promoting appropriate management of diarrhea: a systematic ... · objective: to identify,...

INDIAN PEDIATRICS 627 VOLUME 49__AUGUST 16, 2012

Diarrhea is one of the top five causes of deathamong infants and under-five children inIndia [1], despite the availability of easilyimplementable interventions and existence of

National Guidelines for management at the communitylevel. Oral rehydration therapy (ORT) with oralrehydration salt (ORS) solution remains the cornerstoneof appropriate case management of diarrheal dehydrationand is considered the single most effective strategy toprevent diarrheal deaths in children. However, data fromUnited Nations Children’s Fund (UNICEF) coverageevaluation survey (CES) [2] and the third NationalFamily Health Survey (NFHS-3) [3] show that ORSusage rates are still unacceptable; while unwarranted anti-

Promoting Appropriate Management of Diarrhea: A Systematic Review ofLiterature for Advocacy and Action: UNICEF-PHFI Series on Newbornand Child Health, India

DHEERAJ SHAH, *PANNA CHOUDHURY, PIYUSH GUPTA, #JOSEPH L MATHEW, $TARUN GERA, **SIDDHARTHA GOGIA,##PAVITRA MOHAN, ‡RAJMOHAN PANDA AND ‡SUBHADRA MENON

From the Department of Pediatrics, University College of Medical Sciences (University of Delhi), New Delhi, ¶Advanced PediatricsCenter, PGIMER, Chandigarh; *Child Health Foundation; $Sunder Lal Jain Hospital, New Delhi; **Max Hospital, Gurgaon,Haryana; ##UNICEF, India; and ‡Public Health Foundation of India, New Delhi, India.

Correspondence to:Dr Dheeraj Shah, Associate Professor, Department of Pediatrics, University College of Medical Sciences andGTB Hospital, Dilshad Garden, Delhi 110 095, India. [email protected]

Background: Scaling up of evidence-based management and prevention of childhood diarrhea is a public health priority in India, andnecessitates robust literature review, for advocacy and action.

Objective: To identify, synthesize and summarize current evidence to guide scaling up of management of diarrhea among under-fivechildren in India, and identify existing knowledge gaps.

Methods: A set of questions pertaining to the management (prevention, treatment, and control) of childhood diarrhea was identifiedthrough a consultative process. A modified systematic review process developed a priori was used to identify, synthesize and summarize,research evidence and operational information, pertaining to the problem in India. Areas with limited or no evidence were identified asknowledge gaps.

Results: Childhood diarrhea is a significant public health problem in India; the point (two-weeks) prevalence is 9-20%. Diarrhea accountsfor 14% of the total deaths in under-five children in India. Infants aged 6-24 months are at the highest risk of diarrhea. There is a lack ofrobust nation-wide data on etiology; rotavirus and diarrheogenic E.coli are the most common organisms identified. The current NationalGuidelines are sufficient for case-management of childhood diarrhea. Exclusive breastfeeding, handwashing and point-of-use watertreatment are effective strategies for prevention of all-cause diarrhea; rotavirus vaccines are efficacious to prevent rotavirus specificdiarrhea. ORS and zinc are the mainstay of management during an episode of childhood diarrhea but have low coverage in India due topolicy and programmatic barriers, whereas indiscriminate use of antibiotics and other drugs is common. Zinc therapy given duringdiarrhea can be upscaled through existing infrastructure is introducing the training component and information, education andcommunication activities.

Conclusion: This systematic review summarizes current evidence on childhood diarrhea and provides evidence to inform child healthprograms in India.

Key words: Communication barriers, Diarrhea, National Health Programs, Operations Research, Oral rehydration therapy, Zinc.

SSSSS YYYYY SSSSS TTTTT EEEEE MMMMM AAAAA TTTTT IIIII C C C C C R E V I E WR E V I E WR E V I E WR E V I E WR E V I E W

diarrheal drugs and injections continue to be prescribedfrequently. Moreover, there is lack of knowledge andawareness amongst care providers on how to implementand achieve greater coverage of existing cost-effectiveinterventions.

This systematic review of literature was undertaken toprovide evidence-based guidance for advocacy and actiontowards better management of childhood diarrhea in India.This exercise was a part of the combined initiative ofPublic Health Foundation of India (PHFI), UNICEF India,and a team of independent researchers for advocacy andaction focused on locally relevant issues [4]. The specificobjective was to identify, synthesize and summarize


SHAH, et al. PROMOTING APPROPRIATE MANAGEMENT OF DIARRHEA

evidence pertaining to diarrhea (disease burden, etiology,preventive interventions, client practices and prescriptionpractices) in children aged less than five years, withparticular reference to the Indian context. The reviewfurther aimed to identify knowledge gaps and operationalbottlenecks that plague the current program concerningprevention and control of diarrhea for under-five childrenin India.

METHODS

The methodology for this systematic review has beenpresented earlier [4]. A set of key questions for reviewwere finalized through consensus building, andcategorized as ‘Technical Issues’ and ‘OperationalIssues’ [4]. The details of the search strategy for eachquestion have been presented in WebTable I. Literaturesearches were initially carried out during April 2010, andupdated on 10 January 2012.

RESULTS

1. Diarrheal Morbidity and Mortality in India

According to ‘SRS Report (2009) on Causes of Death(2001-2003)’, diarrhea is the third most common cause ofdeath in under-five children in India, responsible for 14%deaths in this age group [1]. Diarrheal illnesses are theleading causes of childhood deaths beyond infancy; it isresponsible for 24% of the deaths in children aged 1-4years, and 17% of all deaths in children 5-14 years.[1].World Health Organization (WHO) estimates ofmortality from 34 studies published between 1992 and2000 suggest that 4.9 children per 1000 per year indeveloping countries died as a result of diarrheal illnessin the first 5 years of life [5]. This has declined over theyears from 13.6 (1982) and 5.6 (1992) per 1000 per year[6,7]. The decrease was most pronounced in childrenaged under 1 year. Diarrhea accounted for a median of21% of all deaths of children aged under 5 years in theseareas and countries, being responsible for 2.5 milliondeaths per year. Lancet child survival series, using aprediction model, estimated that 22% (14-30%) of allunder-five deaths are attributable to diarrhea in 42countries, where 90% of all under-five deaths occur [8].Most recent prediction modeling data also conclude thatdiarrheal diseases globally are responsible for mostunder-five child deaths beyond neonatal age. This modelpredicts that 14% of under-five child deaths totaling toabout 0.24 million in India occur due to diarrhea [9]. Thisfigure is similar to the SRS verbal autopsy estimates.Further, there was marked regional variation withmortality rate from diarrheal diseases in Central Indiawas three times that in the West. Girls in Central India hadfour times higher diarrheal disease mortality rate

compared to boys in the West [10].

According to NFHS-3 report, 9% of all under-fivechildren were reported to be suffering from diarrhea in last2 weeks [3]. The corresponding figures for NFHS-2 andNFHS-1 were 19.2% and 10%, respectively [11,12].These figures are not truly comparable as these datasetswere obtained from mothers of children with different agegroups (< 4 yr in NFHS-1, < 3 yr in NFHS-2 and < 5 yr inNFHS-3). On comparing the different age groups, theprevalence was more or less similar in NFHS-1 andNFHS-3 whereas it was significantly higher in all agegroups up to three years in NFHS-2 survey. The reasonsfor a worsening trend between NFHS-1 and NFHS-2followed by a decline in NFHS-3 are not clear.Surprisingly, states with some of the worst healthindicators like UP, MP and Rajasthan reported a very lowprevalence of diarrhea in NFHS-1; it increased by 2-3times in NFHS-2. Another surprising finding from NFHS-1 was that almost 50% of the prevalence of diarrhea inprevious two weeks was contributed by prevalence duringlast 24 hours. This may suggest a significant recall bias inusing two week prevalence as an indicator. The second andthird survey did not present any data on 24 hourprevalence. Overall, there appears to be decrease inprevalence from NFHS-2 to NFHS-3. However, all thisdata have to be interpreted with caution as these reportedprevalence levels reflect mother’s perception of the illnessand not the medically certified illness.

