantly different between the two provinces at p < 0.01; ** indicating values that are significantly different between the two
provinces at p < 0.05; * indicating values that are significantly different between the two provinces at p<0.1
Page 154 of 340
Table 4.4. Prevalence of E. coli in On-Farm Water Tests (using ColiplateTM) Used for
Drinking by Source in Thai Binh and An Giang, Vietnam
Water
sources
Presence of E. coli in on-farm water for drinking
An Giang Thai Binh
Positive Total Prevalence Positive Total Prevalence
River with
flocculation 124 131 94.7 - - -
Rain 24 27 88.9 67 92 72.8
Pipe 61 82 74.4 19 36 57.8
Well 3 5 60.0 137 165 83.0
Bottle 19 33 57.6 1 6 16.7
Pipe (stored
in reservoir) 11 20 55.0 - - -
Total 242 298 81.2 224 299 74.9
Table 4.5. Summary Descriptive Statistics for the Dependent E. coli Variables Used in the
Regression Models
Variable
name
Description - Number of E.
coli cfu in: Obs Mean sd Iqr Max a
NE.coli
Drink-rain Rain water used for drinking 150 46.9 74.0 61 232
NE.coli
Drink-pipe Pipe water used for drinking 117 10.1 17.1 13 50
NEcoli
Drink-well Well water used for drinking 156 24.1 40.0 25 213
NE.coli
Drink- bottle
Bottled water used for
drinking 42 8.0 20.1 0 110
NE.coli
Drink-river River water used for drinking 117 12.8 30.5 2 260
NE.coli
Dom-well
Well water used for domestic
purposes 29 360.1 542.3 55 1,800
NE.coli
Dom-pond
Pond water used for
domestic purposes 353
2,596.
6
3,237.
5 1,200
12,40
0
NE.coli
Dom-river
River water used for
domestic use 213 817.8
1,077.
9 300 3,200
NE.coli
Drink
All sources of water used for
drinking 529 12.7 21.0 2 98
NE.coli Dom
All sources of water used for
domestic purposes 576
1,506.
3
1,994.
6 560
10,06
0 Notes: Dom = Domestic, N = Number, a Min values of these variables equal “0”
Page 155 of 340
Table 4.6a. Summary Statistics for the Independent Variables (IVs) Considered for Use in Regression Models
Variable name Description N Mean Sd Min Max Count (1) % (1)
Age Age of the farmers 597 45.9 11.3 17 85 - -
DGender Gender of the farmers (Male = 1; Female =0) 598 - - - - 428 71.6
Schooling Years of attending school 598 6.9 3.2 0 18 - -
DIncome Income under poverty (Poor =1; No poor = 0) 590 - - - - 118 20.0
Children Number of household member under 18 years old (Yes
=1; No = 0) 598 - - - - 421 70.4
Experience Years of farming 598 9.6 8.3 0 50 - -
DOff-farm job Had off-farm jobs (Yes =1; No = 0) 593 - - - - 293 49.4
DFPR-Livestock-
Man
Man primarily responsible for livestock production (Yes
=1; No = 0) 589 - - - 244 41.4
DFPR-Livestock-
Woman
Woman primarily responsible for livestock production
(Yes =1; No = 0) 589 - - - - 91 15.5
DFPR-Fish-
Woman
Woman primarily responsible for fish production (Yes
=1; No = 0) 598 - - - 41 6.9
DFPR-Fish-Man
Man primarily responsible for fish production (Yes =1;
No = 0) 598 - - - - 187 31.3
DFPR-Health-Man
Man primarily responsible for family health production
(Yes =1; No = 0) 591 - - - 131 22.2
Province Thai Binh =1; An Giang = 0) 598 - - - - 300 50.2
DFish prod Having fish production (Yes =1; No = 0) 598 - - - 388 64.9
DRain for animals Using rain water for livestock production (Yes =1; No =
0) 598 - - - - 392 65.6
DPond for animals Using pond water for livestock production (Yes =1; No
= 0) 598 - - - 211 35.3
DPipe for animals Using pipe water for livestock production (Yes =1; No =
0) 598 - - - - 67 11.2
DMulti-agri
activities
Engage in more than three agricultural activities (Yes
=1; No = 0) 503 - - - - 436 86.7
DPoultry AI Poultry infected with AI in the past (Yes =1; No = 0) 598 - - - - 50 8.3
Npigs/yr Number of pigs produced on-farm per year 491 11.3 16.9 0 70 - -
Note: AI = Avian influenza, D = Dummy, FPR = Farmers’ perceived responsibility, and N = Number
Page 156 of 340
Table 4.6b. Summary Statistics for the Independent Variables (IVs) Considered for Use in Regression Models (cont’d)
Variable name Description N Mean Sd Min Max Count (1) % (1)
Independent variables relating to farmers’ perceptions of:
DFPWater quality general water quality in the villages (Good=1; Not good = 0) 597 - - - - 486 81.4
DFPBarr-Cost cost as barriers of WRZD mitigating actions (Yes =1; No = 0) 598 - - - - 239 40.0
DFPRisk rain water
Risk of WRZD transmission from untreated rain water (Yes
=1; No = 0) 550 - - - - 517 94
DFPS-AI-Drink susceptibility to AI from UDW (Yes =1; No = 0) 496 - - - - 469 94.6
DFPS-Diarrhea-Drink susceptibility to diarrhea from UDW (Yes =1; No = 0) 559 - - - - 554 99.1
DFPS-Coliform-Drink susceptibility to coliform from UDW (Yes =1; No = 0) 213 - - - - 213 100.0
DFPS-Parasite-Drink susceptibility to parasites from UDW (Yes =1; No = 0) 404 - - - - 402 99.5
DFPS-AI-Dom susceptibility to AI from UDomW (Low =1; High = 0) 497 - - - - 289 58.2
DFPS-Coliform-Dom susceptibility to coliform from UDomW (Low =1; High = 0) 213 - - - - 146 68.5
DFPS-Diarrhea-Dom susceptibility to diarrhea from UDomW (Low =1; High = 0) 561 - - - - 354 63.1
DFPS-Parasite-Dom Susceptibility to parasites from UDomW (Low =1; High = 0) 405 - - - - 293 72.4
DSatisfy-Drink satisfaction with source of drinking water (Low =1; High = 0) 596 - - - - 244 40.9
DSatisfy-Dom satisfaction with domestic water (Low =1; High = 0) 580 - - - - 398 68.6
DAI heard heard of AI or not (Yes =1; No = 0) 598 - - - - 541 90.5
DFPHarm-Dom harm of UDomW (Low =1; High = 0) 576 - - - - 332 57.6
DFPHarm-Drink harm of UDW (Yes =1; No = 0) 574 - - - - 565 98.4
Other Independent variables:
DDump Waste
Dump waste of livestock production to sources of water for
domestic (Yes =1; No = 0) 547 - - - - 175 32.0
DFE-Livestock
Management Engage in livestock management (Yes =1; No = 0) 598 - - - - 329 55.0
DTest-dom Have tested domestic water in the past (Yes =1; No = 0) 559 - - - - 40 7.2
Notes: AI = Avian influenza, D = Dummy, Dom = Domestic, FP = Farmers’ perceptions of, FPBarr = Farmers’ perceived barriers to taking mitigating
actions, FPS = Farmers’ perceived susceptibility to, UDW = untreated drinking water, UDomW = untreated domestic water
Page 157 of 340
The model for E. coli cfu in pipe water was statistically significant with p < 0.01 and
adjusted R2 = 0.20. Two variables were significantly associated with of E. coli in piped drinking
water including the number of chickens produced (Npoultry/yr, positive, p < 0.05) and the
number of ducks produced (Nducks/yr, p < 0.01). The most interesting model fitting for E. coli
in well water was significant at p < 0.01 and adjusted R2 = 0.35. This model included three
variables that were significantly associated with E. coli cfu in well water used for drinking:
income (DIncome, p<0.05), fish production (DFish prod, p<0.01), and perceived susceptibility to
parasites from using untreated drinking water (DFPS-Parasite-Drink, p < 0.1). The model for
associated factors of E. coli cfu in river water was significant overall with p < 0.05 and adjusted
R2 = 0.07. Three variables were significantly associated with E. coli cfu in river water including
gender (DGender, positive, p < 0.1) and the number of chickens produced (Nducks/yr, p < 0.01)
(Table 4.7).
4.3.4.3. MLR Model - E. coli cfu in water sources for domestic use (DV) vs. a set of IVs
In Table 4.8, the model fitting results of E. coli cfu in pond water for domestic purposes
was significant overall with p < 0.01 and adjusted R2 = 0.05. Three variables were significant
associated factors of E. coli cfu in pond water, specifically: education (Schooling, p < 0.05),
perceived susceptibility to HPAI when using untreated domestic water (DFPS-AI-Dom, p < 0.1),
and satisfaction with current sources of domestic water (DSatisfy-Dom, p < 0.1). The model
fitting results of E. coli cfu in river water for domestic purposes was significant overall with p <
0.01 and adjusted R2 = 0.12. Two variables were positive significantly associated factors of E.
coli cfu in river, specifically the following: occupation other than farming (DOff-farm job, p <
0.1) and number of pigs produced (Npigs/yr, p < 0.01). Three factors negatively and significantly
Page 158 of 340
associated with E. coli in river water for domestic purposes were gender (DGender, p < 0.1),
male farmers playing major roles in livestock production (DFPR-Livestock-Man, p < 0.05), and
perceived susceptibility to parasites from using untreated domestic water (DFPS-Parasite-Dom, p
< 0.01).
4.3.4.4.SUR Model - E. coli cfu in sources of on-farm water used for domestic purposes
Tables 4.9a-c show a SUR model fit of four associated factors of the level of E. coli cfu
in sources of on-farm drinking water in which one variable was significant. This variable was the
perceived level of harm from untreated drinking water (DFPHarm-Drink, negative, p < 0.05).
With respect to E. coli cfu in sources of on-farm domestic water, a more complex model fit was
found with 25 associated factors of E. coli cfu in sources of on-farm domestic water. Only four
factors were significantly associated with the level of E. coli cfu in sources of on-farm domestic
water. These four factors consisted of the following: number of years attended school
(Schooling, p < 0.01), production of fish (DFish prod, p < 0.05), ever had poultry infected with
bird flu (DPoultry AI, p < 0.01), and has heard of bird flu (DAI heard, p < 0.1). Overall, the SUR
model is significant with the p = 0.03 for the Breusch-Pagan test (Table 4.9b).
Page 159 of 340
Table 4.7a. Summary of the Best MLR Model Fittings for E. coli cfu in Each of the Four Sources of Drinking Water in the
Provinces of Thai Binh and An Giang, Vietnam
Independent variables a Rain Pipe Well River
Coef P>|t| Coef P>|t| Coef P>|t| Coef P>|t|
DGender - - -5.64 0.19 -14.64 0.11 14.12* 0.05
Age 0.76 0.17 0.19 0.32 - - - -
Schooling -2.86 0.23 -0.69 0.26 -1.96 0.17 -1.61* 0.07
Experience -0.89 0.22 0.28 0.36 - - - -
DIncome -34.52* 0.09 6.86 0.11 37.48** 0.03 8.55 0.16
Children - - 6.67 0.14 - - - -
DOff-farm job - - 1.03 0.81 -9.85 0.22 - -
Npoultry/yr - - 0.27** 0.03 - - 0.75** 0.01
Nducks/yr - - 0.08*** 0.00 - - -0.02 0.15
Npigs/yr - - 0.10 0.53 - - - -
DFish prod 50.07*** 0.00 - - -70.33*** 0.00 -
DFPR-Livestock-Man - -1.28 0.78 - - - -
DFPR-Livestock-Woman 17.93 0.24 -2.12 0.66 17.83 0.10 - -
DFPWater quality - - -0.52 0.91 - - - -
DFPHarm-Drink 20.33 0.20 6.11 0.21 - - - -
No. of significant variables 3 - 2 - 3 - 3 - Notes: AI = Avian influenza, D = Dummy, FP = Farmers’ perceptions of, FPR = Farmers’ perceived responsibility, FPS = Farmers’ perceived susceptibility to, N
= Number, a Independent Variables described in Table 4.6; (*) significant at p<0.1; (**) significant at p<0.05; (***) significant at p<0.01
Page 160 of 340
Table 4.7b. Summary of the Best MLR Model Fittings for E. coli cfu in Each of the Four Sources of Drinking Water in the
Provinces of Thai Binh and An Giang, Vietnam (cont’d)
Independent variables a Rain Pipe Well River
Coef P>|t| Coef P>|t| Coef P>|t| Coef P>|t|
DFPS-AI-Drink - - 7.27 0.11 -12.48 0.16 - -
DFPS-Diarrhea-Drink -23.75 0.18 -0.94 0.86 16.19 0.10 - -
DFPS-Coliform-Drink -19.00 0.20 0.77 0.88 - - - -
DFPS-Parasite-Drink - - -7.59 0.10 -17.41* 0.07 - -
DSatisfy-Drink -47.07*** 0.00 2.34 0.57 - - -
constant 28.34 0.43 -12.80 0.36 129.68*** 0.00 1.79 0.82
Obs 135 - 90 - 138 - 111 -
Prob > F 0.00*** - 0.01** - 0.00*** - 0.03** -
R-squared 0.24 - 0.37 - 0.39 - 0.11 -
Adj R-squared 0.17 - 0.20 - 0.35 - 0.07 -
Root MSE 66.82 - 14.16 - 44.24 - 30.15 -
No. of significant variables 3 - 2 - 3 - 3 - Notes: AI = Avian influenza, D = Dummy, FP = Farmers’ perceptions of, FPR = Farmers’ perceived responsibility, FPS = Farmers’ perceived susceptibility to, N
= Number, a Independent Variables described in Table 4.6; (*) significant at p<0.1; (**) significant at p<0.05; (***) significant at p<0.01
Page 161 of 340
Table 4.8. Summary of the MLR Best Model Fittings for E. coli in Pond Water and River
Water for Domestic Purposes in the Provinces of Thai Binh and An Giang, Vietnam
Independent variables a
Pond River
Coef P>|t| Coef P>|t|
DGender - - -338.80* 0.08
Age - - 11.06 0.11
Schooling 213.66** 0.02 - -
Experience - - - -
DIncome - - -264.47 0.10
Children - - - -
DOff-farm job - - 264.33* 0.08
Npoultry/yr -17.79 0.10 - -
Nducks/yr -1.03 0.20 - -
Npigs/yr 14.78 0.11 36.97** 0.01
DFish prod - -
DFPR-Livestock-Man - - -368.40** 0.03
DFPR-Livestock-Woman - - - -
DFPWater quality - - - -
DFPHarm-Dom - - -278.42 0.10
DFPS-AI-Dom 1,044.65* 0.05 279.94 0.16
DFPS-Diarrhea-Dom - - - -
DFPS-Coliform-Dom - - - -
DFPS-Parasite-Dom - - -482.03** 0.01
DSatisfy-Dom 1,028.56* 0.09 - -
constant 210.21 0.81 824.62** 0.03
Obs 197 - 188 -
Prob > F 0.02** - 0.00*** -
R-squared 0.08 - 0.16 -
Adj R-squared 0.05 - 0.12 -
Root MSE 3,433.80 - 990.43 -
No. of significant variables 3 - 5 - Notes: AI = Avian influenza, D = Dummy, Dom = Domestic, FP = Farmers’ perceptions of, FPR =
Farmers’ perceived responsibility, FPS = Farmers’ perceived susceptibility to, a Independent Variables
described in Table 4.6; (*) significant at p<0.1; (**) significant at p<0.05; (***) significant at p<0.01
Page 162 of 340
Table 4.9a. SUR Model Estimation of the Two Equations of E. coli cfu in Water Used for
Drinking with Cross-Equation Constraints
[NE.coli Drink]Nducks/yr - [NE.coli Dom]Nducks/yr = 0
chi2( 1) = 1.52; Prob > chi2 = 0.22
Iteration 1: tolerance = 4.18
(…results of iteration 2-5 omitted)
Iteration 6: tolerance = 0.00
Seemingly unrelated regression, iterated
Equation Obs Parms RMSE R-sq chi2 P
NE.coli Drink 243 4 14123.37 0.03 6.74 0.15
NE.coli Dom 243 26 2761.33 0.17 50.29 0.00 Notes: N = Number, Dom = Domestic
Table 4.9b. SUR Model Estimation of the Two Equations of E. coli cfu in Water Used for
Drinking with Cross-Equation Constraints (cont’d)
Correlation matrix of residuals:
NE.coli Drink NE.coli Dom
NE.coli Drink 1
NE.coli Dom 0.14 1
Breusch-Pagan test of independence: chi2(1) = 4.667, Pr = 0.03 Notes: N = Number, Dom = Domestic
Page 163 of 340
Table 4.9c. SUR Model Estimation of the Two Equations of E. coli cfu in Water Used for
Drinking with Cross-Equation Constraints (cont’d)
Coef. Std. Err z P>z [95% Conf. Interval]
NE.coli Drink
DOff-farm job 2,984.05 1,873.74 1.59 0.11 -688.41 6,656.50
Nducks/yr 0.13 0.98 0.13 0.89 -1.80 2.06
DSatisfy-Drink -1,362.50 1,841.32 -0.74 0.46 -4,971.42 2,246.42
DFPHarm-Drink -4,640.21** 2,020.48 -2.30 0.02 -8,600.27 -680.15
_cons 3,334.69 1,966.01 1.70 0.09 -518.61 7,187.99
NE.coli Dom
Age -0.44 19.03 -0.02 0.98 -37.73 36.84
DGender 155.06 493.96 0.31 0.75 -813.09 1,123.20
Schooling 195.10*** 70.11 2.78 0.01 57.68 332.52
Children -354.82 457.96 -0.77 0.44 -1,252.41 542.76
Experience -44.13 29.32 -1.50 0.13 -101.61 13.34
DFPR-Livestock-Woman -947.35 737.80 -1.28 0.20 -2,393.41 498.72
DFPR-Livestock-Man -991.06 543.12 -1.82 0.07 -2,055.56 73.44
DFPR-Fish-Woman -586.72 930.27 -0.63 0.53 -2,410.00 1,236.57
DFPR-Fish-Man 313.86 596.17 0.53 0.60 -854.60 1,482.33
DFPR-Health-Man 291.22 646.41 0.45 0.65 -975.71 1,558.15
DProvince 19.02 697.20 0.03 0.98 -1,347.46 1,385.51
DFish prod 1,210.97** 611.32 1.98 0.05 12.80 2,409.14
DRain for animals 200.38 548.55 0.37 0.72 -874.76 1,275.51
DPond for animals 3.86 451.08 0.01 0.99 -880.25 887.96
DPipe for animals -368.21 687.67 -0.54 0.59 -1,716.01 979.59
DMulti-agri activities -523.38 640.58 -0.82 0.41 -1,778.90 732.14
DPoultry AI 2297.35*** 848.80 2.71 0.01 633.73 3,960.98
Nducks/yr 0.13 0.98 0.13 0.89 -1.80 2.06
Npoultry/yr -5.63 7.98 -0.71 0.48 -21.27 10.01
DFPWater quality -714.74 523.78 -1.36 0.17 -1,741.33 311.85
DFPBarr-Cost 251.82 479.81 0.52 0.60 -688.60 1,192.24
DFPRisk rain water -1,261.25 1,153.12 -1.09 0.27 -3,521.33 998.83
DFPS-Parasite-Dom 57.65 586.64 0.10 0.92 -1,092.14 1207.43
DAI heard -2,676.05* 1,693.47 -1.58 0.11 -5995.18 643.08
DFPHarm-Dom 490.64 493.53 0.99 0.32 -476.66 1457.94
DFE-Livestock Management -358.27 548.16 -0.65 0.51 -1432.65 716.11
_cons 5,502.15*** 2,344.63 2.35 0.02 906.76 10,097.55
Notes: AI = Avian influenza, D = Dummy, Dom = Domestic, FP = Farmers’ perceptions of, FPBarr = Farmers’
perceived barriers to taking mitigating actions, FPR = Farmers’ perceived responsibility, FPS = Farmers’
perceived susceptibility to, * significant at p < 0.05; ** significant at p < 0.1
Page 164 of 340
4.4. Discussion
4.4.1. Sources and frequency of use of on-farm water for drinking and domestic purposes
Farmers had to depend on various sources of water for drinking and domestic purposes.
This result is consistent with other reports on sources of water used in rural areas in Vietnam.
These sources are pond/lake, river/stream/canal, rainwater collection, piped water, hand dug
wells, drilled wells, and bottled water (MoC and MARD, 2000; MoH, 2009b; MoH, 2009a).
Farm observations (visits) and in-depth interviews of the participating farmers supported this
result.
The percentages of farmers who had access to improved sources38 of water for drinking
were 96% in Thai Binh and 99% in An Giang. These percentages were slightly greater than the
World Bank published Vietnamese national percentage of 95% of the rural population having
access to improved water. Moreover, although farmers in the research study used multiple
sources of water for drinking, their frequencies of use were varied among sources of water. This
variation needs to be considered when developing rural water interventions in the studied areas.
In the following paragraphs, I further discuss by province the various sources of on-farm water
used.
In Thai Binh, observations of farm settings and in-depth interviews showed that rain
water was considered a precious high quality source of water. Compared to other sources of
water, stored rainwater was the most used source by farmers for drinking water. Although the
number of Thai Binh farms that reported using drilled well water for drinking was highest
compared to other sources of water, drilled well water was second to stored rainwater in terms of
38 “improved sources of water” refers to pipe water into dwelling, pipe water to yard/plot, public pipe or standpipe,
tube well or borehole, protected dug well, and rainwater (JMP, 2014)
Page 165 of 340
frequency of use. Therefore, rural water interventions should prioritize stored rainwater and
drilled well water as most Thai Binh farmers frequently used these sources. Moreover, they did
not use dug wells very often, and farmers indicated that it was becoming harder and harder to dig
a well that had good quality water. Even for drilled wells, the depth of the wells should be about
70 meters to reach underground water, which is quite deep. The other reason participating
farmers did not favour dug wells was the low quality of water found in them (e.g., bad smelling
water was perceived as being low quality water) if they did not dig deeper wells. Farmers were
also afraid that dug wells were likely to be contaminated with pollution from livestock waste and
pesticides.
In An Giang, on the other hand, the most frequently used source for drinking water was
river/canal water. This reflects the more comprehensive network of river branches and canals in
the Mekong River Delta in An Giang compared to the Red River Delta in Thai Binh (Nguyen,
2007). Specifically, the number of SSI farmers in An Giang who had piped water for drinking
was four times greater versus those in Thai Binh. However, these SSI farmers rarely used water
services due to their economic barriers (i.e. not being able to pay for connecting to the piped
water network or for the monthly fees), as well as the availability of other sources for on-farm
drinking water. Another study also showed limited use of piped water by rural households in the
Mekong Delta due to similar reasons (Wilbers et al., 2014) . In contrast, dug wells and drilled
wells were still common in An Giang with more than two-thirds of the participating research
study farmers reporting having access to and use of both dug and drilled wells for drinking
water. However, the mean frequencies of use for drinking water from either dug or drilled wells
were “rarely”. Consequently, water interventions in An Giang may need to focus on solutions for
using river/canal water for drinking in the short and medium term. In the long term, more
Page 166 of 340
research is needed to explore how to improve accessibility and affordability of piped water
supply networks for rural farmers.
Water sources used for domestic purposes were similar to those used for drinking water
in both provinces. However, the frequencies of using these sources of water for domestic
purposes were different between the two provinces. For example, drilled well water was the most
frequently used water source for domestic purposes in Thai Binh. Meanwhile, in An Giang,
farmers used river water the most for their domestic purposes. In-depth interviews with farmers
revealed that when the same water source was used for both drinking and domestic purposes,
treatment methods such as boiling water and/or flocculating water were used for drinking water
but not for domestic use water. In addition, given the limited supply of on-farm water, it is
understandable that rain water, piped water, and bottled water were not selected by the
participating farmers as the most frequently used sources of water for domestic purposes. In Thai
Binh, the common and most frequently used source of water for domestic purposes was different
from that used for drinking water (i.e., stored rainwater and drilled well water respectively).
Meanwhile, in An Giang, water from rivers/canals was the most common and frequently used
source of water for both drinking and domestic purposes.
The differences between the two provinces in frequencies of use of all water sources for
drinking and domestic purposes were significant (except for bottle and piped water for drinking),
which confirmed the first hypothesis of this research. The hypothesis indicates that frequencies
of use of water sources for drinking and domestic purposes are different between farmers in Thai
Binh and An Giang. For that reason, such differences need to be considered in designing water
related interventions. For example, the mean frequencies of using water from rain, pipes, and
bottles by farmers of both Thai Binh and An Giang ranged between rarely used and frequently
Page 167 of 340
used. On the other hand, Thai Binh participating farmers did not use water from ponds, lakes,
rivers and canals for drinking purposes, while participating farmers in An Giang did use these
drinking water sources although rarely (e.g., for pond water) and occasionally (e.g., for river
water). Participating farmers in An Giang did not use water from hand-dug wells and drilled
wells for drinking water. However, participating farmers in Thai Binh rarely and occasionally
used water from hand-dug wells and drilled wells respectively.
4.4.2. Low quality of on-farm water for both drinking and domestic purposes
Compared to the basic indicators (i.e., pH, turbidity, and E. coli) of drinking water quality
set by the WHO and the Vietnamese MOH, the quality of water on participating farms was low.
The water samples from rain, pipe, wells, and rivers (flocculated) had acceptable levels of pH
(i.e., pH within 6.5 to 8.5). Water samples from bottles had a high pH mean level (i.e., mean pH
of 9.4). Bottled water, stored rainwater, and pipe water had turbidity levels that met the national
standard level of turbidity for drinking water (i.e., maximum turbidity of 2 NTUs). However,
well water and flocculated river water exceeded the maximum turbidity level for drinking water
(i.e., 3.7 NTUs and 7.6 NTUs respectively). Levels of pH and turbidity in water have been
shown to influence the microbial quality of water (WHO and OECD, 2003c). The WHO
recommends that E. coli be used as an indicator organism to assess the microbial safety of rural
water and drinking water should be free from E. coli (WHO and OECD, 2003b). E. coli is a well-
defined member of the Enterobacteriaceae family. There are four main reasons to use E. coli as
an indicator organism. First, it is not practical to conduct isolations of all pathogens. Second,
pathogens from human and animal faeces pose the greatest danger to public health and detecting
faecal contamination in drinking water is important for public health safety. Third, methods of
assessing E. coli are simpler, more available, and more affordable when compared to those of
Page 168 of 340
other microbial agents including potentially pathogenic agents. Fourth, E. coli is a more specific
indicator of faecal contamination compared to other members of the total coliform group. The
presence of E. coli in water indicates contamination with faeces and possible other pathogens in
the water.
Now, let me focus my discussion on the E. coli levels in sources of on-farm drinking
water. Almost 60% of the water samples were greater than the maximum level of E. coli cfu
indicated in the drinking water quality guidelines of Vietnam, Canada, and the WHO (FPTCW,
2013; MoH, 2009b; WHO, 2012). Piped water and bottled water supplied by either local
governments or private companies were authorized and monitored by the local health authorities.
Water from pipes and bottles were expected to meet the national standards of drinking water
quality. However, in this research, it was clear that piped and bottled water used on at least one-
third of the SSI farms did not meet the standards of E. coli (i.e., 30.95% and 41.03% of the
bottled and pipe water samples respectively had E. coli). Another study of piped water in rural
areas of the Mekong Delta also found that E. coli was commonly detected in piped water
(Wilbers et al., 2014). None of the results of the water tests was in conformity with the standards
of drinking water quality set by both the WHO and the Vietnamese MoH. The level of E. coli cfu
in bottled water and piped water used by the participating farmers for drinking can be
classified/scored as an intermediate risk with low action priority (i.e., 1-10 E. coli cfu/100ml)
(WHO, 2012). Meanwhile, the level of E. coli cfu in stored rainwater, well water, and river water
used by the participating farmers can be classified/scored as being a high risk. These
classifications should be combined with corresponding sanitation inspection scores to completely
analyze the risk of water on-farm (WHO, 2012).
Page 169 of 340
In terms of the quality of on-farm water for domestic purposes, the mean pH level of on-
farm domestic water met the national pH standards for domestic water. However, the mean
turbidity and E. coli levels in the sources of on-farm domestic water were much greater than
those in the national standards. Moreover, the mean level of E. coli cfu in river water used for
domestic purposes in Thai Binh was almost four times greater than that in An Giang. This
difference may be one of the reasons why farmers in Thai Binh use river water frequently for
domestic purposes and almost never for drinking water; meanwhile, farmers in An Giang
appeared to use river water frequently for both drinking water and for domestic purposes. The
microbial qualities of water for both drinking water and domestic purposes were significantly
lower than the national standards of water quality and were significantly different between the
two provinces. This difference and the low microbial quality of water on-farm needs to be
considered when designing interventions and/or further research about on-farm water.
Concerning the microbial quality of on-farm drinking water tested by the CHWs/CVWs
using ColiplateTM tests, the percentage of water samples with the presence of E. coli tested by
CHWs/CVWs were greater than those results done by the laboratories. Although, the ColiplateTM
test results produced by CHWs/CVWs supported the two laboratories’ results of E. coli cfu in
water, only the results of water quality tested using laboratory protocol were used to confirm the
second hypothesis of this research. Some factors that need to be considered when explaining the
differences between the level of E. coli measured by CHWs/CVWs and the laboratories are
discussed in the following paragraphs.
First, ColiplateTM tests were highly sensitive and could indicate the presence of as small
as only one E. coli in water samples. Second, the mean time to testing was much lower for the
ColiplateTM test than for the laboratory tests (i.e., 30 to 45 minutes and 3 to 6 hours respectively).
Page 170 of 340
Third, CHWs/CVWs incubated water samples using simple tools available on farms (e.g.,
Styrofoam box, light bulbs, and insulating materials such as used clothing) and the incubations
depended on the weather; this explained the actual longer time it took to incubate water samples
in Thai Binh (colder weather) compared to An Giang (warmer weather). Fourth, technical
persons at the laboratories were to dilute a water sample before incubating it if they thought that
there might be a high level of E. coli cfu in the water sample (based on observing the water
sample and reading notes taken during the water sample collection).
However, we did not train CHWs/CVWs to use the dilution technique due to limited time
and resources for the survey as well as possible contamination of the water samples.
Furthermore, the dilution technique was not addressed in the guidelines of the ColiplateTM testing
procedure. Thus, introducing the dilution technique may enhance the limits of ColiplateTM tests
in the future. Finally, although a ColiplateTM test used a simple procedure and CHWs/CVWs
were trained, most of the CHWs/CVWs had never tested water quality before this research.
Meanwhile, technicians from the national laboratories were highly skilled and experienced in
water testing. Given this, the results of water tests by the laboratory were used for further
analysis (i.e., exploring associated factors of on-farm water quality).
The results of water tests using ColiplateTM test kits were also used for confirming the
water tests using Watercheck (i.e., water tests performed by farmers in an extended research, not
reported in this research), and for studying possible solutions (in another research) to address on-
farm water public health (e.g., ownership, participation, awareness, and behaviour change of
farmers, and community stakeholders).
Page 171 of 340
4.4.3. Associated factors of on-farm water quality
Farm visits during the field study of this research indicated that on-farm water issues
appeared complex and there was a potential correlation between the quality of both drinking and
domestic water within each farm. Farmers used one source of water for multiple purposes and/or
multiple sources of water for one purpose. Furthermore, the dependent variables of interest in
this chapter (e.g., the number of E. coli cfu in on-farm water) have a continuous nature and the
hypothesis of the research in this chapter addressed these associations. Therefore, regression
analysis was an appropriate technique to explore factors that are associated with the level of E.
coli contamination in on farm water for drinking.
Farmers’ income (DIncome)39 was associated with number of E. coli cfu in two out of the
four sources of on-farm water for drinking. Although there are no other studies that look at
exactly the same associations to compare with this interesting finding, some researchers did
indicate the possibly important role economic factors play in rural public health (Duong et al.,
2004). When controlling for other variables in each fitting of the model, two variables that were
negatively associated with the level of E. coli cfu in stored rainwater were income and
satisfaction with the quality of the current sources of drinking water. This indicates that, among
farmers who used stored rainwater for drinking, a poor farmer (i.e., coded as 1 if income is less
than the poverty line) was likely to have a better microbial quality of stored rainwater (i.e., fewer
E. coli cfu counts) than a not poor farmer (i.e., coded as 0 if income is greater than the poverty
line). In addition, if a farmer was satisfied with the quality of stored rainwater (i.e., moving from
39 The variable of income per person per year was dichotomized using the cut-off income level for poverty set by the
Vietnamese government (i.e., less than 4.8 million VND per person per year – equivalent to $244 CAD per person
per year) at the exchange rate of $ CAD = 19,700 on 5th December 2013. An income per person per year below the
cut-off income level (i.e., a poor farm) was coded as 1 and 0 if otherwise (i.e., a not poor farm). I should have named
this variable as DPoverty instead of DIncome to avoid confusion.
Page 172 of 340
code 0 if not satisfied to code 1 if satisfied), the farmer was likely to have better microbial
quality of stored rainwater. From my farm visits, this indication makes sense as I saw poor
farmers often depended on and cared for stored rainwater more than not poor farmers (who may
have more sources of water for drinking other than stored rainwater) did. Furthermore, farmers
were confident in assessing the quality of their stored rainwater and in indicating their levels of
satisfactions with stored rainwater. Having fish production on farms was also a positively
associated factor for the level of E. coli contamination in rain and well water for drinking. This
association may be explained by the reported inappropriate practice of dumping livestock waste
directly into fishponds.
The number of chickens and ducks on farms were additional positive significantly
associated factors of the level of E. coli contamination in piped water. This means farms with
more chickens and ducks were likely to have worse piped water (i.e., higher E. coli cfu in pipe
water) compared to those farms that have fewer chicken and ducks. Incomes on farm per year
and perceptions of the level of susceptibility to parasites from using untreated drinking water
were also associated with the increased E. coli contamination of well water for drinking
purposes. This means that farmers with a poor yearly income who thought that using untreated
drinking water gave them susceptibility to parasites tended to have higher E. coli contamination
in their well water used for drinking. Farmers who attended more years of school tended to have
better on-farm water quality (i.e., lower E. coli cfu in river water for drinking) compared to
farmers who had attended school for fewer years. On the other hand, female-headed farms with
more numbers of chickens had higher levels of E. coli cfu contamination in river water compared
to male-headed farms with fewer chickens.
Page 173 of 340
Yet, farms headed by men which had male farmers primarily responsible for livestock
production tended to have more E. coli contamination of river water used for domestic purposes
compared to farms that were headed by women which had female farmers primarily responsible
for livestock production. Furthermore, having more pigs and an occupation other than farming
and perceived increased susceptibility to parasites from using untreated river water for domestic
purposes tended to increase the E. coli contamination of river water used for domestic purposes.
Farmers who had more years of attending school, perceived susceptibility to HPAI from using
untreated domestic water, and who were satisfied with their current sources of domestic water
tended to have higher levels of E. coli contamination in pond water used for domestic purposes.
The SUR model with cross-equation constraints showed an improvement in the analysis
of the relationship between independents variables and E. coli cfu in drinking and domestic
water sources in the two equations. E. coli cfu in drinking water sources with cross-equation
constraints had the same number of IVs as the associated factors as the model fit without cross-
equation constraints. Farmers having other occupations than farming were likely to have more E.
coli cfu in drinking water sources. Meanwhile, farmers who thought there was harm in using
untreated sources of drinking water were likely to have less E. coli contaminated water compared
to those who did not think there is harm in using untreated drinking water. Improvements of the
SUR model with cross-equation constraints in fitting the levels of E. coli cfu in domestic water
support my expectation of the complexity and a potential correlation between the quality of on
farm water for drinking and for domestic uses. Years of attending school, having fish production,
and past experience with AI in poultry were positive significant associated factors. Meanwhile,
the variables of male farmers primarily responsible for livestock production and having heard of
AI were negative significant associated factors to the levels of E. coli cfu in domestic water.
Page 174 of 340
The results of the SUR model have confirmed the last part of my research hypothesis that
the microbial quality of on-farm water is associated with farmers’ demographics, socioeconomic
status, and their perceptions of risk factors for WRZD transmission. I believe this association
may happen because farmers with different socioeconomic status may have different access to
improved sources of water (e.g. using river water instead of pipe water for drinking).
Furthermore, they may have different levels of resources to protect their sources of water, and
properly treat, distribute and store their water (e.g., insufficient land areas to separate livestock
housing from drinking water sources). Therefore, different demographic and socioeconomic
status may lead to different quality of water on farms. Furthermore, if a farmer does not perceive
that there are risk factors for WRZD transmission, the farmer may not see a need to take actions
to improve the quality of on farm water. The results of the SUR model imply the importance of
considering farmers’ demographics, socioeconomic status, and perceptions to improve the
microbial quality of water on farms.
My research used a cross-sectional design in which on-farm water was assessed at one
point in time. However, water quality may vary over time. For example, in developing countries,
the fecal contamination of drinking water appears to be greater during the wet season (Costyla et
al., 2015). Therefore, in addition to the associated factors of on-farm water quality discussed
above, future studies may be needed to assess contributions of seasonality to water quality on
SSI farms.
4.4.4. Rural water and its regulatory framework on SSI farms in Vietnam
Considering the results and discussion of the multiple sources of low quality on-farm
water and the insights of possible associated factors for on-farm water quality presented in the
previous sections, this section will discuss issues of on-farm water public health and its
Page 175 of 340
corresponding policy framework in Vietnam. The section will provide a brief overview and
discussion of the regulatory framework of rural water supply and sanitation globally and in
Vietnam, focusing on its impact on humans, animals, and their environment at the SSI farming
level.
Globally, drinking water coverage in 2011 is 1% above the Millennium Development
Goal (MDG)’s target. However, it is estimated that 768 million people still relied on unimproved
drinking water resources and the world does not meet the MDG’s sanitation target by half a
billion people (WHO and UNICEF, 2013). Target 7.C of the MDG aims at reducing by 50% the
proportion of people who do not have access to safe drinking water and basic sanitation (UN,
2013). The proportion of the Vietnamese population using improved sources of drinking water
and improved sanitation is estimated to be 76 - 90% and 50 – 75% respectively (WHO and
UNICEF, 2013). However, rural farmers have limited awareness of the water and environmental
sanitation issue with 70% of them using contaminated water and having no latrines (MoC, 2000).
Only 9% of the rural population uses piped-water on their premises compared to 58% of the
urban population.
Regulatory framework and the National Rural Clean Water Supply and Sanitation Strategy up to
2020 (the strategy)
The evolution of water-related policies in Vietnam began decades ago. The long-term
overall framework for the development of rural water supply and sanitation in Vietnam is driven
by a number of policies in the National Rural Clean Water Supply and Sanitation Strategy up to
2020 (MoC, 2000). However, there has been limited progress and many challenges still exist
with room for improvement, especially relating to engaging multiple stakeholders and
empowering the rural community to ensure a clean water supply and adequate sanitation. Many
Page 176 of 340
other countries also face similar challenges and it is suggested that future water-related policies
should recognize and support the role of multiple stakeholders (Lockwood and Smits, 2011;
Ravnborg and Jensen, 2012)
The strategy is one of the key national policies that directly influence SSI famers.
Although this strategy involved multiple stakeholders in its design and development stages, the
level of stakeholder consultation for the strategy phase was not comprehensive and the
participation of grassroots stakeholders (i.e., farmers) was limited. Moreover, although livestock
production is part of the strategic goals, veterinary services were not engaged, nor were linkages
between water and livestock production explicitly addressed in the strategy. At the same time,
research studies have shown that SSI farmers and their animals are at high risk of WRZD and
that water-related policies have direct important impacts on human and animal health, and on
sustainable agriculture.
Therefore, the regulatory framework and the strategy need to be reviewed using trans-
disciplinary approaches (e.g., an EcoHealth approach) to define clearer roles and responsibilities
and to be more inclusive of stakeholders. The strategy should address rural water public health
issues more explicitly to not only enhance safety for the rural water supply, but to maximize use
of the limited water resources to improve the health of humans, animals, and their environment.
It has been recommended that the Vietnamese government consider values of incorporating an
EcoHealth approach to revising, formulating, and implementing government policies addressing
SSI farming and water public health (Hall and Dinh, 2009; Hall and Le, 2013; Gleeson et al.,
2011; Le and Hall, 2011b; Le and Hall, 2011a). It is a basic human right for each person to have
access to safe water, sanitation, and hygiene at home. Moreover, water, sanitation, and hygiene
Page 177 of 340
should be sustainable, and inequalities in access should be eliminated (WHO and UNICEF,
2013).
4.5. Conclusions
This chapter has presented objective assessments of on-farm water quality as well as
associated factors of the quality of on-farm water for drinking and domestic use in the provinces
of Thai Binh and An Giang in Vietnam. The research study revealed that the quality of on-farm
water for both drinking and domestic use was low with most of the water samples contaminated
with E. coli and they did not met the quality standards for drinking and domestic water. Although
testing of E. coli in water is recommended for establishing if a water source is recently
contaminated with faeces, such faecal contamination of water is often intermittent and may not
be revealed in one assessment of a single sample. Therefore, in addition to having simple
affordable tests of E. coli in water, it is important to build capacity for inspecting and improving
household water sanitation, and on-farm water public health management. By doing so, rural
farmers will not only be able to afford and conduct the tests themselves, but also to know when it
is appropriate to take the test and what to do with the test results.
The modeling of the relationships between the objectively assessed quality of on-farm
water and the subjectively assessed on farm water quality by farmers gave some idea of the
factors that were relevant to further explore in the next chapter of this dissertation which
addresses possible associated factors for farmers’ mitigating strategies for WRZD transmission
in the context of SSI farming. The research results have also indicated areas for improving rural
water supply and sanitation and on-farm water and health management through updates, as well
as revising the water regulatory framework/policies on SSI farms in Vietnam. The results of the
research also showed the need to consider rural water testing and training in improving on-farm
Page 178 of 340
water public health and that the microbial assessment of rural water quality can help improve
public health protection (Allen et al., 2010).
Page 179 of 340
References
Allen, M. J., Payment, P. R., and Clancy, J. L. (2010). Rapid microbial methods can improve
public health protection. Journal (American Water Works Association), 102, 44-51.
Bach, A. H. (2007). Equivalence in the translation of Vietnamses cultural words in the book
"Wandering through Vietnamese culture" by Huu Ngoc. Vietnam College of Foreign
Languages (VHUN), Hanoi.
Boere, G. C., Galbraith, C. A., and Stroud, D. A. (2006). Waterbirds Around the World The
Stationery Office, UK.
Brown, J. D., Swayne, D. E., Cooper, R. J., Burns, R. E., and Stallknecht, D. E. (2007).
Persistence of H5 and H7 avian influenza viruses in water. Avian Dis, 51, 285-289.
Bui, X. M. (2010). Duck farming systems and avian influenza in the Mekong Delta of Vietnam.
In Rome: FAO.
Carr, R. M., Blumenthal, U. J., and Mara, D. D. (2004). Guidelines for the safe use of
wastewater in agriculture: revisiting WHO guidelines. Water Sci Technol, 50, 31-38.
Costa, T. P., Brown, J. D., Howerth, E. W., and Stallknecht, D. E. (2011). Variation in viral
shedding patterns between different wild bird species infected experimentally with low-
pathogenicity avian influenza viruses that originated from wild birds. Avian Pathology,
40, 119-124.
Costyla, C., Bain, R., Cronk, R., and Bartram, J. (2015). Seasonal variation of faecal
contamination in drinking water sources in developing countries: A systematic review.
Science of the Total Environment, 541, 333-343.
Page 180 of 340
Dalsgaard, A. (2001). Health Aspects of the Reuse of Wastewater in Agriculture and
Aquaculture in Vietnam. In Wastewater Reuse in Agriculture in Vietnam: Water
Management, Environment and Human Health Aspects International Water Management
Institute.
Dinh, N. P., Long H.P, Tien, N. T. K., Hien N.T, Phong L.H, and et al (2004). Risk factors for
human infection with avian influenza A H5N1, Vietnam. Emerging Infectious Diseases,
12, 1841-1847.
Domanska, B. K., Minta, Z., Smietanka, K., Sylvie, M., and Berg, A. T. (2010). H5N1 High
Pathogenicity Avian Influenza Virus Survival in Different Types of Water. Avian
Diseases Digest, 5, e180-e181.
Dooho, I. R., Martin, W., and Stryhn, H. (2009). Veterinary Epidemiologic Research. (2nd ed.)
VER, Inc..
Duong, D. V., Binns, C. W., and Lee, A. H. (2004). Utilization of delivery services at the
primary health care level in rural Vietnam. Social Science and Medicine, 59, 2585-2595.
Ford, T. E. and Colwell, R. (1996). A Global Decline in Microbiological Safety of Water: A Call
for Action. American Academy of Microbiolog.
FPTCW (2013). Guideline Technical Document: Escherichia coli. Ottawa, Ontario: Federal-
Provincial-Territorial Committee on Drinking Water of the Federal-Provincial-Territorial
Committee on Health and the Environment, Health Canada.
Gilbert, M., Xiao, X., Pfeiffer, D. U., Epprecht, M., Boles, S., Czarnecki, C. et al. (2008a).
Mapping H5N1 highly pathogenic avian influenza risk in Southeast Asia. Proceedings of
the National Academy of Sciences of the United States of America, 105, 4769-4774.
Page 181 of 340
Gilbert, M., Xiao, X., Pfeiffer, D. U., Epprecht, M., Boles, S., Czarnecki, C. et al. (2008b).
Mapping H5N1 highly pathogenic avian influenza risk in Southeast Asia. Proceedings of
the National Academy of Sciences, 105, 4769-4774.
Gleeson, L. J., Nguyen, D. C., Do, P. H., and Nguyen, T. K. O. (2011). Joint Government -
United Nations Programme to Fight Highly Pathogenic Avian Influenza (HPAI): Final
Evaluation Report.
Guan, T. Y. and Holley, R. A. (2003). Pathogen survival in swine manure environments and
transmission of human enteric illness – a review. Journal of Environmental Quality, 32
(2), 392.
Hall, D. C. and Dinh, X. T. (2009). Perceived constraints on small scale poultry in Vietnam.
Asian Livestock.
Hall, D. C. and Le, Q. B. (2013). Impact on human, animal, and ecosystem health and food
security from restructuring of livestock production farming practices in Vietnam. The
First International One Health Congress, Human Health, Animal Health, the Environment
and Global survival, Melbourne, Australia Feb 2011 [On-line]. Available:
http://download.springer.com/static/pdf/638/art%253A10.1007%252Fs10393-010-0376-
0.pdf?originUrl=http%3A%2F%2Flink.springer.com%2Farticle%2F10.1007%2Fs10393-
010-0376-
0andtoken2=exp=1447890688~acl=%2Fstatic%2Fpdf%2F638%2Fart%25253A10.1007
%25252Fs10393-010-0376-
0.pdf%3ForiginUrl%3Dhttp%253A%252F%252Flink.springer.com%252Farticle%252F1
0.1007%252Fs10393-010-0376-
0*~hmac=80c4fbd920ab0106f79b30de3faab0550896d5a6a1a5f310256c3bf160ef14d9
Page 182 of 340
Henning, K. A., Henning, J., Morton, J., Long, N. T., Ha, N. T., and Meers, J. (2009). Farm- and
flock-level risk factors associated with Highly Pathogenic Avian Influenza outbreaks on
small holder duck and chicken farms in the Mekong Delta of Viet Nam. Preventive
Veterinary Medicine, 91, 179-188.
Hooda, P. S., Edwards, A. C., Anderson, H. A., and Miller, A. (2000). A review of water quality
concerns in livestock farming areas. Sci Total Environ, 250, 147-67.
Hubbard, R. K., Newton, G. L., and Hill, G. M. (2004). Water quality and the grazing animal.
Journal of Animal Science, 82, E255-E263.
Huynh, T. T. T., Aarnink, A. J. A., Drucker, A., and Verstegen, M. W. A. (2006). Pig production
in Cambodia, Laos, Philippines, and Vietnam: a review. Asian Journal of Agriculture and
Development, 3, 69-90.
IDRC (2012). Ecohealth Research in Practice. (Insight and Innovation in International
Development ed.) (vols. Innovative Applications of an Ecosystem Approache to Health)
International Development Centre (IDRC); Springer.
Ip, H., Flint, P., Franson, J. C., Dusek, R., Derksen, D., Gill, R., Jr. et al. (2008). Prevalence of
Influenza A viruses in wild migratory birds in Alaska: Patterns of variation in detection at
a crossroads of intercontinental flyways. Virol J, 5, 1-10.
JMP (2014). Improved and unimproved water and sanitation facilities. WHO/UNICEF [On-line].
Available: http://www.wssinfo.org/definitions-methods/watsan-categories/
Le, Q. B. and Hall, D. C. (2011a). Stakeholder cooperation and ecohealth to control Highly
Pathogenic Avian Influenza (HPAI) in Vietnam. Canadian Public Health Association
Annual Conference 2011, Montreal Quebec, 2011 [On-line]. Available:
http://resources.cpha.ca/CPHA/Conf/Data/2011/A11-270ae.pdf
Page 183 of 340
Le, Q. B. and Hall, D. C. (2011b). Using an ecohealth approach to livestock production:
community surveillance of newly emerging infectious diseases (nEIDs), International
Meeting on Emerging Diseases and Surveillance (IMED 2011.
Lin, S., Fend, M. L., Hu, R. G., Liu, R. Y., Wei, M. Y., and Jang, C. (2010). Characteristics of
Nitrogen Cycling in Farm Systems in a Small Watershed of Three Gorges Reservoir
Area, China. Environmental Science [On-line].
Lockwood, H. and Smits, S. (2011). Supporting Rural Water Supply. Practical Action Publishing
Ltd.
Lye, D. J. (2002). Health Risk Associated with Consumption of Untreaed Water from Household
Roof Catchment Systems. JAWRA Journal of the American Water Resources Association,
38, 1301-1306.
Markwell, D. D. and Shortridge, K. F. (1982). Possible waterborne transmission and
maintenance of influenza viruses in domestic ducks. Applied and Environmental
Microbiology, 43, 110-115.
MoC (2000). Vietnam National Rural Clean Water Supply and Sanitation Strategy up to 2020
Ministry of Agriculture and Rural Development- Minsistry of Construction.
MoC and MARD (2000). National Rural Clean Water Supply and Sanitation Strategy up to the
year 2020 Hanoi, Vietnam: Ministry of Construction and Ministry of Agricultrue and
Rural Development of Vietnam.
MoH (2009a). National technical regulation on domestic water quality (QCVN 02: 2009/BYT).
Hanoi: Department of Preventive Medicine and Environment, Ministry of Health,
Socialist Republic of Vietnam.
Page 184 of 340
MoH (2009b). National technical regulation on drinking water quality (QCVN 01: 2009/BYT).
Hanoi: Department of Preventive Medicine and Environment, Ministry of Health,
Socialist Republic of Vietnam.
Morris, R. S. and Jackson, R. (2005). Epidemiology of H5N1 Avian Influenza in Asia and
Implications for Regional Control Rome, Italy: Food and Agriculture Organisation of the
UN.
Nazir, J., Haumacher, R., Ike, A., Stumpf, P., B+¦hm, R., and Marschang, R. E. (2010). Long-
Term Study on Tenacity of Avian Influenza Viruses in Water (Distilled Water, Normal
Saline, and Surface Water) at Different Temperatures. Avian Diseases Digest, 5, e174-
e175.
Nguyen, H. N. (2007). Flooding in Mekong River Delta, Vietnam Hanoi: UNDP.
Nguyen, H. V., Vuong, T. A., Pham, D. P., and Vu, V. T. (2011). Safe use of wastewater in
agriculture and aquaculture (Rep. No. 2). The National Centre of Competence in
Research (NCCR) North-South.
Nguyen, T. V., Kitajima, M., Nguyen Vo, M. H., Matsubara, K., Takizawa, S., Katayama, H. et
al. (2008). Bacterial contamination of raw vegetables, vegetable-related water and river
water in Ho Chi Minh City, Vietnam. Water Science and Technology, 58, 2403-2411.
Pfeiffer, J., Pantin, J. M., To, T. L., Nguyen, T., and Suarez, D. L. (2009). Phylogenetic and
biological characterization of highly pathogenic H5N1 avian influenza viruses (Vietnam
2005) in chickens and ducks. Virus Research, 142, 108-120.
Phan, Q. M. (2010). Epidemiological Studies of Highly Pathogenic Avian Influenza in Vietnam.
Massey University.
Page 185 of 340
Rapport, D. J., Howard, J., Maffi, L., and Mitchell, B. (2006). Avian Influenza and the
Environment: An Ecohealth Perspective.
Ravnborg, H. M. and Jensen, K. M. (2012). The water governance challenge: the discrepancy
between what is and what should be. Water Science and Technology: Water Supply, 12,
799-809.
Roche, B., Lebarbenchon, C., Gauthier-Clerc, M., Chang, C. M., Thomas, F., Renaud, F. et al.
(2009). Water-borne transmission drives avian influenza dynamics in wild birds: The
case of the 2005-2006 epidemics in the Camargue area. Infection, Genetics and
Evolution, 9, 800-805.
Sims, L. D., Domenech, J., Benigno, C., Kahn, S., Kamata, A., Lubroth, J. et al. (2005). Origin
and evolution of highly pathogenic H5N1 avian influenza in Asia. Veterinary Record,
157, 159-164.
Soares, R. J., Ortiz-Pelaez, A., Lai, T. K. L., Dinh, Q., Otte, J., and Pfeiffer, D. (2010).
Associations between attributes of live poultry trade and HPAI H5N1 outbreaks: a
descriptive and network analysis study in northern Vietnam. BMC Veterinary Research,
6, 10.
Stallknecht, D. E., Shane, S. M., Kearney, M. T., and Zwank, P. J. (1990). Persistence of Avian
Influenza Viruses in Water. Avian diseases, 34, 406-411.
Toews, D. W. (2009). Food, Global Environmental Change and Health: EcoHealth to the
Rescue? Mcgill Journal of Medicine, 12, 85-89.
Tran, H. P., Adams, J., Jeffery, J. A. L., Nguyen, Y. T., Vu, N. S., Kutcher, S. C. et al. (2010).
Householder perspectives and preferences on water storage and use, with reference to
dengue, in the Mekong Delta, southern Vietnam. International Health, 2, 136-142.
Page 186 of 340
Trang, D. T., Hien, B. T. T., Molbak, K., Cam, P. D., and Dalsgaard, A. (2007a). Epidemiology
and aetiology of diarrhoeal diseases in adults engaged in wastewater-fed agriculture and
aquaculture in Hanoi, Vietnam. Tropical Medicine and International Health, 12, 23-33.
Trang, D. T., Molbak, K., Cam, P. D., and Dalsgaard, A. (2007b). Incidence of and risk factors
for skin ailments among farmers working with wastewater-fed agriculture in Hanoi,
Vietnam. Transactions of the Royal Society of Tropical Medicine and Hygiene, 101, 502-
510.
Trinh, T. L. (2001). Wastewater and its Treatment System in the Southern Part of Vietnam. In
International Water Management Institute: Proceedings of a Workshop held in Hanoi,
Vietnam 14 March 2001 Southern Institute of Water Resources Research.
Trinh, T. L. and Nguyen, P. L. (2009). Assessment of surface water quality by water quality
index (WQI) at the Cai Sao canal, An Giang province, Vietnam. Livestock Research for
Rural Development [On-line]. Available: http://www.lrrd.org/lrrd23/7/lan23151.htm
UN (2013). Goal 7: Ensure Environmental Sustainability - Target 7.C. United Nations Goals
[On-line]. Available: http://www.un.org/millenniumgoals/environ.shtml
Vu, T. T. H. (2001). Treatment of Domestic Wastewater and its Reuse in Farm Irrigation in the
Red River Delta. In Wastewater Reuse in Agriculture in Vietnam: Water Management,
Environment and Human Health Aspects Center for Irrigation and Water Supply,
Vietnam Institute for Water Resources Research.
Ward, M. P., Maftei, D. N., Apostu, C. L., and Suru, A. R. (2009). Association Between
Outbreaks of Highly Pathogenic Avian Influenza Subtype H5N1 and Migratory
Waterfowl (Family Anatidae) Populations. Zoonoses and Public Health, 56, 1-9.
Page 187 of 340
Webster, R. G., Peiris, M., Chen, H., and Guan, Y. (2006). H5N1 outbreaks and enzootic
influenza. Emerg Infect Dis [serial on the Internet].
WHO (2004). Waterborne Zoonoses: Identification, Causes, and Control. IWA Publishing.
WHO (2005). Avian Influenza A (H5N1) Infection in Humans. New England Journal of
Medicine, 353, 1374-1385.
WHO (2007a). Review of latest available evidence on potential transmission of avian influenza
(H5N1) through water and sewage and ways to reduce the risks to human
health (Rep. No. WHO/SDE/WSH/06.1). Geneva: World Health Organisation: Water,Sanitation
and Health Public Health and Environment.
WHO (2007b). Review of latest available evidence on potential transmission of avian influenza
(H5N1) through water, sewage and ways to reduce the risks to human health (Rep. No.
WHO/SDE/WSH/06.1). Geneva: World Health Organisation: Water, Sanitation and
Health Public Health and Environment.
WHO (2011a). Annex 1: Supporting documentation to the Guidelines. In (pp. 443-448). World
Health Organization.
WHO (2011b). Guidelines for drinking-water quality. (Fourth ed.) World Health Organisation.
WHO (2011c). Microbial aspects. In Guidelines for drinking-water quality (pp. 117-153). World
Health Organization.
WHO (2011d). Water and Public Health. World Health Organization [On-line]. Available:
http://www.who.int/water_sanitation_health/dwq/S01.pdf
WHO (2012). Rapid assessment of drinking water quality - A handbook for implementation.
World Health Organization.
Page 188 of 340
WHO and OECD (2003a). Assessing Microbial Safety of Drinking Water: Improving
Approaches and Methods. IWA Publishing.
WHO and OECD (2003b). Assessing Microbial Safety of Drinking Water: Improving
Approaches and Methods. IWA Publishing.
WHO and OECD (2003c). Assessing Microbial Safety of Drinking Water: Improving
Approaches and Methods. IWA Publishing.
WHO and UNICEF (2013). Progress on sanitation and drinking-water - 2013 update France:
World Health Organization and UNICEF 2013.
Wilbers, G. J., Sebesvari, Z., and Renaud, F. G. (2014). Piped-Water Supplies in Rural Areas of
the Mekong Delta, Vietnam: Water Quality and Household Perceptions. Water
(20734441), 6, 2175-2194.
Yajiam, A. and Kurokura, H. (2008). Microbial risk assessment of livestock-integrated
aquaculture and fish handling in Vietnam. Fisheries Science, 74, 1062-1068.
Yajima, A. and Kurokura, H. (2008). Microbial risk assessment of livestock-integrated
aquaculture and fish handling in Vietnam. Fisheries Science, 74, 1062-1068.
Page 189 of 340
Chapter Five: Factors Associated with Small-Scale Integrated (SSI) Farmers' Engagement
in Mitigating Strategies to Reduce Transmission of Water-Related Zoonotic Diseases
(WRZD) in the Provinces of Thai Binh and An Giang in Vietnam
5.1. Introduction
5.1.1. Rationale
The previous chapters of this dissertation have provided the background for the role of
SSI farms in the provinces of Thai Binh and An Giang in Vietnam. These chapters included the
following: 1) a profile of SSI farms; 2) the perceptions of SSI farmers toward risk factors for the
transmission of water-related zoonotic diseases (WRZD); and 3) the link between water quality
and public health among SSI farms. This background laid the foundation for this chapter in
which I discuss factors that are associated with engaging SSI farmers in implementing mitigating
strategies to reduce transmitting WRZD. In the following paragraphs, I explain some of the key
terms used in this chapter, and the conceptual and theoretical models of the research (see
Appendix H). I also provide a summary of the literature review for this chapter at the end of this
section (i.e., Section 5.1.1). Figure 5.1 additionally presents the structure of this chapter.
There are various definitions for the term mitigation. For example, in the context of a
pandemic influenza, the WHO (2008) has classified mitigation as actions or measures
undertaken beforehand to moderate the effect of an event or to decrease its impact on society and
the environment. Next, strategies and tactics are defined as the “what” and the “how”
components respectively of priority setting in health. Specifically, strategies are defined as action
plans (i.e., the “what”) to achieve health conditions that warrant the most attention and tactics
(i.e., the “how”) are defined as specific actions used to carry out these strategies (Bloom et al.,
2013). The WHO has recommended some key mitigating strategies as useful starting points for
reducing the transmission of WRZD. These mitigating strategies include livestock management,
Page 190 of 340
source water protection, water storage, water treatment and distribution, and point of
use/household40 as recommendations (WHO, 2004)
In my research, I am interested in the types of mitigating strategies SSI farmers engage in
to reduce the transmission of WRZD and the factors associated with their engaging in such
mitigating strategies. I define a mitigating strategy to reduce the transmission of WRZD as a plan
that includes one or more activities a SSI farmer could choose to reduce transmitting WRZD
(Figure 5.1 and Section 5.2.1). In my research, an example of such a mitigating strategy is good
livestock management, which may include actions or tactics (e.g., treat animal waste at source)
that SSI farmers can choose to reduce the transmission of WRZD (Carr and Bartram, 2004;
WHO, 2004). Thus, in this chapter, I refer to WHO’s recommended key mitigating strategies for
reducing the transmission of WRZD as “the mitigating strategies”. Moreover, an SSI farmer is
considered to be engaging in a mitigating strategy to reduce transmission of WRZD if that
farmer takes certain actions included in that mitigating strategy (Section 5.2.1 and Table 5.1).
The WHO defines WRZD as diseases in humans that are spread from animals and are
related to water such as Highly Pathogenic Avian Influenza (HPAI), some parasites, or diseases
caused by a range of bacteria (e.g., pathogenic strains of Escherichia coli such as Shiga toxin-
producing E. coli) (Carr and Bartram, 2004; WHO, 2004). In Vietnam, 40% of farm animals
have diarrhea (Nguyen, 2004), and 31.3% and 14% of diarrheic calves carried at least one
fimbrial gene and genes for Shiga toxin respectively (Nguyen, Vo, and Vu-Khac, 2011). The
WHO has outlined some criteria for determining whether a potential pathogen is a WRZD. These
criteria include three elements: 1) the pathogen must spend part of its life cycle within one or
40 Point of Use/Household, as a mitigating strategy to reduce the transmission of WRZD, refers to actions relating to
conditions of water, sanitation and hygiene (personal, domestic, and food) which can affect the transmission of
WRZD at point of water use (e.g., handwashing) (WHO, 2004)
Page 191 of 340
more animal species; 2) it is probable or conceivable that some life stage of the pathogen will
enter water; and 3) transmission of the pathogen from animals to humans must be through a
water- related route (Moe, 2004).
Figure 5.1 below shows a pictorial representation of the structure of Chapter Five.
Figure 5.1. Outline of Chapter Five
Chapter Five:
Presents factors associated with farmers'
engagement in mitigating strategies of
water- related zoonotic disease (WRZD)
transmission in the Thai Binh and An Giang
provinces of Vietnam
Act
ion
1
Clustered actions into mitigating strategies
SSI farmers’
perceptions of
risk factors
for WRZD
transmission
(from Chapter
Three)
Act
ion
4
Act
ion
..
. Act
ion
n
Act
ion
2
Act
ion
3
Research question 6:
Do SSI farmers engage
in mitigating strategies
to reduce transmission
of WRZD?
Research question 7:
What are factors associated with SSI
farmers’ engagement in mitigating
strategies to reduce transmission of
WRZD?
Livestock
Management
Point of use/
Household
Methods
Water
Storage,
Treatment,
Distribution
Source
Water
Protection
Socioecono
mic factors
of SSI
farms
(from
Chapter
Two)
Water quality
among SSI
farms
(from
Chapter
Four)
Page 192 of 340
The conceptual model of my research is that SSI farmers make decisions, including
management decisions related to animal and human health, based on their socioeconomic
conditions and their impressions of the water quality around them. The theoretical model for my
research is adapted from the theory of planned behaviour and the health beliefs model (Becker,
2012; Payne et al., 2002; Rosenstock et al., 1988a). This adapted theoretical model for my
research indicates that the belief of SSI farmers in threats to their health and/or wellbeing,
combined with their beliefs in both the benefits and barriers to an action, together with their
socioeconomic status are contributing factors of whether they will engage in mitigating strategies
to reduce the transmission of WRZD (see Appendix A). This model is prefaced on the
observation that an individual’s experiences and actions are reflected in his or her perceptions;
therefore, perceptions can be associated with people’s health seeking behaviours (Brewer et al.,
2007; WHO, 2012). For example, the perceptions of individuals about the threat of diseases,
expectations, cues to actions41, and their socioeconomic status may be associated with their
health behaviours (Becker, 1974; Rosenstock et al., 1988b). Therefore, understanding the
associations between the perceptions of SSI farmers towards WRZD transmission, their
socioeconomic status, and their behaviours could be useful for research into and interventions
needed to improve their engagement in mitigating strategies to reduce WRZD transmission.
A review of the related literature indicated that although SSI farmers have a basic
knowledge about proper hygienic practices, they usually employ ineffective practices for hygiene
measures (e.g., hand washing without soap). Misconceptions about risk factors and/or a lack of
knowledge of the cause-effect relationships, habits, and/or not ranking health risks highly have
41 A cue to action is “the stimulus needed to trigger the decision making process to accept a recommended health
action.” Examples of possible cues to actions are chest pains, mass media campaigns, advice from others, a reminder
postcard from a physician, illness of family member or friend, etc.) (BUMC, 2013).
Page 193 of 340
been proposed as links to ineffective practices of hygiene measures (Herbst et al., 2009; Keraita
et al., 2010; Rechenburg and Herbst, 2006). Even though SSI farmers are aware of the health
risks and potential health benefits to be gained by implementing mitigating actions, other factors
such as overall economic impact on work efficiency were thought to be associated with their
assessment of mitigating actions (Keraita et al., 2010). Some studies in Africa (Mekoya et al.,
2008; Adesina and Baidu, 1995; Abdul et al., 2014; Negatu and Parikh, 1999) and South East
Asia (Van Mele and Cuc, 2001; Cramb, 2005; Le et al., 2014a) have explored potential factors
that are associated with rural farmers’ choices of adopting a farming technique or an adaptive
measure for climate change. However, these studies did not address SSI farming and the
transmission of WRZD. Yet, these studies provided some ideas for further research of factors
(e.g., incomes) that may be associated with SSI farmers’ choices. In summary, I have not found
any studies that examine factors associated with SSI farmers’ engagement in mitigating
strategies to reduce WRZD transmission. An understanding of these factors may help
inform/bring scientific evidence for policy advocacy and assist with education/knowledge
relating to SSI farming, rural water quality, and public health. Therefore, in my research, I
explored factors that may be associated with SSI farmers’ engagement in mitigating strategies to
reduce transmission of WRZD.
5.1.2. Research questions and hypothesis
5.1.2.1. Research questions
This chapter addresses the sixth and seventh questions of my research, which are: 1) Do
SSI farmers in the provinces of An Giang and Thai Binh engage in mitigating strategies to
reduce the transmission of WRZD? and 2) What factors are associated with SSI farmers’
engagement in mitigating strategies to reduce the transmission of WRZD?”
Page 194 of 340
5.1.2.2 Research hypothesis
I hypothesized that in the provinces of An Giang and Thai Binh, SSI farmers’
socioeconomic status and their perceptions of risk factors for WRZD transmission are associated
with their engagement in mitigating strategies to reduce transmission of WRZD.
5.2. Methods
This section describes the methods used to identify SSI farmers’ engagement in the
mitigating strategies to reduce transmission of WRZD and the methods used to explore potential
factors that are associated with each of these mitigating strategies. The theoretical model used for
my research guided the selection of these potential factors. The other details of the methods (e.g.,
sample size, studied areas, and questionnaires) used to collect data relating to these potential
factors were presented in Chapters Two, Three, and Four.
5.2.1. Methods used to identify SSI farmers’ engagement in mitigating strategies to reduce
the transmission of WRZD
To identify SSI farmers’ engagement in the mitigating strategies defined in section 5.2.1,
I first collected data pertaining to farmers’ specific choice of actions that were clustered under
the key WHO recommended mitigating strategies (Table 5.1). Then, I created thresholds of
actions to generate variables that reflect farmers’ engagement in each strategy. In these
thresholds, an SSI farmer is considered engaging in a mitigating strategy to reduce transmission
of WRZD if the farmer takes certain actions included in that mitigating strategy (See section
5.3.1.1 for more detailed information). I used questionnaires to collect data from SSI farmers
about their choice of actions to mitigate WRZD transmission. Table 5.1 provides detailed
descriptions of the mitigating strategies to decrease WRZD transmission and the corresponding
actions which were adapted and selected based on the WHO’s recommendations (WHO, 2004).
Page 195 of 340
The field data collection focused on the actions clustered under the mitigating strategies. I
asked SSI farmers to indicate their choice of actions (Table 5.1) to reduce transmission of
WRZD. Their answers were either yes or no indicating whether they did or did not use each of
the actions that were included in each of the mitigating strategies. Given the collected data on the
corresponding actions, I used the thresholds mentioned above to create variables that reflected
SSI farmers’ engagement (yes or no) in mitigating strategies to decrease WRZD transmission.
These binary variables helped to inform the second part of the hypothesis. An SSI farmer’s
engagement in a mitigating strategy was coded as “1” and “0” otherwise (See section 5.3.1.1 for
more detailed information). I used descriptive statistics to summarize these binary variables.
Section 5.2.2 shows the methods used to explore factors that are associated with these binary
variables (i.e., factors that are associated with SSI farmers’ engagement in mitigating strategies
of WRZD transmission).
Page 196 of 340
Table 5.1. Small-Scale Integrated (SSI) Farmers’ Mitigating Strategies and Their
Corresponding Actions in Reducing Risk Factors for Water-Related Zoonotic Disease
(WRZD) Transmission (*)
Mitigating strategies Corresponding actions (i.e., tactics)
Livestock management (LM)
strategy
- Vaccinate domestic animals
- Ensure hygienic rearing conditions for
animals
- Treat animal waste at source
- Exclude animals from catchment basin
- Stop using antibiotics as growth promoters
Source water protection (SWP)
strategy
- Treat wastewater
- Treat human and animal waste at source
- Use buffer zones
- Protect wells/springs/groundwater
- Cover drinking water reservoirs
- Repair leaking sanitation systems
- Treat human/animal waste prior to use as
fertilizers
Water storage, treatment, and
distribution (WSTD) strategy
- Change household water treatment
technologies
- Change/use disinfectants
- Repair household water distribution
- Repair leaking household sewers
Point of use/household (POU)
strategy
- Ensure adequate household sanitation
- Invest in improved water supply
- Protect house hold water sources and food
supply from animals
- Wear protective equipment (e.g., mouth
mask) (*) Adapted from the WHO’s recommendations (WHO, 2004).
5.2.2. Methods used to explore factors that are associated with SSI farmers’ engagement in
mitigating strategies
I used a univariate inferential statistical analysis technique to explore factors that are
associated with SSI farmers’ engagement in the mitigating strategies and to test the research
hypothesis. Stata 13 was used to assist in the data analysis (Stata, 2013). More specifically, in my
Page 197 of 340
research, a probit regression technique (hereafter referred to as probit analysis) was used to
examine specific factors (explanatory variables) that may be associated with SSI farmers’
engaging in mitigating strategies to reduce WRZD transmission (outcome variables). Probit
analysis is commonly used for analyzing binary outcome variables and exploring how each
factor affects the probability of an outcome event occurring (Greene, 2008b; Long and Freese,
2014).
The univariate probit analysis model is specified accordingly as follows: assuming “i” is
an engagement in a mitigating strategy for each farmer, the equation describing a latent variable
that corresponds to SSI farmers’ engagement in mitigating strategies is written as:
Yi* = β0+ βi Xi + ɛi ; with i = 1,2,3,.. n (i)
Yi = {0, 1} (ii)
Yi* is an unobserved variable representing the latent utility or propensity of the Yi under
consideration. Yi is an observed dummy variable reflecting SSI farmers’ engagement in a
mitigating strategy. Yi equals 1 if Yi* >0 (i.e., a farmer used a mitigating strategy) and Yi equals
0 otherwise.
Xi is a vector of observed explanatory variables (e.g., perceptions and water quality) believed to
be relevant to the consideration of the ith equation.
βi is a vector of unknown coefficients to be estimated.
ɛi is an error term, which is normally distributed with mean 0 and variance 1.
Dependent (outcome) variables included SSI farmers’ engagement in mitigating
strategies to reduce transmission of WRZD coded as dichotomous variables. The mitigating
strategies explored in my research (see section 5.1.1 and Table 5.2) were coded as “0” if SSI
farmers did not engage in a mitigating strategy and “1” otherwise. A set of explanatory variables
Page 198 of 340
of interest was specified based on the conceptual and theoretical models and literature review of
my research. These explanatory variables included characteristics of SSI farmers and their farms
(e.g., age, income, and livestock production) and SSI farmers’ perceptions (e.g., perceived
barriers to actions). Descriptive statistics were used to summarize the explanatory variables used
in the univariate probit analysis.
Probit analysis can show rates of changes in magnitudes and levels of significance of the
associations between the explanatory variables and an outcome variable. In my research,
estimates of individual probit coefficients of the explanatory variables as well as the
corresponding marginal probability effects (i.e., marginal effects) at the means of explanatory
variables were used to report the results of the modeling factors that are associated with SSI
farmers’ engagement in mitigating strategies. A probit coefficient (βi) is a change in the “z-
score” of Yi* given a one unit change in Xi, holding all other explanatory variables constant at
their means. Meanwhile, a marginal effect is the change in Pr (Yi = 1) for a one unit change in Xi,
given that all other explanatory variables are held constant at their means (see the Discussion
section for a more detailed interpretation of probit coefficients and marginal effects). The
significance level was set a priori at p-value of 0.1 or less which is a common significance level
(Hall et al., 2003; JAE, 2014) .
5.3. Results
This section presents the results of my research in this chapter including 1) SSI farmers’
engagement in mitigating strategies for reducing WRZD transmission and their corresponding
actions, and 2) factors that are associated with SSI farmers’ engagement in mitigating strategies
to decrease WRZD transmission. Appendix G provides an additional summary of information
that was presented in Chapter Two and in Chapter Three and was used as part of the data
Page 199 of 340
analysis in this chapter. The additional information includes summaries of the socio-
demographic characteristics of SSI farmers, SSI farmers’ perceptions of risk factors for WRZD
transmission, and SSI farmers’ livestock production.
5.3.1. SSI farmers’ actions and engagement in mitigating strategies to reduce WRZD
transmission
5.3.1.1 Livestock management (LM) actions
There were five actions used to indicate SSI farmers’ engagement in livestock
management strategies to mitigate WRZD transmission (Table 5.1, Figure 5.2). These actions
included vaccinating domestic animals, treating animal waste at source, discontinuing using
antibiotics as a growth promoter, ensuring hygienic rearing conditions for animals, and excluding
animals from catchment basins. The results showed significant differences between two
provinces in terms of the numbers of SSI farmers employing these actions. The majority of the
SSI farmers in the province of Thai Binh engaged in all five livestock management actions to
mitigate the transmission of WRZD, with vaccination being the most common and excluding
animals from the catchment basin being the least common. In the province of An Giang,
vaccination was the only action used by a majority of the SSI farmers to mitigate the
transmission of WRZD. Each of the remaining actions was employed by a third or less of the SSI
farmers, with exclusion of animals from the catchment basin again being the least common..
5.3.1.2. Source water protection (SWP) actions
SSI farmers in two provinces indicated seven actions that they took to protect sources of
water on their farms (SWP). These actions included 1) treating wastewater, 2) using a buffer
zone, 3) covering the drinking water reservoir, 4) treating animal and human waste prior to its
use as fertilizer, 5) treating animal and human waste at source, 6) protecting well/spring/ground
water, and 7) repairing a leaking sewage system. These actions reflected SSI farmers’
Page 200 of 340
engagement in a SWP strategy to mitigate WRZD transmission. However, significantly fewer
SSI farmers in An Giang took these actions compared to those in Thai Binh. The greatest number
of SSI farmers in both provinces selected “Covering the drinking water reservoir” as a key risk
mitigating action (Figure 5.3).
5.3.1.3.Water storage treatment and distribution (WSTD) actions
SSI farmers in both provinces indicated that they took four key actions regarding storage,
treatment, and distribution of water to mitigate WRZD transmission (Figure 5.4). These actions
included 1) spending money on household treatment technology, 2) repairing household water
storage, 3) changing/using disinfectants, and 4) repairing household sewers. There was
significantly (p < 0.01) fewer SSI farmers in An Giang compared to those in Thai Binh who took
each of these actions. Among the four actions, the greatest number of SSI farmers in An Giang
chose was repairing household water storage. Meanwhile, repairing the household sewer instead
of the household water storage was the most common action taken by SSI farmers in Thai Binh.