More recent data from UNICEF 10-district surveyreport that 19.8% children (2-59 months) suffered fromdiarrhea in the two weeks preceding the survey [13].However, this UNICEF data was from some districts instates with higher child mortality rates, and therefore notrepresentative for the whole country. CountrywideUNICEF coverage evaluation survey reported 14.3% ofchildren (< 2 yrs) to be suffering from diarrhea in the twoweeks preceding the survey [2].

The maximum risk of diarrhea was in the age group of6-12 months across all the surveys [2,3,11,12]. Blood wasreported to be present in stools in around 10% of thediarrheal episodes in all the national health surveys[3,11,12]. Proportion of diarrheal episodes with visibleblood in stools was least in infants (2% in < 6 mo and 5% in6-12 mo; NFHS-3). No major rural/urban or genderdifferences were found, except for dysentery, which wasmore common in rural children. Prevalence of dysenteryalso had a negative correlation with mother’s educationalstatus and family’s wealth or standard of living index [2,3].

Very few studies have attempted to find the incidenceof diarrhea among children through well-designedlongitudinal studies. Analysis of 27 studies from



developing countries, including three from India,suggested that the median incidence of diarrhea for under-five children was 3.2 episodes per child-year [5], withmaximum incidence at 6-12 months (4.8 episodes perchild year). More recent data suggested that a child (<4-6years) suffers from average of 2-3 episodes of diarrhea peryear [14-16]. Web Table II presents the summary of allsuch studies conducted in India.

Conclusion and Comments: BOX 1

Knowledge gaps and directions for future research:Reliable information regarding the contribution ofdiarrhea to total deaths is missing because of absence ofmedical certification of cause of death in majority of cases.The risk factors and profile of children dying of diarrhea inthe community are not known. Whether most diarrhealdeaths in the community are occurring in outbreaksituations or occurring in isolation? Large scale incidencestudies are not available. Most of the studies, and even thenational databases, have estimated the incidence andprevalence of diarrhea in under-five children. Similar datain neonates and older children are scarce.

2. Etiology of Childhood Diarrhea in India

There is no large-scale nationally representativecommunity based study in last two decades regardingetiology or trends of diarrhea from India. Most studieshave been either designed to specifically evaluate thecontribution of a single etiological agent or are based onisolation of microbes from stool samples of hospitalizedpatients without their clinical details. Moreover, the

frequent presence of enteric pathogens in healthychildren from developing countries makes it moredifficult to determine their true etiological role incausation of diarrhea. The seasonality of some agentsalso preclude valid analysis from studies reporting caseswithin a short duration.

A multicentric (China, India, Mexico, Myanmar andPakistan), hospital-based study in 1991 evaluated theetiology of acute diarrhea (including those with presenceof blood or mucus in stools) in children (0-35 months)[20]. The study attempted to eliminate the role ofseasonality by enrolling comparable number of patientseach week throughout the two-year study-period. Thestudy also evaluated the presence of microorganisms inhealthy controls to further increase the validity. An entericpathogen could be detected in 68% cases. The organismsmore prevalent in cases than in controls were rotavirus,Shigella spp. and enterotoxigenic E. coli (ETEC). FromIndia (N=5,862), rotavirus was detected in 18%,Enterotoxigenic E. coli (ETEC) in 14% and Shigella spp.in 20%. Enteropathogenic E. coli (EPEC), Salmonellaspp., Clostridium jejuni were equally prevalent in casesand controls. Vibrio cholerae was isolated from 2% ofcases (0.2% of controls) whereas E. histolytica wasdetected only in 0.1% of cases (0.2% of controls).Rotavirus was isolated most frequently during the firstyear of life whereas Shigella spp. was the most commonisolate in children aged 12-35 months.

A recent study [21] amongst subjects (includingunder-five children) hospitalized with acute diarrhea,reported rotavirus (48%), diarrheogenic E. coli, Vibrio(19% each), Giardia (14%), adenovirus (12%), andCryptosporidium (11%) to be the most commonorganisms isolated from under-five children. Further, thisstudy documented mixed infections in a substantial (48%)proportion of children.

There has not been any similar attempt to evaluate theetiology of diarrhea from the community. Most studies haveconcentrated on the role of a specific etiological agent,especially rotavirus, in causation of diarrhea (WebTableIII). It is apparent from these studies that rotavirus is themost frequent etiological agent, being responsible for about15-30% of episodes in hospitalized children, and 7-15% incommunity. Almost all episodes of rotaviral diarrhealeading to hospitalization are reported in children less thantwo years of age. In an analytical review of 46 Indianstudies, rotavirus positivity was detected in 20% ofhospitalizations, 35% of neonatal infections, 15% ofsymptomatic infections in the community, and 23% ofnosocomial infections. The modeling data extrapolating theprevalence of rotavirus in diarrhea hospitalizations to total

BOX 1

• Diarrheal illnesses are the third most commoncause of death amongst under-five children, andis the leading cause beyond infancy.

• A child below five years of age suffers fromaverage of 2-3 episodes of diarrhea per year.

• The point-prevalence (last two weeks) of diarrheaamong under-five children is about 9-20%.

• A child is at maximum risk of diarrhea between6-12 mo of age.

• About 10% of the diarrheal episodes are dysentry;there has been no change in this trend over lasttwo decades.

• There are no major rural/urban or genderdifferences in prevalence of diarrhea.

• Dysentry is more common in rural areas, and inchildren from poorer families.

• There seems to be a positive trend of decreasein diarrheal morbidity and mortality.



deaths are not valid as most diarrhea deaths occur incommunity, and probably in malnourished children.Among diarrhea hospitalizations, the commonestG types were G1 and G2 while commonest P types were P8,P6 and P4.

Diarrheagenic E. coli, esp. Enteroaggregative E. coli,is the most common bacterial pathogen isolated in moststudies. However, it has also been isolated frequently incontrols. Vibrio cholerae predominantly occurs inoutbreaks, and most affected children are 2-5 yr olds. V.cholerae O1 is the predominant serogroup amongdiarrheal patients. The high isolation rate of Vibriocholerae from studies utilizing methodology of analysis ofsamples sent to the microbiology department for culture islikely to be related to the patient profile as it is likely thatsamples would be sent more commonly in children havingrice watery stools and severe dehydration to specificallylook for V. cholerae.

Conclusions and Comments: BOX 2

Knowledge gaps and directions for future research: Thereis need for large scale community based studies evaluatingthe etiology of diarrhea, rather than concentrating onidentification of one pathogen. Studies need to determineetiology of diarrhea, and correlates of diarrheal deaths inmalnourished children compared to well-nourishedchildren. Studies need to evaluate the contribution ofrotavirus and other factors in causation of diarrheal deaths.However, as deaths from diarrhea primarily occur insettings where access to medical care is limited, collectionof appropriate clinical specimens to perform amicrobiologic evaluation is extremely challenging.

3. National Guidelines for Diarrhea Control

In India, National diarrhea control program (CDD) wasimplemented from 1980 as a part of Sixth Five Year Plan(1980-85) with the primary thrust of improving theknowledge and practices of appropriate casemanagement among caretakers and health care providers,and primary objective of preventing deaths due todehydration. This program was integrated within ChildSurvival and Safe Motherhood (CSSM) program, whichevolved into Reproductive and Child Health (RCH-I andRCH-II) programs. These programs with more fundingand stronger management systems included integratedmanagement of childhood illness (IMCI) as a centralstrategy for child health and survival. Under IMCI,frontline workers and health professional are trained inintegrated management of newborns and sick children,including for diarrhea. In addition, RCH-II proposed tostrengthen availability of ORS at community levelthrough making them available at Aanganwadis, sub-centers and through alternate approaches such as socialmarketing and Public Distribution System. It also talks ofimproving families’ practices on home management ofdiarrhea through Behavior Change Communication. Themainstay of the case management approach during acutediarrhea included oral rehydration therapy (ORT),continued breastfeeding, continued semisolid feeding inchildren older than 6 months of age and use ofappropriate antimicrobials in cholera and bloodydiarrhea. Although there has been a decrease in diarrhealmortality over last three decades, the ORS use ratesduring diarrhea have hardly changed. The practices ofcontinued feeding and increase in fluids during diarrheaalso are sub-optimal.

The current Government of India guidelinesrecommend low osmolarity ORS, zinc and continuedfeeding of energy dense feeds in addition to breastfeedingfor management of diarrhea [41,42]. Antimicrobials arerecommended only for gross blood in stools or Shigellapositive culture, cholera, associated systemic infection, orsevere malnutrition. The first line antibiotic recommendedfor treatment of dysentery is oral ciprofloxacin [41,42].Measures for prevention of diarrhea include promotingexclusive breastfeeding, use of safe water, handwashing,food safety, safe disposal of excreta, and immunizationagainst measles. These recommendations take intoaccount new research findings while building on pastrecommendations. Two recent advances in managingdiarrheal disease: (i) reduced osmolarity oral rehydrationsalts (ORS) containing lower concentrations of glucoseand salt, and (ii) zinc supplementation as part of thetreatment; have the potential to further reduce diarrhealmorbidity and mortality by reducing the duration and

BOX 2

• There is no large-scale comprehensivecommunity based study regarding etiology ofdiarrhea from India.

• Rotavirus is responsible for about 15-30% ofepisodes in hospitalized children; 7-15% incommunity. Almost all episodes leading tohospitalization are reported in children < 2 years.It is also the most commonly isolated agent inneonatal infections.

• Diarrheogenic E. coli is the most commonbacterial pathogen.

• Shigella spp. is responsible for about 10-20% ofepisodes. Visible blood is present in stools in halfto two-thirds of these episodes.

• Vibrio cholerae predominantly occurs inoutbreaks. Most affected children are 2-5 yr olds.



severity of diarrheal episodes and lower their incidence.

National Expert Group formulated by the Ministry ofHealth, Government of India proposed that a singleuniversal ORS solution containing sodium 75 mmol/L andglucose 75 mmol/L, and with osmolarity 245 mOsmol/Lwas acceptable for all ages and all types of diarrhea. Therevised formulation was approved by the Drug Controllerof India and the Government formally launched it in June2004. Similarly, based on the WHO/UNICEF/IAPrecommendations, the Ministry of Health, Government ofIndia has recommended that 20 mg of elemental zincshould be given to all children older than 6 months withdiarrhea, and should be started as soon as diarrhea startsand continued for a total period of 14 days [41,42].Children aged 2-6 months should be advised 10 mg perday of elemental zinc for 14 days.

Zinc dispersible tablets and liquid formulations arenow available widely in the private sector. Zinc dispersibletablets are also being supplied in the RCH kits. Efforts areon for making them available at most peripheral centers,and as over the counter (OTC) drug [41]. Currently, thereare no studies available on the operationalization of zinctreatment through government sector.

Knowledge gaps and directions for future research: Thecurrent diarrhea management guidelines mainly discussthe technical issues with very little emphasis onoperationalization. The key strategies for IEC, anddistribution system for reduced osmolarity ORS, zinc andciprofloxacin have not been given adequate emphasis.Although National Guidelines recommend ciprofloxacinfor dysentery and resistant cholera cases, there are issuesrelated to its approval by Drug Controller General of Indiafor use in children.

4. Preventive Interventions

A. Breastfeeding

Breastfeeding has the potential of preventing 13% ofunder-five deaths in developing countries [8,43]. WebTable IV presents a summary of studies reporting effectof breastfeeding on diarrhea morbidity or mortality. Apooled analysis of six studies from developing countries(Brazil, The Gambia, Ghana, Pakistan, the Philippines,and Senegal) documented a significant reduction indiarrhea related mortality with breastfeeding [51]. In thefirst 6 months of life, protection against diarrhea wassubstantially greater (OR 6·1 [95% CI 4·1–9·0]) than forinfants aged 6–11 months (OR 1·9 [95% CI 1·2–3·1]).Protection was highest when maternal education was low.

The benefits of breastfeeding are greater when it isexclusive. In a study from Bangladesh [50], partial or no

breastfeeding was associated with a 3.94-fold higher riskof deaths attributable to diarrhea in comparison toexclusive breastfeeding in the first few months of life.Subsequent studies from all settings have consistentlydocumented the protective effect of exclusivebreastfeeding on diarrhea morbidity (incidence andprevalence).

The evidence in favor of breastfeeding is ofintermediate quality as randomized controlled trialsdirectly evaluating breastfeeding versus no breastfeedingare not available due to ethical reasons. However,observational studies and breastfeeding promotion trialsconclusively support the protective role of exclusivebreastfeeding in prevention of diarrheal morbidity andmortality.

Conclusions and Comments: BOX 3

B. Handwashing

As pathogens causing diarrheal diseases are mostlytransmitted through the feco-oral route, handwashing isproposed as an important prevention strategy.Epidemiological evidence shows that the important riskbehaviors that encourage human contact with fecal matterinclude lack of handwashing after defecation, afterhandling feces, and before handling food. Handwashingaims to decontaminate the hands and prevent crosstransmission. Washing with soap and water removespathogens mechanically as well as by chemicalmicrobicidal action. Hand washing may requireinfrastructural, cultural, and behavioural changes, whichtake time to develop, as well as substantial resources (egtrained personnel, community organization, provision ofwater supply and soap).

Several systematic reviews have attempted to addressthe issue of prevention of diarrhea by promotinghandwashing interventions (Web Table V). The CochraneReview [52] included all randomized controlled trialswhich assessed interventions promoting handwashingafter defecation or after disposal of children’s feces andbefore preparing or handling foods. These activities

BOX 3

• Exclusive breastfeeding for six months reducesdiarrheal morbidity and mortality.

• The preventive effect of breastfeeding on diarrheaincidence is most marked among children fromlower socio-economic classes.

• The preventive effect is highest in the first sixmonths but continues beyond six months ifbreastfeeding is continued along withcomplementary foods.