Page 201 of 340
Figure 5.2. Livestock Management Actions
Figure 5.3. Source Water Protection Actions
0
100
200
300
Thai Binh An Giang
Vaccination of domestic animals (p<0.01) Ensure hygienic rearing conditions for animals (p<0.01)
Treat animal waste at source (p<0.01) Exclude animals from catchment basin (p<0.01)
Stop using antibiotics (p<0.01)
Num
ber
of fa
rme
rs
Graphs by Provinces
Livestock management actions0
100
200
300
Thai Binh An Giang
Treat wastewater (p< 0.01) Treat human/animal waste at source (p<0.01)
Use buffer zones (p<0.01) Protect well/springs/ground water (p<0.01)
Cover drinking water reservoirs (p<0.01) Repair leaking in sewage system (p<0.01)
Treat animal/human wastes prior to use as fertilizers (p=0.01)
Num
ber o
f far
mer
s
Graphs by Provinces
Source water protection actions
Page 202 of 340
Figure 5.4. Water Storage, Treatment, and Distribution Actions
5.3.1.4. Point of use/household (POU) actions
Four key actions at a POU level used to mitigate transmission of WRZD were the
following: having adequate household sanitation, protecting water sources and food sources for
animals, investing in water supply improvement, and wearing protective equipment. The results
showed that there were significantly greater numbers of SSI farmers in Thai Binh compared to
An Giang who selected each of these four actions. Ensuring adequate household sanitation was
the most common POU action in both provinces. However, the actions that were taken by the
least number of SSI farmers were different between the two provinces – wearing protective
equipment (in the province of An Giang) and protecting water and food sources for animals (in
the province of Thai Binh) (See Figure 5.5).
050
100
150
200
250
Thai Binh An Giang
Spend money in household treatment technology (p<0.01) Change/use disinfectants (p<0.01)
Repair household water storage (p<0.01) Repair household sewer (p<0.01)
Nu
mbe
r o
f fa
rmer
s
Graphs by Provinces
Water storage distribution and treatment actions
Page 203 of 340
Figure 5.5. Point of Use/Household Actions
5.3.1.5. SSI farmers’ engagement in mitigating strategies to reduce WRZD transmission
Sections 5.3.1.2-4 provided detailed information of the actions that SSI farmers took to
reduce the transmission of WRZD. In this section, these actions were clustered under the four
broad mitigating strategies to reduce the transmission of WRZD using a certain thresholds. The
World Health Organization recommends these mitigating strategies to control and prevent
WRZD in the context of rural areas (WHO, 2004).
Table 5.2 below presents a description of thresholds used to cluster farmers’ actions
under the four mitigating strategies and create variables that reflect SSI farmers’ engagement in
these mitigating strategies to reduce transmission of WRZD. An SSI farmer was considered to
engage in a mitigating strategy if that farmer took certain actions. For example, an SSI farmer
was considered “engaged in LM strategy” if that farmer took three or more of the five LM
actions and “not engaged in LM strategy” otherwise. The thresholds (i.e., number of actions)
were defined based on the goodness-of-fit of the models and the number of observations. The
table also provides summary statistics for SSI farmers’ engagement in the mitigating strategies.
050
100
150
200
250
Thai Binh An Giang
Adequate household sanitation (p<0.01) Invest in improving water supply (p<0.01)
Protect water source and food source for animals (p<0.01) Wear protective and filter water use (p<0.01)
Num
ber o
f far
mer
s
Graphs by Provinces
Point of use/household actions
Page 204 of 340
Table 5.2. Summary Statistics of Variables Indicating SSI Farmers’ Engagement in
Mitigating Strategies for WRZD Transmission
Variable name Description - SSI farmers' engagement in:
Count ( =
“1”)
% ( =
“1”)
DFE-Livestock
Management
Livestock management as a mitigating
strategy (LM strategy) with “1” if SSI
farmers took three or more of the five
actions included in the LM strategy and
“0” otherwise. 329 55
DFE-Water
Protection
Source water protection as a mitigating
strategy (SWP strategy) with “1” if SSI
farmers took four or more of the seven
actions included in the SWP strategy and
“0” otherwise. 299 50
DFE-Point of
Use
Point of use/household method as a
mitigating strategy (POU strategy) with
“1” if SSI farmers took two or more of the
four actions included in the POU strategy
and “0” otherwise. 312 52
DFE-Water
Distribution
Water storage, treatment, and distribution
as a mitigating strategy (WSTD strategy)
with “1” if SSI farmers took three or more
of the four actions included in the WSTD
strategy and “0” otherwise. 325 54 Notes: D = Dummy, FE = Farmers’ engagement in a mitigating strategy.
5.3.2. Factors associated with SSI farmers’ engagement in the mitigating strategies to reduce
transmission of WRZD
5.3.2.1. Results of probit analyses of SSI farmers’ engagement in mitigating strategies to reduce
WRZD transmission
Table 5.3a-c shows the estimated probit coefficients of the factors that are associated with
each of the four mitigating strategies that SSI farmers engaged in to mitigate WRZD
transmission. Overall, the probit analyses revealed that the models for the four key mitigating
strategies were significant at p<0.1 with McFadden's R2 from 0.4440 to 0.6214, and the
percentages of correct predictions from 83% to 89% (Table 5.5). The models for SSI farmers’
engaging mitigating strategies to reduce WRZD transmission had 14,13, 12, and 14 significant
factors in LM (i.e., DFE-Livestock Management variable), SWP (i.e., DFE-Water Protection
Page 205 of 340
variable), POU (i.e., DFE-Point of Use variable), and SDT (i.e., DFE-Water Distribution
variable) respectively (Table 5.3a-c). “SSI farmers’ income” (i.e., DIncome variable) appeared to
be negatively significantly associated with SSI farmers’ engagement in all four mitigating
strategies while “Female SSI farmers primarily responsible for family health” (i.e., DFPR-
Health-Woman variable) was positively significantly associated with three mitigating strategies.
5.3.2.2. Marginal probability effects of the factors that are associated with SSI farmers’
engagement in mitigating strategies to reduce WRZD transmission
In this section, more intuitive results of probit analyses are presented (i.e., a marginal
probability effect, or a rate of change of the probability of each of the dependent variables as per
a unit change in an explanatory variable at mean values of other explanatory variables). A
marginal probability effect shows not only levels of significance and signs but also the changes
(in percentage) of the magnitudes of the association between a factor and SSI farmers’
engagement in a mitigating strategy holding other explanatory variables constant at their means.
Table 5.5a shows that the changes in the magnitudes of the associations ranged from as
low as negative 1% to as great as positive 44% for the probability of a per unit change in each of
the factors that are associated with a mitigating strategy, while holding other factors at their
means. Having an additional chicken increases by 1% the probability of a farmer engaging in
good livestock management as a mitigating strategy to decrease WRZD transmission (Table
5.5b). Meanwhile, SSI farmers who considered “worry of WRZD transmission” a trigger for
their actions had an increase of 44%in the probability of engaging in the LM mitigating strategy
(Table 5.5b).
Page 206 of 340
Table 5.3a. Maximum Likelihood Estimates of Factors that are Associated with Small-Scale Integrated (SSI) Farmers’
Engagement in Mitigating Strategies of Water Related Zoonotic Diseases (WRZD) Transmission Using Probit Analyses
Variables DFE-Livestock Management DFE-Water Protection DFE-Water Distribution DFE-Point of Use
Estimates a Estimates Estimates Estimates
_cons -2.3361* b -2.5403* -3.5636* -1.5002*
(0.6613) c (0.6032) (0.6888) (0.4377)
DProvince - - - 0.9727*
- - - (0.2585)
Schooling 0.0696* 0.0979* 0.0679* -
(0.0362) (0.0334) (0.0350) -
Experience 0.0232 0.0231* 0.0237 0.0100
(0.0175) (0.0145) (0.0154) (0.0137)
DChildren 0.1499 - 0.0016 0.2206*
(0.1141) - (0.1111) (0.0940)
DFPR-Health-Man 0.5871* 0.1698 0.3444 -
(0.3298) (0.2885) (0.3020) -
DFPR-Health-Woman 1.3567* 1.0419* 1.4946* -
(0.3761) (0.3267) (0.3856) -
DFPR-Livestock-Man - - - 0.1190
- - - (0.1938)
DFPR-Livestock-Woman - - - 0.5163*
- - - (0.2963)
DIncome -0.4408* -0.4861* -0.8940* -0.7309*
(0.3121) (0.3087) (0.3080) (0.2653)
Npoultry/yr 0.0252* 0.0032 0.0040 -
(0.0079) (0.0036) (0.0039) -
Notes: D = Dummy, Dom = Domestic, FE = Farmers’ engagement in, FPS = Farmers’ perceived susceptibility to, FPT = Farmers’ perceived trigger of taking
mitigating actions, N = Number, a Estimated values of individual coefficients of explanatory variables – each coefficient represents a change in the “z-score” of
Yi* given a one unit change in Xi; b An asterisk (*) indicates a statistical significance at the level of p<0.10; c Values in brackets are standard errors
Page 207 of 340
Table 5.3b. Maximum Likelihood Estimates of Factors that are Associated with Small-Scale Integrated (SSI) Farmers’
Engagement in Mitigating Strategies of Water Related Zoonotic Diseases (WRZD) Transmission Using Probit Analyses
(cont’d.)
Variables DFE-Livestock Management DFE-Water Protection DFE-Water Distribution DFE-Point of Use
Estimates a Estimates Estimates Estimates
Npigs/yr 0.0435* b 0.0223* 0.0249* 0.0074
(0.0106) c (0.0068) (0.0087) (0.0063)
Ncattle/yr -0.0378 -0.0004 - -
(0.0419) (0.0074) - -
DFPS-AI-Dom 0.5902* 0.0034 -0.2308 -0.1185
(0.2706) (0.2177) (0.2299) (0.1947)
DFPS-AI-Wastewater 0.4498 0.8853* 0.8076* 0.7799*
(0.3640) (0.3265) (0.3343) (0.2936)
DAI Severe -0.3098 -0.3062 0.7775* -
(0.4752) (0.4256) (0.4296) -
DFPSafe-Drink -1.0939* b - - -0.7434*
(0.4534) c - - (0.3696)
DFPBarr-Cost -1.0431* -0.6013* -0.7807* -0.7700*
(0.2527) (0.2138) (0.2261) (0.1938)
DFPBarr-Not know 0.1992 0.3410 0.7508* -
(0.2547) (0.2226) (0.2397) -
DFPBarr-Peer -0.0301 -0.9428* 0.1689 -
(0.5323) (0.4213) (0.4978) -
DFPBen-Action -0.3946 -0.0880 0.5566* -
(0.3321) (0.3006) (0.3207) -
DFPE-Livestock Management 0.9814* 0.5686* 0.7246* 0.8607*
(0.2707) (0.2119) (0.2361) (0.2971)
Notes: AI = Avian Influenza, D = Dummy, Dom = Domestic, FE = Farmers’ engagement in, FPE = Farmers’ perceived self-efficacy in, FPS = Farmers’
perceived susceptibility to, FPT = Farmers’ perceived trigger of taking mitigating actions, N = Number, a Estimated values of individual coefficients of
explanatory variables – each coefficient represents a change in the “z-score” of Yi* given a one unit change in Xi; b An asterisk (*) indicates a statistical
significance at the level of p<0.10; c Values in brackets are standard errors
Page 208 of 340
Table 5.3c. Maximum Likelihood Estimates of Factors that are Associated with SSI Farmers’ Engagement in Mitigating
Strategies of Water Related Zoonotic Diseases (WRZD) Transmission using Probit Analyses (cont’d.)
Variables DFE-Livestock Management DFE-Water Protection DFE-Point of Use DFE-Water Distribution
Estimates a Estimates Estimates Estimates
DFPE-Water Protection - - -0.4701* -
- - (0.2788) -
DFPT-Health Advice 0.4292* b 0.4429* 0.6408* 0.4929*
(0.2336) c (0.2138) (0.1857) (0.2251)
DFPT-Loss Income -0.1377 0.1484 - 0.5378*
(0.2957) (0.2475) - (0.2785)
DFPT-Peer 0.8522* 1.0932* 0.3115 0.8266*
(0.3426) (0.2992) (0.2322) (0.3285)
DFPT-Worry 1.3433* 0.6463* 0.6737* 0.2645
(0.3008) (0.2332) (0.2120) (0.2484)
LogE.coli Drinkd - 0.0551 -0.0003 0.0919
- (0.0475) (0.0446) (0.0531)
LogE.coli Dom -0.0874* - -0.0710* -
(0.0477) - (0.0417) -
Turbidity Drink - -0.0016 - 0.0018
- (0.0092) - (0.0090)
Turbidity Domestic - -0.0049* - -0.0048*
- (0.0029) - 0.0030
Model chi-squaree 313.4700* 196.4400* 237.1500* 215.1200*
McFadden's R 2 f 0.6214 0.4440 0.4537 0.4964
% correct predictions 0.8862 0.8349 0.8307 0.8380
Notes: D = Dummy, Dom = Domestic, FE = Farmers’ engagement in, FPE = Farmers’ perceived self-efficacy in, FPT = Farmers’ perceived trigger of taking
mitigating actions, a Estimated values of individual coefficients of explanatory variables– each coefficient represents a change in the “z-score” of Yi* given a
one unit change in Xi; b An asterisk (*) indicates a statistical significance at level of p<0.10; c Values in brackets are standard errors; d E.coli Drink and E.coli
Dom were transformed into LogE.coli Drink and LogE.coli Dom using a natural log transformation; e Test of the overall model fit with null hypothesis indicating
that all individual coefficients of the model equal zero; f This is a pseudo R squared statistic calculated by 1 minus the ratio of the log-likelihood of the full model
(i.e., with all explanatory variables) to the log-likelihood of the intercept model (i.e., without explanatory variables)
Page 209 of 340
Table 5.4. Predicted Frequency of Mitigating Strategies
DFE-Livestock
Management
DFE-Water
Protection DFE-Point of Use DFE-Water Distribution
Classified D ~D Total D ~D Total D ~D Total D ~D Total
+ 187 19 206 150 28 178 170 34 204 163 26 189
- 23 140 163 25 118 143 30 144 174 28 104 132
Total 210 159 369 175 146 321 200 178 378 191 130 321
Sensitivity 89.05% 85.71% 85.00% 85.34%
Specificity 88.05% 80.82% 80.90% 80.00%
Positive predictive value 90.78% 84.27% 83.33% 86.24%
Negative predictive value 85.89% 82.52% 82.76% 78.79%
False + rate for true ~D 11.95% 19.18% 19.10% 20.00%
False - rate for true D 10.95% 14.29% 15.00% 14.66%
False + rate for classified + 9.22% 15.73% 16.67% 13.76%
False - rate for classified - 14.11% 17.48% 17.24% 21.21%
Correctly classified 88.62% 83.49% 83.07% 83.18% Classified + if predicted Pr (D) >= .5
True D defined as DFE-Livestock Management! = 0; DFE-Water Protection = 0; DFE-Point of Use = 0; DFE-Water Distribution = 0.
Notes: D = Dummy, FE = Farmers’ engagement in
Page 210 of 340
Table 5.5a. Marginal Effects (at Means) of Factors Associated with SSI Farmers’
Engagement in Mitigating Strategies for WRZD Transmission Using Probit Analyses
Variables
DFE-Livestock
Management
DFE-Water
Protection DFE-Point of Use
DFE-Water
Distribution
Marginal effect a Marginal effect Marginal effect Marginal effect
DProvince - - 0.3860* -
- - 0.1025 -
Schooling 0.0226* 0.0383* - 0.0229*
(0.0119) b (0.0131) - (0.0118)
Experience 0.0075 0.0090 0.0040 0.0080
(0.0058) (0.0057) (0.0054) (0.0052)
DChildren 0.0486 - 0.0875* 0.0005
(0.0371) - (0.0373) (0.0374)
DFPR-Health-
Man 0.1961* 0.0675 - 0.1248
(0.0995) (0.1141) - (0.1043)
DFPR-Health-
Woman 0.3371* 0.3546* - 0.3588*
(0.0668) (0.0887) - (0.0606)
DFPR-Livestock-
Man - - 0.0474 -
- - (0.0771) -
DFPR-Livestock-
Woman - - 0.1984* -
- - (0.1076) -
DIncome -0.1430* -0.1903* -0.2901* -0.3007*
(0.1028) (0.1214) (0.1056) (0.1066)
Npoultry/yr 0.0082* 0.0013 - 0.0013
(0.0024) (0.0014) - (0.0013)
Npigs/yr 0.0141* 0.0087* 0.0029 0.0084*
(0.0031) (0.0026) (0.0025) (0.0027)
Ncattle/yr -0.0123 -0.0001 - -
(0.0138) (0.0029) - -
DFPS-AI-Dom 0.1914* 0.0013 -0.0470 -0.0776
(0.0882) (0.0852) (0.0772) (0.0774)
DFPS-AI-
Wastewater 0.1459 0.3465* 0.3095* 0.2716*
(0.1197) (0.1281) (0.1167) (0.1145)
DAI Severe -0.1005 -0.1198 - 0.2615*
(0.1549) (0.1666) - (0.1431)
Notes: AI = Avian Influenza, D = Dummy, Dom = Domestic, FE = Farmers’ engagement in, FPR = Farmers’
perceived main responsibility, FPS = Farmers’ perceived susceptibility to, FPT = Farmers’ perceived trigger of
taking mitigating actions, N = Number, a Marginal probabilities of changes in dependent variables as per changes in
explanatory variables at mean values of other explanatory variables; b Values in brackets are standard errors; (*)
indicates a statistical significance at the level of p<0.10
Page 211 of 340
Table 5.5b. Marginal Effects (at Means) of Factors that are Associated with Small-Scale
Integrated (SSI) Farmers’ Engagement in Mitigating Strategies of WRZD Transmission
Using Probit Analyses
Variables
DFE-Livestock
Management
DFE-Water
Protection
DFE-Point of
Use
DFE-Water
Distribution
Marginal effect a Marginal effect Marginal effect Marginal effect
DFPSafe-Drink -0.3547* - -0.2950* -
(0.1543) - (0.1470) -
DFPBarr-Cost -0.3383 -0.2353* -0.3056* -0.2626*
(0.0826) (0.0837) (0.0771) (0.0763)
DFPBarr-Not know 0.0646 0.1335 - 0.2525*
(0.0831) (0.0874) - (0.0806)
DFPBarr-Peer -0.0098 -0.3690* - 0.0568
(0.1727) (0.1650) - (0.1671)
DFPBen-Action -0.1280 -0.0345 - 0.1872*
(0.1097) (0.1177) - (0.1069)
DFPE-Livestock
Management 0.3183* 0.2225* 0.3416* 0.2437*
(0.0867) (0.0830) (0.1176) (0.0787)
DFPE-Water Protection - - -0.1866 -
- - (0.1106) -
DFPT-Health Advice 0.1392* 0.1733* 0.2543* 0.1658*
(0.0767) (0.0838) (0.0737) (0.0770)
DFPT-Loss Income -0.0446 0.0581 - 0.1809*
(0.0956) (0.0969) - (0.0932)
DFPT-Peer 0.2764* 0.4278* 0.1236 0.2780*
(0.1120) (0.1175) (0.0922) (0.1094)
DFPT-Worry 0.4356* 0.2530* 0.2673* 0.0890
(0.1026) (0.0915) (0.0842) (0.0836)
LogE.coli Drink - 0.0216 -0.0001 0.0309*
- (0.0186) (0.0177) (0.0178)
LogE.coli Dom -0.0284* - -0.0282* -
(0.0157) - (0.0166) -
Turbidity Drink - -0.0006 - 0.0006
- (0.0036) - (0.0030)
Turbidity Domestic - -0.0019* - -0.0016*
- (0.0011) - (0.0010)
Notes: D = Dummy, Dom = Domestic, FE = Farmers’ engagement in, FP = Farmers’ perceptions of, FPBarr =
Farmers’ perceived barriers to taking mitigating actions, FPBen = Farmers’ perceived benefit of, FPS = Farmers’
perceived susceptibility to, FPT = Farmers’ perceived trigger of taking mitigating actions, a Marginal probabilities of
changes in dependent variables as per changes in explanatory variables at mean values of other explanatory
variables; b Values in brackets are standard errors; (*) indicates a statistical significance at the level of p<0.10
Page 212 of 340
5.4. Discussion
In this discussion section, I start with a discussion of the results of SSI farmers’ choices
of actions and their engaging in mitigating strategies to reduce the transmission of WRZD
(section 5.4.1). Then, I unpack the important results of my research – the factors that are
associated with SSI farmers’ engagement in mitigating strategies to reduce the transmission of
WRZD (section 5.4.2). Next, the limitations of my research will be outlined towards the end of
this section (section 5.4.3).
5.4.1. SSI farmers’ choices of actions and their engagement in mitigating strategies to reduce
the transmission of WRZD
On-farm water quality and its link to public health are complex due to the diversity of
WRZDs (Dufour, 2004). This presents a challenge for public health as each WRZD may require
different strategies to mitigate it, depending on the pathogen involved and the source of the
problem (i.e., LM, SWP, WSDT, and POU) (WHO, 2004). Each of these key strategies,
recommended by the WHO, requires a number of actions/tactics to mitigate WRZD transmission
(Carr and Bartram, 2004). Observations of SSI farms during the field survey outlined in my
research indicated that SSI farmers usually took more than one action with various combinations
and frequencies of actions to mitigate the transmission of WRZD. My research uncovered
interesting results concerning SSI farmers’ choices of specific actions together with their broad
engagement in various strategies (outlined in Table 5.1) to mitigate the transmission of WRZD.
5.4.1.1. Wide differences in choices of specific actions to mitigate WRZD transmission
The SSI farmers in the research study selected a wide range of actions (clustered under
the WHO recommended mitigating strategies) to help them reduce the transmission of WRZD.
Specifically, more SSI farmers in Thai Binh than in An Giang chose actions to mitigate the
Page 213 of 340
transmission of WRZD. These greater numbers are supported by the results of the qualitative
component of my research (e.g., in-depth interviews), which established that in Thai Binh SSI
farmers and the general population were culturally more conservative due to a more difficult
economy as well as colder weather conditions compared to those in An Giang. Economic factors
may also explain why more SSI farmers in Thai Binh compared to An Giang took actions to
mitigate WRZD transmission.
The results in my research from analyzing the actions that SSI farmers choose are similar
to those of other studies and reports. For example, the great number of SSI farmers in my
research who indicated they vaccinate domestic animals is in keeping with local veterinary
reports which indicate that vaccinations of domestic animals against key WRZD (e.g., H5N1) are
maintained (Nguyen, 2014). For some other actions, however, my research results differ from
those of other studies. For example, most of the participating Thai Binh SSI farmers in my
research (81%) indicated they had adequate household sanitation based on their subjective
assessment, while only one-third (35%) of An Giang farmers estimated they did. This differs
from the Ministry of Health’s (MoH) statistics which showed that few households (18.5%) had
adequate household sanitation (MoH, 2006). Some possible explanations for this difference in
results include that 1) SSI farmers in my research subjectively assessed their household
sanitation conditions while the MOH’s experts objectively assessed the household sanitation
conditions of farms, and 2) my research target groups (i.e., SSI farmers) were more specific than
those who were part of the MoH research (i.e., general rural farmers). However, this difference
gives us some ideas about possible gaps between the perceptions and knowledge SSI farmers
Page 214 of 340
have relating to adequate household sanitation – i.e., SSI farmers may think they have adequate
household sanitation but this may not be the case.
5.4.1.2. SSI farmers’ broad engagement in mitigating strategies to reduce WRZD transmission
The results of SSI farmers’ engagement in implementing mitigating strategies to reduce
WRZD transmission are interesting for two reasons. First, they show that more than half of SSI
farmers do engage in the mitigating strategies to reduce WRZD transmission (Table 5.2), each of
which combines some specific actions to reduce the transmission of WRZD. Again, this reflects
the complexity of the interface between livestock production, on-farm water quality, and public
health issues. Second, although it is necessary to look at each mitigating strategy as a
combination of various actions to reduce the transmission of WRZD (Carr and Bartram, 2004;
WHO, 2004), I was unable to find any studies that explored the ways SSI farmers engage in
mitigating WRZD transmission as a collection of actions. Thus, the results of my research could
be the basis for future studies regarding further measurements of SSI farmers’ mitigating
strategies to reduce WRZD transmission (e.g., an objective assessment of farmers’ engagement
and/or adoption of mitigating strategies to reduce transmission of WRZD).
5.4.2. Factors associated with SSI farmers’ engagement in mitigating strategies to reduce
transmission of WRZD
As indicated earlier in this chapter, the main objective of my research was to explore
factors that are associated with SSI farmers’ engagement in mitigating strategies to reduce the
transmission of WRZD (Table 5.1). Therefore, in this section I focus on discussing the factors
that are associated with SSI farmers’ engagement in the mitigating strategies. However, prior to
discussing associations, it is necessary to note that, compared to other regressions (e.g., linear
Page 215 of 340
regression or logit regression), the interpretation of probit coefficients and marginal effects is
less straightforward.
A negative coefficient indicated that an increase in an explanatory variable is associated
with a decrease in the predicted probability of the dependent variable. On the other hand, a
positive value coefficient indicated that an increase in an explanatory variable is associated with
an increase in the predicted probability of the dependent variable. A marginal probability effect
at the means presented the rate of change in the probability of a farmer engaging in a mitigating
strategy for a per unit change in an explanatory variable, holding all other explanatory variables
at their means. More specifically, for a continuous explanatory variable (Xi), a marginal
probability effect was a rate of change in the probability of a binary dependent variable (i.e.,
engagement in a mitigating strategy), given a unit change in that continuous explanatory variable
at mean values of other explanatory variables. This rate of change was expressed in a percentage.
For a binary explanatory variable (Xi) with two values (i.e., “0” or “1”), a marginal probability
effect was the difference between the probability of a dependent variable given Xi = 0 and the
probability of that dependent variable given Xi = 1, holding all other explanatory variables
constant at their means. In addition to indicating significant levels and signs of the associations
between the explanatory variables on SSI farmers’ engagement in mitigating strategies, marginal
probability effects show the change in magnitudes of the associations.
A standard maximum likelihood estimating procedure was used to estimate the individual
coefficients of the explanatory variables (Greene, 2008a). With respect to how well the probit
model fits the data, a direct assessment of the effectiveness of the probit model in modeling the
mitigating strategies was used. In the direct assessment, the probit model fits were measured by
Page 216 of 340
computing predicted probabilities of SSI farmers’ engagement in mitigating strategies, and then
comparing it with the actual engagement in mitigating strategies. Given the non-linear feature of
the probit model, marginal probability effect analysis is normally used to have a more direct
interpretation of the coefficients of the explanatory variables (Greene, 2008b). The goodness-of-
fit measured by model chi-square, McFadden's R2, and percentages of correct predictions of all
probit models (concordance) for the four key mitigating strategies indicated that selected
explanatory variables in the probit models contributed to explaining the variation in SSI farmers’
engagement in the mitigating strategies.
Overall, the results in this chapter supported the hypothesis that SSI farmers’
socioeconomic status and their perceptions of risk factors for WRZD transmission are associated
with their engagement in mitigating strategies to reduce the transmission of WRZD. The
following section will explore in detail how the results of each mitigating strategy supported the
hypothesis.
5.4.2.1. Factors associated with SSI farmers’ engagement in LM as a mitigating strategy to
reduce WRZD transmission
The factors significantly associated with SSI farmers’ engagement in a LM mitigating
strategy for reducing WRZD transmission were related to the demographics of farm households,
livestock production, SSI farmers’ perceptions, and the quality of water on SSI farms. In the
following paragraphs, some of the most important factors that are significantly associated with
SSI farmers’ engagement in a LM mitigating strategy will be discussed.
SSI farmers’ years of attending school, as expected, have a positive effect on their
engagement in a LM mitigating strategy. For those SSI farmers who have an additional year of
attending school, the marginal probability effects on their engagement in a LM mitigating
Page 217 of 340
strategy increases to reduce transmission of WRZD. Although I did not find studies that
specifically explored associations between education and SSI farmers’ engagement in a LM
mitigating strategy, some studies of SSI farmers in Vietnam found education to be an important
factor in influencing their choices for adaptive climate change measures (Le et al., 2014b). The
number of livestock (i.e., chickens and pigs) on farms also has a positive effect on SSI farmers’
engagement in a LM mitigating strategy. This means that as a SSI farmer gains one more
chicken or pig, the probability of taking on a LM mitigating strategy increases, assuming all
other explanatory variables are held constant at their means. Therefore, enhancing rural
education and encouraging larger scale production of chickens and pigs may contribute to
improving SSI farmers’ engagement in LM to reduce WRZD transmission.
The negative sign of the association between the log42 number of E.coli cfu in domestic
water and SSI farmers’ engagement in a LM mitigating strategy to reduce WRZD transmission
means that an SSI farmer has less probability of taking on a LM mitigating strategy for one
greater log number of E.coli cfu in domestic water assuming all other explanatory variables are
held constant at their means. Research in the U.S. also reported an association with a negative
sign between the log E. coli cfu number in farm water and farm management (Park et al., 2014).
However, caution is needed when comparing my research and the U.S. research due to the
differences in the context of the two studies (e.g., different operational definitions of farm
management). It appears that a high E. coli cfu count in domestic water may provide a warning
sign of a decrease in farmers’ engagement in LM strategy.
42 Natural logarithm of the number of E.coli cfu in domestic water variable
Page 218 of 340
5.4.2.2. Factors associated with SSI farmers’ engagement in SWP as a mitigating strategy to
reduce WRZD transmission
As indicated in the results section, 11 explanatory variables/factors were significantly
associated with SSI farmers’ engagement in a SWP mitigating strategy. These factors were again
related to the demographics of farm households, livestock production (e.g., number of pigs), and
SSI farmers’ perceptions. The results indicate that SSI farmers who perceived costs and peer-
pressure as barriers to their engagement in risk mitigating strategies were less likely to engage in
a SWP mitigating strategy compared to those who did not. These results agree with other studies
that have shown the associations between economic factors and SSI farmers’ behaviour (Duong
et al., 2004; Keraita et al., 2010). These results also confirmed my expectation, prior to the
research, of associations with negative signs between economic factors such as costs and SSI
farmers’ behaviours. This mean that a high cost related to a mitigating strategy may prevent SSI
farmers from engaging in that strategy. Conversely, SSI farmers with more years of attending
school, more years of farming, or more pigs were more likely to engage in a SWP mitigating
strategy. Thus, economic development policies and programs targeting SSI farmers would
contribute to more fully engaging them in protecting sources of water to mitigate WRZD
transmission. Moreover, SSI farmers with less formal education and fewer years of farming may
need more support than other SSI farmers in order to engage them more in source water
protection to mitigate WRZD transmission.
Discrete factors significantly associated with SSI farmers’ engagement in a SWP
mitigating strategy include the following: men playing a major role in family health on farms;
the perceived susceptibility to AI from contacting untreated wastewater; one’s perceived self-
efficacy in taking on source water protection actions; advice from health workers as a trigger;
Page 219 of 340
perceived pressure from peer SSI farmers as a trigger; and perceived worry about being affected
by HPAI as a trigger. As SSI farmers move from “No” to “Yes” in these discrete factors, the
corresponding probabilities of taking on SWP as a mitigating strategy increases.
No other studies were found that specifically explored factors that are associated with SSI
farmers’ engagement in SWP. However, social factors were reported to influence a farmer’s
commitment to pro-environment action in general (Michel-Guillou and Moser, 2006), and
suggestions were made with respect to the need to study triggers of farmers’ adoption of
mitigating strategies (Keraita et al., 2010). I suggest that rural policies and programs in Vietnam
addressing gender equality, farmers’ perceptions of health risk factors, and community health
workers’ capacity should be prioritized to help improve source water protection among SSI
farmers in mitigating WRZD transmission.
5.4.2.3. Factors associated with SSI farmers’ engagement in POU as a mitigating strategy to
reduce WRZD transmission
The factors significantly associated with SSI farmers’ engagement in a POU mitigating
strategy to reduce WRZD transmission were also related to geographic regions, demographics of
farm households, SSI farmers’ perceptions, and the quality of water on SSI farms. Some of these
factors will be discussed in the following paragraph.
It was not surprising that factors positively and significantly associated with SSI farmers’
engagement in a POU mitigating strategy were related to province, having children in a
household, gender roles in livestock production, perceptions, and advice from health workers.
The probabilities for taking on a POU mitigating strategy increase among SSI farmers with these
Page 220 of 340
factors compared to those who did not. Meanwhile, income per person per year43, the perceived
safety of untreated drinking water, the perceived cost as a barrier to taking actions, and the log44
number of E.coli cfu in domestic water, as expected, were significantly and negatively associated
with SSI farmers’ engagement in POU mitigating strategies to reduce transmitting WRZD. A one
unit increase in each of these factors, holding other factors constant at their means, decreases the
probability of taking on a POU mitigating strategy. This means, for example, that when a farm
becomes poor (i.e., move up from code 0 to code 1), the probability of that farm taking on a POU
mitigating strategy decreases. Policies and programs that address POU strategies in mitigating
WRZD transmission among SSI farmers should, for example, tailor their programs by province
or combine them with poverty reduction programs.
5.4.2.4. Factors associated with SSI farmers’ engagement in WSTD as a mitigating strategy to
reduce WRZD transmission
A number of factors relating to the socioeconomic status of SSI farm households, number
of livestock produced, SSI farmers’ perceptions, and the quality of water on SSI farms were
associated with SSI farmers’ engagement in a WSTD mitigating strategy. Some of these factors
are discussed in detail in the following paragraphs.
Economic factors such as income or the perceived cost of taking mitigating actions as a
barrier to taking action once again are associated with SSI farmers’ engagement (or not) in a
WSTD mitigating strategy to lessen WRZD transmission. This means SSI farmers who were
43 The variable of income per person per year was dichotomized using the cut-off income level for poverty set by the
Vietnamese government (i.e., less than 4.8 million VND per person per year – equivalent to $244 CAD per person
per year) at the exchange rate of $ CAD = 19,700 on 5th December 2013. An income per person per year below the
cut-off income level (i.e., a poor farm) was coded as 1 and 0 if otherwise (i.e., a not poor farm). I should have named
this variable as DPoverty instead of DIncome to avoid confusion. 44 The variable E.coliDom was transformed into LogE.coliDom (i.e., number of E.coli cfu in domestic water) using
a natural log transformation
Page 221 of 340
living under the poverty line or who perceived costs as barriers to their mitigating strategies have
lower probabilities of engaging in a WSTD mitigating strategy compared to those who were not
poor or did not perceive cost to be a barrier. These results, as expected, align with other studies
that have shown the associations between economic factors and SSI farmers’ behaviours (Duong
et al., 2004; Keraita et al., 2010).
On the other hand, an additional year of attending school, one more pig, or one greater
log number of E. coli cfu in drinking water are positively significantly associated with SSI
farmers’ engagement in a WSTD strategy or for increasing the corresponding probabilities of
SSI farmers engaging in a WSTD mitigating strategy. Moreover, compared to a SSI farm that
has shared responsibility for family health between men and women, a farm with women
primarily responsible for family health increases the probability of engaging in a WSTD strategy.
Therefore, policies and programs addressing SSI farmers’ WSTD strategies to mitigate WRZD
transmission should prioritize issues such as poverty reduction and gender equality.
5.4.2.5. Common factors associated with SSI farmers’ engagement across four mitigating
strategies to reduce WRZD transmission
It is interesting to notice some factors that were consistently associated with the four
mitigating strategies. For example, perceived self-efficacy in engaging in a LM mitigating
strategy and advice from health workers were positively significantly associated with all four
mitigating strategies. Six factors were significantly associated with three mitigating strategies:
years of attending school, female farmers playing the main roles in family health, the perceived
susceptibility to HPAI from using untreated wastewater, the number of pigs, cost as a barrier to
taking action, and peer pressure. This consistency suggests that in order to improve SSI farmers’
engagement in mitigating strategies to reduce WRZD transmission, it may be important to
Page 222 of 340
consider the above significant factors in developing water-related programs and policies
targeting SSI farmers.
Other studies also suggested that when designing programs and developing policies with
respect to the behavioural actions taken by rural farmers, it may be important to consider rural
farmers’ engagement in strategies to mitigate infectious diseases (OECD, 2012). In addition to
consistency, there were mixtures of factors that were positively and negatively associated with
SSI farmers’ mitigating strategies. These mixtures, as expected, are supported by previous
studies which indicated that demographic factors and traditional practices may facilitate or
hinder their actions (OECD, 2012), and only focusing on health-related factors is insufficient for
SSI farmers to engage in a mitigating strategy for disease transmission (Keraita et al., 2010).
5.5. Conclusions
In this chapter, I have provided the background, the methods, and the corresponding
results for my research addressing SSI farmers’ choices of actions as well as their broad
engagement in mitigating strategies to reduce WRZD transmission. My research results from the
previous chapters were used in this chapter to model factors that are associated with SSI farmers’
engagement in individual mitigating strategies to decrease WRZD transmission.
SSI farmers in Thai Binh and An Giang indicated that they have a wide range of actions
to choose from to reduce WRZD transmission. These actions were used to provide a more in-
depth understanding of SSI farmers’ engagement in four key mitigating strategies to decrease
WRZD transmission as recommended by the WHO. The following strategies were included:
livestock management; source water protection; water storage, treatment, and distribution; and a
point of use/household strategy. The research in this chapter has modeled factors associated with
Page 223 of 340
SSI farmers’ engagement in these mitigating strategies. The results, showing a combination of
various factors that are associated with SSI farmers’ engagement in each of the mitigating
strategies, have confirmed my study’s hypothesis that SSI farmers’ socioeconomic status and
their perceptions of risk factors for WRZD transmission are associated with their engagement in
mitigating strategies to reduce the transmission of WRZD.
In conclusion, policies and programs relating to rural water and public health and the
mitigating strategies to reduce WRZD transmission should focus on a combination of factors
such as SSI farmers’ perceived self-ability to take livestock management actions as well as to
receive advice from health workers. It is also important to pay attention to other factors that
significantly and positively contribute to three out of the four mitigating strategies to lessen
WRZD transmission including education, women’s roles in family health, perceived
susceptibility to HPAI from using untreated wastewater, pig production, and peer pressure on
SSI farmers. Additionally, attention must be paid to economic factors such as perceived cost as a
barrier, which was the only factor that was significantly and negatively associated with SSI
farmers’ engagement across the four mitigating strategies of reducing the transmission of
WRZD.