included promotion strategies such as small groupdiscussions and larger meetings, multimediacommunication campaigns with posters, radio/TVcampaigns, leaflets, comic books, songs, slide shows, useof T-shirts and badges, pictorial stories, dramas, andgames. All institution-based trials were from high incomecountries (except one from China) with adequate provisionof water supply and soap. Community-based trialsincluded five cluster-randomized controlled trials thatused entire communities (generally villages orneighborhoods, except one trial, which used households)as units of randomization. These trials were conducted inlow-and middle-income countries in Asia (N=4) andAfrica (N=1). In three out of these five trials, soap wasprovided to the community by researchers. Three trialsevaluated hand washing as the only intervention, and othertwo trials used multiple hygiene interventions thatincluded hand washing with soap (the type of soap used isnot described). Participants were mainly mothers orcaregivers as well as children. The outcome of diarrheawas assessed only in children in all the trials included inthis review. From Institution-based trials, the incidence ofdiarrhea was assessed in 7711 children aged less thanseven years in 161 day-care centres and 87 schools in theeight trials. The two trials that adjusted for clustering andconfounders, showed a reduction in the incidence ofdiarrhea by 39% (IRR 0.61, 95% CI 0.40 to 0.92). The fivetrials with rate ratios that did not adjust for clustering alsoshowed a benefit from the intervention. In Communitybased trials, the intervention reduced the incidence ofdiarrhea by 32% (IRR 0.68, 95% CI 0.52 to 0.90; 4 trials)in trials that adjusted for clustering and confounders. Thesingle trial which did not adjust for clustering effects alsoshowed a similar benefit. The reduction in the risk ofdiarrhea was greater in the trials which provided soap tothe communities (IRR 0.49, 95% CI 0.39 to 0.62) than inthe trials that did not provide soap and promoted multiplehygiene interventions. Overall, the trials documented asignificant benefit of interventions promotinghandwashing in the institutions or communities [52]. It isimportant to note that the control group in most casesreceived quite frequent monitoring, which may itself haveinfluenced hand washing behavior. This might havelowered the estimate of the quantum of benefit fromhandwashing. Overall, this review provided strongevidence that handwashing interventions reduce diarrhealmorbidity by about one-third. However, most trials in thisreview had short-term follow up, and it is unclear if theirlevel of effectiveness would be maintained if they werescaled up to larger regions with less intensive monitoringover a longer time-period.

In another systematic review of published studies(cohort, case-control and RCTs) assessing the water

sanitation and hygiene interventions to reduce diarrhea inthe less developed countries, all studied interventions werefound to reduce significantly the risks of diarrheal illness[53]. Most of the interventions had a similar degree ofimpact on diarrheal illness, with the relative risk estimatesranging between 0·63 and 0·75 [53]. Hygieneinterventions, including promotion of handwashingreduced diarrheal illness (RR 0·63, 95% CI 0·52–0·77; 11studies). Re-analysis of the data after exclusion of poorquality studies lowered the risk further (RR 0·55, 95% CI0·40–0·75). Hygiene interventions were typically of twotypes, those concentrating on health and hygieneeducation, and those that actively promoted handwashing(usually alongside education messages). In general,education was aimed at the mothers, although the outcomewas measured in children. Separate meta-analysesexamining the effectiveness of each of these specificinterventions resulted in pooled estimates of RR of 0·56(0·33–0·93) and 0·72 (0·63–0·83) for the effects ofhandwashing and education, respectively. In water qualityinterventions, point-of-use water treatment was found tobe more effective than targeting the water source [53].Multiple interventions (consisting of combined water,sanitation, and hygiene measures) were not more effectivethan interventions with a single focus. This is likely to bethe result of piecemeal implementation of more ambitiousintervention programs, which may result in an overall lackof focus or lack of sufficient attention on importantcomponents such as handwashing and water treatment atsource.

Curtis and Cairncross [54] reported a systematicreview focusing on handwashing studies as opposed togeneral hygiene interventions and also combined resultsfrom developed and less developed countries. The pooledrelative risk of diarrheal disease associated with notwashing hands was 1·88 (95% CI 1·31–2·68), implyingthat handwashing could reduce diarrhea risk by 47%.When all studies, when only those of high quality, andwhen only those studies specifically mentioning soap werepooled, risk reduction ranged from 42–44% [54].

Almost all studies included in the systematic reviewsutilized soap (plain or anti-bacterial) for handwashing ordid not report the proper technique. The only studyreporting use of ash or soil in handwashing did notdocument any benefit of same in the reduction of diarrhea.

In the Lancet series of publications related to childsurvival [43], meta-analysis indicated that the relative riskof diarrhea with hand washing is 0.70 (95% CI 0.56-0.89).

The long term follow-up data from these studies aremostly not available. In an earlier study in Karachi,Pakistan, households that received free soap and



handwashing promotion for 9 months reported 53% lessdiarrhea than controls [55]. Eighteen months after theintervention ended, these households were enrolled in afollow-up study to assess sustainability of handwashingbehavior. Upon re-enrollment, mothers in householdsoriginally assigned to the intervention were more likely tohave a place with soap and water to wash hands, and morelikely to demonstrate the correct procedure [55]. However,in the ensuing 14 months, former intervention householdsreported a similar proportion of person-days with diarrhea(1.59% versus 1.88%, P = 0.66) as controls [55]. Althoughintervention households showed better handwashingtechnique after 2 years without intervention, their soappurchases and diarrhea experience was not significantlydifferent from controls.

Conclusions and Comments: Box 4

Knowledge gaps and directions for future research:Although the interventions promoting handwashing andother hygiene measures clearly show a reduction indiarrheal risk in the short term, the sustainability ofhandwashing-behavior in the communities requireinvestigation. Also, the evidence of reduction in risk ofdiarrheal illnesses after scaling up the intervention inuncontrolled situations is not available. India-specificdata on handwashing are scarce.

C. Vaccination

Rotavirus vaccine

In a recent systematic review of randomized controlledtrials [56], available rotavirus vaccines were efficaciousin reducing rotavirus-specific diarrhea by 72-73% at oneyear and 62-67% at two years following the vaccination.The efficacy in preventing severe rotaviral diarrhea wasgreater (80-93% at one year and 84-89% at two years)(Table I). There was no significant effect of one type ofvaccine (Rotarix) on all-cause diarrhea whereas the otherbrand (RotaTeq) resulted in reduction of all-cause

diarrhea by about half. There was no effect of eithervaccine on mortality. The only study from India includedin this review was an immunogenecity and safety study[67]. This study documented that the seroconversion rateone month after receiving two doses of vaccine was58.3% [95% CI: 48.7-67.4] as against 6.3% [95% CI:2.5-12.5] in the placebo group. Efficacy in terms ofreduction of diarrheal episodes or any other functionalparameter was not assessed in this study.

Well-controlled effectiveness studies of rotavirusvaccine are non-existent. Studies from South Americancountries and USA documented a reduction in rotaviraldiarrhea incidence, and associated hospitalizations, afterintroducing rotavirus vaccine [57-63]. Data from Mexico[61,63] documented a small reduction in diarrhea relatedmortality before and after introduction of rotavirusvaccine. However, these studies have limited validity dueto a large time gap between the two comparison groups.The cost-efficacy for reducing mortality is likely to beextremely low as about 11,900 vaccinations were requiredto prevent one diarrheal death in the Mexican study [61].Recent data from Brazil also suggest that the effectivenessof vaccination program may reduce over the years [57] asvaccinated children had more severe episodes 18 monthsafter vaccination although all-cause diarrhea episodeswere more severe in unvaccinated children in the first yearof age [57]. Data from USA also documented a 15%increase in number of rotavirus positive cases in thesecond year over the immediate year following thevaccination program [62].

Podewils, et al. [68] projected that a universalrotavirus vaccination program could avert about 0.1million deaths, 1.4 million hospitalizations, and 7.7million outpatient visits among an Asian birth cohort till itreaches 5 years of age. These estimates for rotaviral deathsare crude, and do not represent a true picture as thesemodels assume that rotavirus would be responsible forabout one-fourth of diarrheal deaths in India. There is nodirect estimate for this figure, and it merely reflects therotavirus detection rates in hospitalized children in bigcities. Most deaths due to diarrhea occur in communitieswhere medical care is limited whereas etiologic picturederived from in-hospital deaths due to diarrhea would bebiased toward agents that are less likely to respond tomedical care. Moreover, most deaths occur inmalnourished children where the role of rotavirus incausation of diarrhea is not clear, and there is likely to besignificant role of systemic infections and other co-morbidities [69].

Cholera vaccine

Most vaccines against cholera have not been found to be

BOX 4

• Interventions promoting handwashing can reducediarrheal episodes by about one-third.

• Washing hands with soap and water is the mosteffective hygiene intervention to prevent diarrhea.

• Water treatment at point-of-use also reduces therisk of diarrhea by about one-third.

• Programs addressing multiple components (e.g.hygiene, sanitation, water quality and source) arenot more effective than those focusing onindividual components such as handwashing andwater treatment at source.



TA

BL

E I

SY

STE

MA

TIC

RE

VIE

WS A

ND

ST

UD

IES*

OF

VA

CC

INE

S IN

PR

EV

EN

TIO

N O

F D

IAR

RH

EA

Stud

y Ye

arP

arti

cipa

nts

Stud

y ty

peR

esul

tsC

omm

ent

[Ref

.]

Rot

avir

usR

otar

ixS

oare

s-W

eise

r,17

5,94

4 ch

ildr

enSy

stem

atic

RR

of d

iarr

hea:

0.2

8 ( 9

5% C

I 0.1

7 to

0.4

8; 1

1,12

1 pa

rtic

ipan

ts,

All

the

34 tr

ials

(15

unpu

blis

hed)

et a

l. 20

11 [5

6](a

ge 1

mo-

6 m

o) fr

omR

evie

w6

tria

ls) d

urin

g fi

rst y

ear

and

RR

0.3

3 ( 9

5% C

I 0.2

1 to

0.5

0; 7

293

wer

e sp

onso

red

by c

ompa

nies

34 R

CTs

com

pari

ngof

RC

Tspa

rtic

ipan

ts, 5

tria

ls) d

urin

g se

cond

yea

r of f

ollo

w-u

p.m

anuf

actu

ring

or l

icen

cing

rota

viru

s va

ccin

es w

ith

RR

for s

ever

e ep

isod

es a

fter

one

yea

r: R

R 0

.20

(95%

CI 0

.11

to 0

.35;

resp

ecti

ve v

acci

nes.

Rep

orti

ng o

fpl

aceb

o. R

otar

ix (2

6 tr

ials

;35

,004

par

tici

pant

s, 7

tria

ls) a

nd a

fter

two

year

s: R

R 0

.16

tria

l met

hods

was

poo

r in

mos

t. T

he99

,841

par

tici

pant

s, (9

5% C

I 0.1

2 to

0.2

2; 3

2,10

6 pa

rtic

ipan

ts, 7

tria

ls).

allo

cati

on c

once

alm

ent,

blin

ding

and

Rot

aTeq

(8 tr

ials

, 76,

103

Rot

aTeq

anal

ysis

wer

e un

clea

r or i

nade

quat

epa

rtic

ipan

ts).

Mos

t tri

als

RR

for d

iarr

hea

afte

r one

yea

r : 0

.27

(95%

CI 0

.22

to 0

.33;

in m

ost s

tudi

es.

had

mul

tipl

e si

tes,

oft

en76

14 p

arti

cipa

nts,

4 tr

ials

) and

dur

ing

the

seco

nd y

ear:

RR

0.3

8 ( 9

5%in

sev

eral

cou

ntri

es. T

heC

I 0.2

6 to

0.5

5; 1

569

part

icip

ants

, 1 tr

ial)

only

tria

l fro

m In

dia

was

RR

for s

ever

e ep

isod

es: f

irst

yea

r 0.0

7 (9

5% C

I 0.0

1 to

0.5

0; 1

485

a sa

fety

stu

dy.

part

icip

ants

, 2 tr

ials

) and

for t

wo

year

s: R

R 0

.11

(95%

CI 0

.03

to 0

.47;

1569

par

tici

pant

s, 1

tria

l).

Rot

arix

was

not

bet

ter t

han

plac

ebo

in re

duci

ng th

e nu

mbe

r of c

ases

of

all-

caus

e di

arrh

ea b

ut R

otaT

eq re

duce

d th

e nu

mbe

r of c

ases

of a

ll-c

ause

diar

rhea

at o

ne y

ear

(RR

0.4

1, 9

5% C

I 0.2

8 to

0.6

0; 1

030

part

icip

ants

).N

o st

atis

tica

lly

sign

ific

ant d

iffe

renc

e in

the

num

ber o

f dea

ths

in c

hild

ren

usin

g R

otar

ix o

r Rot

aTeq

.

Vie

ira,

et a

l.Tw

o co

hort

of 2

50 c

hild

ren

Coh

ort

The

mea

n nu

mbe

rs o

f all

-cau

se d

iarr

hea

epis

odes

/chi

ld i

n th

e fi

rst y

ear

A v

ery

low

(4.

9%) p

reva

lenc

e of

rota

viru

s20

11 [

57]

(vac

cina

ted

and

unva

cci-

stud

yw

ere

sim

ilar

(0.8

7 an

d 0.

84) i

n va

ccin

ated

and

unv

acci

nate

d ch

ildr

en.

posi

tivi

ty w

as s

een

in b

oth

grou

psna

ted)

in N

orth

east

Bra

zil

Dur

ing

the

seco

nd y

ear,

the

num

ber o

f epi

sode

s/ch

ild

decr

ease

d to

0.5

2in

dica

ting

cha

ngin

g di

seas

e ep

idem

iolo

gyan

d 0.

42. A

ll-c

ause

dia

rrhe

a ep

isod

es w

ere

mor

e se

vere

in u

nvac

cina

ted

rath

er th

an e

ffec

t of a

ny v

acci

nati

onch

ildre

n in

the

firs

t yea

r of a

ge w

here

as th

ese

wer

e m

ore

seve

re in

prog

ram

vacc

inat

ed c

hild

ren

18 m

onth

s af

ter v

acci

nati

on.

Mol

to, e

t al.

Dia

rrhe

a-as

soci

ated

His

tori

cal

Aft

er th

e va

ccin

e in

trod

ucti

on, t

here

was

a d

ecre

ase

in d

iarr

hea-

Lik

ely

cont

ribu

tion

of o

vera

ll20

11 [

58]

hosp

ital

izat

ions

am

ong

cont

rol

asso

ciat

ed h

ospi

tali

zati

ons

of 2

2% in

firs

t yea

r and

37%

in s

econ

dde

velo

pmen

t and

bet

ter d

iarr

hea

unde

r-fi

ve c

hild

ren

in(B

efor

e-af

ter)

year

.m

anag

emen

t pra

ctic

es. I

ncre

ase

in20

07 a

nd 2

008

vs.

stud

y v

acci

nati

on c

over

age

coul

d be

the

reas

onm

ean

of 2

003-

05 fr

omfo

r bet

ter r

esul

ts in

sec

ond

year

.fi

ve h

ealt

h re

gion

s in

Pan

ama.

Qui

ntan

ar-

Dia

rrhe

a-re

late

d ho

spit

ali-

His

tori

cal

Dia

rrhe

a-re

late

d ho

spit

aliz

atio

ns d

ecre

ased

by

11%

in fi

rst y

ear a

nd b

yM

axim

um d

ecli

ne in

hos

pita

liza

tion

was

Sol

ares

, et a

l.za

tion

s du

ring

the

2008

cont

rol

40%

in s

econ

d ye

ar.

seen

in in

fant

s.20

11 [

59]

and

2009

rota

viru

s se

ason

s(B

efor

e-af

ter)

wit

h th

e m

edia

n of

200

3-st

udy

06 in

hos

pita

ls o

f Mex

ico

cont

d...



de P

alm

a et

al.