A solid understanding SSI farmers’ choice of actions and factors that are associated with
their engaging in (or not engaging in) mitigating strategies to decrease WRZD transmission has
important implications for future research, interventions, and policy advocacy. Such
understanding will help in changing farmers’ behaviours, reducing rural poverty, implementing
effective water policies and mixed agriculture policies, as well as improving animal and
environmental health services in Vietnam.
Page 224 of 340
Page 225 of 340
References
Abdul, H. A., Ayamga, M., and Donkoh, S. A. (2014). Smallholder adoption of soil and water
conservation techniques in Ghana. African Journal of Agricultural Research, 9, 539-546.
Adesina, A. A. and Baidu, F. J. (1995). Farmers' perceptions and adoption of new agricultural
technology: evidence from analysis in Burkina Faso and Guinea, West Africa.
Agricultural Economics, 13, 1-9.
Becker, M. H. (1974). The Health belief model and personal health behavior. Thorofare, N.J. :
Slack, c1974.
Becker, M. H. (2012). The Health Belief Model and Personal Health Behavior. (4 ed.) (vols. 2).
Bloom, D. E., Cafiero, E. T., and Chu, M. (2013). Setting health priority: strategy versus tactics.
PSI Impact Managzine.
Brewer, N. T., Chapman, G. B., Gibbons, F. X., Gerrard, M., McCaul, K. D., and Weinstein, N.
D. (2007). Meta-Analysis of the Relationship Between Risk Perception and Health
Behavior: The Example of Vaccination. Health Psychology, 26, 136-145.
BUMC (2013). The Health Belief Model. BU School of Public Health [On-line]. Available:
http://sphweb.bumc.bu.edu/otlt/MPH-Modules/SB/SB721-Models/SB721-Models2.html
Carr, R. and Bartram, J. (2004). The control envelope and risk management. In (pp. 67-92). IWA
Publishing.
Cramb, R. A. (2005). Farmers' strategies for managing acid upland soils in Southeast Asia: An
evolutionary perspective. Agriculture, Ecosystems and Environment, 106, 69-87.
Page 226 of 340
Dufour, A. (2004). Water-related zoonotic disease impacts - geographical prevalence. In
Waterborne Zoonoses: Identification, Causes, and Control (pp. 91-150). London: WHO
and IWA.
Duong, D. V., Binns, C. W., and Lee, A. H. (2004). Utilization of delivery services at the
primary health care level in rural Vietnam. Social Science and Medicine, 59, 2585-2595.
Greene, W. H. (2008a). Maximum Likelihood Estimation. In Econometric Analysis (6th ed.,
Pearson Prentice Hall.
Greene, W. H. (2008b). Models for Discrete Choice. In Econometric Analysis (6th Edition ed.,
Pearson Prentice Hall.
Hall, D. C., Knight, T. O., Coble, K. H., Baquet, A. E., and Patrick, G. F. (2003). Analysis of
Beef Producers' Risk Management Perceptions and Desire for Further Risk Management
Education. Review of Agricultural Economics, 25, 430-448.
Herbst, S., Benedikter, S., Koester, U., Phan, N., Berger, C., Rechenburg, A. et al. (2009).
Perceptions of water, sanitation and health: a case study from the Mekong Delta,
Vietnam. Water Science and Technology, 60, 699-707.
JAE (2014). Author Guidelines: Online Manuscript Submission. Wiley Online Library [On-line].
Available: http://onlinelibrary.wiley.com/journal/10.1111/(ISSN)1477-
9552/homepage/ForAuthors.html
Keraita, B., Drechsel, P., Seidu, R., Amerasinghe, P., Cofie, O., and Konradsen, F. (2010).
Harnessing Farmers' Knowledge and Perceptions for Health-Risk Reduction in
Wastewater-Irrigated Agriculture. In P.Drechsel, C. A. Scott, L. Raschid-Sally, M.
Redwood, and A. Bahri (Eds.), Wastewater irrigation and health : assessing and
Page 227 of 340
mitigating risk in low-income countries (pp. 337-353). London: Earthscan Publications
Ltd.
Le, D. H., Li, E., Nuberg, I., and Bruwer, J. (2014a). Farmers's assessments of private adaptive
measures to climate change and influential factors: a study in the Mekong Delta,
Vietnam. Nat Hazards, 71, 385-401.
Le, D. H., Li, E., Nuberg, I., and Bruwer, J. (2014b). Farmers' assessments of private adaptive
measures to climate change and influential factors: a study in the Mekong Delta,
Vietnam. Nat Hazards, 71, 385-401.
Long, J. S. and Freese, J. (2014). Regression Models for Categorical Dependent Variables Using
Stata. (Third ed.) Texas: Stata Press College Station.
Marteau, T. M., Hollands, G. J., and Fletcher, P. C. (2012). Changing Human Behavior to
Prevent Disease: The Importance of Targeting Automatic Processes. Science, 337, 1492-
1495.
Mekoya, A., Oosting, S. J., Fernandez, R. S., and Akke, V. J. (2008). Farmers' perceptions about
exotic multipurpose fodder trees and constraints to their adoption. Agroforestry Systems,
73, 141-153.
Michel-Guillou, E. and Moser, G. (2006). Commitment of farmers to environmental protection:
From social pressure to environmental conscience. Journal of Environmental Psychology,
26, 227-235.
Moe, C. L. (2004). What are the criteria for determining whether a disease is zoonotic and water
related. In World Health Organization (WHO). Waterborne Zoonoses: Identification,
Causes and Controls (pp. 27-45). London: IWA Publishing.
Page 228 of 340
MoH (2006). The survey on environmental sanitation in rural Vietnam Hanoi: Ministry of
Health.
Neal, D., Vujcic, J., Hernandez, O., and Wood, W. (2014). Handwashing and the Science of
Habit. Sanitation Updates [On-line]. Available:
https://sanitationupdates.files.wordpress.com/2014/12/neal-handwashing2014.pdf
Negatu, W. and Parikh, A. (1999). The impact of perception and other factors on the adoption of
agricultural technology in the Moret and Jiru Woreda (district) of Ethiopia. Agricultural
Economics, 21, 205-216.
Nguyen, V. D. (2014). Results of the Control and Prevention of Animal Diseases in early 2014
Thai Binh: Thai Binh Department of Veterinary.
Nguyen, T. D., Vo, T. T., and Vu-Khac, H. (2011). Virulence factors in Escherichia coli isolated
from calves with diarrhea in Vietnam. Journal of Veterinary Science, 12, 159-164.
Nguyen, V. Q. (2004). The role of Salmonella and E. coli in diarrhea syndrome of cattle and
calves in provinces of Central Vietnam and the production of antibodies for treatment of
diarrhea disease caused by Salmonella and E. coli. National Institute of Veterinary
Research, Hanoi.
OECD (2012). Farmer Behaviour, Agricultural Management and Climate Change. OECD
Publishing.
Park, S., Navratil, S., Gregory, A., Bauer, A., Srinath, I., Szonyi, B. et al. (2014). Farm
Management, Environment, and Weather Factors Jointly Affect the Probability of
Spinach Contamination by Generic Escherichia coli at the Preharvest Stage. Appl Environ
Microbiol, 80, 2504-2515.
Page 229 of 340
Payne, N., Jones, F., and Harris, P. (2002). The impact of working life on health behavior: The
effect of job strain on the cognitive predictors of exercise. Journal of Occupational
Health Psychology, 7, 342-353.
Rabbi, S. and Dey, N. (2013). Exploring the gap between hand washing knowledge and practices
in Bangladesh: a cross-sectional comparative study. BMC Public Health, 13, 89.
Rechenburg, A. and Herbst, S. (2006). Hygienic Aspects of Sanitation and Water in the Rural
Areas of the Mekong Delta, Vietnam (Rep. No. "Prosperity and Poverty in a Globalised
World Challenges for Agricultural Research"). University of Bonn, Institute for Hygiene
and Public Health, Germany.
Rosenstock, I. M., Strecher, V. J., and Becker, M. H. (1988a). Social Learning Theory and the
Health Belief Model. Health Education and Behavior, 15, 175-183.
Rosenstock, I. M., Strecher, V. J., and Becker, M. H. (1988b). Social Learning Theory and the
Health Belief Model. Health Education and Behavior, 15, 175-183.
Stata (2013). Stata Users' Guide Release 13. Stata: Data Analysis amd Statistical Software [On-
line]. Available: http://www.stata.com/manuals13/u.pdf
Trinh, T. L. and Nguyen, P. L. (2009). Assessment of surface water quality by water quality
index (WQI) at the Cai Sao canal, An Giang province, Vietnam. Livestock Research for
Rural Development [On-line]. Available: http://www.lrrd.org/lrrd23/7/lan23151.htm
Van Mele, P. and Cuc, N. T. T. (2001). Farmers' Perceptions and Practices in Use of
Dolichoderus thoracicus (Smith) (Hymenoptera: Formicidae) for Biological Control of
Pests of Sapodilla. Biological Control, 20, 23-29.
Page 230 of 340
Vindigni, S. M., Riley, P. L., and Jhung, M. (2011). Systematic review: handwashing behaviour
in low- to middle-income countries: outcome measures and behaviour maintenance.
Tropical Medicine and International Health, 16, 466-477.
WHO (2004). Waterborne Zoonoses: Identification, Causes, and Control. IWA Publishing.
WHO (2012). Health Risk Perception World Health Organisation Western Pacific Region.
Page 231 of 340
Chapter Six: Conclusion
6.1. Overview
This dissertation has presented the original results of my Ph.D research, which presents
the research story of a commonly used mixed agricultural model in the provinces of Thai Binh
and An Giang in Vietnam. This research story began with a developed socioeconomic profile of
the small-scale integrated (SSI)45 farming model in these two provinces followed by an
assessment of SSI farmers’ perceptions of risk factors for transmitting WRZD46 and their choice
of actions to mitigate the transmission of WRZD. The research story also included the results of
determining the sources of water used on farms and an objective assessment of the quality of
water used on farms. Furthermore, the research story presented factors that are associated with
the quality of water on SSI farms and factors that are connected to SSI farmers’ engagement in
mitigating strategies to reduce the transmission of WRZD.
The dissertation is structured using a manuscript-based format. Besides the introduction
and conclusion chapters, I presented the research results as a collection of four manuscript-based
chapters that were prepared as a basis for submission to peer-review journals preferably in the
field of agricultural economics, water quality, and public health. Although each manuscript-
based chapter is a stand-alone manuscript, I arranged the chapters of this dissertation in a logical
and continuous order that presented the results in answer to my overall research questions. This
45Small-scale integrated (SSI) farming is defined as farming practices in which farmers combine some forms of
livestock and are directly or indirectly dependent on agriculture (Hall et al., 2006; Irini and Rapsomanikis, 2005;
Pica-Ciarmarra et al., 2011). 46 In this research, I used the WHO’s definition of WRZD as diseases in humans that come from animals and are
related to water such as Highly Pathogenic Avian Influenza, some parasites, or diseases caused by a range of
bacteria (e.g., pathogenic strains of Escherichia coli) (WHO, 2004). Criteria for determining WRZD include: 1) the
pathogen must spend part of its life cycle within one or more animal species; 2) it is probable or conceivable that
some life stage of the pathogen will enter water; and 3) transmission of the pathogen from animals to humans must
be through a water-related route (Moe, 2004)
Page 232 of 340
presentation began with an understanding of the profile of the SSI farming model followed by
SSI farmers’ perceptions of risk factors for the transmission of WRZD and their on-farm water
quality, and ended with an understanding of factors that are associated with both on-farm water
quality and farmers’ mitigating strategies to reduce the transmission of WRZD.
In this chapter, I integrate the research results presented in the four manuscript-based
chapters with an emphasis on how the chapters build upon each other. The integration begins
with a summary of the literature review concerning the SSI farming model (i.e., Section 6.2)
followed by a summary of the research questions, hypotheses, and methods (Section 6.3). In
section 6.4, I summarize the contributions of my research to scientific knowledge with respect to
the SSI farming model, policies, and practices in the context of rural water quality and public
health. Section 6.5 provides a review of challenges and limitations of the research. The last two
sections (Section 6.6-7) provide overall recommendations for future research and policy
advocacy with respect to the SSI farming model, rural water quality and public health, and my
final thoughts respectively.
6.2. Summary of the Literature Review
The SSI farming model is a commonly used model for mixed agricultural practice in
Vietnam and has been promoted as a model in rural areas in Vietnam for a couple of decades.
This model followed a policy shift from a centrally planned and subsidized system to a market
oriented system in Vietnam, and it will continue to be an important part of the agricultural
system in Vietnam in the future. In this model, farmers use an integrated farm management
system to improve their farming practices and to achieve high economic efficiency. The model
has proven to be successful in helping farmers make efforts to optimize their on-farm resources
and to generate food and income. In this model, there are interactions among farmers, their
Page 233 of 340
animals, and the water environment on SSI farms. However, these interactions have the potential
for adverse health impacts such as the transmission of WRZD. My review of the literature found
no studies that explore factors that are associated with on-farm water quality and farmers’
engagement in mitigating strategies to reduce the transmission of WRZD. However, I have
studied these factors in my research using the following research questions, hypothesis, and
methods.
6.3. Research Questions, Hypotheses, and Methods
The results of my research have addressed the following overall research questions:
1. What is the socioeconomic profile of SSI farmers in the Thai Binh and An Giang
provinces of Vietnam?
2. Is the socioeconomic profile of SSI farmers in the provinces of Thai Binh and An Giang
consistent with an EcoHealth approach in the context of SSI farming?
3. What are the perceptions of SSI farmers concerning the risk factors for transmitting
WRZD in the provinces of Thai Binh and An Giang?
4. What sources of water do SSI farmers use for drinking and domestic purposes47 and what
are the corresponding basic indicators48 of water quality for those sources in the
provinces of Thai Binh and An Giang?
5. How does the quality of on-farm water that SSI farmers in the provinces of Thai Binh and
An Giang use for drinking and domestic purposes relate to their perceptions about both
water quality and risk factors for transmitting WRZD?
47 In my research, drinking water is defined as “water used for direct drinking or food processing”. Domestic water
is defined as “water use for domestic purposes but nor for direct drinking or processing food” (MoH, 2009b; MoH,
2009a) 48 The WHO recommends Escherichia coli (E. coli) colony forming unit (cfu), turbidity, and pH as basic microbial
and related indicators for assessing rural water quality (WHO, 2012)
Page 234 of 340
6. Do SSI farmers in the provinces of Thai Binh and An Giang engage in mitigating
strategies to reduce transmitting WRZD?
7. What factors are associated with SSI farmers’ engagement in mitigating strategies to
reduce the transmission of WRZD in the provinces of Thai Binh and An Giang?
Further to the description and exploration of the profile of SSI farms and farmers’
perceptions of risk factors for transmission of WRZD in Chapter Two and Three respectively,
four key hypotheses were formulated and tested in Chapter Four and Five. These hypotheses
were:
1. The frequencies of use of water sources for drinking and domestic purposes among
farmers in Thai Binh are different from those in An Giang.
2. The sources of water use for drinking and domestic purposes49 in both Thai Binh and An
Giang farms are of low quality.
3. SSI farmers’ socioeconomic status and perceptions of risk factors for transmission of
WRZD are associated with the microbial content of water on farms.
4. SSI farmers’ socioeconomic status and their perceptions of risk factors for transmission
of WRZD are associated with their engagement in mitigating strategies to reduce the
transmission of WRZD.
In my research, I have used a combination of qualitative and quantitative methods to
describe the profile of the SSI farmers and to explore their perceptions about risk factors for and
actions needed to reduce the transmission of WRZD. Furthermore, I used various modelling
techniques to explore associating factors of on-farm water quality (i.e., multiple linear regression
49 Drinking water is defined as “water used for direct drinking or processing food” (MoH, 2009b). Domestic water is
defined as “water used for domestic use but not for direct drinking or processing food” (MoH, 2009a).
Page 235 of 340
and seeming unrelated regression technique) and associating factors for farmers’ engagement in
mitigating the transmission of WRZD (i.e., Probit regression).
6.4. Contribution to Knowledge
6.4.1. An understanding of SSI farming model in the context of rural water and public health
This dissertation has described the profile of the SSI farming model in the provinces of
Thai Binh and An Giang in the context of rural water and public health. This description focused
on the socioeconomic profile of the SSI farms, SSI farmers’ perceptions of water quality and
their perceptions about risk factors for transmitting WRZD, SSI farmers’ choices of actions and
their engagement in mitigating strategies to reduce the transmission of WRZD, and the quality of
drinking water used on SSI farms.
The developed socioeconomic profile of SSI farmers in Thai Binh and An Giang in
Chapter Two helped to present a picture of the SSI farming model in the two provinces by
focusing on demographic characteristics, livestock production, and farm economics. Chapter
Two also provided insight into the SSI farming model with respect to understanding the
ecological and health implications of the EcoHealth approach to SSI farming and into reducing
the unintended of consequences of SSI farming. Chapter Three provided a more specific
understanding of the SSI farming model concerning SSI farmers’ perceptions of risk factors for
transmission of WRZD. Chapter Three also demonstrated the possible important roles these
perceptions can play in improving the health of farmers and their animals, and in providing
guidance for policy formulations. Chapter Four (the first part)50 presented an understanding of
the sources and the quality of water used on SSI farms. The low microbial and related quality of
50 The second part of Chapter Four, factors that are associated with the microbial quality of water used on SSI farms,
is summarized in the next section - Section 6.4.2
Page 236 of 340
water for drinking compared to the guidelines of clean water supply, sanitation, and water quality
assessment of the WHO and the Government of Vietnam suggested possible implications for
rural water testing, and on-farm water public health training and management. The first part51 of
Chapter Five presented an understanding of SSI farmers’ choices of actions to reduce risk factors
for the transmission of WRZD. The following sub-sections (i.e., 6.4.1.1-4) provide a summary of
key contribution to knowledge this research study makes with respect to an understanding of the
profile of the SSI farming model in the context of rural water and public health.
6.4.1.1. A socioeconomic profile of the SSI farming practice in Thai Binh and An Giang
Participating SSI farmers in the study had limited education, but most of them had
attended primary school. Almost 80% of SSI farmers in both provinces had attended secondary
or lower schools. The average SSI family size was four people with fewer female farmers mainly
responsible for livestock production, fish production, and family health compared to male
famers. The proportion of the SSI farmers who lived in poverty is one in five. This proportion is
five times greater in An Giang compared to Thai Binh, and is two times greater than the general
poverty proportion in rural areas of Vietnam (GSO Vietnam, 2013) The mean annual number of
livestock produced per farm was as follows: 17 chickens, 87 ducks, 13 pigs, and two cattle.
Moreover, although SSI farmers were still practicing integrated livestock and aquaculture
production, they did not usually refer to this as a model of farming practice.
The SSI farming model demonstrated a truly complex environment in which interactions
take place between SSI farmers, their livestock, and the on-farm water environment. This
complexity was reflected in the following key results: SSI farmers used animals’ manure as
51 The second part of Chapter Five which discussed, factors that are associated with SSI farmers’ engagement in
mitigating strategies to reduce transmission of WRZD, are summarized in the next section - Section 6.4.2
Page 237 of 340
fertilizer and food for plants and fish; livestock production can bring both income and food for
SSI farmers; and products from gardens can be used as food for both fish and animals, as well as
provide an income source for SSI farmers. There were also geographic variations to SSI farming.
Yet, the interactions that take place on SSI farms have potential linkages to increasing the risk
factors for transmitting WRZD. However, SSI farmers have also had only limited participation in
the design, implementation, and monitoring of policies relating to livestock production, rural
water supplies and sanitation although they are the owners and implementers of the SSI farming
model.
Gender and social equity among SSI farming appeared to be a concern where poverty
existed among the participating SSI farmers in the two provinces. SSI household farms had a
greater mean family size than the Vietnamese 2009 national average. Consequently, it is
understandable that farmers are finding additional income from off-farm seasonal work.
However, there are potential social challenges for SSI farmers when they look for off-farm
seasonal work given their limited education level, younger age, poor working conditions, and far
distance from home with limited social benefits.
The SSI farming model was initially introduced to enhance the sustainability of farmers’
livelihoods, but several issues may influence the sustainability of the model and the health of
farmers and their livestock. Examples of these issues include low income, off-farm work among
farmers, low water quality on farms, lack of farmers’ participation in influencing related public
policies, and lack of capacity to face challenges and competitiveness when Vietnam joins the
WTO. In reality, these issues may worsen the undesirable consequences of SSI farming instead
of improving them.
Page 238 of 340
The description of the SSI farming model can be used as a basis for future studies of this
model. An EcoHealth approach is good for community development and there are examples of
applying the EcoHealth approach in academic research, policy advocacy, and community
development (Nguyen, et al.; Hall, Dinh, Vu, Tran, and Le, 2012). When using an EcoHealth
perspective and taking into consideration the increased transmission of WRZD, it is suggested
that the SSI farming model needs to be reviewed to ensure that it will not be a simple agricultural
economic model in which food and income are outputs. Although this SSI farming model
reflects some of the principles of ecosystem approaches to health (EcoHealth), the model should
be a more holistic EcoHealth-based model in which the health of both farmers and their livestock
are interwoven with on-farm production and the protection of the water environment, and where
health is considered an equally important output of the model. Understanding and improving the
SSI farming model using an EcoHealth approach may help with managing the risk factors
associated with the transmission of WRZD.
6.4.1.2. SSI farmers’ limited perceptions of risk factors for transmission of WRZD
Participating SSI farmers perceived the overall on-farm water conditions in their villages
to be adequate. They were also aware of the potential harm to health when using untreated
sources of water, and assessed water quality subjectively and primarily based on impressions and
previous experiences. Most of the participating SSI farmers considered drinking untreated water
and handling untreated domestic water as being a moderate to high risk factor for the
transmission of Highly Pathogenic Avian Influenza (HPAI), diarrhea, coliform bacteria, and
parasites. If infected, they believed that HPAI, diarrhea, and coliform bacteria would be severe or
very severe. Economic factors were perceived by most of the participating SSI farmers as both
barriers and benefits to taking mitigating actions against the transmission of WRZD. With
Page 239 of 340
respect to perceived self-efficacy, most of the participating SSI farmers believed that they have a
moderate or high ability to conduct mitigating actions against the transmission of WRZD. The
top three triggers/cues for taking mitigating actions were media, worries about being infected,
and advice from health workers.
6.4.1.3. SSI farmers’ engagement in mitigating strategies to reduce transmission of WRZD
SSI farmers in Thai Binh and An Giang indicated they have wide differences in their
choice of actions to reduce the risk factors for transmitting WRZD. Farmers’ choices of actions
provided an in-depth understanding of their engagement in the four key mitigating strategies52 of
transmission of WRZD as recommended by the WHO. These strategies included livestock
management; source water protection; water storage, treatment and distribution; and point of
use/household strategies. More than half of the SSI farmers in the study did engage in the four
mitigating strategies to reduce the risk of transmitting WRZD that was associated with on-farm
water use.
All results of SSI farms in the context of rural water and public health provided ideas for
formulating hypotheses with respect to modeling factors that are associated with on-farm water
quality and farmers’ engagement in mitigating the risks of transmitting WRZD (i.e., the second
parts of Chapter Four and Chapter Five).
6.4.1.4.Low quality of water used on SSI farms for both drinking and domestic purposes
The microbial quality was low for on-farm water for both drinking and domestic use, and
most of the water samples contaminated with E. coli did not met the quality standards for
52 In my research, I defined a mitigating strategy to reduce transmission of WRZD as a plan that includes one or
more activities a SSI farmer could choose to reduce transmission of WRZD (see figure 5.1 and section 5.2.1 in
Chapter Five). An example of a mitigating strategy to reduce transmission of WRZD in my research is good
livestock management which may include actions or tactics (e.g., treat animal waste at source) that SSI farmers
could choose to reduce transmission of WRZD (Carr and Bartram, 2004; WHO, 2004).
Page 240 of 340
drinking and domestic water. Although testing of E.coli in water is recommended to determine if
a water source has recently been contaminated with faeces, such faecal contamination of water is
often intermittent and might not be revealed in one assessment of a single sample. Therefore, in
addition to simple affordable tests for SSI farmers of E. coli in water, it is important to build
farmers’ capacity for inspecting and improving their household water sanitation and on-farm
water public health management. By doing so, SSI farmers will not only be better able to afford
to conduct the tests, but also to know the appropriate time and circumstances for taking
corresponding measures following water testing results.
6.4.2. An understanding of factors associated with on-farm water quality and SSI farmers’
engagement in mitigating strategies to reduce the transmission of WRZD
The description of SSI farms in the context of rural water and public health were used to
model factors that are associated with the objectively assessed on-farm water quality and factors
that are associated with SSI farmers’ engagement in mitigating strategies to reduce transmission
of WRZD. The modelling results have possible implications for rural water supply and
sanitation, on-farm public health management, and for policy advocacy in improving the health
of farmers, their livestock, and their water environment.
6.4.2.1. Factors that are associated with on-farm water quality
With respect to drinking water quality, farmers who had greater income per person per
year and were not satisfied with the quality of their current sources of drinking water were likely
to have better stored rainwater (i.e., lower E. coli cfu in water) compared to those who had lower
income and were happy with the current quality of their drinking water. In terms of piped water
for drinking, farmers who thought that using untreated water for drinking brings susceptibly to
parasite infection were likely to have better piped water (i.e., lower E. coli cfu in piped water)
Page 241 of 340
compared to those who thought that using piped water does not bring susceptibly to parasite
infection. With regard to well water for drinking, it is likely that farms that were headed by
females, involved in fish production, and perceived susceptibility to parasites when using
untreated water for drinking had better piped water (i.e., lower E. coli cfu in pipe water). This
was in comparison to farms headed by men and farmers who thought that using well water did
not bring susceptibly to parasitic infection. The quality of river water for drinking was associated
with gender, years of attending school, and number of chickens produced.
In terms of domestic water, the microbial (E. coli) quality of river water used for
domestic purposes was associated with gender, occupations other than farming, number of pigs,
men primarily responsible for livestock production, and perceived susceptibility to parasites from
untreated domestic water. Years of attending school, numbers of chickens, perceptions of
susceptibility to HPAI from untreated domestic water, and satisfaction with current sources of
domestic water were associated with E. coli in pond water for domestic purposes.
The modeling of factors that are associated with the quality of on-farm water gave some
ideas of factors that may be relevant to explore further possible associating factors for farmers’
mitigating strategies to reduce the transmission of WRZD in the context of SSI farming. The
results of my research have indicated there is room for improving rural water supply and
sanitation, and on-farm water and health management through updates and revision of the water
regulatory framework/policies for SSI farms in Vietnam. The results of my research also showed
a need to consider rural water testing and training in improving on-farm water public health.
Page 242 of 340
6.4.2.2. Factors that are associated with SSI farmers’ engagement in mitigating strategies to
reduce the transmission of WRZD
The four mitigating strategies included the following: livestock management; source
water protection; water storage, treatment and distribution; and point of use/household. More
than half of the SSI farmers in the study did engage in these four mitigating strategies to reduce
risk factors for the transmission of WRZD associated with on-farm water use. There were
combinations of various factors that contributed to SSI farmers’ engagement in each of the four
mitigating strategies to reduce transmission of WRZD. The significant factors that positively
contribute to SSI farmers engaging in the four mitigating strategies were their self-ability to take
livestock management actions and to receive advice from health workers. It is also important to
pay attention to other factors that contributed significantly and positively to three out of four of
the mitigating strategies to reduce the transmission of WRZD including years of attending
school, women’s role in family health, perceived susceptibility to HPAI from using untreated
wastewater, pig production, and peer pressure on SSI farmers. On the other hand, it is important
to recognize perceived cost as a barrier as it was the only one significant negative contributing
factor to SSI farmers engaging across the four mitigating strategies to reduce the transmission of
WRZD.
6.5. Limitations
Some limitations must be taken into consideration when interpreting my research results.
First, although my research had a relatively large sample size, my research only focused
on SSI farming in the context of the targeted river deltas. VAC is also a dynamic model with
much variation, even within a geographical region (Dang et al., 2005; Devendra and Thomas,
2002b). For example, seven VAC farming systems were commonly practiced in the Mekong
Page 243 of 340
Delta (Dang et al., 2005). Furthermore, although some frameworks exist that identify SSI
farming (Dixon et al., 2004; Nguyen, 2014a), no universal criteria are in place to precisely define
SSI farming (Heidhues and Bruntrup, 2003). Thus, there may be considerable variation in
aspects such as the number of livestock, the volume of fish, and the volume of crops that
individuals produce who are identified as SSI farmers. These variations thus make my research
results more difficult to generalize for all regions. Therefore, for SSI farming in other regions of
Vietnam (e.g., irrigated lowlands or rain fed uplands), the more specific context of these regions
must be considered and/or further research may be needed before generalizing the results from
my research. Multiple criteria should be used to examine and interpret information relating to the
profile and on-farm production of SSI farming.
Second, in terms of study design, my research used a cross-sectional study design.
Therefore, factors relating to farmers’ perceptions, socioeconomic status, and farmers’ choices of
actions were assessed at a particular point-in-time. Thus, these factors as well as choices of
actions may change over time (Marteau et al., 2012; Vindigni et al., 2011). Further, the
association between these factors and farmers’ mitigating strategies was identified by giving the
existing farmers’ perceptions and mitigating strategies, which were based on their subjective
self-assessments, which may not meet the standard(s) for proper actions. Moreover, my research
did not objectively assess whether farmers have the correct knowledge about rural water quality
and transmission of WRZD or whether or not they undertake proper actions that meet a certain
standard. At the same time, gaps may exist between not only perceptions but also knowledge
(Rabbi and Dey, 2013), habit (Neal et al., 2014) and actions. Therefore, factors used in modeling
in my research as well as the corresponding directions and magnitude of associations may
change considerably when other factors such as standards, quality of mitigating strategies
Page 244 of 340
applied, timeline, knowledge, and habits are applied or considered. Future studies using study
designs that take into account these other factors are recommended to understand better the
complexity of farmers’ engagement in strategies to mitigate the transmission of WRZD.
Finally, the four mitigating strategies used in this study were modelled separately with a
focus on perceptions of risk factors for the transmission of WRZD and farmers’ socioeconomic
status as possible explanatory variables of interest. However, farmers’ engagement in these
mitigating strategies may be more complex and may relate to each other because all mitigating
strategies take place on farms by farmers. Therefore, farmers’ engagement in one mitigating
strategy may be related to their engagement in other mitigating strategies. As a result, it may be
valuable to estimate farmers’ engagement in all mitigating strategies jointly by using a
multivariate regression analysis (e.g., multivariate probit modelling technique). The literature
review undertaken for this study found that no previous study explored factors that are associated
with SSI farmers’ mitigating strategies to reduce the transmission of WRZD with respect to on-
farm water issues. Consequently, the approach in my research was to focus on modeling
individual mitigating strategies separately by using a univariate probit regression technique.
Beyond the context of the research results reported in this dissertation, I plan to use more
advanced techniques (e.g., multivariate probit regression) to explore further the complexity of
water quality, public health, and farmers’ behaviours/decision making process in mitigating
strategies to reduce the transmission of WRZD.
6.6. Overall Recommendations
Use an Ecosystem approach to health in Vietnam in the context of SSI farming:
An EcoHealth approach is considered to be an excellent approach for community
development around the world and good examples of the application of the EcoHealth approach
Page 245 of 340
exist in Vietnam. An EcoHealth approach considers both human and animal health as well as
environmental health through equitable and sustainable methods (IDRC, 2012). An EcoHealth
approach to the SSI farming model for both farmers and policy makers is recommended for
consideration in Vietnam. Revising the SSI farming model and related policies with respect to an
EcoHealth approach may ensure that the model will not be merely a simple agricultural
economic model in which food and income are outputs. Instead, it should be a more holistic
EcoHealth-based model in which the health of farmers and animals is interwoven with on-farm
livestock production and the protection of the environment, and these are considered to be an
equally important output of the model.
Implementing a transdisciplinary approach: The complexity of SSI farming requires
the knowledge and involvement of various disciplines such as veterinary health, public health,
water security, the private sector, and so on, to not only improve farming practice and generate
income but also to improve the health of farmers and their animals and to avoid polluting the
environment.
More active and engaged community participation: To optimize the limited water
resources available for farmers and to reduce the negative socio-economic and health impacts on
the rural population, it is important to enhance community participation to a greater extent not
only from policy makers but also from farmers themselves in the policy making process. This
enhancement could help to more explicitly address the issues relating to small-scale integrated
farming such as water, health, and small-scale livestock production.
Improve rural water quality: In designing rural water interventions and developing
rural water policies, it is important to consider the fact that SSI farmers depended on multiple
sources of water on farms for drinking and domestic purposes, and that there are differences in
Page 246 of 340
frequencies of use between provinces (i.e., Thai Binh and An Giang). The microbial quality of
water used on SSI farms is also low and this challenge needs to be addressed. In addition to
having simple affordable tests of E. coli in water, it is also important to build the capacity of SSI
farmers and the government for inspecting and improving household water sanitation, and for
on-farm water public health management. By doing so, rural farmers will not only be able to
afford and conduct the tests, but also know when it is appropriate to take the test and what to do
with the test results. Therefore, any efforts to improve the microbial quality of water used on SSI
farms should focus on providing better education for SSI farmers, improving farmers’ income,
increasing awareness among SSI farmers of the potential harm caused by using untreated water,
and sharing the experiences of farms with fish production and farms that have had past
experiences with HPAI in poultry production.
Increase farmers ‘engagement in mitigating strategies to reduce WRZD
transmission: Multiple factors are associated with SSI farmers’ engagement in mitigating
strategies to reduce the transmission of WRZD. Policies and programs to increase farmers’
engagement in the four mitigating strategies to reduce transmission of WRZD should focus on
key factors such as improving SSI farmers’ perceived self-ability in taking livestock
management actions and increasing their access to advice from health/veterinary workers. It is
also important to improve farmers’ education, promote women’s role in family health, increase
awareness of susceptibility to HPAI from using untreated wastewater, encourage larger scale pig
production, and increase peer pressure among SSI farmers. Furthermore, it is important to
improve farmers’ income while also offering affordable services relating to rural water quality
and livestock management for SSI farmers.
Page 247 of 340
6.7. Conclusions/Final Thoughts
This dissertation has provided insights into the SSI farming model in Vietnam with
respect to livestock production, on-farm water quality, and public health, and also into an
EcoHealth approach. Results of my research have potential implications for future research,
interventions, and policy advocacy with respect to rural water policies, mixed agriculture
policies, and delivery of animal and environmental health services. The EcoHealth approach has
been considered a valuable and sustainable approach for community development. In Vietnam,
there are some examples of the EcoHealth approach in academic research, policy advocacy, and
community development. The SSI farming model in Vietnam needs to be reviewed and studied
further using an EcoHealth approach and considering risks for increased transmission of WRZD.
An EcoHealth approach may help SSI farmers to enhance their public health water management
and turn the SSI farming model into a holistic integrated farming model, which not only
maximizes beneficial impacts but also reduces any adverse health impacts of the model. In such
a model, the health of farmers, animals and the environment should be interwoven with livestock
production and be considered an equally important output of the SSI farming model.
Page 248 of 340
References
Carr, R. and Bartram, J. (2004). The control envelope and risk management. In (pp. 67-92). IWA
Publishing.
Dang, K. N., Le, T. D., Nguyen, V. S., and Verdengem, M. (2005). Development of "VAC"
Integrated Farming Systems in the Mekong Delta, Vietnam - A View of a System and a
Participatory Approach.
Devendra, C. and Thomas, D. (2002). Smallholder farming systems in Asia. Agricultural
Systems, 71, 17-25.
Dixon, J., Tanyeri-Abur, A., and Wattenbach, H. (2004). Framework for analyzing impacts of
globalization on smallholders. Food and Agriculture Organization of the United Nations
(FAO) [On-line]. Available: http://www.fao.org/docrep/007/y5784e/y5784e02.htm
GSO Vietnam (2013). General poverty rate by residence and by region and Year - Rural. GSO
Vietnam [On-line]. Available: http://www.gso.gov.vn/default_en.aspx?tabid=783
Hall, D. C., Thao, T. D., Minh, D. V., and Lien, L. V. (2006). Competitiveness of the livestock
sector in Vietnam World Bank paper. WB-EASRD and FAO-TCIP.
Heidhues, F. and Bruntrup, M. (2003). Subsistence agriculture in Central and Eastern Europe:
how to break the vicious circle? (vols. 22) Halle (Saale): Leibniz Institute of Agricultural
Development in Central and Eastern Europe.
IDRC (2012). Ecohealth Research in Practice: Innovative Applications of an Ecosystem
Approaches to Health. International Development Centre (IDRC); Springer.
Page 249 of 340
Irini, M. and Rapsomanikis, G. (2005). Contribution of Livestock to Household income in
Vietnam: A household typology based analysis (Rep. No. PPLPI Working Paper N.21). A
Living from Livestock, FAO.
Marteau, T. M., Hollands, G. J., and Fletcher, P. C. (2012). Changing Human Behavior to
Prevent Disease: The Importance of Targeting Automatic Processes. Science, 337, 1492-
1495.
Moe, C. L. (2004). What are the criteria for determining whether a disease is zoonotic and water
related. In World Health Organization (WHO). Waterborne Zoonoses: Identification,
Causes and Controls (pp. 27-45). London: IWA Publishing.
MoH (2009). National technical regulation on domestic water quality (QCVN 02: 2009/BYT).
Hanoi: Department of Preventive Medicine and Environment, Ministry of Health,
Socialist Republic of Vietnam.
MoH (2009). National technical regulation on drinking water quality (QCVN 01: 2009/BYT).
Hanoi: Department of Preventive Medicine and Environment, Ministry of Health,
Socialist Republic of Vietnam.
Neal, D., Vujcic, J., Hernandez, O., and Wood, W. (2014). Handwashing and the Science of
Habit. Sanitation Updates [On-line]. Available:
https://sanitationupdates.files.wordpress.com/2014/12/neal-handwashing2014.pdf
Nguyen, M. T. N. (2014). Translocal Householding: Care and Migrant Livelihoods in a Waste-
trading Community of Vietnam's Red River Delta. Development and Change, 45, 1385-
1408.