323

chil

dren

Mat

ched

Eff

ecti

vene

ss o

f tw

o do

ses

of v

acci

nati

on a

gain

st d

iarr

hea

requ

irin

g 2

010

[60]

(< 2

yr)

adm

itte

d w

ith

case

-con

trol

hosp

ital

adm

issi

on w

as 7

6%. A

dmis

sion

s fo

r dia

rrhe

a am

ong

chil

dren

rota

vira

l dia

rrhe

a an

dst

udy.

unde

r 5 d

ecli

ned

by 4

0% in

200

8 an

d by

51%

in 2

009

from

the

prev

acci

ne96

9 he

alth

y m

atch

edye

ar 2

006.

cont

rols

Ric

hard

son,

Dia

rrhe

a re

late

dH

isto

rica

lR

educ

tion

in d

iarr

hea-

rela

ted

mor

tali

ty fr

om a

n an

nual

med

ian

ofL

ikel

y co

ntri

buti

on o

f ove

rall

et a

l. 20

10m

orta

lity

in u

nder

-co

ntro

l18

.1/1

00,0

00 c

hild

ren

at b

asel

ine

to 1

1.8/

100,

000

chil

dren

in 2

008

(rat

e d

evel

opm

ent a

nd b

ette

r dia

rrhe

a[6

1]fi

ve c

hild

ren

duri

ng(B

efor

e-af

ter)

redu

ctio

n, 3

5% [9

5% C

I, 2

9 to

39;

P<

0.00

1].

man

agem

ent p

ract

ices

. Abs

olut

e ri

sk20

08 a

nd 2

009

stud

yre

duct

ion

of d

eath

is 6

.3/1

00,0

00ro

tavi

ral s

easo

nsch

ildr

en; N

NT

158

74 fo

r pre

vent

ing

one

vs. b

asel

ine

of 2

003-

06de

ath.

Wit

h 74

% c

over

age,

cor

rect

edle

vels

.N

NT

119

04

CD

C 2

009

Rot

avir

us p

osit

ivit

yH

isto

rica

l20

07-0

8 an

d 20

08-0

9 se

ason

s w

ere

shor

ter a

nd la

ter t

han

the

med

ian

Set

ting

of d

evel

oped

cou

ntry

.[6

2 in

200

8-09

vs.

200

0-06

cont

rol

duri

ng 2

000-

2006

. The

200

8-09

sea

son

had

15%

mor

e po

siti

ve ro

tavi

rus

(pre

-vac

cine

) in

US

A(B

efor

e-af

ter)

test

resu

lts

than

the

2007

-08

seas

on.

stud

y

Esp

arz,

et a

l.M

orta

lity

due

toH

isto

rica

lF

rom

200

0-07

, dea

ths

due

to a

cute

dia

rrhe

a in

und

er-f

ive

chil

dren

Lik

ely

cont

ribu

tion

of o

vera

ll20

09 [

63]

diar

rhea

in u

nder

-co

ntro

ldr

oppe

d 42

%. D

iarr

hea

mor

tali

ty d

ecre

ased

bet

wee

n 20

06-0

7; 1

5.8%

deve

lopm

ent a

nd b

ette

r dia

rrhe

afi

ve c

hild

ren

in(B

efor

e-af

ter)

in <

1 yr

and

22.

7% in

1-4

yea

r old

.m

anag

emen

t pra

ctic

es.

Mex

ico

stud

y

Cho

lera

Sin

clai

r,F

our t

rial

s fr

om P

eru,

Syst

emat

icP

rote

ctiv

e ef

fica

cy o

f ora

l cho

lera

vac

cine

s in

und

er-f

ive

chil

dren

was

Sig

nifi

cant

het

erog

enei

ty o

f vac

cine

et a

l. 20

11B

angl

ades

h an

d In

dia

Rev

iew

of

38%

(95%

CI 2

0% to

53%

; fou

r tri

als)

in c

ompa

riso

n to

old

er c

hild

ren

prep

arat

ions

. Vac

cine

eff

icac

y to

o lo

w[6

4]re

port

ed o

utco

mes

of

RC

Tsan

d ad

ults

(Eff

icac

y 66

%, 9

5% C

I 57%

to 7

3%).

Pro

tect

ive

effi

cacy

for r

outi

ne u

se. M

ay fi

nd u

se in

pla

ces

vacc

ine

effi

cacy

indi

min

ishe

d af

ter t

wo

year

s an

d w

as lo

st in

the

four

th y

ear o

f fol

low

-up.

wit

h hi

gh c

hole

ra e

ndem

icit

y.ch

ildr

enN

o cl

inic

ally

sig

nifi

cant

incr

ease

in a

dver

se e

vent

s.

Gra

ves,

et a

l.St

udie

s en

roll

ing

Syst

emat

icIn

und

er-f

ive

chil

dren

, vac

cine

eff

icac

y w

as 5

5% (R

R 0

.45,

95%

0.3

5M

ost o

f the

se v

acci

nes

are

not i

n cl

inic

al20

10 [

65]

chil

dren

and

adu

lts

Rev

iew

of

to 0

.59;

250

,941

par

tici

pant

s) in

firs

t yea

r and

17%

(RR

0.8

3, 9

5%us

e an

ymor

e m

akin

g re

sults

of t

his

revi

eww

ho h

ad re

ceiv

edR

CTs

CI 0

.52

to 1

.31;

42,

039

part

icip

ants

) in

the

seco

nd y

ear o

f fol

low

up.

redu

ndan

t.in

ject

able

cho

lera

vacc

ine

vs. p

lace

boor

no

inte

rven

tion

Mea

sles

Chi

ldre

n ag

ed 9

-24

Lon

gitu

dina

lA

ttac

k ra

tes

of d

iarr

hea

in im

mun

ized

chi

ldre

n (1

.6/c

hild

/yr)

was

no

Mea

sles

vac

cina

tion

had

no

impa

ct o

nR

edda

iah,

mon

ths

in R

ural

stud

ydi

ffer

ent t

o th

at in

the

non-

imm

uniz

ed (1

.5/c

hild

/yr)

. The

mea

ndi

arrh

eal m

orbi

dity

.et

al.

1993

Har

yana

dura

tion

of d

iarr

hea

in b

oth

grou

ps w

as 2

.3 d

ays.

The

pre

vale

nce

of[6

6]di

arrh

ea in

imm

uniz

ed a

nd n

on-i

mm

uniz

ed w

as 3

.85

and

3.67

resp

ectiv

ely.

*Stu

dies

incl

uded

in s

yste

mat

ic r

evie

ws

have

bee

n de

-dup

lica

ted

in p

repa

ring

this

tabl

e



very effective. Injectable vaccines are no longer inclinical use, and their protective efficacy lasted only forone to two years [65]. A recent systematic review of oralcholera vaccines reported a protective efficacy of 38%(95% CI 20% to 53%; four trials) in under-five children.This was lesser in comparison to older children andadults. Moreover, the efficacy gradually declined overtwo years, and was completely lost after three years. AnIndian trial included in this review reported 67% efficacyof a modified killed-whole-cell oral vaccine againstclinically significant cholera in an endemic setting [70].The vaccine has the potential to be used in communitieswith the highest risk of cholera epidemics.

Measles vaccine

Although measles vaccination is promoted as a diarrheaprevention strategy, there are not enough studies tosupport its role. A single published study on the role ofmeasles vaccine in prevention of diarrhea did notdocument any benefit [66]. Attack rate of diarrhea inimmunized children (1.6/child/yr) was no different to thatin the non-immunized (1.5/child/yr).

Conclusions: Box 5

Knowledge gaps and directions for future research: Asmost data regarding rotavirus related deaths and rotavirusvaccine in India are based on modeling methods andefficacy studies in other settings, respectively, thesecannot be used to start large scale vaccination programsin India. There is an urgent need to generate India specificdata on rotavirus vaccine effectiveness by evaluatingfunctionally important outcomes such as incidence/prevalence of diarrhea and related mortality. The effect ofan operational program on other vaccines’ coverage, andalso on other child survival interventions need to becarefully evaluated in any future studies. Communityacceptance, cost factors and logistics also need to bestudied carefully. Utility and acceptability of choleravaccine need to be studied in outbreak situations.