Page 250 of 340
Pica-Ciarmarra, U., Tasciotti, L., Otte, J., and Zezza, A. (2011). Livestock Assets, Livestock
Income and Rural Households: Cross-Country Evidence from Household Surveys (Rep.
No. ESA Working Paper No. 11-17). Rome: World Bank, FAO, AU-IBAR, ILRI.
Rabbi, S. and Dey, N. (2013). Exploring the gap between hand washing knowledge and practices
in Bangladesh: a cross-sectional comparative study. BMC Public Health, 13, 89.
Vindigni, S. M., Riley, P. L., and Jhung, M. (2011). Systematic review: handwashing behaviour
in low- to middle-income countries: outcome measures and behaviour maintenance.
Tropical Medicine and International Health, 16, 466-477.
WHO (2004). Waterborne Zoonoses: Identification, Causes, and Control. IWA Publishing.
WHO (2012). Rapid assessment of drinking water quality - A handbook for implementation.
World Health Organization.
Page 251 of 340
APPENDIX A: THEORETICAL MODEL
Page 252 of 340
APPENDIX B: QUESTIONNAIRES USED DURING THE FIELD STUDY
B.1. Questionnaires in Vietnamsese:
1. Mã số hộ nông dân trả lời bảng hỏi
2. Giới tính?
1. Nam
2. Nữ
3. Khác
3. Ngày tháng năm sinh của Quý vị là gì?
__________
4. Quý vị dân tộc gì?
1. Kinh
2. Khác: ghi rõ
5. Trình độ học vấn cao nhất của Quý vị là gì?
6. Gia đình Quý vị có bao nhiêu nhân khẩu?
Page 253 of 340
7. Số nhân khẩu dưới 18 tuổi?
8. Quý vị đã làm nghề vừa chăn nuôi vừa thả cá, trồng trọt được bao nhiêu năm rồi?
9. Ngoài nghề nông, Quý vị còn làm nghề gì khác?
10. Nhìn chung, ai là người chịu trách nhiệm CHÍNH về chăn nuôi gia súc gia cầm, thả cá, và
vấn đề sức khỏe gia đình?
Nam Nữ Cả hai Không
chắc chắn
Từ chối
không trả
lời
Chăn nuôi gia súc gia cẩm ❏ ❏ ❏ ❏ ❏
Page 254 of 340
Thả cá ❏ ❏ ❏ ❏ ❏
Sức khỏe gia đình ❏ ❏ ❏ ❏ ❏
11. Vị trí của Quý vị với chủ hộ là gì?
1. Chủ hộ
2. Vai trò khác
12. Quý vị nhìn nhận về điều kiện môi trường nước nói chung của thôn mình như thế nào?
1. Rất tốt
2. Tốt
3. Bình thường
4. Kém
5. Rất kém
6. Không có ý kiến
13. Quý vị có cho rằng nguồn nước chưa qua xử lý của quý vị (chưa lọc hoăc chưa đun sôi)
có an toàn để uống không?
1. Có
2. Không
14. Quý vị có xử lý nước uống và nước sinh hoạt của mình không?
Lọc Đun sôi Cả lọc
và đun
sôi
Cách
khác
Không
xử lý
Không
có ý
kiến
Nước uống ❏ ❏ ❏ ❏ ❏ ❏
Nước sinh hoạt ❏ ❏ ❏ ❏ ❏ ❏
15. Có vấn đề gì đã xảy ra trong quá khứ khiến Quý vị lọc, đun sôi (hoăc cả hai) nước uống?
1. Không
2. Có: hãy ghi rõ vấn đề gì_____________________________
Page 255 of 340
16. Nếu có (lọc, đun sôi, cả hai) thì theo quý vị nước đã lọc và/hoặc đã đun sôi có an toàn để
uống không?
Có phải cả lọc và đun sôi Chỉ lọc Chỉ đun sôi Không
17. Hãy điền vào mức độ nguy hại cho sức khỏe của những nguồn nước sau (chưa lọc hoặc
chưa đun sôi).
Cực kỳ
hại
Nguy hại
cao
Nguy hại
vừa
Nguy hại
thấp
Không
nguy hại
Không
có ý kiến
Nước uống (chưa xử lý) ❏ ❏ ❏ ❏ ❏ ❏
Nước sinh hoạt (chưa xứ lý) ❏ ❏ ❏ ❏ ❏ ❏
Nước ao hồ (chưa xử lý) ❏ ❏ ❏ ❏ ❏ ❏
Nước đóng chai ❏ ❏ ❏ ❏ ❏ ❏
18. Bây giờ tôi muốn hỏi quý vị về cúm chim/cúm gia cầm: Quý vị đã bao giờ nghe đến cúm
chim/cúm gia cầm chưa?
1. Có, đã nghe đến cúm chim/cúm gia cầm
2. Không, chưa bao giờ nghe đến cúm chim/cúm gia cầm
19. Hãy điền vào khả năng nhiễm cúm chim từ việc sử dụng những nguồn nước sau (chưa
đun sôi hoặc chưa lọc)
Cực cao Cao Vừa Thấp Không
có khả
năng
Không
có ý kiến
Nước uống (chưa xử lý) ❏ ❏ ❏ ❏ ❏ ❏
Nước sinh hoạt (chưa xứ lý) ❏ ❏ ❏ ❏ ❏ ❏
Nước ao hồ (chưa xử lý) ❏ ❏ ❏ ❏ ❏ ❏
Nước thải chăn nuôi, gia cầm
(chưa xử lý) ❏ ❏ ❏ ❏ ❏ ❏
20. Nếu nhiễm cúm chim, Quý vị nghĩ mức độ nặng của bệnh sẽ như thế nào?
1. Rất nặng
2. Nặng
3. Vừa
Page 256 of 340
4. Nhẹ
5. Không sao
6. Không có ý kiến
21. Gia cầm của Quý vị đã từng bị tiêu hủy do dịch bệnh cúm gia cầm tại địa phương chưa?
1. Chưa từng bị hủy
2. Đã từng bị hủy
3. Không nhớ
4. Không có ý kiến
22. Quý vị sẽ làm gì nếu phát hiện gia súc, gia cầm, chim bị ốm hoặc chết?
1.Bỏ qua
2.Tham khảo ý kiến các nông dân khác
3.Thông báo cho trưởng thôn
4.Chôn tiêu hủy
5.Báo cho cán bộ phụ trách thú y địa phương
6.Sử dụng như thức ăn (cho người hoặc gia súc gia cầm)
7.Không có ý kiến
8.Khác: đề nghị nêu rõ
23. Hiện tại gia cầm của Quý vị đã được tiêm phòng dịch cúm gia cầm chưa?
1. Tiêm rồi
2. Chưa tiêm
3. Không có ý kiến
24. Quý vị đã bao giờ nghe đến vi trùng tồn tại trong nước như vi rút, vi khuẩn, giun sán
chưa?
1. Đã nghe qua
2. Chưa bao giờ
25. Hãy điền vào khả năng mắc bệnh tiêu chảy từ việc sử dụng những nguồn nước sau (chưa
lọc hoặc chưa đun sôi).
Rất cao Cao Vừa Thấp Không Không có ý kiến
Nước uống (chưa xử lý) ❏ ❏ ❏ ❏ ❏ ❏
Nước sinh hoạt (chưa xứ lý) ❏ ❏ ❏ ❏ ❏ ❏
Nước ao hồ (chưa xử lý) ❏ ❏ ❏ ❏ ❏ ❏
Nước thải chăn nuôi, gia cầm
(chưa xử lý) ❏ ❏ ❏ ❏ ❏ ❏
Page 257 of 340
26. Nếu mắc tiêu chảy, Quý vị nghĩ mức độ nặng của bệnh sẽ như thế nào?
1. Rất nặng
2. Nặng
3. Vừa
4. Nhẹ
5. Không sao
6. Không có ý kiến
27. Một trong những loại vi trùng có thể tìm thấy trong nước ở nông trại và đôi khi gây nguy
hại là E.coli. Đây là loại vi khuẩn nhỏQuý vị đã bao giờ nghe đến E.coli hay bệnh do
nhóm coliform gây ra chưa?
1. Có, chuyển sang câu kế tiếp và tiếp tục
2. Không, bỏ qua câu kế tiếp và tiếp tục
28. Điền khả năng nhiễn bệnh nhóm coliform từ nguồn nước sau (chưa đun sôi hay lọc).
Rất cao Cao Vừa Thấp Không Không có ý
kiến
Nước uống (chưa xử lý) ❏ ❏ ❏ ❏ ❏ ❏
Nước sinh hoạt (chưa xứ lý) ❏ ❏ ❏ ❏ ❏ ❏
Nước ao hồ (chưa xử lý) ❏ ❏ ❏ ❏ ❏ ❏
Tiếp xúc nước thải chăn nuôi,
gia cầm (chưa xử lý) ❏ ❏ ❏ ❏ ❏ ❏
29. Nếu Quý vị mắc bện từ nhóm coliform, mức độ nặng của bệnh sẽ thế nào?
1. Rất nặng
2. Nặng
3. Vừa
4. Nhẹ
5. Không sao
6. Không có ý kiến
30. Một nguồn khác có thể gây bệnh từ nước của nông trại và đôi khi gây nguy hại là giun
sán.Qúy vị đã bao giờ nghe đến giun sán chưa?
1. Có, trả lời câu kế tiếp và tiếp tục
2. Chưa, bỏ qua 2 câu kế tiếp và tiếp tục
Page 258 of 340
31. Điền khả năng nhiễm giun sán từ những nguồn nước sau (chưa đun sôi hoặc lọc).
Rất cao Cao Vừa Thấp Không Không có ý
kiến
Nước uống (chưa xử lý) ❏ ❏ ❏ ❏ ❏ ❏
Nước sinh hoạt (chưa xứ lý) ❏ ❏ ❏ ❏ ❏ ❏
Nước ao hồ (chưa xử lý) ❏ ❏ ❏ ❏ ❏ ❏
Tiếp xúc nước thải chăn nuôi,
gia cầm (chưa xử lý) ❏ ❏ ❏ ❏ ❏ ❏
32. Nếu quý vị mắc bệnh giun sán, mức độ nặng nhẹ sẽ như thế nào?
1. Rất nặng
2. Nặng
3. Vừa
4. Nhẹ
5. Không sao
6. Không có ý kiến
33. Mức độ quan trọng/nghiêm trọng của các vấn đề liên quan đến việc sử dụng nước trong
việc giảm nguy cơ mắc bệnh truyền nhiễm là gì?
Cực kỳ
quan trọng
Quan
trọng cao
Vừa Ít quan
trọng
Không quan
trọng
Không có ý
kiến
Nguồn nước ❏ ❏ ❏ ❏ ❏ ❏
Không dùng chung
nguồn nước người và
vật nuôi
❏ ❏ ❏ ❏ ❏ ❏
Khoảng cách nguồn
nước và vật nuôi ❏ ❏ ❏ ❏ ❏ ❏
Thiếu nước ❏ ❏ ❏ ❏ ❏ ❏
Lượng mưa ❏ ❏ ❏ ❏ ❏ ❏
Chất lượng ❏ ❏ ❏ ❏ ❏ ❏
Vấn đề khác ❏ ❏ ❏ ❏ ❏ ❏
34. Quan điểm hiện tại của Quý vị về tầm quan trọng của việc sử dụng nước đúng cách trong
việc giảm nguy cơ các bệnh truyền nhiễm cũng như cúm gia cầm như thế nào?
1. Rất quan trọng
2. Quan trọng
Page 259 of 340
3. Quan trọng vừa
4. Quan trọng ít
5. Không quan trọng
6. Không có ý kiến
35. Về tổng thể, mức độ hài lòng với nguồn nước sau mà Quý vị đang sử dụng là gì?
Hoàn toàn hài
lòng
Hài lòng
cao
Hài lòng
vừa
Ít hài
lòng
Không
hài lòng
Không có ý
kiến
Để uống ❏ ❏ ❏ ❏ ❏ ❏
Cho SXNN ❏ ❏ ❏ ❏ ❏ ❏
36. Xin hãy cho biết suy nghĩ của Quý vị về mức độ nguy cơ mắc các bệnh truyền nhiễm từ
những nguồn nước sau đây?
Nguy cơ
rất cao
Nguy cơ
cao
Nguy cơ
vừa
Nguy cơ
ít/thấp
Không
có nguy
cơ
Không
có ý kiến
Mưa ❏ ❏ ❏ ❏ ❏ ❏
Hồ ❏ ❏ ❏ ❏ ❏ ❏
Sông ❏ ❏ ❏ ❏ ❏ ❏
Kênh ❏ ❏ ❏ ❏ ❏ ❏
Giếng đào ❏ ❏ ❏ ❏ ❏ ❏
Giếng khoan ❏ ❏ ❏ ❏ ❏ ❏
Nguồn khác ❏ ❏ ❏ ❏ ❏ ❏
37. Những yếu tố nào dưới đây có tác động đến cách xác định chất lượng nước của Quý vị?
1. Dựa vào ấn tượng tạo ra bởi bất cứ một chất nào trong nước và nước
SXNN, mùi, vị, mầu
2. Theo kinh nghiệm
3. Theo quan niệm về rủi ro
4. Thái độ đối với việc ô nhiễm hóa chất đối với nước
5. Những biểu hiện liên quan đến nguồn nước
6. Quen với nguồn nước cụ thể
7. Sự tin tưởng vào nhà cung cấp vào nguồn nước, nước thải
8. Truyền thông
Page 260 of 340
9. Không có ý kiến
10. Các yếu tố khác: cụ thể
38. Quý vị có hoặc lấy thông tin về các bệnh truyền nhiễm từ nguồn nào dưới đây và mức độ
tin cậy của những nguồn nay ra sao?
Rất tin
cậy
Tin cậy
cao
Khá tin
cậy
It tin cậy Không
tin cậy
Không
có ý kiến
Ti vi ❏ ❏ ❏ ❏ ❏ ❏
Báo chí/ tờ rơi ❏ ❏ ❏ ❏ ❏ ❏
Loa đài ❏ ❏ ❏ ❏ ❏ ❏
Gia đình/bạn bè/nông dân khác ❏ ❏ ❏ ❏ ❏ ❏
Cán bộ/ trung tâm y tế ❏ ❏ ❏ ❏ ❏ ❏
Cán bộ trung tâm thú y ❏ ❏ ❏ ❏ ❏ ❏
Nguồn khác ❏ ❏ ❏ ❏ ❏ ❏
39. Theo Quý vị, việc áp dụng biện pháp dự phòng có giúp chống lại các bệnh truyền nhiễm
không?
1. Không
2. Có, ví dụ như rửa tay
3. Có, ví dụ như cô lập nguồn nhiễm (người, động vật, nước)
4. Có, ví dụ như đeo các thiết bị bảo hộ (khẩu trang)
5. Có, ví dụ như tiêm phòng vắc- xin
6. Có, ví dụ như dùng thuốc
7. Có, ví dụ như nấu chín/không ăn thức ăn tươi sống
8. Không có ý kiến
9. Biện pháp khác
40. Co lợi ích nào khác (ngoài sức khỏe) đã khuyến khích Quý vị hành động phòng chống
các bệnh truyền nhiễm? Đề nghị nêu rõ
Page 261 of 340
41. Co lợi ích nào đã khuyến khích Quý vị KHÔNG hành động phòng chống các bệnh truyền
nhiễm lây qua đường nước có nguồn gốc từ động vật? Đề nghị nêu rõ.
42. Quý vị KHÔNG hành động phòng chống bệnh truyền nhiễm do các yếu tố nào sau đây?
1. Không đủ điều kiện kinh tế để thực hiện
2. Không biết cách (kỹ năng)
3. Không biết (hiểu biết)
4. Hàng xóm không hành động
5. Quá bận với những việc khác
6. Không có ý kiến
7. Yếu tố khác
43. Yếu tố nào sau đây tác động/ảnh hưởng đến việc Quý vị ra quyết định thực hiện hành
động phòng chống các bệnh truyền nhiễm?
1. Hàng xóm đã làm tương tự
2. Khuyến cáo từ cán bộ sức khỏe cộng đồng, cán bộ thú y chuyên trách
3. Báo đài
4. không bận với những việc khác
5. Lo lắng về việc bị lây nhiễm hoặc bị ảnh hưởng bởi bệnh truyền nhiễm
6. Mất thu nhập
7. Không có ý kiến
8. Yếu tố khác
44. Phần này, tôi sẽ phỏng vấn Quý vị về đặc điểm hoạt động chăn nuôi gia súc gia cầm, thủy
sản, và trồng trọt của Quý vị.Quý vị có những hoạt động sản xuất nông nghiệp nào?
1. Trồng lúa
2. Sắn
Page 262 of 340
3. Tre
4. Ngô
5. Lạc/đậu
6. Rau
7. Hoa quả
8. Fish production
9. Livestock production
45. Ba nguồn thông tin quan trọng nhất quý vị nhận được về quản lý gia trại là gì? (điền số 1,
2, 3 of 9):
Nông dân khác __________
Thành viên gia đình __________
Báo in, tài liệu tuyên truyền __________
Đài, tivi __________
Cán bộ chuyên trách xã __________
Tập huấn __________
Công ty bán hàng nông nghiệp __________
Tổ chức phi chính phủ __________
Nguồn khác __________
46. Quý vị có loại hình chăn nuôi (gia súc, gia cầm, thả cá) nào
Ước tính số con
(bé/con) thương có một
vụ
Ước tính số con
(lớn/trưởng thành)
thường có một vụ
Ước tính số kg
thường có một vụ
Ghi chú
Thả cá
Gà
Vịt
Lợn
Dê
Gia cầm
khác
Gia súc
khác
Vật nuôi
khác
47. Đối với NƯỚC UỐNG, Quý vị dùng từ những nguồn nước nào và tần suất sử dụng
Page 263 of 340
những nguồn này như thế nào?
Rất thường
xuyên
Thường
xuyên
Thỉnh
thoảng
Ít khi Không
bao giờ
Không có ý kiến
Ao/hồ ❏ ❏ ❏ ❏ ❏ ❏
Sông/kênh ❏ ❏ ❏ ❏ ❏ ❏
Nước mưa ❏ ❏ ❏ ❏ ❏ ❏
Nước máy ❏ ❏ ❏ ❏ ❏ ❏
Giếng đào ❏ ❏ ❏ ❏ ❏ ❏
Giếng khoan ❏ ❏ ❏ ❏ ❏ ❏
Nước mua (đóng
chai/bình) ❏ ❏ ❏ ❏ ❏ ❏
Nguồn khác ❏ ❏ ❏ ❏ ❏ ❏
48. Đối với nước SXNN, Quý vị dùng từ những nguồn nước nào và tần suất sử dụng những
nguồn này như thế nào?
Rất thường
xuyên
Thường
xuyên
Thỉnh
thoảng
Ít khi Không
bao giờ
Không có ý
kiến
Ao/hồ ❏ ❏ ❏ ❏ ❏ ❏
Sông/kênh ❏ ❏ ❏ ❏ ❏ ❏
Nước mưa ❏ ❏ ❏ ❏ ❏ ❏
Nước máy ❏ ❏ ❏ ❏ ❏ ❏
Giếng đào ❏ ❏ ❏ ❏ ❏ ❏
Giếng khoan ❏ ❏ ❏ ❏ ❏ ❏
Nước mua ❏ ❏ ❏ ❏ ❏ ❏
Nguồn khác ❏ ❏ ❏ ❏ ❏ ❏
49. Gia cầm của Quý vị có bị nhiễm cúm gia cầm trong đợt dịch năm 2004?
1. Có
2. Không
3. Không nhớ
4. Không có ý kiến
50. Gia cầm của Quý vị có bị nhiễm cúm gia cầm trong vòng 12 tháng qua không?
1. Có
2. Không: Bỏ qua 2 câu kế tiếp
Page 264 of 340
3. Không nhớ
4. Không có ý kiến
51. Chất thải chăn nuôi của Quý vị có được xả trực tiếp (không qua xử lý) vào nguồn nước
dùng trong sản xuất nông nghiệp không?
1. Không
2. Có xả trực tiếp vào ao, hồ, đầm thả cá
3. Có xả trực tiếp vào kênh, rạch, sông, ngòi
4. Không có ý kiến
5. Có xả trực tiếp vào nơi khác: ghi rõ____________________
52. Thu nhập đầu người hàng năm của quý vị là khoảng bao nhiêu?
1. < 4.8 triệu đồng
2. 4.8-10 triệu đồng
3. 10-20 triệu đồng
4. 20-40 triệu đồng
5. >40 triệu đồng
6. Không có ý kiến
53. Quý vị bảo vệ bản thân khỏi nguy cơ mắc bệnh truyền nhiễm khi tiếp xúc với nước
SXNN như thế nào?
1. Mặc đồ bảo hộ
2. Đeo găng tay bảo vệ tay
3. Rửa chân tay sau khi làm việc ngoài đồng ruộng
4. Không bảo vệ
5. Không có ý kiến
6. Dùng biện pháp khác
54. Quý vị có áp dụng những biện pháp/hành động dưới đây không?
1. Tiêm phòng cho vật nuôi
2. Đảm bảo điều kiện chăn nuôi sạch sẽ đạt tiêu chuẩn cho vật nuôi
3. Xử lý chất thải vật nuôi tại nguồn
4. Cách ly vật nuôi khỏi các vùng đánh bắt
5. Không sử dụng kháng sinh làm chất tăng trọng
6. Không áp dụng biện pháp nào nêu trên
7. Không có ý kiến
Page 265 of 340
55. Tầm quan trọng của những biện pháp sau trong việc tránh nguy cơ mắc các bệnh truyền
nhiễm là gì?QUẢN LÝ GIA SÚC GIA CẦM
Rất quan
trọng
Quan
trọng
Quan
trọng
vừa
Ít quan
trọng
Không
quan trọng
Không
có ý kiến
Tiêm phòng cho vật nuôi ❏ ❏ ❏ ❏ ❏ ❏
Đảm bảo điều kiện chăn nuôi
sạch sẽ đạt tiêu chuẩn cho vật
nuôi
❏ ❏ ❏ ❏ ❏ ❏
Xử lý chất thải vật nuôi tại
nguồn ❏ ❏ ❏ ❏ ❏ ❏
Cách ly vật nuôi khỏi các vùng
đánh bắt ❏ ❏ ❏ ❏ ❏ ❏
Không sử dụng kháng sinh làm
chất tăng trọng ❏ ❏ ❏ ❏ ❏ ❏
56. Quý vị có áp dụng những biện pháp/hành động dưới đây không?
1. Xử lý nước thải
2. Xử lý chất thải động vật/con người tại nguồn
3. Sử dụng vùng đệm (ví dụ: bèo)
4. Bảo vệ, che chắn nguồn nước từ giếng đào/giếng khoan/suối
5. Che/chắn/đậy nguồn hoặc bể đựng nước uống
6. Sửa chữa các chỗ thất thoát trong hệ thống nước thải
7. Xử lý chất thải từ động vật và người trước khi sử dụng làm phân bón
8. Không áp dụng biện pháp nào nêu trên
9. Không có ý kiến
57. Tầm quan trọng của những biện pháp sau trong việc tránh nguy cơ mắc các bệnh truyền
nhiễm là gì?BẢO VỆ NGUỒN NƯỚC
Rất quan
trọng
Quan
trọng
Quan
trọng vừa
Ít quan
trọng
Không
quan trọng
Không có ý
kiến
Xử lý nước thải ❏ ❏ ❏ ❏ ❏ ❏
Xử lý chất thải động vật/con
người tại nguồn ❏ ❏ ❏ ❏ ❏ ❏
Sử dụng vùng đệm (ví dụ:
bèo) ❏ ❏ ❏ ❏ ❏ ❏
Page 266 of 340
Bảo vệ, che chắn nguồn
nước từ giếng đào/giếng
khoan/ suối
❏ ❏ ❏ ❏ ❏ ❏
Che/chắn/đậy nguồn hoặc bể
đựng nước uống ❏ ❏ ❏ ❏ ❏ ❏
Sửa chữa các chỗ thất thoát
trong hệ thống nước thải ❏ ❏ ❏ ❏ ❏ ❏
Xử lý chất thải từ động vật
và người trước khi sử dụng
làm phân bón
❏ ❏ ❏ ❏ ❏ ❏
58. Quý vị có áp dụng những biện pháp/hành động dưới đây không?
1. Che bể nươc hoặc chứa nước vào thùng, bể
2. Đầu tư vào hệ thống xử lý nước sinh hoạt gia đình
3. Thay/sử dụng chất tẩy rửa nước
4. Sửa chữa đồ chứa nước trong gia đình
5. Sửa chữa hệ thống cống thoát của gia đình
59. Tầm quan trọng của những biện pháp sau trong việc tránh nguy cơ mắc các bệnh truyền
nhiễm là gì?LƯU TRỮ, XỬ LÝ, VÀ PHÂN PHỐI NƯỚC
Rất quan
trọng
Quan
trọng
Quan
trọng
vừa
Ít quan
trọng
Không
quan trọng
Không
có ý
kiến
Chứa nước vào thùng, bể ❏ ❏ ❏ ❏ ❏ ❏
Đầu tư vào hệ thống xử lý nước
sinh hoạt gia đình ❏ ❏ ❏ ❏ ❏ ❏
Thay/sử dụng chất tẩy rửa nước ❏ ❏ ❏ ❏ ❏ ❏
Sửa chữa đồ chứa nước trong
gia đình ❏ ❏ ❏ ❏ ❏ ❏
Sửa chữa hệ thống cống thoát
của gia đình ❏ ❏ ❏ ❏ ❏ ❏
60. Quý vị có áp dụng những biện pháp/hành động dưới đây không?
1. Nâng cao vệ sinh cá nhân và trong gia đình (ví dụ rửa tay xà
phòng)
2. Vệ sinh gia trại hợp lý/xử lý chất thải gia dinh tốt
3. Đầu tư nâng cấp nguồn nước
4. Phòng các lây nhiễm chéo
5. Nâng cao vệ sinh thực phẩm (ví dụ rửa rau thịt)
Page 267 of 340
6. Nấu chín/không ăn đồ ăn sống
7. Bảo vệ nước và nguồn cung cấp thức ăn cho động vật
8. Trao đổi thông tin về nguy cơ đối với các thành viên khác trong gia
đình, cộng đồng
9. Đeo thiết bị bảo hộ (như khẩu trang) hay dùng nước lọc
10. Dùng thuốc kháng sinh
61. Tầm quan trọng của những biện pháp này trong việc tránh nguy cơ mắc các bệnh truyền
nhiễm là gì?TẠI ĐIỂM SỬ DỤNG (HỘ GIA ĐÌNH)
Rất quan
trọng
Quan
trọng
Quan trọng
vừa
Ít quan
trọng
Không
quan trọng
Không có
ý kiến
Nâng cao vệ sinh cá nhân và
trong gia đình (ví dụ rửa tay
xà phòng)
❏ ❏ ❏ ❏ ❏ ❏
Xử lý chất thải gia đình tốt ❏ ❏ ❏ ❏ ❏ ❏
Đầu tư nâng cấp nguồn nước ❏ ❏ ❏ ❏ ❏ ❏
Phòng các lây nhiễm chéo ❏ ❏ ❏ ❏ ❏ ❏
Nâng cao vệ sinh thực phẩm
(ví dụ rửa rau thịt) ❏ ❏ ❏ ❏ ❏ ❏
Nấu chín/không ăn đồ ăn
sống ❏ ❏ ❏ ❏ ❏ ❏
Bảo vệ nước và nguồn cung
cấp thức ăn cho động vật ❏ ❏ ❏ ❏ ❏ ❏
Trao đổi thông tin về nguy
cơ đối với các thành viên
khác trong gia đình, cộng
đồng
❏ ❏ ❏ ❏ ❏ ❏
Đeo thiết bị bảo hộ (như
khẩu trang) hay dùng nước
lọc
❏ ❏ ❏ ❏ ❏ ❏
Dùng thuốc kháng sinh ❏ ❏ ❏ ❏ ❏ ❏
62. Quý vị thường xuyên xử lý chất thải chăn nuôi như thế nào?
1. Thu lại
2. Thải ra sông/hồ/cánh đồng
3. Đổ ở bãi thải gần nhà
4. Chôn
5. Đốt
6. Khác: Đề nghị cho biết cụ thể
Page 268 of 340
7. Không có ý kiến
63. Quý vị có sử dụng chất thải của người, nước thải/chất thải chăn nuôi cho việc thả cá
không?
1. Có
2. Không: Bỏ qua 2 câu kế tiếp
3. Không có ý kiến
Water testing results questions:
pH levels:
64. What are pH levels in on-farm stored rainwater for drinking?
65. What are pH levels in on-farm pipe water for drinking?
66. What are pH levels in on-farm well water for drinking?
67. What are pH levels in on-farm bottled water for drinking?
68. What are pH levels in on-farm river water (fl.) for drinking?
69. What are pH levels in on-farm stored rainwater for domestic purposes?
70. What are pH levels in on-farm well water for domestic purposes?
71. What are pH levels in on-farm river water (fl.) for domestic purposes?
72. What are pH levels in on-farm pond water for domestic purposes?
Turbidity levels:
73. What are turbidity levels in on-farm stored rainwater for drinking?
74. What are turbidity levels in on-farm pipe water for drinking?
75. What are turbidity levels in on-farm well water for drinking?
76. What are turbidity levels in on-farm bottled water for drinking?
77. What are turbidity levels in on-farm river water (fl.) for drinking?
78. What are turbidity levels in on-farm stored rainwater for domestic purposes?
79. What are turbidity levels in on-farm well water for domestic purposes?
80. What are turbidity levels in on-farm river water (fl.) for domestic purposes?
81. What are turbidity levels in on-farm pond water for domestic purposes?
E. coli cfu levels:
82. What are E. coli cfu levels in on-farm stored rainwater for drinking?
83. What are E. coli cfu levels in on-farm pipe water for drinking?
84. What are E. coli cfu levels in on-farm well water for drinking?
85. What are E. coli cfu levels in on-farm bottled water for drinking?
86. What are E. coli cfu levels in on-farm river water (fl.) for drinking?
87. What are E. coli cfu levels in on-farm stored rainwater for domestic purposes?
88. What are E. coli cfu levels in on-farm well water for domestic purposes?
89. What are E. coli cfu levels in on-farm river water (fl.) for domestic purposes?
90. What are E. coli cfu levels in on-farm pond water for domestic purposes?
Page 269 of 340
B.2. Questionnaires in English:
1. Code of the participating farmer
2. What is your gender?
1. Male
2. Female
3. Other
3. When were you born? ___________(dd/mm/yy)
4. What is your ethnicity?
1. Kinh
2. Other: specify
5. What was your highest school attended?
6. How many people are there in your house?
7. How many kids (at the age under 18) are there in your house?
Page 270 of 340
8. How many years have you been an integrated farmer?
9. What is your occupation other than farming?
10. Overall, who is primarily responsible for livestock production and health of the family?
Issue Main responsibility
Man Woman Both Not sure Do not want to
answer
Livestock production
Fish production
Family health
11. What is your relationship with other household members?
1. Head of the household
2. Other roles
12. How do you see the general environmental water condition of your village?
1. Excellent
2. Good
3. Adequate
4. Poor
5. Very poor
6. Do not want to answer
Page 271 of 340
13. Do you think that your source of untreated water (not boiled or filtered) for drinking is safe?
1. Yes
2. No
14. Do you treat your drinking or household water?
Filter Boil Both boil and
filter
Other
(state)
No
treatment
Do not
want to
answer
Drinking
water ❏ ❏ ❏ ❏ ❏ ❏
Household
water ❏ ❏ ❏ ❏ ❏ ❏
15. If yes (both, or filter, or boil) then do you treat because of a past event?
1. No
2. Yes (please state what event) _________________________________
16. If yes (both, filter, boil) then do you think your filtered and/or boiled drinking water is safe to
drink?
Yes, but must do both Filter only Boil only No
17. Please check the level of harmfulness to your health of the following sources of water (not
boiled or filtered).
Extreme High Moderate Low No effect No
opinion
Consuming drinking water
(not treated)
Handling household water
(not treated)
Handling pond water (not
treated)
Bottled water
18. Now I want to ask you about bird flu/avian flu. Have you heard of bird flu/avian flu?
1. Yes, I have heard of bird flu/avian flu
2. No, I have not heard of bird flu/avian flu
Page 272 of 340
19. Please check the level of susceptibility of bird flu from the following sources of water (not
boiled or filtered).
Extreme High Moderate Low No
effect
No
opinion
Drinking water (not treated)
Handling household water (not
treated)
Handling pond water (not
treated)
Handling livestock or poultry
waste water (not treated)
20. If you were to get bird flu, how severe a disease do you think it would be?
1. Very severe
2. Severe
3. Moderate
4. Mild
5. Trivial
6. No opinion
21. Was your poultry destroyed because of an outbreak of bird flu (HPAI) in your province?
1. No, our poultry was never destroyed
2. Yes, our poultry was destroyed
3. Do not remember
4. No opinion
22. What do you do when you observe unusual number of sick/death your poultry/livestock?
1. Ignore
2. Consult other farmers
3. Notify the village leader
4. Burn/Bury
5. Notify the local para-veterinarians
6. Use as food (for livestock or human)
7. Nothing
8. Other, please specify
23. Currently, is your poultry being vaccinated because of the bird flu (HPAI) outbreak?
1. Yes vaccinated
2. No, not vaccinated yet
3. No opinion
Page 273 of 340
24. Have you ever heard about the microbes that may exist in water such as viruses, bacteria, and
helminths?
1. Yes, I have
2. No, I have not
25. Please check the level of susceptibility to diarrhea from the following sources of water (not
boiled or filtered).
Extrem
e
Hig
h
Moderat
e
Low No
effect
No
opinion
Drinking water (not
treated)
Handling household water
(not treated)
Handling pond water (not
treated)
Handling livestock or
poultry waste water (not
treated)
26. If you were to get diarrhea, how severe a disease do you think it would be?
1. Very severe
2. Severe
3. Moderate
4. Mild
5. Trivial
6. No opinion
27. One kind of microbes that can be found in farm water and sometime dangerous is E. coli.
Have you heard of E. coli or coliform disease?
1. Yes, move on to the next question and continue
2. No, skips the next question and continue
28. Please check the level of susceptibility to coliform disease from the following sources of
water (not boiled or filtered).
Extrem
e
Hig
h
Modera
te
Lo
w
No
effect
No opinion
Drinking water (not treated)
Handling household water
(not treated)
Handling pond water (not
treated)
Page 274 of 340
Handling livestock or
poultry waste water (not
treated)
29. If you were to get coliform disease, how severe a disease do you think it would be?
1. Very severe
2. Severe
3. Moderate
4. Mild
5. Trivial
6. No opinion
30. Another source of disease that can be found in water on a farm and sometimes can be
harmful is parasites such as ascarids or helminths. Have you heard of the parasites called
ascarids or helminths?
1. Yes, move on to the next question and continue
2. No, skips 2 next questions and continue
31. Please check the level of susceptibility of disease from parasites from the following sources
of water (not boiled or filtered).
Extreme High Moderate Low No
effect
No opinion
Drinking water (not
treated)
Handling household water
(not treated)
Handling pond water (not
treated)
Handling livestock or
poultry waste water (not
treated)
32. If you were to get disease from parasites, how severe a disease do you think it would be?
1. Very severe
2. Severe
3. Moderate
4. Mild
5. Trivial
6. No idea
Page 275 of 340
33. How important/serious are the issues of water usage to the reduction of risk factors for
WRZDs (e.g., AI)?
Extremel
y
important
High
important
Moderate
important
Low
important
Not
importa
nt
No
opinion
Source of water
Not share between
human and livestock
Closeness of water
source and livestock
Shortage of water
Rainfall
Water quality
Other
issues________
34. Here is the scale representing the important of proper water usage in reducing risk factors for
WRZDs as well as AI. On which step of the scale do you feel you personally stand now?
1. Extremely important
2. High important
3. Moderate important
4. Low important
5. Not important
6. No opinion
35. Overall, are you satisfied or dissatisfied with the current source of water and wastewater that
you are using for yourself, your livestock, and your crop?
Totally
satisfied
Highly
satisfied
Moderate
satisfied
Low
satisfied
Dissatis
fied
No
opinion
Yourself
(drinking)
livestock, crop
36. Please indicate your thought about the level of risk of WRZDs of the following sources of
water
Extremely
risk
High
risk
Moderate
risk
Low
risk
No risk No
opinion
From the rain
Lake
River
Page 276 of 340
Canal
Dug well
Drilled well
Other (please specify)
37. What factors drive your thoughts of the quality of water and wastewater?
1. Based on impression produced by any substance on the water and
wastewater, taste, or smell, and also on the organism as a whole
(organoleptic)
2. Previous experiences
3. Risk perceptions
4. Attitudes towards water chemical contamination
5. Contextual cues relating to the source of water supply
6. Familiarities with specific water properties
7. Trust in suppliers/source
8. Media
9. No opinion
10. Other factors _please specify
38. What are the sources of information relating to WRZDs as well as AI that you get and how
reliable are these sources?
Extr
eme
High Moder
ate
Low No No
opinion
TV
Newspapers
Radio
Family/friends/other farmers
Government (CHCs)
Vet/Doctor
Other source_ please
specify__________________
_______
39. Do you think applying preventive actions will protect against the WRZDs?
1. No
2. Yes, such as hand washing
3. Yes, Isolation
4. Yes, Wearing protective (e.g., mouth mask, filter water use)
5. Yes, Vaccination
6. Yes, Taking antiviral
Page 277 of 340
7. Yes, Cook food/not eating raw food
8. No opinion
9. Other (please name it) ______________
40. Are there benefits (other than health) that ENCOURAGE you to take actions will protect
against the WRZDs?
Please indicate_____________________________________________________________
41. Are there benefits that encourage you NOT to take actions will protect against the WRZDs?
Please indicate_____________________________________________________________
42. What are the barriers that prevent you from undertaking action that can protect against the
WRZDs?
1. Cannot afford the cost of actions
2. Do not know how to take actions
3. Do not understand the need to take action
4. Influence by neighbours who do not take actions
5. Do not have time for taking action, too busy with other things
6. No opinion
7. Other reason (please explain it)
______________________________________________
43. What are the triggers/cues to that influence your decision to undertake action that can protect
against the WRZDs?
1. Neighbours do the same
2. Advice from health officer, veterinarians
3. Radio/newspaper
4. Not busy with other tasks
5. Worry about being infected or affected by EID
6. Lost income
7. No opinion
8. Other reason (please explain it)
______________________________________________
44. What are the agricultural production types that you have?
1. Rice
2. Cassava
3. Bamboo
Page 278 of 340
4. Corn
5. Peanuts
6. Vegetables (state)
7. fruit (state)
8. Fish production
9. Livestock production
45. What are your top three sources of farm management information (put number 1, 2, 3 of 9 ):
Fellow farmers
Family members
Newspaper or other printed media
Radio or TV
Commune expert
Training program
Feed supplier
NGO
Other (state)
46. What types of livestock do you normally have in your farm and its respective number?
Estimated number per
season (YOUNG)
Estimated number per
season (ADULT)
Estimated production
volume (kg) per season
Note
Fish
Chicken
Duck
Pig
Goat
Other poultry
Other cattle
Other pet
47. What type of water sources does your family have access to and uses the most for drinking?
Most of the
time
Frequently Occasionally Rarely Never No
opini
on
Pond/ lake
River/ stream/canal
Rain water collection
Page 279 of 340
Government
supply/piped
Hand dug wells
Drilled wells
Buying water
Other sources
_________
48. What type of water sources does your family have access to and uses the most for your
poultry and other livestock?
Most of
the time
Frequentl
y
Occasiona
lly
Rarely Never No
opini
on
Pond/ lake
River/ stream/canal
Rain water collection
Government
supply/piped
Hand dug wells
Drilled wells
Buying water
Other sources
_________
49. Was your poultry infected with bird flu (HPAI) during the HPAI outbreak of 2004?
1. Yes
2. No
3. Do not remember
4. No opinion
50. Was your poultry infected with bird flu (HPAI) during the last 12 months?
1. Yes
2. No, skip next 2 questions
3. Do not remember
4. No opinion
51. Do you dump waste from livestock production directly to the source of water use for
agriculture production?
1. No
2. Yes, directly to pond/lake
Page 280 of 340
3. Yes, directly to canal/river/creek
4. No opinion
5. Yes, directly to other places. Please
specify__________________________________
52. What is your total on-farm income per person per year?
1. < 4.8 million VND
2. 4.8-10 million VND
3. 10-20 million VND
4. 20-40 million VND
5. >40 million VND
6. No opinion
53. How do you protect yourself from the risk of WRZDSs?
1. Wear protection cloth
2. Wear glove to protect hands
3. Wash hands and feet after field work
4. No protection
5. No opinion
6. Other, please explain
54. Please indicate if you have been employing one or more of the following livestock
management methods
1. Vaccination of domestic animals
2. Ensure hygienic rearing conditions for animals
3. Treatment of animal waste at source
4. Exclusion of animals from catchment basin
5. Stop using antibiotics as growth promoters
6. None of the above
7. No opinion
55. How important are these methods to avoid being transmitted with WRZDSs from water and
wastewater source? LIVESTOCK MANAGEMENT
Very
importan
t
Highly
Important
Moderat
e
importan
t
Low
importan
t
Not
important
No
opinio
n
Vaccination of
domestic animals
Page 281 of 340
Ensure hygienic
rearing conditions
for animals
Treatment of animal
waste at source
Exclusion of
animals from
catchment basin
Stop using
antibiotics as
growth promoters
56. Please indicate if you have been employing one or more of the following source water
protection methods
1. Wastewater treatment
2. Human/animal waste treatment at source
3. Using buffer zones (e.g., vegetable)
4. Protect of well/springs/ground water
5. Protect Cover drinking water reservoir
6. Repair leaking in sewage system
7. Treatment of animal/human wastes prior to use as fertilizers
8. None of the above
9. No opinion
57. How important the methods below are to avoid being transmitted with WRZDSs from water
and wastewater source? SOURCE WATER PROTECTION
Very
important
Highly
Important
Moderat
e
importan
t
Low
importan
t
Not
important
No
opinion
Wastewater treatment
Human/animal waste
treatment at source
Using buffer zones
(e.g., vegetable)
Protect of
well/springs/ground
water
Cover drinking water
reservoir
Repair leaking in
sewage system
Page 282 of 340
Treatment of
animal/human wastes
prior to use as
fertilizers
58. Please indicate if you have been employing one or more of the following mitigation practices
in relating to water storage treatment and distribution
1. Cover water storage reservoir or store water in confined tanks
2. Spend money in household water treatment technology
3. Change/use disinfectants
4. Repair household water storage
5. Repair household sewer
59. How important the methods below are to avoid being transmitted with WRZDSs from water
and wastewater source? WATER STORAGE TREATMENT ANF DISTRIBUTION
Very
importa
nt
Highly
Importa
nt
Moderat
e
importa
nt
Low
importa
nt
Not
importa
nt
No
opinion
Cover water storage
reservoir or store water
in confined tanks
Spend money in
household water
treatment technology
Change/use
disinfectants
Repair household water
storage
Repair household sewer
60. Please indicate if you have been employing one or more of the following mitigation
strategies/practices/behaviours
1.Improve personal, domestic and food hygiene (e.g. hand washing)
2.Adequate household sanitation
3.Invest in improving water supply
4.Prevent cross-contamination
5.Improve food hygiene
6.Cook foods/ now eating raw food
7.Protect water source and food source for animals
Page 283 of 340
8.Communicate risk to other members of household, community
9.Wear protective (e.g., mouth mask), filter water use
10. Taking antiviral
61. How important it is to avoid being transmitted with WRZDSs from water and wastewater
source? POINT OF USE/HOUSEHOLD
Very
importa
nt
Highly
Importa
nt
Moderat
e
importan
t
Low
importa
nt
Not
important
No
opinio
n
Improve personal,
domestic and food
hygiene (e.g. hand
washing)
Adequate household
sanitation
Invest in improving
water supply
Prevent cross-
contamination
Improve food hygiene
Cook foods/ now eating
raw food
Protect water source
and food source for
animals
Communicate risk to
other members of
household, community
Wear protective (e.g.,
mouth mask), filter
water use
Taking antiviral
62. How does your household usually handle livestock waste?
1.Collect
2.Dump in river/lake/rice field
3.Dump in a site near by
4.Bury
5.Burn
6.Other Please specify
7.No opinion
Page 284 of 340
63. Do you use waste from livestock production for fish production?
1. Yes
2. No, skip the next 2 questions
3. No opinion
Water testing results questions:
pH levels:
64. What are pH levels in on-farm stored rainwater for drinking?
65. What are pH levels in on-farm pipe water for drinking?
66. What are pH levels in on-farm well water for drinking?
67. What are pH levels in on-farm bottled water for drinking?
68. What are pH levels in on-farm river water (fl.) for drinking?
69. What are pH levels in on-farm stored rainwater for domestic purposes?
70. What are pH levels in on-farm well water for domestic purposes?
71. What are pH levels in on-farm river water (fl.) for domestic purposes?
72. What are pH levels in on-farm pond water for domestic purposes?
Turbidity levels:
73. What are turbidity levels in on-farm stored rainwater for drinking?
74. What are turbidity levels in on-farm pipe water for drinking?
75. What are turbidity levels in on-farm well water for drinking?
76. What are turbidity levels in on-farm bottled water for drinking?
77. What are turbidity levels in on-farm river water (fl.) for drinking?
78. What are turbidity levels in on-farm stored rainwater for domestic purposes?
79. What are turbidity levels in on-farm well water for domestic purposes?
80. What are turbidity levels in on-farm river water (fl.) for domestic purposes?
81. What are turbidity levels in on-farm pond water for domestic purposes?
E. coli cfu levels:
82. What are E. coli cfu levels in on-farm stored rainwater for drinking?
83. What are E. coli cfu levels in on-farm pipe water for drinking?
84. What are E. coli cfu levels in on-farm well water for drinking?
85. What are E. coli cfu levels in on-farm bottled water for drinking?
86. What are E. coli cfu levels in on-farm river water (fl.) for drinking?
87. What are E. coli cfu levels in on-farm stored rainwater for domestic purposes?
88. What are E. coli cfu levels in on-farm well water for domestic purposes?
89. What are E. coli cfu levels in on-farm river water (fl.) for domestic purposes?
90. What are E. coli cfu levels in on-farm pond water for domestic purposes?
Page 285 of 340
APPENDIX C: CONSENT SCRIPTS
C.1. Consent Scripts in Vietnamese
PHIÊU TINH NGUYÊN THAM GIA ĐIÊU TRA
A. Phiếu tình nguyện tham gia điều tra _Dành cho Bảng hỏi và lấy mẫu nước xét nghiệm:
Phiếu này phải được phỏng vấn viên đọc cho người tham gia
Xin chào và xin cảm ơn Quý vị đã đến dự cuộc thảo luận nhóm hôm nay! Tên tôi là….. tôi là
thành viên của nhóm điều tra thí điểm. Thay mặt nhóm điều tra, tôi xin kính mời các cô/bác
anh/chi tham gia buổi thảo luận nhóm hôm nay. Đây là một phần của điều tra thí điểm mang tên
“Động vật, nước, và y tế công cộng tại Việt Nam”. Việc tham gia điều tra thí điểm này là hoàn
toàn mang tính tình nguyện, Quý vị có quyền quyết định có tham gia buổi thảo luận nhóm này
hay không. Điều tra thí điểm nhằm kiểm tra các công cụ được thiết kế cho điều tra nhằm xác
định những chỉ số cơ bản của chất lượng nước sử dụng tại nông hộ (ví dụ như độ đục và vi khuẩn
chỉ định); tìm hiểu loại hình chăn nuôi, thả cá trong nông hộ quy mô nhỏ; ghi nhận những suy
nghĩ và hành động của nông dân liên quan đến việc sử dụng nước, nước thải cũng như nguy cơ
lây truyền bệnh qua nước; xác định mối liên hệ giữa những chỉ số nêu trên với suy nghĩ và
phương cách của nông dân.
Bảng hỏi và lấy mẫu nước xét nghiệm là một trong các hoạt động của điều tra thí điểm tại xã.
Trong bảng hỏi này, tôi sẽ hỏi các nhóm câu hỏi liên quan đến cá nhân hộ gia đình và nông trại
của Quý vị như 1) thông tin về cá nhân và hộ gia đình; 2) quan niệm và thái độ liên quan đến các
yếu tố ảnh hưởng đến sức khỏe của việc sử dụng nước và nước thải; 3) các loại hình chăn nuôi,
thả cá và trông trọt tại nông trại; 4) thu nhập có được từ chăn nuôi, thả cá và trông trọt tại nông
trại; và 5) các hành động nhằm loại bỏ, giảm thiểu tần suất, mức độ tiếp xúc với nguy cơ mắc các
bênh truyên nhiễm cũng như cúm gia cầm liên quan đến việc sử dụng nước và nước thải. Tôi sẽ
ghi lai câu trả lời của Quý vị. Dự kiến mất khoảng 45 phút để hoàn thành bảng hỏi này. Liên
quan đến phần lấy mẫu xét nghiệm, đại diện của TT Y tế dự phòng Thái Bình sẽ xin phép lấy
mẫu nước uống và nước sinh hoạt của nông trại để xét nghiệm về các chỉ sô cơ bản của chất
lượng nước. Tôi đề nghị Quý vị chỉ dẫn chúng tôi đến nguồn nước uống và nguồn nước sinh hoạt
được dùng nhiều nhất tại nông hộ cho người, vật nuôi, và trồng trọt. Tất cả các câu trả lời và
thông tin thu thập đươc sẽ được giữ bí mật riêng tư. Tuy nhiên những hoạt động vi phạm pháp
luật nếu có sẽ phải báo cáo với chính quyền địa phương. Nếu các Quý vị có bất cứ câu hỏi nào
liên quan đến điều tra thí điểm này tôi sẽ vui lòng trả lời trong khả nằng có thể.
Dự kiến, điều tra này sẽ giúp chỉnh sửa công cụ điều tra cho phù hợp trước khi đươc sử dụng
trong một điều tra quy mô lớn hơn nhằm giúp hiểu biết tốt hơn về mối liên hệ giữa quan niệm về
chất lượng nước sử dụng tại nông trại và việc giảm thiểu nguy cơ mắc bệnh truyền nhiễm trong
nông hộ chăn nuôi quy mô nhỏ ỏ Thái Bình, từ đó cung cấp thông tin và khuyến nghị cho việc
thiết kể và thực hiện chương trình, chính sách phù hợp mới mối quan tâm, nhu cầu cũng như
điều kiện kinh tế của hộ nông dân do đó giúp cải thiện sức khỏe nông dân, vật nuôi và môi
trường một cách bền vững. Các thông tin cá nhân thu thập được sẽ được đảm bảo giữ bí mật
riêng tư. Mã số sẽ được dùng cho thảo luận. Chỉ có tôi mới biêt đó là mã số gì. Quý vị có thể
Page 286 of 340
quyết định không tham trả lời bất kỳ câu hỏi nào hoặc ngừng tham gia điều tra tại bất cứ thời
điểm nào nếu muốn. Nếu ai có câu hỏi gì liên quan đến điều tra thí điểm này thì có thê liên hệ
BS. Lê Bá Quỳnh theo sổ điện thoại 0165 247 1058.
Xin trân trọng cảm ơn!
Page 287 of 340
B. Phiếu tình nguyện tham gia điều tra _Dành cho Phỏng vấn sâu, Quan sát:
Phiếu này phải được phỏng vấn viên đọc cho người tham gia
Xin chào và xin cảm ơn các Quý vị đã đến dự cuộc thảo luận nhóm hôm nay! Tên tôi là….. tôi là
thành viên của nhóm điều tra thí. Thay mặt nhóm điều tra, tôi xin kính mời các cô/bác anh/chi
tham gia buổi thảo luận nhóm hôm nay. Đây là một phần của điều tra thí điểm mang tên “Động
vật, nước, và y tế công cộng tại Việt Nam”. Việc tham gia điều tra thí điểm này là hoàn toàn
mang tính tình nguyện, cô/bác anh/chi có quyền quyết định có tham gia buổi thảo luận nhóm này
hay không. Điều tra thí điểm nhằm kiểm tra các công cụ được thiết kê cho điều tra nhằm xác
định những chỉ số cơ bản của chất lượng nước sử dụng tại nông hộ (ví dụ như độ đục và vi khuẩn
chỉ định); tìm hiểu loại hình chăn nuôi, thả cá trong nông hộ quy mô nhỏ; ghi nhận những suy
nghĩ và hoạt động của nông dân liên quan đến việc sử dụng nước, nước thải cũng như nguy cơ
lây truyền bệnh qua nước; xác định mối liên hệ giữa những chỉ số nêu trên với suy nghĩ và
phương cách của nông dân.
Phỏng vấn sâu, quan sát nông trại là một trong các hoạt động của điều tra thí điểm tại xã. Trong
phần phỏng vấn sâu tôi sẽ hỏivà thảo luận với Quý vị vấn đề đến 1) quan niệm và thái độ liên
quan đến các yếu tố ảnh hưởng đến sức khỏe của việc sử dụng nước và nước thải; 2) các loại
hình chăn nuôi, thả cá và trông trọt tại nông trại; 4) thu nhập có được từ chăn nuôi, thả cá và
trồng trọt tại nông trại; và 5) các hành động nhằm loại bỏ, giảm thiểu tần suất, mức độ tiếp xúc
với nguy cơ mắc các bênh truyên nhiễm cũng như cúm gia cầm liên quan đến việc sử dụng nước
và nước thải. Tôi sẽ ghi lai câu trả lời và phần chia sẻ ý kiến của Quý vị. Dự kiến mất khoảng 30
phút để hoàn thành phần phỏng vấn sâu. Sau khi hoặc trong khi phỏng vấn sâu tôi đề nghị được
quan sát nông trại của Quý vị, ghi chép lại và chụp ảnh nếu cần. Tất cả các câu trả lời của các
Quý vị và thông tin thu thập sẽ được giữ bí mật riêng tư. Tuy nhiên những hoạt động vi phạm
pháp luật nếu có sẽ phải báo cáo với chính quyền địa phương. Nếu các Quý vị có bất cứ câu hỏi
nào liên quan đến điều tra thí điểm này tôi sẽ vui lòng trả lời trong khả nằng có thể. Tổng thời
gian dành cho Phỏng vấn sâu, quan sát nông trại dự kiến khoảng 1 giờ đồng hồ.
Dự kiến, kết quả của điều tra này sẽ giúp chỉnh sửa công cụ điều tra cho phù hợp trước khi đươc
sử dụng trong một điều tra quy mô lớn hơn nhằm giúp hiểu biết tốt hơn về mối liên hệ giữa quan
niệm về chất lượng nước sử dụng tại nông trại và việc giảm thiểu nguy cơ mắc bệnh truyền
nhiễm trong nông hộ chăn nuôi quy mô nhỏ ỏ Thái Bình, từ đó cung cấp thông tin và khuyến
nghị cho việc thiết kể và thực hiện chương trình, chính sách phù hợp mới mối quan tâm, nhu cầu
cũng như điều kiện kinh tế của hộ nông dân do đó giúp cải thiện sức khỏe nông dân, vật nuôi và
môi trường một cách bền vững
Các thông tin cá nhân thu thập được sẽ được đảm bảo giữ bí mật riêng tư. Mã số sẽ được dùng
cho thảo luận. Chỉ có tôi mới biêt đó là mã số gì. Các Quý vị có thể quyết định không tham trả
lời bất kỳ câu hỏi nào hoặc ngừng tham gia điều tra tại bất cứ thời điểm nào nếu muốn. Nếu ai có
câu hỏi gì liên quan đến điều tra thí điểm này thì có thê liên hệ BS. Lê Bá Quỳnh theo sổ điện
thoại 0165 247 1058.
Xin trân trọng cảm ơn!
Page 288 of 340
C. Phiếu tình nguyện tham gia điều tra _Dành cho Thảo luận nhóm:
Phiếu này phải được phỏng vấn viên đọc cho người tham gia
Xin chào và xin cảm ơn Quý vị đã đến dự cuộc thảo luận nhóm hôm nay! Tên tôi là….. tôi là
thành viên của nhóm điều tra thí điểm. Thay mặt nhóm điều tra, tôi xin kính mời các Quý vị
tham gia buổi thảo luận nhóm hôm nay. Đây là một phần của điều tra thí điểm mang tên “Động
vật, nước, và y tế công cộng tại Việt Nam”. Việc tham gia điều tra thí điểm này là hoàn toàn
mang tính tình nguyện, Quý vị có quyền quyết định có tham gia buổi thảo luận nhóm này hay
không. Điều tra thí điểm nhằm kiểm tra các công cụ được thiết kê cho điều tra nhằm xác định
những chỉ số cơ bản của chất lượng nước sử dụng tại nông hộ (ví dụ như độ đục và vi khuẩn chỉ
định); tìm hiểu loại hình chăn nuôi, thả cá trong nông hộ quy mô nhỏ; ghi nhận những suy nghĩ
và hoạt động của nông dân liên quan đến việc sử dụng nước, nước thải cũng như nguy cơ lây
truyền bệnh qua nước; xác định mối liên hệ giữa những chỉ số nêu trên với suy nghĩ và phương
cách của nông dân.
Thảo luận nhóm là một trong các hoạt động của điều tra thí điểm tại xã. Trong phần thảo luận
nhóm này, tôi sẽ đề nghị Quý vị thảo luận cùng nhau về những chủ đề chính dưới đây:
1. Những vấn đề gì liên quan đến nước và chất thải ảnh hưởng nhiều nhất đến sinh kế, chăn
nuôi, thả cá của các Quý vị, và tại sao đó là vấn đê đáng quan tâm?
2. Những vấn đề đó xảy ra thường xuyên như thế nào trong quá khứ?
3. Nguyên nhân của những vấn đề đó là gì?
4. Hậu quả của các vấn đề đó như thế nào đối với bản thân, gia đình, và chăn nuôi thả cá
của các Quý vị?
5. Quý vị đã đối phó/ giải quyết những vấn đề đó như thế nào?
6. Ai có thể giúp Quý vị giải quyết vấn đề này?
7. Các Quý vị cần những sự giúp đỡ gì, như thế nào?
8. Nếu các vấn đề nêu trên xảy ra trong tương lai, Các Quý vị sẽ giải quyết thế nào?
9. Có những chính sách, chương trình, quy định gì của các tổ chức khác giúp giải quyết vấn
đề nêu trên? Nếu có, đó là gì và theo Quý vị có hiệu quả không?
Tôi đề nghị các Quý vị thảo luận với nhau thay vì với tôi. Tôi sẽ chỉ điều khiển cuộc thảo luận,
ghi chép. Dự kiến thảo buổi thảo luận nhóm không kéo dài quá 2 giờ đồng hồ và sẽ được ghi âm.
Đề ghị các Quý vị lưu ý là không có ý kiến nào là sai hay đúng, và khuyến khích các Quý vị chia
sẻ ý kiến khác nhau nhưng cần tôn trọng ý kiến người khác mặc dù có đồng ý hay không. Tất cả
những thông tin trong cuộc thảo luận sẽ được giữ bí mật. Trừ trường hơp những hoạt động vi
phạm pháp luật nếu có sẽ phải báo cáo với chính quyền địa phương. Nếu các Quý vị có bất cứ
câu hỏi nào liên quan đến điều tra thí điểm này tôi sẽ vui lòng trả lời trong khả nằng có thể. Dự
kiến, điều tra này sẽ giúp chỉnh sửa công cụ điều tra cho phù hợp trước khi đươc sử dụng trong
một điều tra quy mô lớn hơn nhằm giúp hiểu biết tốt hơn về mối liên hệ giữa quan niệm về chất
lượng nước sử dụng tại nông trại và việc giảm thiểu nguy cơ mắc bệnh truyền nhiễm trong nông
hộ chăn nuôi quy mô nhỏ ỏ Thái Bình, từ đó cung cấp thông tin và khuyến nghị cho việc thiết kể
và thực hiện chương trình, chính sách phù hợp mới mối quan tâm, nhu cầu cũng như điều kiện
kinh tế của hộ nông dân do đó giúp cải thiện sức khỏe nông dân, vật nuôi và môi trường một
cách bền vững
Page 289 of 340
Các thông tin cá nhân thu thập được sẽ được đảm bảo giữ bí mật riêng tư. Mã số sẽ được dùng
cho thảo luận. Chỉ có tôi mới biêt đó là mã số gì. Quý vị có thể quyết định không tham trả lời bất
kỳ câu hỏi nào hoặc ngừng tham gia điều tra tại bất cứ thời điểm nào nếu muốn. Nếu ai có câu
hỏi gì liên quan đến điều tra thí điểm này thì có thê liên hệ BS. Lê Bá Quỳnh theo sổ điện thoại
0165 247 1058.
Xin trân trọng cảm ơn!
Page 290 of 340
C.2. Consent Scripts in English
A. Consent script for Questionnaire and water sample collection
This consent script is required to be read to participants by enumerators/field investigators.
Greetings everybody and thank you for agreeing to meet with me! My name is … I am a member
of the research team. On behalf of the research team, I am inviting you to participate in the research
titled Ecosystem approaches to improved water and farm health in Vietnam with assistant from
the commune and district collaborators in Thai Binh and An Giang provinces. Your participation
in this research is highly appreciated and voluntary, so it is your decision to choose to participate
or not. The research is interested in determining the basic level of on-farm water quality (i.e.,
turbidity and total coliforms/fecal coliform indicators); examining the types of fish and livestock
raised on small scale farms in Thai Binh and An Giang; recording farmers’ perceptions and actions
with respect to water and wastewater usage and risk of disease to animals and humans from water;
and determining the associations between these indicators and the presence of fish and livestock,
and the recorded farmers’ perceptions and actions. As part of the research, I will ask you some
questions about 1) the general description of yourself and your household; 2) your attitudes,
perception, and your actions relating to risk factors of WRZD transmission; 3) your types of
livestock, fish, and crop; 4) your income from different livestock, fish, and crop; and 5) your
actions to eliminate or reduce the frequency, magnitude, or severity of exposure to risks, or
minimization of the potential impact of a risk of emerging infectious diseases as well as avian
influenza transmission relating to water and wastewater usage. I will write down your answers to
these questions in the questionnaire. It will take approximately 45 minutes of your time to answer
these questions. All information in your answers to the questionnaire will be kept confidential.
However, please note that I will have to report to local authority if any illegal or abuse observed.
Please feel free to ask questions anytime if you have any. I will be happy to explain the research
in detail if you wish.
It is envisaged that this research will bring about better understanding of associations between the
perceptions of water and wastewater usage and the reduction of risk factor of infectious diseases
amongst small scale farmers in Thai Binh and An Giang province and from that a larger scale
study is expected to follow to obtain further information and provide appropriate recommendations
for better design and implementation of programs/policies that are more suitable to the concerns
of small scale livestock production farmers, reflect the need, perceptions of farmers and their
socioeconomic status, their environment, and therefore can help improving health of farmers,
livestock, and environment sustainably (in the long run).
There is a procedure in place to make sure that all personal information will be kept private and
confidential. A code will be assigned to each questionnaire and only I will be able to identify which
code is for whom. You may choose not to answer any of the questions and/or decide to discontinue
your participation any time if you are no longer wish to participate in the research without any
penalty. If you have any questions, concerns about the research, please contact Dr. Quynh Ba Le
@ 0165 247 1058. Thank you for your participation!
Page 291 of 340
B. Consent script for in-depth interview and observation
This consent script is required to be read to participants by enumerators/field investigators.
Greetings everybody and thank you agreeing to meet with me! My name is …I am a member of
the research team. On behalf of the research team, I am inviting you to participate in the research
titled Ecosystem approaches to improved water and farm health in Vietnam. Your participation
in this research is highly appreciated and voluntary, so it is your decision to choose to participate
or not.
The research is interested in determining the basic level of on-farm water quality (i.e., turbidity
and total coliforms/fecal coliform indicators); examining the types of fish and livestock raised on
small scale farms in Vietnam; recording farmers’ perceptions and actions with respect to water
and wastewater usage and risk of disease to animals and humans from water; and determining the
associations of these indicators with the presence of fish and livestock, and the recorded farmers’
perceptions and actions. As part of the research, I will ask you some questions, observe your farm
and undertake in-depth interview, and take water sample from your farm for testing. The questions
will be about 1) the general description of yourself and your household); 2) your attitudes,
perception, and your actions relating to risk factors of WRZD transmission; 3) your type of
livestock, fish, and crop; 4) your income from different livestock, fish, and crop; and 5) your
actions to eliminate or reduce the frequency, magnitude, or severity of exposure to risks, or
minimization of the potential impact of a risk of emerging infectious diseases as well as avian
influenza transmission relating to water and wastewater usage. It will take about two and 30
minutes for the in-depth interview. The in-depth interview will be recorded. The total time for an
in-depth interview and observation will be about 1 hour.
It is envisaged that this research will bring about better understanding of associations between the
perception of water and wastewater usage and the reduction of risk factors for infectious diseases
amongst small scale farmers in Thai Binh and An Giang provinces and from that a larger scale
study is expected to follow to obtain further information and provide appropriate recommendations
for better design and implementation of programs/policies that are more suitable to the concerns
of small scale farmers, reflect the need, perceptions of farmers and their socioeconomic status,
their environment, and therefore can help improving health of farmers, livestock, and environment
sustainably (in the long run). There is a procedure in place to make sure that all personal
information will be kept private and confidential. A code will be assigned to each interview and
water sample and only I will be able to identify which code is for whom. You may choose not to
answer any of the questions and/or decide to discontinue your participation any time if you no
longer wish to participate in the research without any penalty. If you have any questions, concerns
about the research, please contact Dr. Quynh Ba Le @ 0165 247 1058.
Thank you for your participation!
Page 292 of 340
C. Consent script for focus group discussion
This consent will be used. This oral consent script is required to be read to participants by
enumerators/field investigators.
Greetings everybody and thank you for being here today for the group discussion! My name is …I
am a member of the research team. On behalf of the research team, I am inviting you to participate
in the research titled Ecosystem approaches to improved water and farm health in Vietnam. Your
participation in this research is highly appreciated and voluntary, so it is your decision to choose
to participate or not. The research is interested in determining the basic level of on-farm water
quality (i.e., turbidity and total coliforms/fecal coliform indicators); examining the types of fish
and livestock raised on small scale farms in Vietnam; recording farmers’ perceptions and actions
with respect to water and wastewater usage and risk of disease to animals and humans from water;
and determining the associations of these indicators with the presence of fish and livestock, and
the recorded farmers’ perceptions and actions. As part of the research, I will ask you to discuss the
following questions as a group:
1. Which water and wastewater related problems have been the most urgent for your
livelihood, livestock/fish production and why?
2. How often is the problem in comparison with the past?
3. What are the causes of the problem?
4. What are the consequence of the problem to you, your family, and your livestock?
5. How did you deal with the problem?
6. Who do you think should help you in this situation?
7. What kind of help did/do you need?
8. How will you deal with the problem if it happens again?
9. Are there any policy/regulation/ program from other organization to address this
problem?
I want you to talk to each other rather than to me. I will be moderating and taking notes of the
discussion, which will take about two hours and will be recorded. There is no right or wrong
comments, please respect but do not hesitate to disagree with what others have said or to share
different opinion. All information relating to the group discussion will be kept confidential.
However, please note that I will have to report to local authority if any illegal or abuse observed.
Please feel free to ask questions anytime if you have any. I will be happy to explain the research
in detail if you wish. It is envisaged that this research will bring about better understanding of
associations between the perception of water and wastewater usage and the reduction of risk factor
of infectious diseases amongst small scale farmers in Thai Binh and An Giang province and from
that a larger scale study is expected to follow to obtain further information and provide appropriate
recommendations for better design and implementation of programs/policies that are more suitable
to the concerns of small scale farmers, reflect the need, perceptions of farmers and their
socioeconomic status, their environment, and therefore can help improving health of farmers,
livestock, and environment sustainably (in the long run).
Page 293 of 340
There is a procedure in place to make sure that all personal information will be kept private and
confidential. A code will be assigned to each interview and water sample and only I will be able
to identify which code is for whom. You may choose not to answer any of the questions and/or
decide to discontinue your participation any time if you no longer wish to participate in the research
without any penalty. If you have any questions, concerns about the research, please contact Dr.
Quynh Ba Le @ 0165 247 1058.
Thank you for your participation!
Page 294 of 340
APPENDIX D: ETHICS APPROVALS
D.1. University of Calgary Ethics ID: REB13-0470
- By the Conjoint Faculties Research Ethics Boards
D.2. Hanoi School of Public Health Ethics Approval: No. 151/2013/YTCC_HD3
- By the Chair of the Ethical Review Board for Biomedical Research, Hanoi School of
Public Health
D.3. Hanoi School of Public Health Ethics Approval: No. 008/2014/YTCC_HD3
- By the Chair of the Ethical Review Board for Biomedical Research, Hanoi School of
Public Health
Page 295 of 340
APPENDIX E: CHAPTER TWO’S APPENDIX
Table 2.5 (Chapter Two) shows the final results of young fish production without outliers
and stratified over the different levels of education and different levels of income among the
participating farmers. There were changes in the mean numbers of young fish production by
strata of education and income. More specifically, in Thai Binh, the participating farmers who
attended either secondary school or post-high school and had incomes of 10 to 40 million VND
per person per year were the farmers who produced the greatest mean number of young fish per
year. The stratification confirmed that the data collected from the wealthy group of farmers
relating to young fish production were outliers in the young fish production variable.
Tables E.1-3 provided additional detailed analysis of data on fish production in Thai Binh
and An Giang.
In Table E.1, I present a brief description of the fish production among those participating
farmers who produced fish. This table shows the great amounts of variation (Std. Dev) in young
fish production. The corresponding coefficients of variation of fish production in Thai Binh and
An Giang were 5.7 and 9.5 respectively. Therefore, in Table E.2, I have used median, 25
percentile, and 75 percentile to assist the presentation of data on annual fish production. Table
E.2 also shows the results without the outliers identified in Section 2.2.4. However, the
coefficients of variation of young fish production were still high 5.1 and 4.0 in Thai Binh and An
Giang respectively. Therefore, to describe the numbers of young fish produced per year in the
participating farms more clearly, I have presented these numbers of young fish produced per year
over different strata of income, education, and both income and education in Table E.3 (with
outliers) and Table 2.5 (in the main text of Chapter Two, without outliers).
Page 296 of 340
Table E.1. Annual Fish Production on Small-Scale (SSI) Farms in the Provinces of Thai
Binh and An Giang in Vietnam
Provinces Mean Std. Dev.
Thai Binh
Young fish a 24,238 138,592
Adult fish b 2,060 3,585
Fish c 1,773 2,391
An Giang
Young fish 43,400,000 411,000,000
Adult fish 1,117 17,333
Fish (kg) 1,087 12,232
Total
Young fish 28,700,000 334,000,000
Adult fish 1,454 14,056
Fish (kg) 1,393 9,246
a = Young fish produced (in number); b = Adult fish produced (in number); and c = Fish produced (in kg)
Among the participating farmers who produced fish at the time of the survey, farmers in
Thai Binh and An Giang estimated the number of young fish, the number of adult fish, and the
volume of fish in kg they normally had per season (Table E.2). I did not ask type of fish in the
questionnaires; however, it appeared from our observations, in-depth interviews, and group
discussion that the majority of the participating farms raised various “trash fish and weed fish”53.
53 This term was used by the participating farmers to indicate the various types of local fish in small mixed-fish
production not for commercial purpose.
Page 297 of 340
A few farms had fattening ponds, either finishing ponds, or hatching ponds mostly for Pangasius
bocourti fish in An Giang province. The number of fish farms at the time of the survey in Thai
Binh was about four times greater than that in An Giang (240 and 58 respectively) (Table E.2)
Table E.2. Quantity and Kilograms of Fish Produced per Year on Small-Scale Integrated
(SSI) Farms in the Provinces of Thai Binh and An Giang, Vietnam (without outliers)
Variable Obs % Mean Median
Std.
Dev. Min Max
25th
percentile
75th
percentile
Thai Binh
Young
fish 151 50% 4,852 1,000 13,095 20 100,000 500 3,000
Adult fish 166 55% 2,041 700 3,589 30 30,000 300 2,500
Fish (kg) 240 80% 1,779 700 2,395 5 10,000 1,700 2,000
An Giang
Young
fish 58 19% 8,159 800 68,011 3 100,000 100 3,600
Adult fish 36 12% 9,311 450 49,210 3 300,000 35 1,000
Fish (kg) 51 17% 6,296 30 28,797 2 150,000 10 250
Total
Young
fish 209 35% 5,769 1,000 15,857 3 1,000,000 500 3,000
Adult fish 202 34% 3,337 600 21,252 3 300,000 300 2,000
Fish (kg) 291 49% 2,570 625 12,495 2 150,000 200 2,000
Given the identified outliers in the variable of young fish production (Table E.2), more
detailed results of young fish production was stratified over the different levels of education and
different levels of income among the participating farmers (Table E.3). The outliers were
included in the results presented in this table. Most of the participating farmers who produced
young fish were those who had levels of education from primary school to high school. In An
Giang, it appeared that those participating farmers who had more years of attending school and
greater income produced the greatest mean numbers of young fish per year. In Thai Binh, the
participating farmers who attended up to primary school and had incomes of 10 to 20 million
Page 298 of 340
VND per person per year were the farmers who produced the greatest mean number of young
fish per year. However, the standard deviations of the number of young fish produced in these
farms were still several time greater than the corresponding means. Therefore, data in table 2.5 in
the main text of Chapter Two is presented in another way (i.e., without outliers).
Page 299 of 340
Table E.3. Quantity of Young Fish Produced per Year on Small-Scale Integrated (SSI)
Farms in the Provinces of Thai Binh and An Giang in Vietnam Stratified over Education
and Income (with outliers)
Provinces/
income b
Year of school attended a
1 2 3 4 5
Thai Binh
< 4.8 - 3,800.0 2,500.0 - -
- (5,377.7) (4,203.6) - -
4.8-10 - 2,469.2 35,335.9 933.6 520.0
- (2,593.1) (171,343.9) (896.5) (678.8)
10-20 - 108,600.0 5,146.1 86,516.7 14,190.0
(313,441.2) (17,010.5) (287,694.7) (21,759.3)
20-40 - 1,000.0 10,000.0 1,000.0 4,000.0
- n.ac (11,742.0) n.a n.a
>40 - 10,000.0 1,508.3 800.0 200.0
- n.a (985.1) (244.9) n.a
An Giang
< 4.8 3 1,307.4 75.0 129.5 -
- (2,001.0) (35.4) (247.0) -
4.8-10 - 1,357.6 3,100.0 1,850.0 -
- (1,544.8) (3,489.3) (2,333.4)
10-20 - 18,006.1 6,000.0 7,006.7 -
- (27,239.7) n.a (11,263.2) -
20-40 - 14,000.0 1,667Md 2.9M -
- (8,485.3) (2,080M) (4.5M) -
>40 - 3,000M - 2,500M -
n.a - (3,530M) - (.) Numbers in brackets are standard deviations; a 1 = no school; 2 = primary school; 3 = secondary school; 4 = high
school; and 5 = post-high school; b Income was reported in millions of Vietnamese dong (VND); c No s.d. is reported
where only one observation exists; d Means of young fish production greater than one million are reported in
millions denoted by M.
Page 300 of 340
APPENDIX F: CHAPTER FOUR’S APPENDICES
F.1. Brief Ecology of Escherichia coli and WHO’s Recommendations for Assessing Rural
Water Quality
E. coli, a member of the coliform bacteria, is part of the Enterobacteriaceae family and is
a facultative anaerobic, Gram-negative, non-spore forming, rod-shape bacterium. E. coli is
naturally found in the intestinal tract of humans and warm-blooded animals at a concentration of
approximately 109 cells per gram of faeces (Edberg et al., 2000). E. coli is not naturally found in
soil and water or on plants (Winfield and Groisman, 2003). Surface water and ground water
(under direct influence of the surface water) are commonly contaminated with human and animal
faeces. A presence of E. coli in a source of water indicates that the source is contaminated with
faecal materials from humans and/or animals and is microbiologically not safe for drinking. E.
coli is generally sensitive to environmental stressors and does not survive in the environment for
as long as protozoans and some viruses do. E. coli can survive for 4–12 weeks in water with
moderate micro flora and a temperature of 15–18°C (Edberg et al., 2000). Most waterborne E.
coli isolates produce enzyme β-glucuronidaze (Fricker et al., 2008) which is used to facilitate the
detection of E. coli in water (FPTCW, 2013).