D. Vitamin A

Results from a systematic reviews (Table II) indicated

that vitamin A supplementation has no consistentprotective effect on the incidence of diarrhea or diarrhea-related mortality in neonates and infants less than 6months. However, there was some evidence of benefit inchildren aged 6-59 months in low and middle, incomecountries. A systematic review examining the role ofvitamin A given during measles episode documented asignificant reduction in duration of diarrhea [76].However, even this review did not document anyreduction in incidence of diarrhea. Overall, the evidencerelated to benefit of vitamin A in prevention of diarrhea isconflicting, and thus it is not recommended as a diarrheaprevention strategy, except in case of measles.

Conclusions

Vitamin A supplementation does not reduce incidence ofdiarrhea or diarrhea-related mortality in neonates andchildren < 6 months but there is a benefit in children aged6-59 months.

E. Zinc

A large body of evidence from India and other developingcountries shows important therapeutic benefits with zincadministration during an episode of diarrhea [77]. Inaddition, many studies have examined the role of zincsupplementation as a preventive strategy to reducediarrhea morbidity in the subsequent months (Table III).

A meta-analysis of 17 randomized controlled trials ofzinc supplementation (for >3 months) for under-fivechildren reported fewer episodes of diarrhea (rate ratio:0.86), and significantly fewer attacks of severe diarrhea ordysentery (rate ratio: 0.85), and persistent diarrhea (rateratio: 0.75) with zinc supplementation in comparison toplacebo [81]. Zinc-supplemented children also hadsignificantly fewer total days with diarrhea (rate ratio:0.86). The review concluded that zinc supplementationreduced significantly the frequency and severity ofdiarrhea and the duration of diarrheal morbidity. However,the relatively limited reduction in morbidity, and thepresence of significant heterogeneity and publication biasindicate the need for larger, high-quality studies to identifysub-populations most likely to benefit. Other recentsystematic reviews [78,79] concluded that there is asignificant reduction in the incidence and prevalence ofdiarrhea with zinc supplementation but effect on mortalityreduction is minimal. Zinc supplementation given duringneonatal period also does not seem to reduce diarrhealmorbidity in next one year [80].

Conclusions

Zinc supplementation for at least 2 weeks leads to 15-25% fewer episodes of diarrhea in under-five children in

BOX 5

• Rotavirus vaccine is efficacious in preventionagainst severe rotaviral diarrhea (80-93%) andany rotaviral diarrhea (62-73%).

• There is no well controlled study evaluating theeffectiveness of rotavirus vaccine as a publichealth strategy implemented in India or similarsettings.

• Existing cholera vaccines are not effective enoughfor implementation on a large scale level.



TA

BL

E I

I S

YST

EM

AT

IC R

EV

IEW

S O

F V

ITA

MIN

A IN

PR

EV

EN

TIO

N O

F D

IAR

RH

EA

Stud

y Ye

arP

arti

cipa

nts

Stud

y ty

peR

esul

ts

Gog

ia, e

t al.

Infa

nts

(< 6

mo)

from

6 s

tudi

esSy

stem

atic

Rev

iew

No

redu

ctio

n in

risk

of d

iarr

hea

(poo

led

risk

rati

o 1.

02; 9

5% C

I 0.9

9 to

1.0

6, P

= 0

.19;

248

0220

11 [

71]

cond

ucte

d in

low

and

mid

dle

of ra

ndom

ized

and

part

icip

ants

, 6 s

tudi

es) o

r dia

rrhe

a re

late

d m

orta

lity

(RR

1.0

1; 9

5% C

I 0.7

2 to

1.4

1; 4

7998

inco

me

coun

trie

squ

asi-

rand

omiz

ed tr

ials

part

icip

ants

, 7 s

tudi

es) d

urin

g fi

rst y

ear o

f lif

e.

May

o-W

ilso

nC

hild

ren

aged

6 m

o- 5

yrs

who

Syst

emat

ic R

evie

w o

fSe

ven

tria

ls re

port

ed a

28%

redu

ctio

n in

mor

tali

ty a

ssoc

iate

d w

ith

diar

rhea

(0.7

2, 0

.57

to 0

.91)

.et

al.

2011

[72]

rece

ived

syn

thet

ic o

ral v

itm

inR

CTs

Vit

amin

A s

uppl

emen

tati

on w

as a

lso

asso

ciat

ed w

ith

a re

duce

d in

cide

nce

of d

iarr

hoea

(0.8

5, 0

.82

A s

uppl

emen

ts o

r pla

cebo

/no

to 0

.87)

inte

rven

tion

.

Imda

d, e

t al.

Chi

ldre

n <

5 y

ears

of a

geSy

stem

atic

Rev

iew

of

Vita

min

A s

uppl

emen

tati

on re

duce

d di

arrh

ea s

peci

fic

mor

tali

ty b

y 30

% [R

R 0

.70;

95

% C

I 0.5

8-20

11 [

73]

RC

Ts0.

86] i

n ch

ildr

en 6

-59

mon

ths.

No

sign

ific

ant b

enef

it w

as s

een

in c

hild

ren

<6

mo

of a

ge.

Gog

ia, e

t al.

neon

ates

(<1

mo)

rece

ivin

gSy

stem

atic

Rev

iew

of

No

redu

ctio

n in

risk

of d

iarr

hea

wit

h ne

onat

al v

itam

in A

sup

plem

enta

tion

.20

09 [

74]

prop

hyla

ctic

vita

min

AR

CTs

(inc

ludi

ng q

uasi

or

supp

lem

enta

tion

from

6 tr

ials

clus

ter)

from

dev

elop

ing

coun

trie

s

Gro

tto,

et a

l.9

RC

Ts d

eali

ng w

ith

mor

bidi

tySy

stem

atic

revi

ew o

fN

o co

nsis

tent

ove

rall

pro

tect

ive

effe

ct o

n th

e in

cide

nce

of d

iarr

hea

(RR

, 1.0

0; 9

5% C

I, 0

.94-

1.07

)20

03 [

75]

from

dia

rrhe

a in

6 m

o- 7

yr o

lds

RC

Ts

D'S

ouza

, et a

l.V

itam

in A

for p

reve

ntin

gS

yste

mat

ic R

evie

w o

fN

o si

gnif

ican

t red

ucti

on in

the

inci

denc

e of

dia

rrhe

a. T

here

was

a s

igni

fica

nt d

ecre

ase

in th

e20

02 [

76]

mor

bidi

ty in

mea

sles

6m

o-R

CTs

dura

tion

of d

iarr

hea.

13yr

s (4

92 v

it A

, 536

pla

cebo

)

the subsequent 2-3 months. There is no benefit ofproviding zinc in the neonatal period.

5. Health Care Practices in Management of Diarrhea(WebTable VI)

A. Care-Seeking

In NFHS-3 [3], treatment for diarrhea was sought from ahealth provider for 60% under-five children. Theproportion of mothers who sought care was similar toNFHS-2 (63%) and NFHS-1 (61%) data [11,12]. Urbanchildren, boys, children of educated mothers and childrenin households belonging to the higher wealth quintileswere more likely than other children to be taken to ahealth facility or provider for advice or treatment. In theUNICEF ten district survey [13] and CES [2], almostthree-fourths of caregivers reported seeking care outsidethe home for diarrhea in children. In the Ministry ofHealth and Family Welfare-UNICEF rural India survey[83] conducted almost 20 years back, the proportion whosought care from a health provider was 65%. From thesenationwide surveys, it is apparent that two-thirds to three-fourths of mothers seek care outside home when theirchild (below the age of five years) suffers from diarrhea.It is striking to note that care-seeking behavior does notseem to have changed over last twenty years, with almost30-40% children never seeking care outside home

Few surveys have provided information about theprofile of health workers visited by mothers. In theUNICEF ten-district survey (2009) [13], a large majority(79% total, 82% urban and 77% rural) of mothers soughttreatment from private medical sector. Private serviceproviders (doctor) and private hospitals were the mostcommon sources of consultations. Only 22% of motherssought care from public health sector. Among the publicsector health facilities; government hospitals, PHCs,CHCs, and rural hospital were more popular. Anganwadiand sub-centres were least used for diarrhea treatment. Inthe UNICEF coverage evaluation survey [2], 48% motherssought care from private sector and 21% from governmentsector during an episode of diarrhea in their child.