The Ministry of Health (MoH) of Vietnam developed its national technical regulations on
drinking and domestic water quality (MoH, 2009b; MoH, 2009a) based on the World Health
Organization guidelines for assessing rural water quality (WHO, 2011a; WHO, 2011b; WHO
and OECD, 2003a). The technical regulations and guidelines from the MoH and WHO were used
to guide the assessments of on-farm water quality in this research and are referred to hereafter as
the recommendations for assessing rural water quality. These recommendations include
Page 301 of 340
microbiological and physicochemical indicators of rural water quality. Escherichia coli or E. coli
was used as an indicator organism of faecal contamination of water and possible presence of
other harmful bacteria and viruses. Levels of pH and turbidity were used as these indicators are
positively correlated with pathogen removal in water.
Figures F.1 summarizes the main characteristics of the coliform group and the
corresponding recommendations in assessing microbial water quality based upon information
obtained from the World Health Organization and the Organization for Economic Co-operation
and Development (WHO and OECD, 2003a). There are several reasons why E. coli is
recommended as an indicator organism for assessing microbial safety of rural water. First of all,
it is not practical to conduct isolations of all pathogens in water due to the large number existing
pathogens, its uneven distribution in water, and required resources for the isolation (Payment and
Pintar, 2006). Secondly, pathogens from human and animal faeces pose the greatest danger to
public health compared to other contaminants that are regularly found in water. This makes the
ability to detect faecal contamination in drinking water important for public health safety
(FPTCW, 2013). Thirdly, current methods of assessing E. coli have been simpler, more
available, and more affordable than in late 1980s (Edberg et al., 2000). Fourthly, E. coli is
considered a more specific indicator of faecal contamination compared to other members of the
total coliform bacteria group, and can be rapidly identified and easily quantified in water. E. coli
has a similar lifespan, comparable to that of other enteric bacterial pathogens, does not usually
multiply in the environment, and can be detected even if it is greatly diluted due to the great
number of E. coli that are excreted in faeces (FPTCW, 2013; WHO, 2011b; WHO and OECD,
2003b).
Page 302 of 340
In addition to assessing E. coli level in on-farm water, other WHO recommended basic
physical indicators of on-farm water quality (i.e., level of pH and turbidity) were also assessed.
Levels of pH and turbidity in water have been shown to influence microbial quality of water
(WHO and OECD, 2003b).
Figure F.1. The Coliform Group
Source: This figure is created using information from WHO and OECD, 2003.
WHO recommends pH and turbidity levels in water to be included as parts
Total coliforms
Thermotol--erant
coliforms
E. coli
Main characteristics Recommendations
- Include Escherichia, Citrobacter, Enterobacter,
and Klebsiella genera
- Include lactose fermenting bacteria (e.g.,
Enterobacteria cloacae and Citrobacter freundii) which can be found in both faeces,
environment, and drinking water.
- Include members of genera (e.g., Budvicia, and
Rahnella) that are never found in mammalian
faeces - Oxidase negative, ferment lactose at 35-37oC
with the production of acid, gas, and aldehyde within 24-48 hours
- Possess β-galactosidase
- Not an index of
faecal
contamination or of health risk
- Basic information
on source water quality
- Easy to detect and
enumerate in water
- Account for 60% to 90% of the total coliforms
group
- Include primarily the genus of Escherichia (Hall, 1999), and to lesser extent Klebsiella,,
Enterobacter, and Citrobacter
- May originate from organically enriched water
and does not necessarily suggest faecal contamination except E. coli
- Ferment lactose at 44-45oC
- Less reliable index
of faecal contamination than
E. coli but
acceptable for
water-quality
examination when
on other method is available
- A well-defined member of the family
Enterobacteriaceae
- Possession of β-galactosidase β-glucuronidase
- Most strains of E. coli grow at 44-45 oC,
ferments lactose and mannitol with production of
acid and gas and produces indole from
tryptophan
- Rapid and reliable tests for E. coli are available.
- Found in water and environment subject to
recent faecal contamination
- Abundant in human and animal faeces, fresh
faeces may attain concentration of 109 of E. coli per gram.
- Widely preferred
as an index of
faecal contamination
Page 303 of 340
of the core set of microbial and related indicators in assessing and monitoring water quality
(WHO, 2012).
F.2. Protocols Used for Assessing On-Farm Water Quality
Two protocols were used in this research to assess the quality of on-farm water. These
protocols included a protocol for laboratories to assess on-farm water quality (i.e., pH, turbidity,
and E. coli) and a protocol for community health/veterinary workers (CHWS/CVWs) to quantify
E. coli. This appendix provides detailed information of these protocols.
F.2.1. Protocol for Laboratories to Assess Basic Microbial and Related Water Quality
This section elaborates the first protocol used for collecting and testing water quality in
the participating SSI farms. Both drinking water54 and domestic water55 were assessed using the
basic indicators and methods recommended by WHO. This protocol was used by two national
microbiological laboratories of the Vietnam National Institute of Veterinary Research and the Ho
Chi Minh City Institute of Public Health - to assess the basic indicators of on-farm water quality.
The basic indicators included: 1) number of E. coli colony forming units (cfu) in on-farm water
(using membrane filtration method); 2) level of pH of on-farm water; and 3) level of turbidity of
on-farm water. The research team collaborated with two national microbiological laboratories -
one in the north and one in the south of Vietnam - to collect and test water samples in Thai Binh
and An Giang respectively. Both laboratories followed the same national laboratory protocol for
assessing water quality which conforms to WHO/UNICEF water testing standards (MoH, 2009b;
MoH, 2009a).
54 Drinking water is defined as “water used for direct drinking or processing food” (MoH, 2009a) 55 Domestic water is defined as “water used for domestic use but not for direct drinking or processing food” (MoH
and MARD, 2011)
Page 304 of 340
Basic indicators of water quality assessed included: 1) E. coli colony forming unit
(cfu)/100ml in on-farm water (using membrane filtration method); 2) pH of on-farm water; and
3) turbidity of on-farm water. Membrane filtration (MF) method is one of the two common
methods used to detect and quantify coliform bacteria in water (i.e., “multiple fermentation
tube/most probable number method” and “membrane filter/membrane filtration method”. The
MF method was selected because of its advantages over the multiple fermentation tube in
assessing large number of water samples. Examples of the advantages include more rapid
yielding results given relatively large numbers of samples, less labor intensive, results obtained
directly by colony count (Bartram and Pedley, 1996; TCVN 6187, 2009; WHO, 2011a).
With respect to collection of water samples, each national microbiological laboratory sent
a technical team to the participating farms to collect two water samples. In each province, 34
(11%) and 63 (21%) farms in Thai Binh and An Giang respectively had duplicated water
samples used for testing of E. coli, pH, and turbidity. For duplicated samples, mean results of
two samples were reported. At each of the participating farm, two water samples of 200ml each
were collected for testing of the three basic indicators - one sample was for drinking water
testing and the other sample was for domestic water testing. Collected samples of water were
treated within 6-8 hours of sampling. The collection, storage, and transportation of the water
samples followed the national guidance on water quality and sampling (TCVN 6663, 2011).
With regards to the MF method, the participating laboratories followed the technical
guidelines for water quality testing of Vietnam MoH (TCVN 6187, 2009). These guidelines
conformed to standards for detection and enumeration of faecal indicator organisms in water by
WHO and ISO (i.e., ISO 9301 - 12:1990). Four key steps of the MF method conducted include
Page 305 of 340
filtration, incubation, confirmatory tests, and calculation of results. The number of E. coli colony
forming units (cfu) in each 100ml of sample water was calculated using the following formula:
𝐶(𝑐𝑓𝑢 𝑖𝑛 𝑉 𝑚𝑙) =𝐴 × 𝑁
𝐵
C: Number of E. coli cfu confirmed in 100ml of sample water
A: Number of presumptive E. coli cfu that were positive with Indole test
B: Number of presumptive E. coli cfu transferred for further incubation in Trypone Soy Agar
(TSA) medium
N: Number of presumptive E. coli cfu on filtered incubated membranes with Lactose TTC agar
medium
V: Volume filtered sample water
The national laboratories followed the national technical standards and guidelines for
determination of pH level using pH meter (TCVN 6492, 2011; IHPH, 2013a) and followed the
EPA method 180.1 (Powell, 2007) for determination of turbidity by Nephelometry unit using a
portable turbidity meter (i.e., Hatch Model 2100P) (EPA, 1993; IHPH, 2013b). The field
investigator tested the protocol in a pilot study of 30 farms before using it in the full research.
F.2.2. Protocol for Community Health/Veterinary Workers to Quantify E. coli
The second protocol was for community health/veterinary workers CHW/CVWs to
collect and test drinking water samples of the participating SSI farms. The protocol was color-
changed quantification of total coliforms and E. coli bacteria in water using ColiplateTM tests
(Bluewater Biosciences, 2013). ColiplateTM is an immunofluorescence plate test of water
produced by Bluewater Biosciences Inc. Canada. The use of ColiplateTM tests were supported
financially by a doctoral research award (IDRA) provided by the International Development
Page 306 of 340
Research Centre (IDRC, 2012). This protocol conforms to EPA based test regulatory guidelines
(Bluewater Biosciences, 2013). I selected and trained twenty CHW/CVWs in 1) water sample
collection procedure following the Vietnam Ministry of Health guidelines for collecting water
samples for microbial (MoH, 2009b) and 2) using the ColiplateTM testing procedure (e.g., testing
steps and handling storage, cleanup and disposal of the test kits) . For each farm, a 96-well
immunofluorescence ColiplateTM plate test was used by CVW (in the province of Thai Binh) and
CHW (in the province of An Giang) to quantify coliform bacteria and E. coli in on-farm water.
The field investigator tested ColiplateTM tests before being used by the CHWs/CVWs. Water
quality tests were carried out on fresh water samples collected on site in the communities.
The procedure used in this survey for testing on-farm water is described as follows. The
CHWs/CVWs collected water samples on farms, stored it in cool Styrofoam boxes, and
transported the collected water samples back to commune health/veterinary stations. Within six
hours since the water samples were collected, CHWs/CVWs poured the water samples into the
provided 96-well ColiplateTM plates and started incubating the water samples in Styrofoam boxes
in four days. Each CHW/CVW was instructed to check and take notes in a given paper form if
any of the 96 wells in each of the ColiplateTM plates with the sample water turns blue during the
four-day incubation. One water sample was used in one 96-well ColiplateTM plate. ColiplateTM
tests have varied incubation time depending on incubation temperatures - from 24 hours (at 35-
37oC) to 60 hours (at 20-25oC). After incubation, the CHWs/CVWs reported the numbers of
wells with blue water in each plate. These blue wells indicate the presence of coliform bacteria.
To quantify E. coli, CHWs/CVWs used ultraviolet (UV) lamps, provided by the research, to light
on the blue colored wells in a dark room. All wells that reflected the UV light with fluorescent
Page 307 of 340
light were the wells that have E. coli. CHWs/CVWs counted the numbers of blue wells with
fluorescent light reflection. The numbers of blue wells and blue wells with fluorescent light
reflection were used to demine the number of colony-forming units (cfu) of coliform bacteria and
E. coli respectively using a Most Probable Number (MPN) table.
F.3. Detailed Model Building Strategies for MLR and SUR
Regression analyses (Greene, 2008a; Greene, 2008b) were used to model the association
between the objectively assessed on-farm water quality and its potential farm related associated
factors. Sections below present detailed strategies used for building MLR and SUR models and
detailed model fitting/selecting procedures used in each model for analysing research data.
F.3.1. Modeling building strategies
The overall strategy for building models for this chapter was to start with a multiple
linear regression (MLR) model, then use a ‘seemingly unrelated regression’ (SUR) model
(Greene, 2008a) to analyze associated factors of the quality of on-farm drinking water and on-
farm domestic water. Within each of the MLR and SUR models, four specific model building
strategies (Dohoo et al., 2003; Greene, 2008a) were used: specifying a full model; specifying
selection criteria for independent variables; specifying selection strategies for fitting model; and
analysing the data using my knowledge and experiences of water public health in Vietnam and
statistical software.
F.3.2. Multiple linear regression (MLR) model
I used MLR models to start exploring associations between objectively assessed quality
of various sources of on-farm water for drinking and domestic use (i.e., E. coli cfu in various on-
farm water sources – the dependent variables) and subjectively assessed water quality and other
Page 308 of 340
appropriate on-farm factors (e.g., farmers’ perceptions - potential independent variables). In each
MLR model, a set of appropriate independent variables were regressed against a dependent
variable. The initial potential set of independent variables was determined based on the
conceptual and theoretical model of this research (presented in Chapter Three), the hypotheses of
this research, and the literature. Tables F.1a-d presented pairwise correlations between the
independent variables.
Four selection procedures were used in the MLR models. First, in a full simultaneous
procedure, a dependent variable (e.g., E. coli in stored rainwater for drinking water) was
predicted from the full set of potential independent variables simultaneously (e.g., gender, age,
and years of attending school). Second, in a stepwise regression procedure, I used a syntax that
started with backward elimination then forward selection. After the elimination of each variable
(i.e., the highest non-significant variable with p-value > a predefined α), all the eliminated
variables were checked to see if any of them should be included using forward selection criterion
(i.e., the lowest significant variable p-value <= α). A significance level of α < or => 0.15 was
used for adding and deleting new variables to/from the model respectively. Third, in a forward
selection procedure, Stata was commanded to start a model with only an intercept then
selectively added variables that meet the criterion that I specified (i.e., p-value < a specific α
level). At each adding step, a variable with the lowest significant p-value (p-value <= a
recommended level of α = 0.25) will enter the model. The adding variable process continued
until no more variables meet the adding criteria. Fourth, in a backward elimination procedure,
steps were reversed compared to that of the forward selection procedure. The backward
elimination procedure started with a maximum model then variable(s) that meet the elimination
Page 309 of 340
criteria were removed. In particular, at each step, a variable with the highest non-significant p-
value (p-value > α) was removed from the model. A recommended significance level of α = 0.25
was used for removing variables from the model.
With respect to fitting the models and determining independent variables that need to be
retained in the final models, I considered two sets of criteria including 1) my research interest
and a priori to be potential confounders and 2) p-values, F-test, and Adjusted R squared before
presenting the results of the final fittings of the MLR model.
The MLR model used in this Chapter is written as the following:
Yi = β0+ βi Xi + ɛi; with i = 1, 2, 3... n
Yi is a continuous dependent variable representing on-farm water quality, considering
bundle i. Each Y is the numbers of E .coli cfu in 100 ml of a source of on-farm water.
Xi is a vector of IVs (continuous and/or dichotomous) believed to be relevant to the
consideration of ith equations (i = 1, 2, 3... n).
βo is a intercept or constant to be estimated,
βi is a vector of unknown coefficients describing the direction and magnitude of the effect
of an Xi on Yi after controlling for the effects of other Xi; βi to be estimated,
ɛi is error term
F.3.3. Seemingly Unrelated Regression (SUR) Model
I assumed residuals of the two equations associating the quality of on-farm drinking
water and on-farm domestic water from other factors are related. Therefore, in addition to the use
of the MLR model, I used a SUR model to explore systematically the associations between a
dependent variable and the corresponding different set of independent variables. The assumption
Page 310 of 340
of correlated residuals was based on the facts that the SSI farming practice was used among the
participating SSI farmers and that both quality of on-farm drinking water variables and quality of
on-farm domestic water variables come from the same dataset. This assumption was tested when
fitting the SUR model. In addition, some potential factors that may be associated with quality of
on-farm drinking water may not be the same for quality of on-farm domestic water. More
specifically, in the SUR model, two MLR equations were considered simultaneously to explore
associated factors (potential independent variables) of E. coli cfu in on-farm drinking water as
well as domestic water (Long and Freese, 2014). Potential IVs in these two equations were
expected not to be the same. The first MLR equation is for E. coli cfu in on-farm drinking water
verse a set of potential IVs. The second MLR equation is for E. coli cfu in on-farm domestic
water verse other set of Independent variables.
The model for the two equation is written as below with j = 1… m MLR equations:
yj = x’j βj + uj
These two equations are stacked into an SUR model:
[
𝑦1..
𝑦𝑚
] = [𝑥1 0 ⋯ 0⋮ ⋱ ⋮0 ⋯ 𝑥𝑚
] [
β1..
βm
] + [
u1..
um
]
The error terms are assumed to have zero mean and to be independent across farmers and
homoscedastic. For a given farm, the errors are correlated across equations:
E(uij uij’|𝑋) = σjj’ and σjj’ ≠ 0 where j ≠ j’
Cross-equation restrictions were tested, and constraints were imposed for IVs that were
identified in the fitting of both equations and were not significantly different from each other.
The SUR model was re-estimated.
Page 311 of 340
Table F.1a. Pairwise Correlation between Independent Variables used for Regression
Models
Independent variables a Age DGender Schooling DIncome DChildren
Experie
nce DOff-farm job
Age 1.0
DGender 0.1 1.0
Schooling -0.1* 0.0 1.0
DIncome -0.1* -0.1* -0.2* 1.0
DChildren -0.4* 0.0 0.0 0.* 1.0
Experience 0.3* 0.0 0.1* -0.2* -0.1* 1.0
DOff-farm job -0.2* -0.1* 0.0 0.2* 0.1* 0.0 1.0
DFPR-Livestock-Man -0.1 0.2* 0.1 -0.1 0.1* 0.0 0.0
DFPR-Livestock-Woman 0.0 -0.4* 0.0 0.0 0.0 0.1 0.1
DFPR-Fish-Woman 0.0 -0.3* 0.0 0.0 -0.1 0.0 0.0
DFPR-Fish-Man 0.0 0.1 0.2* -0.2* 0.0 0.1 0.0
DFPR-Health-Man 0.0 0.2* 0.0 -0.1 0.0 0.1* -0.1*
DProvince 0.1* -0.1 0.4* -0.4* -0.3* 0.3* -0.1*
DFish prod 0.0 -0.2* 0.3* -0.2* -0.2* 0.1* 0.0
DRain for animals 0.1* 0.1* 0.0 -0.1 -0.1 0.1* 0.0
DPond for animals 0.0 0.0 0.0 0.0 -0.1 0.0 0.0
DPipe for animals 0.0 -0.1* 0.0 0.3* 0.0 -0.1* 0.1
DMulti-agri activities 0.0 0.1* 0.0 -0.2* 0.0 0.1 -0.1*
DPoultry AI -0.1 0.1 0.0 0.2* 0.1* -0.2* -0.1
Npigs/yr 0.0 0.0 0.2* -0.2* 0.0 0.0 -0.1*
Nducks/yr 0.1 0.1* 0.0 -0.1* -0.2* -0.1 -0.1*
Npoultry/yr 0.0 0.0 0.2* -0.2* -0.1* 0.0 0.0
DFPWater quality 0.0 -0.1* 0.1 -0.2* 0.0 0.1* 0.0
DFPBarr-Cost 0.0 -0.2* 0.0 0.1* 0.1* 0.0 0.2*
DFPRisk rain water 0.0 0.0 0.0 -0.2* 0.0 0.1 0.0
DFPS-AI-Drink 0.0 -0.1* 0.0 0.0 0.0 0.0 0.0
DFPS-Diarrhea-Drink 0.0 0.0 0.1 0.0 -0.1 0.0 0.0
DFPS-Coliform-Drink . . . . . . .
DFPS-Parasite-Drink 0.0 0.0 0.1 -0.1 0.0 -0.1 0.0
DFPS-AI-Dom 0.1 0.1* 0.0 0.0 0.0 -0.1 0.0
DFPS-Coliform-Dom 0.1 0.3* -0.1* -0.2* -0.1 -0.1 -0.1
DFPS-Diarrhea-Dom 0.1 0.2* 0.1* 0.0 0.0 0.0 0.0
DFPS-Parasite-Dom 0.0 0.2* -0.1 0.0 0.0 -0.1 -0.1
DSatisfy-Drink 0.1 0.1* -0.2* 0.0 0.0 -0.2* 0.0
DSatisfy-Dom 0.1* 0.2* -0.2* -0.2* -0.1 -0.1* 0.0
DAI heard 0.1* 0.0 0.3* -0.1 -0.1* 0.1* 0.0
DFPHarm-Dom 0.0 0.1* 0.1* 0.0 0.0 0.0 0.0
DFPHarm-Drink 0.1 0.0 0.0 -0.1 0.0 0.0 0.0
DDump Wastes -0.2* -0.1* 0.1* 0.1* 0.1* -0.1 0.1*
DLivestock Management 0.1 0.0 0.3* -0.4* -0.1* 0.2* -0.1*
DTest-dom 0.0 0.1* -0.1 0.1 0.0 -0.1 -0.1* a See Tables 4.6 in chapter Four for descriptions of the independent variables
Page 312 of 340
Table F.1b. Pairwise Correlation between Independent Variables used for
Regression Model (cont’d)
Independent
variables a
DFPR-
Livest
ock-
Man
DFPR-
Livestock-
Woman
DFPR-Fish-
Woman
DFPR-
Fish-Man
DFPR-
Health-Man
DProvin
ce DFish prod
DFPR-Livestock-
Man 1.0
DFPR-Livestock-
Woman -0.4* 1.0
DFPR-Fish-Woman -0.2* 0.6* 1.0
DFPR-Fish-Man 0.5* 0.0 -0.2* 1.0
DFPR-Health-Man 0.5* -0.1* -0.1* 0.4* 1.0
DProvince -0.1 0.1 0.1* 0.3* 0.0 1.0
DFish prod -0.1* 0.1 0.1* 0.3* -0.1 0.7* 1.0
DRain for animals 0.1 -0.1 0.0 0.1 0.1 0.1* 0.1*
DPond for animals -0.1 0.0 0.0 0.1 0.1* 0.2* 0.3*
DPipe for animals -0.2* 0.1 0.0 -0.1* -0.1* -0.1 0.1
DMulti-agri
activities 0.0 -0.1 0.0 0.0 0.0 0.2* 0.3*
DPoultry AI 0.1* -0.1 -0.1 0.0 0.0 -0.2* -0.1*
Npigs/yr -0.1 0.0 0.0 0.1* 0.0 0.5* 0.3*
Nducks/yr -0.1* 0.0 0.0 -0.1 -0.1 0.3* 0.2*
Npoultry/yr 0.0 0.0 0.0 0.2* 0.0 0.4* 0.3*
DFPWater quality -0.1* 0.1 0.1 0.1* -0.1* 0.3* 0.3*
DFPBarr-Cost 0.1* 0.0 -0.1 0.0 -0.1* -0.2* -0.1
DFPRisk rain water 0.1 0.0 0.0 0.0 0.0 0.1* 0.1
DFPS-AI-Drink 0.0 0.1 0.0 -0.1* 0.0 -0.1 -0.1
DFPS-Diarrhea-
Drink 0.0 0.0 0.0 0.0 0.1 0.0 0.0
DFPS-Coliform-
Drink . . . . . . .
DFPS-Parasite-Drink 0.1 -0.1 -0.1* 0.1 0.0 0.0 0.0
DFPS-AI-Dom -0.1* 0.0 0.1 -0.2* -0.1* 0.1 -0.1
DFPS-Coliform-
Dom -0.2* -0.1* 0.0 -0.1* 0.0 0.0 0.0
DFPS-Diarrhea-Dom 0.1 0.0 0.1* 0.1 0.0 0.2* 0.1
DFPS-Parasite-Dom -0.1 -0.1 0.1 -0.1 -0.1 0.0 0.0
DSatisfy-Drink 0.0 0.0 0.0 -0.2* 0.0 -0.3* -0.3*
DSatisfy-Dom 0.0 0.0 0.0 -0.1* -0.1 -0.1 -0.1*
DAI heard 0.1* 0.0 0.1 0.2* 0.1 0.3* 0.2*
DFPHarm-Dom 0.0 0.0 0.0 0.0 0.0 0.2* 0.1
DFPHarm-Drink 0.1 0.0 -0.1 0.0 0.0 0.0 0.0
DDump Wastes 0.1* 0.0 0.0 0.3* 0.0 0.3* 0.2*
DFPE-Livestock
Management 0.0 0.0 0.1 0.2* 0.0 0.6* 0.4*
DTest-dom 0.1 -0.1* -0.1 0.0 0.1 -0.2* -0.2* a See Tables 4.6 in chapter Four for descriptions of the independent variables
Page 313 of 340
Table F.1c. Pairwise Correlation between Independent Variables used for
Regression Models (cont’d)
Independent
variables a
DRain
for
animals
DPond for
animals
DPipe for
animals
DMulti-
agri
activities DPoultry AI Npigs/yr
Nducks/
yr
DRain for animals 1.0
DPond for animals 0.3* 1.0
DPipe for animals 0.0 0.2* 1.0
DMulti-agri
activities 0.1 0.2* -0.2* 1.0
DPoultry AI -0.1* 0.0 0.1* -0.1 1.0
Npigs/yr 0.1* 0.3* 0.0 0.2* -0.1* 1.0
Nducks/yr 0.1* 0.3* 0.0 0.1* -0.1* 0.3* 1.0
Npoultry/yrs 0.1* 0.1* -0.1* 0.1* -0.1* 0.5* 0.2*
DFPWater quality 0.1 0.3* 0.0 0.3* -0.2* 0.2* 0.1*
DFPBarr-Cost 0.0 -0.1* 0.1 -0.2* 0.0 -0.2* -0.2*
DFPRisk rain
water 0.1* 0.0 -0.1* 0.1* -0.1* 0.1 0.0
DFPS-AI-Drink 0.1* 0.1* 0.0 -0.1 0.0 0.0 -0.1*
DFPS-Diarrhea-
Drink 0.1* 0.0 0.0 0.0 0.0 0.0 0.0
DFPS-Coliform-Drink . . . . . . .
Independent variables a
DRain for
animals
DPond for
animals
DPipe
for
animals
DMulti-
agri
activities
DPoultr
y AI Npigs/yr Nducks/yr
DFPS-Parasite-Drink 0.0 0.0 0.0 0.0 0.0 0.0 0.0
DFPS-AI-Dom 0.2* 0.0 -0.1* 0.1* 0.1 0.2* 0.1*
DFPS-Coliform-Dom 0.3* 0.4* 0.0 0.3* 0.0 0.3* 0.3*
DFPS-Diarrhea-Dom 0.3* 0.2* -0.1 0.2* 0.0 0.3* 0.2*
DFPS-Parasite-Dom 0.3* 0.2* 0.0 0.2* 0.0 0.2* 0.2*
DSatisfy-Drink -0.1* -0.2* -0.1* -0.1 0.1 -0.1* 0.1
DSatisfy-Dom 0.0 0.0 -0.1* 0.0 -0.1* 0.0 0.1*
0.2* 0.1* 0.1 -0.1 0.0 0.1* 0.1*
DAI heard 0.3* 0.2* 0.0 0.1* 0.0 0.2* 0.2*
DFPHarm-Dom 0.0 0.0 0.0 0.0 0.0 -0.1* 0.0
DFPHarm-Drink 0.0 0.1* 0.1* 0.1* 0.0 0.1 0.0
DDump Wastes 0.1* 0.1 -0.2* 0.1* -0.1* 0.3* 0.2*
DFPE-Livestock
Management 0.0 0.2* 0.0 0.0 0.1* -0.1* -0.1
DTest-dom -0.1 -0.2* -0.1* -0.1 0.1 -0.1* 0.1
Independent variables a
DFPS-
Parasite-
Drink
DFPS-
AI-Dom
DFPS-
Coliform-
Dom
DFPS-
Diarrhea-
Dom
DFPS-
Parasite-
Dom
DSatisf
y-Drink
DSatisfy
-Dom
DFPS-Parasite-Drink 1.0
DFPS-AI-Dom 0.1 1.0
DFPS-Coliform-Dom . 0.6* 1.0
DFPS-Diarrhea-Dom 0.1* 0.6* 0.7* 1.0
DFPS-Parasite-Dom 0.0 0.6* 0.8* 0.7* 1.0
DSatisfy-Drink -0.1 0.1* 0.1 0.0 0.0 1.0 a See Tables 4.6 in chapter Four for descriptions of the independent variables
Page 314 of 340
Table F.1d. Pairwise Correlation between Independent Variables used for
Regression Models (cont’d)
Independent variables a
Npoultry/
yr
DFPWater
quality
DFPBarr
-Cost
DFPRisk
rain water
DFPS-
AI-
Drink
DFPS-
Diarrhea-
Drink
DFPS-
Coliform-
Drink
Npoultry/yr 1.0
DFPWater quality 0.1* 1.0
DFPBarr-Cost 0.0 -0.1 1.0
DFPRisk rain water 0.0 0.0 0.0 1.0
DFPS-AI-Drink 0.0 0.0 0.1 0.1* 1.0
DFPS-Diarrhea-Drink 0.0 0.0 0.0 0.1* 0.2* 1.0
DFPS-Coliform-Drink . . . . . . .
DFPS-Parasite-Drink 0.0 0.0 0.0 0.2* 0.2* 0.8* .
DFPS-AI-Dom 0.0 0.0 -0.2* 0.1 0.2* 0.1* .
DFPS-Coliform-Dom 0.1 0.0 -0.4* 0.1 0.0 . .
DFPS-Diarrhea-Dom 0.2* 0.1 -0.2* 0.1* 0.1* 0.1* .
DFPS-Parasite-Dom 0.1* 0.1* -0.4* 0.1 0.0 -0.1 .
DSatisfy-Drink -0.1 -0.3* -0.1 0.0 0.0 -0.1 .
DSatisfy-Dom 0.0 -0.1* 0.0 0.1* 0.0 0.1 .
DAI heard 0.2* -0.1 0.1* 0.1 0.0 0.0 .
DFPHarm-Dom 0.2* 0.0 -0.2* 0.1* 0.1 0.1 .
DFPHarm-Drink 0.0 0.0 0.0 0.1* 0.1* 0.2* .
DDump Wastes 0.1* 0.1* 0.1 0.0 -0.1 0.1 .
DFPE-Livestock
Management 0.3* 0.2* -0.1* 0.1* -0.1 0.0 .
DTest-dom -0.1* -0.1* 0.0 0.0 0.1 0.0 .
Independent
variables a
DFPS-
Parasite-
Drink
DFPS-AI-
Dom
DFPS-
Coliform-
Dom
DFPS-
Diarrhea-
Dom
DFPS-
Parasite-
Dom
DSatisfy-
Drink
DSatisfy-
Dom
DSatisfy-Dom 0.0 0.1 0.0 0.0 0.0 0.3* 1.0
DAI heard 0.0 0.0 0.1 0.2* 0.2* -0.1* 0.0
DFPHarm-Dom 0.1 0.5* 0.6* 0.6* 0.6* 0.0 0.0
DFPHarm-
Drink 0.3* 0.1* 0.1 0.1 0.0 0.0 0.0
DDump Wastes 0.0 0.0 -0.2* 0.0 0.0 -0.2* -0.1*
DFPE-Livestock
Management 0.0 0.1* 0.1 0.2* 0.0 -0.2* 0.0
DTest-dom 0.0 0.0 0.0 0.0 0.0 -0.1* 0.1*
Independent
variables a DAI heard
DFPHarm-
Dom
DFPHarm-
Drink
DDump
Wastes
DLivestock
Management
DTest-
dom
DAI heard 1.0
DFPHarm-Dom 0.2* 1.0
DFPHarm-
Drink 0.0 0.1* 1.0
DDump Wastes 0.2* 0.0 0.0 1.0
DFPE-Livestock
Management 0.2* 0.1* -0.1 0.0 1.0
DTest-dom 0.0 0.0 0.0 0.0 0.0 1.0 a See Tables 4.6 in chapter Four for descriptions of the independent variables
Page 315 of 340
F.3.4. Detailed Results of Sources of Water Used and Corresponding Frequencies of Use
among SSI Farmers in the Provinces of Thai Binh and An Giang
Table F.2. Mean Frequency of Use of On-Farm Water for Drinking in the Provinces of
Thai Binh and An Giang, Vietnam
Province/Sources of
water
Mean frequency of use for drinking by source of water (*)
Obs Mean Std. Dev. Min Max
Thai Binh
Rain 196 3.4 1.3 1 5
Drilled well 231 3.2 1.3 1 5
Bottled 169 2.4 1.1 1 5
Pipe 61 2.4 1.4 1 5
Dug well 40 1.7 1.0 1 4
Pond 143 1.0 0.2 1 3
River/Canal 143 1.0 0.2 1 3
An Giang
River/Canal 297 3.0 1.3 1 5
Pipe 243 2.7 1.8 1 5
Bottled 280 2.5 1.2 1 5
Rain 277 2.4 1.0 1 5
Pond 271 1.3 0.7 1 4
Drilled well 219 1.0 0.2 1 2
Dug well 217 1.0 0.1 1 2 (*) 1 = Never, 2 = Rarely, 3 = Sometimes, 4 = Often, to 5 = Very often
Page 316 of 340
Table F.3. Mean Frequency of Use of On-Farm Water for Domestic Purposes in the
Provinces of Thai Binh and An Giang, Vietnam
Province/
Sources of water Drink rank+
Mean frequency of use for domestic purposes (*)
Obs Mean Std. Dev. Min Max
Thai Binh
Drilled well 2 251 3.5 1.2 1 5
River/Canal 7 251 3.2 1.2 1 5
Pond 6 209 2.9 1.4 1 4
Rain 1 256 2.5 0.9 1 5
Dug well 5 38 1.9 1.1 1 4
Pipe 4 52 1.8 1.0 1 4
Bottled 3 108 1.2 0.5 1 3
An Giang
River/Canal 1 288 3.8 1.0 1 5
Rain 4 267 2.1 1.0 1 4
Bottled 3 254 1.5 0.8 1 4
Pond 5 260 1.4 0.8 1 4
Pipe 2 245 1.4 0.9 1 5
Drilled well 6 227 1.1 0.6 1 5
Dug well 7 226 1.0 0.2 1 3
(*) 1 = Never, 2 = Rarely, 3 = Sometimes, 4 = Often, to 5 = Very often
(+) Drink rank = the ranking for drinking for these water sources (see Table 4.1, Chapter Four)
F.4. Detailed Results of Basic Microbial and Related Indicators of On-Farm Water
F.4.1. Quality of On-Farm Water Used for Drinking
The mean results of pH, turbidity, and E. coli cfu/100ml of sources of on-farm drinking
water are presented for both provinces aggregately (Table F.4). The mean pH levels of these
sources of water were from 6.8 to 7.2 (with an exception of bottled water, which had a mean pH
of 9.4). These sources of water (except river water) had mean turbidity levels that were below 5
Nephelometric Turbidity Unit (Powell, 2007). All these sources of on-farm water used for
drinking were contaminated with E. coli. The mean number of E. coli colony-forming units
Page 317 of 340
(cfu)/100ml in the sources of on-farm water used for drinking ranged from eight to 46.9. Bottled
water had the lowest mean while stored rainwater had the highest.
Table F.4. Mean pH, Turbidity, and E. coli cfu of On-Farm Water Used for Drinking in the
Provinces of Thai Binh and An Giang, Vietnam
Variable N Mean Std. Dev. Min Max
pH
Rain water 148 6.8 3.0 4.9 42.8
Pipe water 118 7.2 2.6 6.0 32.6
Well water 172 6.9 5.3 5.5 76.3
Bottled water 43 9.4 14.0 6.4 99.2
River water (fl.) 119 7.0 0.4 5.8 9.6
Turbidity (measured in Nephelometric Turbidity Unit)
Rain water 148 0.9 0.9 0.0 3.2
Pipe water 114 1.7 2.7 0.0 18.2
Well water 172 3.7 9.2 0.0 72.4
Bottled water 43 0.8 0.4 0.3 1.5
River water (fl.) 120 7.6 20.9 0.2 167.0
E. coli (measured in number of E. coli colony-forming units per 100 mL of sample water)
Rain water 150 46.9 74.0 0.0 232.0
Pipe water 117 10.1 17.1 0.0 50.0
Well water 156 24.1 40.0 0.0 213.0
Bottled water 42 8.0 20.1 0.0 110.0
River water (fl.) 117 12.8 30.5 0.0 260.0
F.4.2. Quality of On-Farm Water Used for Domestic Purposes
The mean levels of pH in well domestic water in both provinces were within the range of
6.8 – 7.3 (Table F.5). The mean turbidity level was greatest in river water (i.e., 61.4) compared
to that in well water and pond water (i.e., 15.3 and 39.9 respectively). Pond water was used the
most for domestic purposes and had much greater mean numbers of E. coli cfu (i.e., 2596.6)
compared to that in well water and river water (i.e., 348.1 E. coli cfu and 817.8 E. coli cfu
respectively).
Page 318 of 340
Table F.5. Mean pH, Turbidity, and E. coli cfu of On-Farm Water Used for Domestic
Purposes in the Provinces of Thai Binh and An Giang, Vietnam
Variable N Mean Std. Dev. Min Max
pH
Well water 31 6.8 1.2 0.8 7.9
Pond water 354 6.7 0.7 0.6 8.5
River water 208 7.3 0.4 6.0 8.1
Turbidity (measured in Nephelometric Turbidity Units)
Well water 29 15.3 37.4 0.0 185.0
Pond water 352 39.9 27.1 0.0 155.0
River water 211 61.4 42.6 5.6 300.0
E. coli (measured in number of E. coli colony-forming units per 100 mL of sample water)
Well water 30 348.1 536.9 0.0 1,800.0
Pond water 353 2,596.6 3,237.5 0.0 12,400.0
River water 213 817.8 1,077.9 0.0 3,200.0
F.4.3. Detailed Results of Drinking Water Quality Assessed by CHW/CVW using ColiplateTM
Table F.6. Mean MPN a of E. coli cfu Using ColiplateTM Tests in On-Farm Water for
Drinking by Sources and by Number of Incubation Day in the Provinces of Thai Binh and
An Giang, Vietnam
Province/Water source
Incubation time (day)
1 2 2.5 3 4
Thai Binh
Rain - - - 47.3 19.5
Pipe - - - 11.8 7.0
Well - - - 27.5 46.9
Bottle - - - 2.7 -
An Giang
Rain - 77.5 950.1 105.2 -
Pipe 87.0 124.7 1,685.6 109.2 34.6
Well - 18.0 - - -
Bottle 39.0 5.9 - 56.9 -
River with flocculation 59.1 461.9 704.8 94.8 151.0
Pipe (stored in reservoir) - 32.6 - 132.0 - a MPN: Most Probable Number (MPN) of E. coli colony-forming units per 100mL sample water
Page 319 of 340
F.5. Detailed Results of MLR Models for E. coli in Sources of Water On Farms
Table F.7a. Summary of the Best MLR Model Fittings for E. coli cfu in Stored Rainwater for Drinking by Selection
Procedures in the Provinces of Thai Binh and An Giang, Vietnam
Independent variables
a Fullb Stepc Forwd Bacwe
Coef P>|t| Coef P>|t| Coef P>|t| Coef P>|t|
DGender -5.66 0.76 - - - - - -
Age 0.79 0.22 0.76 0.17 1.01* 0.05 0.76 0.17
Schooling -0.37 0.88 -2.86 0.23 - - -2.86 0.23
Experience -1.02 0.20 -0.89 0.22 -1.08 0.13 -0.89 0.22
DIncome -37.80* 0.09 -34.52* 0.09 -28.20 0.14 -34.52* 0.09
DChildren -13.49 0.36 - - - - - -
DOff-farm job 11.14 0.42 - - - - - -
Npoultry/yr 0.04 0.74 - - - - - -
DFish prod - - 50.07** 0.00 48.70** 0.00 50.07** 0.00
DFPR-Livestock-Man 7.63 0.63 - - - - - -
DFPR-Livestock-
Woman 25.80 0.20 17.93 0.24 - - 17.93 0.24
DFPWater quality -7.55 0.73 - - - - - -
DFPHarm-Drink 13.90 0.42 20.33 0.20 - - 20.33 0.20
DFPS-AI-Drink -12.28 0.44 - - - - - -
Note: AI = Avian Influenza, D = Dummy, FPS = farmers’ perceived susceptibility to, FP = Farmers’ perceptions of, a
Independent Variables described in Table 4.6; bFull = full simultaneous regression model; cStep = Stepwise selection
regression model; dForw = forward selection regression model; eBacw = backward selection regression model; (*) significant at
p<0.1; (**) significant at p<0.01
Page 320 of 340
Table F.7b. Summary of the Best MLR Model Fittings for E. coli cfu in Stored Rainwater for Drinking by Selection
Procedures in the Provinces of Thai Binh and An Giang, Vietnam (cont’d)
Independent variables a Fullb Stepc Forwd Bacwe
Coef P>|t| Coef P>|t| Coef P>|t| Coef P>|t|
DFPS-Diarrhea-Drink -38.00* 0.06 -23.75 0.18 - - -23.75 0.18
DFPS-Coliform-Drink -14.26 0.38 -19.00 0.20 -25.12* 0.06 -19.00 0.20
DFPS-Parasite-Drink 6.16 0.74 - - - - -
DSatisfy-Drink -49.65 0.00 -47.07** 0.00 -44.88** 0.00 -47.07** 0.00
constant 74.16 0.15 28.34 0.43 0.30 0.99 28.34 0.43
Obs 135 - 135 - 135 - 135 -
Prob > F 0.04* - 0.00** - 0.00** - 0.00** -
R-squared 0.20 - 0.24 - 0.21 - 0.24 -
Adj R-squared 0.09 - 0.17 - 0.17 - 0.17 -
Root MSE 70.31 - 66.82 - 66.96 - 66.82 -
Number of significant
variables 2 - 3 4 3 Notes: D = Dummy, FPS = farmers’ perceived susceptibility to, a Independent Variables described in Table 4.6; bFull = full simultaneous regression model; cStep = Stepwise selection regression model; dForw = forward selection regression model; eBacw = backward selection regression model; (*) significant at
p<0.1; (**) significant at p<0.01
Page 321 of 340
Table F.8a. Summary of the Best MLR Model Fittings for E. coli cfu in Pipe Water for Drinking by Selection Procedures in
the Provinces of Thai Binh and An Giang Provinces, Vietnam
Independent
variables a
Full b Step c Forw d Bacw e
Coef P>|t| Coef P>|t| Coef P>|t| Coef P>|t|
DGender -5.64 0.19 -5.24* 0.10 - - -5.24* 0.10
Age 0.19 0.32 0.19 0.24 - - 0.19 0.24
Schooling -0.69 0.26 -0.66 0.23 - - -0.66 0.23
Experience 0.28 0.36 - - - - - -
DIncome 6.86* 0.11 5.26 0.15 5.04 0.67 5.26 0.15
DChildren 6.67 0.14 6.16 0.13 - - 6.16 0.13
DOff-farm job 1.03 0.81 - - - - - -
Npoultry/yr 0.27** 0.03 0.29*** 0.01 0.37*** 0.00 0.29 0.01
Nducks/yr 0.08* 0.00 0.08*** 0.00 0.06*** 0.00 0.08*** 0.00
Npigs/yr 0.10 0.53 - - - - - -
DFish prod - - - - - - - -
DFPR-
Livestock-Man -1.28 0.78
-
- - - - -
DFPR-
Livestock-
Woman -2.12 0.66
-
- - - - -
DFPWater
quality -0.52 0.91
-
- - - - - Notes: D = Dummy, FPS = farmers’ perceived responsibility to, FP = Farmers’ perceptions of, N = Number, a Independent Variables described in Table 4.6; bFull = full simultaneous regression model; cStep = Stepwise selection regression model; dForw = forward selection regression model; eBacw = backward
selection regression model; (*) significant at p<0.1; (**) significant at p<0.01
Page 322 of 340
Table F.8b. Summary of the Best MLR Model Fittings for E. coli cfu in Pipe Water for Drinking by Selection Procedures in
the Provinces of Thai Binh and An Giang, Vietnam (cont’d)
Independent
variables a
Full b Step c Forw d Bacw e
Coef P>|t| Coef P>|t| Coef P>|t| Coef P>|t|
DFPHarm-
Drink 6.11 0.21 5.47 0.21 - - 5.47 0.21
DFPS-AI-Drink 7.27* 0.11 7.00 0.06 - - 7.00 0.06
DFPS-Diarrhea-
Drink -0.94 0.86
-
- - - - -
DFPS-Coliform-
Drink 0.77 0.88
-
- - - - -
DFPS-Parasite-
Drink -7.59* 0.10 -6.16* 0.11 - - -6.16* 0.11
DSatisfy-Drink 2.34 0.57 - - - - - -
constant -12.80 0.36 -9.60 0.38 - - -9.60 0.38
Obs 90 - 90 - 90 - 90 -
Prob > F 0.01*** - 0.00*** - 0.00*** - 0.00*** -
R-squared 0.37 - 0.35 - 0.26 - 0.35 -
Adj R-squared 0.20 - 0.27 - 0.23 - 0.27 -
Root MSE 14.16 - 13.56 - 13.84 - 13.56 -
Number of
significant
variable 5 4 2 3 Notes: AI = Avian Influenza, D = Dummy, FPS = farmers’ perceived susceptibility to, FP = Farmers’ perceptions of, a Independent Variables described in
Table 4.6; bFull = full simultaneous regression model; cStep = Stepwise selection regression model; dForw = forward selection regression model; eBacw =
backward selection regression model; (*) significant at p<0.1; (**) significant at p<0.01
Page 323 of 340
Table F.9. Summary of the Best MLR Model Fittings for E. coli cfu in Well Water for Drinking by Selection Procedures in the
Provinces of Thai Binh and An Giang, Vietnam
Independent
variables a
Full b Step c Forw d Bacw e
Coef P>|t| Coef P>|t| Coef P>|t| Coef P>|t|
DGender 1.90 0.89 -17.25* 0.06 -14.64* 0.11 -17.25* 0.06
Age -0.58 0.30 - - - - - -
Schooling -4.04** 0.04 - - -1.96 0.17 - -
Experience 0.07 0.93 - - - - - -
DIncome 60.98*** 0.00 48.17*** 0.01 37.48** 0.03 48.17*** 0.01
DChildren -0.08 0.99 -9.71 0.24 - - -9.71 0.24
DOff-farm job -14.85 0.15 -10.47 0.18 -9.85 0.22 -10.47 0.18
Npoultry/yr - - - - - - -
Npigs/yr -0.22* 0.09 - - - - - -
DFish prod - -76.53*** 0.00 -70.33*** 0.00 -76.53*** 0.00
DFPR-Livestock-Man -0.12 0.99 - - - - - -
DFPR-Livestock-
Woman 26.33* 0.07 18.13* 0.09 17.83* 0.10 18.13* 0.09
DFPWater quality -2.91 0.88 18.54 0.16 - - 18.54 0.16 Notes: D = Dummy, FPS = farmers’ perceived responsibility to, N = Number, a Independent Variables described in Table 4.6; bFull = full simultaneous regression
model; cStep = Stepwise selection regression model; dForw = forward selection regression model; eBacw = backward selection regression model; ere = removed
from the model; (*) significant at p<0.1; (**) significant at p<0.01
Page 324 of 340
Table F.10. Summary of the Best MLR Model Fittings for E. coli cfu in Well Water for Drinking by Selection Procedures in
the Provinces of Thai Binh and An Giang, Vietnam
Independent
variables a
Full b Step c Forw d Bacw e
Coef P>|t| Coef P>|t| Coef P>|t| Coef P>|t|
DFPHarm-Drink 11.84 0.40 - - - - - -
DFPS-AI-Drink -9.28 0.43 -13.21 0.14 -12.48* 0.16 -13.21 0.14
DFPS-Diarrhea-
Drink 19.64 0.18 17.13* 0.09 16.19* 0.10 17.13* 0.09
DFPS-Coliform-
Drink -8.66 0.45 - - - - - -
DFPS-Parasite-Drink -26.73** 0.05 -19.40** 0.05 -17.41** 0.07 - -
DAI severe - - - - - -19.40** 0.05
DSatisfy-Drink -1.82 0.87 - - - - - -
constant 100.55** 0.02 110.97*** 0.00 129.68*** 0.00 110.97*** 0.00
Obs 115 - 138 - 138 - 138 -
Prob > F 0.00*** - 0.00*** - 0.00*** - 0.00*** -
R-squared 0.34 - 0.40 - 0.39 - 0.40 -
Adj R-squared 0.23 - 0.35 - 0.35 - 0.35 -
Root MSE 49.19 - 44.18 - 44.24 - 44.18 -
Number of significant
variables 5 - 6 - 7 - 7 - Notes: AI = Avian Influenza, D = Dummy, FPS = farmers’ perceived susceptibility to, a Independent Variables described in Table 4.6; bFull = full simultaneous
regression model; cStep = Stepwise selection regression model; dForw = forward selection regression model; eBacw = backward selection regression model; ere =
removed from the model; (*) significant at p<0.1; (**) significant at p<0.01
Page 325 of 340
Table F.11a. Summary of the Best MLR Model Fittings for E. coli cfu in River Water for Drinking by Selection Procedures in
the Provinces of Thai Binh and An Giang Provinces, Vietnam
Independent
variables a
Full b Step c Forw d Bacw e
Coef P>|t| Coef P>|t| Coef P>|t| Coef P>|t|
DGender
No sig
nifican
t variab
le found in
this fu
ll model fittin
g
pro
cedure
14.12** 0.05 14.12** 0.05 14.12** 0.0
Age - - - - - -
Schooling -1.61* 0.07 -1.61* 0.07 -1.61 0.07
Experience - - - - - -
DIncome 8.55 0.16 8.55 0.16 8.55 0.16
DChildren - - - - - -
DOff-farm job - - - - - -
Npoultry/yr 0.75*** 0.01 0.75* 0.01 0.75*** 0.01
Nducks/yr -0.02 0.15 -0.02 0.15 -0.02 0.15
Npigs/yr - - - - - -
DFish prod - - - - - -
DFPR-Livestock-Man - - - - - -
DFPR-Livestock-
Woman - - - - - -
DFPWater quality - - - - - - Notes: D = Dummy, FPS = farmers’ perceived main responsibility, N = Number, a Independent Variables described in Table 4.6; bFull = full simultaneous
regression model; cStep = Stepwise selection regression model; dForw = forward selection regression model; eBacw = backward selection regression model; ere
= removed from the model; (*) significant at p<0.1; (**) significant at p<0.01
Page 326 of 340
Table F.11b. Summary of the Best MLR Model Fittings for E. coli cfu in River Water for Drinking by Selection Procedures in
the Provinces of Thai Binh and An Giang, Vietnam (cont’d)
Independent
variables a
Full b Step c Forw d Bacw e
Coef P>|t| Coef P>|t| Coef P>|t| Coef P>|t|
DFPHarm-Drink
No sig
nifican
t variab
le found in
this fu
ll model
fitting p
roced
ure
- - - - - -
DFPS-AI-Drink - - - - - -
DFPS-Diarrhea-Drink - - - - - -
DFPS-Coliform-Drink - - - - - -
DFPS-Parasite-Drink - - - - - -
DSatisfy-Drink - - - - - -
constant 1.79 0.82 1.79 0.82 1.79 0.82
Obs 111 111 - 111 -
Prob > F 0.03** - 0.03** - 0.03** -
R-squared 0.11 - 0.11 - 0.11 -
Adj R-squared 0.07 - 0.07 - 0.07 -
Root MSE 30.15 - 30.15 - 30.15 -
Number of significant
variables 0 3 3 2 Notes: D = Dummy, FPS = farmers’ perceived susceptibility to, FP = Farmers’ perceptions of, a Independent Variables described in Table 4.6; bFull = full
simultaneous regression model; cStep = Stepwise selection regression model; dForw = forward selection regression model; eBacw = backward selection
regression model; ere = removed from the model;(*) significant at p<0.1; (**) significant at p<0.01
Page 327 of 340
Table F.12a. Summary of the Best MLR Model Fittings for E. coli cfu in River Water for Domestic Purposes by Selection
Procedures in the Provinces of Thai Binh and An Giang, Vietnam
Independent
variables a
Full b Step c Forw d Bacw e
Coef P>|t| Coef P>|t| Coef P>|t| Coef P>|t|
DGender -488.94* 0.07 -338.80* 0.08 -328.88* 0.09 -338.80* 0.08
Age 9.58 0.23 11.06* 0.11 10.86 0.12 11.06* 0.11
Schooling 7.65 0.76 - - - - - -
Experience 9.13 0.39 - - - - - -
DIncome -181.19 0.33 -264.47* 0.10 -266.40 0.12 -264.47* 0.10
DChildren -17.04 0.94 - - - - - -
DOff-farm job 267.70* 0.10 264.33* 0.08 265.29* 0.08 264.33* 0.08
Npoultry/yr - - - - - - - -
Nducks/yr - - - - - - - -
Npigs/yr 38.07*** 0.01 36.97*** 0.01 37.54*** 0.01 36.97*** 0.01
DFish prod -31.03 0.87 - - - - - -
DFPR-Livestock-
Man -404.43*** 0.03 -368.40** 0.03 -365.00** 0.03 -368.40** 0.03
DFPR-Livestock-
Woman -263.49 0.39 - - - - - - Notes: D = Dummy, FPS = farmers’ perceived responsibility to, FP = Farmers’ perceptions of, N = Number, \a Independent Variables described in Table
4.6; bFull = full simultaneous regression model; cStep = Stepwise selection regression model; dForw = forward selection regression model; eBacw =
backward selection regression model; *) significant at p<0.1; (**) significant at p<0.01
Page 328 of 340
Table F.12b. Summary of the Best MLR Model Fittings for E. coli cfu in River Water Used for Domestic Purposes by Selection
Procedures in the Provinces of Thai Binh and An Giang, Vietnam (cont’d)
Independent
variables a
Full b Step c Forw d Bacw e
Coef P>|t| Coef P>|t| Coef P>|t| Coef P>|t|
DFPWater quality -37.27 0.85 - - - - - -
DFPHarm-Dom -233.61 0.26 -278.42* 0.10 -233.06 0.24 -278.42* 0.10
DFPS-AI-Dom 312.53 0.13 279.94 0.16 295.27 0.14 279.94 0.16
DFPS-Diarrhea-Dom -95.02 0.69 - - -102.39 0.65 - -
DFPS-Coliform-
Dom 101.43 0.78 - - - - - -
DFPS-Parasite-Dom -504.85** 0.02 -482.03*** 0.01 -447.77** 0.03 -482.03*** 0.01
DSatisfy-Dom 186.27 0.36 - - - - - -
constant 847.70 0.12 824.62** 0.03 832.28** 0.02 824.62** 0.03
Obs 189 - 188 - 188 - 188 -
Prob > F 0.01*** - 0.00*** - 0.00*** - 0.00*** -
R-squared 0.18 - 0.16 - 0.17 - 0.16 -
Adj R-squared 0.09 - 0.12 - 0.12 - 0.12 -
Root MSE 1008.80 - 990.43 - 992.64 - 990.43 -
Number of
significant variables 5 - 9 - 6 - 8 - Notes: AI = Avian Influenza, D = Dummy, Dom = Domestic, FPS = farmers’ perceived susceptibility to, FP = Farmers’ perceptions of, a Independent
Variables described in Table 4.6; bFull = full simultaneous regression model; cStep = Stepwise selection regression model; dForw = forward selection
regression model; eBacw = backward selection regression model; *) significant at p<0.1; (**) significant at p<0.01
Page 329 of 340
Table F.13a. Summary of the Best MLR Model Fittings for E. coli in Pond Water for Domestic Purposes by Selection
Procedures in the Provinces of Thai Binh and An Giang, Vietnam
Independent variables a
Full b Step c Forw d Bacw e
Coef P>|t| Coef P>|t| Coef P>|t| Coef P>|t|
DGender 14.97 0.98 - - - - - -
Age -7.28 0.73 - - - - - -
Schooling 118.27 0.13 213.66** 0.02 213.66** 0.02 213.66** 0.02
Experience -41.08 0.14 - - - - - -
DIncome
-
1,139.09* 0.11 - - - - - -
DChildren 268.30 0.55 - - - - - -
DOff-farm job 719.17* 0.08 - - - - - -
Npoultry/yr -7.18 0.18 -17.79* 0.10 -17.79* 0.10 -17.79* 0.10
Nducks/yr - -1.03 0.20 14.78* 0.11 -1.03 0.20
Npigs/yr - 14.78* 0.11 -1.03 0.20 14.78* 0.11
DFish prod -266.54 0.68 - - - - - -
DFPR-Livestock-Man -56.70 0.90 - - - - - -
DFPR-Livestock-
Woman -383.79 0.52 - - - - - -
DFPWater quality -701.91 0.23 - - - - - -
DFPHarm-Dom 352.23 0.53 - - - - - - Notes: D = Dummy, FPR = farmers’ perceived responsibility to, FP = Farmers’ perceptions of, a Independent Variables described in Table 4.6; bFull = full
simultaneous regression model; cStep = Stepwise selection regression model; dForw = forward selection regression model; eBacw = backward selection regression
model; (*) significant at p<0.1; (**) significant at p<0.01
Page 330 of 340
Table F.13b. Summary of the Best MLR Model Fittings for E. coli in Pond Water for Domestic Purposes by Selection
Procedures in the Provinces of Thai Binh and An Giang, Vietnam (cont’d)
Independent
variables a
Full b Step c Forw d Bacw e
Coef P>|t| Coef P>|t| Coef P>|t| Coef P>|t|
DFPS-AI-Dom 211.36 0.68 1,044.65** 0.05 1,044.65** 0.05 1,044.65** 0.05
DFPS-Diarrhea-
Dom 206.70 0.72 - - - - - -
DFPS-Coliform-
Dom 544.62 0.29 - - - - - -
DFPS-Parasite-
Dom -488.10 0.39 - - - - - -
DSatisfy-Dom 619.99 0.17 1,028.56* 0.09 1,028.56* 0.09 1,028.56* 0.09
constant 2,427.87 0.12 210.21 0.81 210.21 0.81 210.21 0.81
Obs 299 - 197 - 197 - 197 -
Prob > F 0.11 - 0.02 - 0.02 - 0.02 -
R-squared 0.08 - 0.08 - 0.08 - 0.08 -
Adj R-squared 0.03 - 0.05 - 0.05 - 0.05 -
Root MSE 3,285.70 - 3,433.80 - 3,433.80 - 3,433.80 -
Number of
significant
variables 2 - 5 - 5 - 5 - Notes: AI = Avian Influenza, D = Dummy, Dom = Domestic, FPS = Farmers’ perceived susceptibility to, a Independent Variables described in Table 4.6; bFull =
full simultaneous regression model; cStep = Stepwise selection regression model; dForw = forward selection regression model; eBacw = backward selection
regression model; (*) significant at p<0.1; (**) significant at p<0.01
Page 331 of 340
F.6. Detailed Results of SUR Model for E. coli in Sources of Water On Farms
The best fit of the SUR model for estimating two equations of E. coli cfu in sources of
drinking water and sources of domestic water is presented in Tables F.14-16. Overall, the two
equations for E. coli cfu in drinking and domestic water were significant at p < 0.1. The SUR
model fitting for E. coli cfu in drinking water has four associated factors of the levels of E. coli
cfu in drinking water among which one was positive significant predictor. The SUR model
fitting for E. coli cfu in domestic water has five significant associated factors. Variable
Nducks/yr on farm showed up as a factor in both equations with different coefficients and p-
values. Table F.15 shows the correlation matrix on the residuals of the two equations for E. coli
cfu in drinking and domestic water. The Breusch-Pagan test of independence shows that the
residuals of the two equations were significantly related (p < 0.05). Therefore, the use of SUR is
justified.
Table F.14. SUR Model Estimation of the Two Equations of E. coli cfu in Water Used for
Drinking
Iteration 1: tolerance = 0.488822
(… results of iteration 2-5 omitted)
Iteration 6: tolerance = 3.22E-08
Seemingly unrelated regression, iterated
Equation Obs Parms RMSE R-sq chi2 P
NE.coli Drink 243 4 14078.81 0.03 8.26 0.08
NE.coli Dom 243 26 2761.06 0.17 50.45 0.00
Table F.15. SUR Model Estimation of the Two Equations of E. coli cfu in Water Used for
Drinking (cont'd)
Correlation matrix of residuals:
NE.coli Drink NE.coli Dom
NE.coli Drink 1
NE.coli Dom 0.14 1
Breusch-Pagan test of independence: chi2(1) = 4.60, Pr = 0.03
Page 332 of 340
Table F.16. SUR Model Estimation of the Two Equations of E. coli cfu in On-Farm Water
Coef. Std. Err. z P>z [95% Conf. Interval]
NE.coli Drink
DOff-farm job 3163.55** 1873.55 1.69 0.09 -508.54 6835.64
NDucks/yr 5.51 4.48 1.23 0.22 -3.26 14.28
DSatisfy-Drink -1706.41 1856.55 -0.92 0.36 -5345.19 1932.37
DFPHarm-Drink -3422.72 2243.79 -1.53 0.13 -7820.47 975.04
_cons 1870.37 2292.16 0.82 0.42 -2622.18 6362.92
NE.coli Dom
Age 0.11 19.03 0.01 1.00 -37.19 37.41
DGender 160.58 494.00 0.33 0.75 -807.66 1128.81
Schooling 195.04* 70.12 2.78 0.01 57.62 332.46
DChildren -354.00 457.98 -0.77 0.44 -1261.63 543.64
Experience -45.37 29.34 -1.55 0.12 -102.88 12.14
DFPR-Livestock-Woman -941.67 737.85 -1.28 0.20 -2387.82 504.49
DFPR-Livestock-Man -987.56** 543.15 -1.82 0.07 -2052.13 77.00
DFPR-Fish-Woman -619.64 930.68 -0.67 0.51 -2443.75 1204.46
DFPR-Fish-Man 301.12 596.29 0.50 0.61 -867.59 1469.83
DFPR-Health-Man 275.55 646.56 0.43 0.67 -991.68 1542.78
DProvince 43.29 697.51 0.06 0.95 -1323.81 1410.39
DFish prod 1217.73* 611.38 1.99 0.05 19.45 2416.00
DRain for animals 204.62 548.58 0.37 0.71 -870.59 1279.82
DPond for animals 21.48 451.33 0.05 0.96 -863.11 906.07
DPipe for animals -365.64 687.70 -0.53 0.60 -1713.51 982.23
DMulti-agri activities -523.14 640.61 -0.82 0.41 -1778.72 732.44
DPoultry AI 2281.77* 848.94 2.69 0.01 617.88 3945.66
NDucks/yr -0.01 0.99 -0.01 0.99 -1.95 1.93
NPoultry/yr -5.40 7.98 -0.68 0.50 -21.05 10.26
DFPWater quality -720.46 523.83 -1.38 0.17 -1747.14 306.22
DFPBarr-Cost 230.33 480.15 0.48 0.63 -710.75 1171.40
DFPRisk rain water -1266.58 1153.19 -1.09 0.28 -3516.79 1003.62
DFPS-Parasite-Dom 68.45 586.73 0.12 0.91 -1081.53 1218.42
DAI heard -2682.34** 1693.55 -1.58 0.10 -6001.65 636.97
DFPHarm-Dom 508.12 493.76 1.03 0.30 -459.63 1475.87
DFPE-Livestock
Management -361.33 548.19 -0.66 0.51 -1435.76 713.10
_cons 5472.15 2344.87 2.33 0.02 876.28 10068.02
Notes: AI = Avian Influenza, D = Dummy, Dom = Domestic, FPS = Farmers’ perceived susceptibility to, FP =
Farmers’ perceptions of, * significant at p < 0.05; ** significant at p < 0.1
Page 333 of 340
The variable of number of ducks (Nducks/yr) was presented in the two individual
equations for E. coli cfu in drinking water and domestic water respectively (Table F.16).
Therefore, a test of constraint for the Nducks/yr variable was conducted. The test of cross-
equation constraint showed the Nducks/yr variable is not significantly different from the two
individual equations at p = 0.22 .The SUR model was run again with the cross-equation
constraint of the Nducks/yr variable.
Page 334 of 340
APPENDIX G: CHAPTER FIVE’S APPENDICES
G.1. Summary Statistics of Variables Used in the Probit Models
Table G.1. Summary Statistics of Demographic Variables Used in Probit Models of Factors
Associated with Small-Scale Integrated (SSI) Farmers’ Engagement in Mitigating
Strategies to Reduce WRZD Transmission
Variable Description Mean s.d. Count (=1) % (=1)
DProvince
Two provinces in the research (1 = Thai
Binh; 0 = An Giang) - - 300 50.2
Age Age of the participating SSI farmers 45.9 11.3 - -
Schooling Years of attending school 6.9 3.2 - -
Experience Years of farming 9.6 8.3 - -
DChildren
Having household members under 18 years
old (1= Yes; 0 = No) - - 421 70.4
Mainly responsible by for:
DFPR-Health-
Man family health by man (1 = Yes; 0 = No) - - 131 22.2
DFPR-Health-
Woman family health by women (1 = Yes; 0 = No) - - 104 17.6
DFPR-Livestock-
Man
Mainly responsible for livestock production
by gender (1 = Yes; 0 = No) - - 244 41.4
DFPR-Livestock-
Woman
Mainly responsible for livestock production
by gender (1 = Yes; 0 = No) - - 91 15.5
DIncome
On-farm DIncome per person per year
below and above poverty line (1= Yes; 0 =
No) - - 118 20.0
Number of livestock by type/year
Npoultry/yr chickens 24.3 21.2 - -
Nducks/yr ducks 36.2 47.9 - -
Npigs/yr pigs 18.9 18.2 - -
Ncattle/yr cattle 3.1 2.1 - -
Number of E. coli cfu in:
NE.coli Drink on-farm drinking water 12.8 21.0 - -
NE.coli Dom on-farm domestic water 1,506.3 1,994.6 - -
LogE.coli Drink on-farm drinking water (log transformed) 1.9 2.0 - -
LogE.coli Dom on-farm domestic water (log transformed) 6.2 2.1 - -
Presence/Absence of E. coli cfu in:
DE.coli Domestic
on-farm domestic water (1 = Presence; 0 =
Absence) - - 574 96.0
DE.coli Drink
on-farm drinking water (1 = Presence; 0 =
Absence) - - 314 58.7
Turbidity levels in:
Turbidity drinking on-farm drinking water 1.1 0.9 - -
Turbidity
domestic on-farm domestic water 43.6 28.6 - -
Notes: D = Dummy, N = Number, FPR = Farmers’ perceived main responsibility
Page 335 of 340
G.2. Summary Statistics of SSI Farmers’ Perceptions Related Variables Used in the Probit Models
Table G.2. Summary Statistics of SSI Farmers’ Perceptions Related Variables Used in Probit Models to Explore Factors
Associated with SSI Farmers’ Engagement in Mitigating Strategies to Reduce WRZD Transmission
Variable name Description – farmers’ perceptions of: Count (=”1”) % (=”1”)
DFPS-AI-Drinking
susceptibility to avian influenza of untreated drinking water (1 = highly susceptible, 0 = low
susceptible) 313 52.3
DFPS-AI-Dom
susceptibility to avian influenza of untreated domestic water (1 = highly susceptible, 0 = low
susceptible) 289 48.3
DFPS-AI-Wastewater susceptibility to avian influenza of untreated wastewater (1 = highly susceptible, 0 = low susceptible) 435 88.0
DFPS-AI-Pond susceptibility to avian influenza of pond water (1 = highly susceptible, 0 = low susceptible) 401 83.7
DAI severe severity of avian influenza if infected water (1 = very severe, 0 = not very serve) 481 80.4
DFPSafe-Drink safety of untreated water for drinking (1 = Yes, it is safe; 0 = No, it is not safe) 54 9.0
DFPBarr-Cost cost is a barrier to taking actions to reduce WRZD transmission (1 = Yes; 0 = No) 239 40.0
DFPBarr-Not know “do not know” as a barrier to taking actions (1 = Yes; 0 = No) 232 38.9
DFPBarr-Not understand “do not understanding” as a barrier to taking actions to reduce WRZD transmission (1 = Yes; 0 = No) 121 20.2
DFPBarr-Peer peer pressure as a barrier to taking actions to reduce WRZD transmission (1 = Yes; 0 = No) 28 4.7
DFPBarr-Busy busy as a barrier to taking actions to reduce WRZD transmission (1 = Yes; 0 = No) 151 25.3
DFPBen-Action benefit of taking actions to reduce WRZD transmission (1 = Yes; 0 = No) 452 77.9
DFPBen-No Action benefit of not taking actions to reduce WRZD transmission (1 = Yes; 0 = No) 132 24.1
DFPE-Livestock Management self-capacity in managing livestock (1 = Strong; 0 = Weak) 263 44.4
DFPE-Point of Use self-capacity in using point of use methods (e.g., washing hand with soap) (1 = Strong; 0 = Weak) 250 44.9
DFPE-Water Protection self-capacity in protecting source water (1 = Strong; 0 = Weak) 255 42.8
DFPE-Water Distribution self-capacity in water storage, treatment, distribution (1 = Strong; 0 = Weak) 239 40.4
DFPT-Health Advice advice from health workers as a trigger of actions to reduce WRZD transmission (1 = Yes; 0 = No) 295 49.3
DFPT-Loss Income loss DIncome as a trigger of actions to reduce WRZD transmission (1 = Yes; 0 = No) 165 27.6
DFPT-Media media as a trigger of actions to reduce WRZD transmission (1 = Yes; 0 = No) 405 67.7
DFPT-Peer peer pressure as a trigger of actions to reduce WRZD transmission (1 = Yes; 0 = No) 101 16.9
DFPT-Worry worry as a trigger of actions to reduce WRZD transmission (1 = Yes; 0 = No) 374 62.5
Notes: D = Dummy, FPS = Farmers’perceived susceptibility, AI = Avian Influenza, FPSafe = Farmers’ perceived safety of, FPBarr = Farmers’ perceived
barriers of, FPBen = Farmers’ perceived benefits of, FPE = Farmers’ perceived self-efficacy, FPT = Farmers’ perceived triggers of
Page 336 of 340
References
Bartram, J. and Pedley, S. (1996). Microbial Analysis. In J.Bartram and R. Ballance (Eds.),
Water Quality Monitoring - A Practical Guide to the Design and Implementation of
Freshwater Quality Studies and Monitoring Programmes ( World Health Organization.
Bluewater Biosciences (2013). Coliplate (TM): A quantitative 24-hour test for Coliforms and E.
coli. Bluewater Biosciences [On-line]. Available:
http://bluewaterbiosciences.com/index.php?main_page=pageandid=16
Dohoo, I., Martin, W., and Stryhn, H. (2003). Veterinary Epidemiologic Research.
Edberg, S. C., Rice, E. W., Karlin, R. J., and Allen, M. J. (2000). Escherichia coli: the best
biological drinking water indicator for public health protection. Journal of Applied
Microbilogy, 88, 106S-116S.
EPA (1993). Method 180.1: Determination of Turbidity by Nephelometry. EPA [On-line].
Available:
http://water.epa.gov/scitech/methods/cwa/bioindicators/upload/2007_07_10_methods_me
thod_180_1.pdf
FPTCW (2013). Guideline Technical Document: Escherichia coli. Ottawa, Ontario: Federal-
Provincial-Territorial Committee on Drinking Water of the Federal-Provincial-Territorial
Committee on Health and the Environment, Health Canada.
Fricker, C. R., DeSarno, M., Warden, P. S., and Eldred, B. J. (2008). False negative ß-D-
glucuronidase reactions in membrane lactose glucuronide agar medium used for the
simultaneous detection of coliforms and Escherichia coli from water. Letters in Applied
Microbiology, 47, 539-542.
Page 337 of 340
Greene, W. H. (2008a). System of Regression Equations. In Econometric Analysis (6th Edition
ed., pp. 252-280). Pearson Prentice Hall.
Greene, W. H. (2008b). The Classical Multiple Linear Regression Model. In Econometric
Analysis (6th Edition ed., pp. 8-17). Pearson Prentice Hall.
Hall, D. C., Thao, T. D., Minh, D. V., and Lien, L. V. (2006). Competitiveness of the livestock
sector in Vietnam World Bank paper. WB-EASRD and FAO-TCIP.
Hall, T. A. (1999). BioEdit: a user-friendly biological sequence alignment editor and
analysisprogram for Windows 95/98/NT. Nucleic Acids Symposium Series, 95-98.
IDRC (2012). Ecohealth Research in Practice. (Insight and Innovation in International
Development ed.) (vols. Innovative Applications of an Ecosystem Approache to Health)
International Development Centre (IDRC); Springer.
IHPH (2013a). Technical Guidelines: Procedure for Determination of pH in Water. Institute of
Hygiene and Public Health Ho Chi Minh City.
IHPH (2013b). Technical Guidelines: Procedure for Determination of Turbidity in Water.
Institute of Hygiene and Public Health Ho Chi Minh City.
Irini, M. and Rapsomanikis, G. (2005). Contribution of Livestock to Household income in
Vietnam: A household typology based analysis (Rep. No. PPLPI Working Paper N.21). A
Living from Livestock, FAO.
Knudsen, L. G., Pham, D. P., Nguyen, H. T., Helle, S., Jensen, P. K., Dalsgaard, A. et al. (2008).
The fear of awful smell: risk perceptions among farmers in Vietnam using wastewater
and human excreta in agriculture. The Southeast Asian journal of tropical medicine and
public health, 39, 341-352.
Page 338 of 340
Long, J. S. and Freese, J. (2014). Regression Models for Categorical Dependent Variables Using
Stata. (Third ed.) Texas: Stata Press College Station.
MoH (2009a). National technical regulation on domestic water quality (QCVN 02: 2009/BYT).
Hanoi: Department of Preventive Medicine and Environment, Ministry of Health,
Socialist Republic of Vietnam.
MoH (2009b). National technical regulation on drinking water quality (QCVN 01: 2009/BYT).
Hanoi: Department of Preventive Medicine and Environment, Ministry of Health,
Socialist Republic of Vietnam.
MoH and MARD (2011). The Vietnam Integrated National Operational Program on Avian
Inlfuenza, Pandemic Preparedness and Emerging Infectious Diseases (APEID) 2010-
2015: Strengthening responses and improving prevention through a One Health
approach. Hanoi: Vietnam Ministry of Agriculture and Rural Development and Ministry
of Health.
Payment, P. and Pintar, K. (2006). Waterborne Pathogens: A Critical Assessment of Methods,
Results and Data Analysis. Journal of Water Science, 19, 233-245.
Pica-Ciarmarra, U., Tasciotti, L., Otte, J., and Zezza, A. (2011). Livestock Assets, Livestock
Income and Rural Households: Cross-Country Evidence from Household Surveys (Rep.
No. ESA Working Paper No. 11-17). Rome: World Bank, FAO, AU-IBAR, ILRI.
Powell, C. (2007). The Perception of Risk and Risk Taking Behavior: Implications for Incident
Prevention Strategies. Wilderness and environmental medicine [On-line]. Available:
http://linkinghub.elsevier.com/retrieve/pii/S1080603207701965?showall=true
TCVN 6187 (2009). Water quality: Detection and enumeration of coliform organisms
thermotolerant coliform organisms and presumptive Escherichia coli. Part 2: Membrane
Page 339 of 340
filtration method. Directorate for Standards, Metrology and Quality, Ministry of Science
and Technology of the Socialist Republic of Vietnam [On-line]. Available:
http://www.tcvn.vn/sites/head/en/tim-kiem-tieu-chuan.aspx?sk=andcat=tcvn
TCVN 6492 (2011). Water quality. Determination of pH. Directorate for Standards, Metrology
and Quality, Ministry of Science and Technology of the Socialist Republic of Vietnam
[On-line]. Available: http://www.tcvn.vn/sites/head/en/tim-kiem-tieu-
chuan.aspx?sk=andcat=tcvn
TCVN 6663 (2011). TCVN 6663 2011 - ISO 5667-1:2006: Water quality - Sampling - Part 1:
Guidance on the design of sampling programmes and sampling techniques. Directorate
for Standards, Metrology and Quality, Ministry of Science and Technology of the
Socialist Republic of Vietnam [On-line]. Available:
http://www.tcvn.vn/sites/head/en/tim-kiem-tieu-chuan.aspx?sk=andcat=tcvn
Trinh, T. L. and Nguyen, P. L. (2009). Assessment of surface water quality by water quality
index (WQI) at the Cai Sao canal, An Giang province, Vietnam. Livestock Research for
Rural Development [On-line]. Available: http://www.lrrd.org/lrrd23/7/lan23151.htm
Ward, H., Toledano, M. B., Shadick, G., Davies, B., and Elliott, P. (2012). Oxford Handbook of
Epidemiology for Clinicians. (1 ed.) Oxford University Press.
WHO (2011a). Guidelines for drinking-water quality. (Fourth ed.) World Health Organisation.
WHO (2011b). Microbial aspects. In Guidelines for drinking-water quality (pp. 117-153). World
Health Organization.
WHO (2012). Rapid assessment of drinking water quality - A handbook for implementation.
World Health Organization.
Page 340 of 340
WHO and OECD (2003a). Assessing Microbial Safety of Drinking Water: Improving
Approaches and Methods. IWA Publishing.
WHO and OECD (2003b). Assessing Microbial Safety of Drinking Water: Improving
Approaches and Methods. IWA Publishing.
Winfield, M. D. and Groisman, E. A. (2003). Role of nonhost environments in the lifestyles of
Salmonella and Escherichia coli. Applied and Environmental Microbiology, 67, 3678-
3694.