Similarly, in the MoHFW-UNICEF rural survey(1990) [83], 83% of mothers sought treatment fromprivate practitioners, mostly practicing allopathy (notnecessarily licensed). Only 12% sought care fromgovernment health center or health worker. The profile ofrural doctors was also assessed in this survey. It was foundthat 62% had no medical qualification and only 3% had aMBBS degree.

In view of the foregoing, it is apparent that anyintervention to improve the diarrhea managementpractices must target the private doctors, including



TA

BL

E I

II S

YST

EM

AT

IC R

EV

IEW

S O

F Z

INC IN

PR

EV

EN

TIO

N O

F C

HIL

DH

OO

D D

IAR

RH

EA

Aut

hor

Year

[R

ef.]

Par

tici

pant

sSt

udy

type

Res

ults

Com

men

t

Pat

el, e

t al.

RC

Ts re

port

ing

use

of z

inc

Met

a-an

alys

isZ

inc

supp

lem

enta

tion

resu

lted

in 1

9% re

duct

ion

in p

reva

lenc

eH

eter

ogen

eity

of s

tudi

es w

as20

11 [

78]

supp

lem

enta

tion

for p

reve

ntio

nof

RC

Tsan

d 9%

redu

ctio

n in

inci

denc

e of

dia

rrhe

a. It

als

o re

duce

d re

curr

ent

an is

sue.

Qua

lity

ass

essm

ent

of c

hild

hood

dia

rrhe

a d

iarr

hea

by 2

8%.

Eff

ects

on

prev

enti

on o

f per

sist

ent d

iarr

hea,

and

met

hodo

logy

of r

epor

ting

dys

ente

ry o

r mor

tali

ty w

ere

not s

igni

fica

nt.

of t

his

revi

ew is

not

rigo

rous

.

Yak

oob,

et a

l.C

Ts o

n zi

nc s

uppl

emen

tati

onM

eta-

anal

ysis

In th

e ra

ndom

eff

ect m

odel

, the

re w

as n

o si

gnif

ican

t eff

ect o

f20

11 [

79]

Rfo

r chi

ldre

n (3

mo-

5yr)

inan

d M

odel

ling

prev

enti

ve z

inc

on d

iarr

hea-

spec

ific

mor

tali

ty (R

R =

0.8

2; 9

5%de

velo

ping

cou

ntri

es a

nd it

sC

I: 0

.64,

1.0

5) o

r all

cau

se m

orta

lity

(RR

= 0

.91;

95%

CI:

0.8

2,ef

fect

on

mor

talit

y1.

01).

Aft

er a

pplic

atio

n of

CH

ER

G ru

les,

a 1

3% re

duct

ion

indi

arrh

ea m

orta

lity

was

see

n.

Gul

ani,

et a

l.R

CTs

and

qua

si-R

CTs

eva

luat

ing

Syst

emat

icN

o si

gnif

ican

t eff

ect o

n th

e nu

mbe

r of e

piso

des

of d

iarr

hea

(RR

Onl

y tw

o tr

ials

repo

rted

dia

rrhe

a20

11 [

80]

effe

ct o

f neo

nata

l zin

cR

evie

w 0

.87,

95%

CI 0

.65-

1.16

; 2 tr

ials

) or n

umbe

r of c

hild

ren

havi

ngre

late

d ou

tcom

e. D

ata

inad

equa

tesu

pple

men

tati

ondi

arrh

ea (R

R 0

.97,

95%

CI 0

.90-

1.04

, 2 tr

ials

) dur

ing

firs

t yea

r.fo

r any

def

init

e co

nclu

sion

.

Agg

arw

al, e

t al.

RC

Ts o

f zin

c su

pple

men

tati

on fo

rSy

stem

atic

Zin

c su

pple

men

tati

on le

d to

few

er e

piso

des

of d

iarr

hea

Lim

ited

redu

ctio

n20

07 [

81]

> 3

mon

ths

cond

ucte

d in

und

er-f

ive

Rev

iew

of

(RR

: 0.8

6) a

nd s

igni

fica

ntly

few

er a

ttack

s of

sev

ere

diar

rhea

or

Het

erog

enei

tych

ildr

enR

CTs

dyse

nter

y (R

R: 0

.85)

, per

sist

ent d

iarr

hea

(RR

: 0.7

5, a

nd lo

wer

Pub

lica

tion

bia

sdu

rati

on o

f dia

rrhe

a (R

R: 0

.86)

Bhu

tta,

et a

l.R

CTs

dea

ling

wit

h zi

nc s

uppl

e-Sy

stem

atic

Poo

led

OR

s fo

r dia

rrhe

al in

cide

nce

and

prev

alen

ce w

ere

0.82

15-2

5% re

duct

ion

in in

cide

nce

1999

[82

]m

enta

tion

in <

5 y

rs. A

t lea

st 5

0%re

view

of

(95%

CI 0

.72

to 0

.93)

and

0.7

5 (9

5% C

I 0.6

3 to

0.8

8)of

dia

rrhe

a in

the

shor

t-te

rmof

RD

A fo

r at l

east

2 w

eeks

RC

Tsre

spec

tive

ly.

Res

ults

sim

ilar

in s

hort

-cou

rse

and

cont

inuo

us tr

ials

wit

h zi

nc s

uppl

emen

tati

on

unlicenced and unqualified practitioners.

B. Oral Rehydration Therapy

National Diarrheal Disease Control Program has beenoperational in the country since 1980s, and oralrehydration therapy (ORT) is the key element.Recently, the Government of India introduced the lowosmolarity Oral Rehydration Solution (ORS),recommended by WHO for the management ofdiarrhea. Zinc has also been approved as an adjunct toORS for the management of diarrhea. Emphasis is alsoplaced on continued feeding, including breastfeedingduring diarrhea. One major goal of this program is toincrease awareness among mothers and communitiesabout the causes and treatment of diarrhea. ORSpackets are made widely available and mothers aretaught how to use them.

NFHS-3 asked mothers of under-five children whosuffered from diarrhea within two weeks preceding thesurvey, a series of questions about feeding practicesduring diarrhea, the treatment of diarrhea, and theirknowledge and use of ORS. Sixty percent of themothers reported consulting a health care providerduring the episode of diarrhea [3]. However, only 26%of children used ORS. Another disturbing fact was thatcaregivers of only one in ten children gave increasedfluids during diarrhea. Twenty-seven percent ofchildren were given less to drink, 10 percent were givenmuch less to drink, and 4 percent were not givenanything to drink, resulting in 4 in 10 children withdiarrhea having their fluids decreased while sufferingfrom diarrhea. More than half (57 percent) of childrenreceived neither oral rehydration therapy nor increasedfluids when sick with diarrhea. Use of ORS and ORTwas even less likely among children living in rural areas,children of mothers with little or no education, andchildren belonging to households in the lower wealthquintiles. These figures indicate poor implementation ofproper diarrhea treatment practices not only amongmothers but also among health-care providers.

All three National family health surveys [3,11,12]also assessed the knowledge of mothers of under-fivechildren regarding ORS. NFHS-3 reported that 73% ofwomen knew about ORS packets. Knowledge of ORSpackets was considerably higher among urban mothers(86%) than rural mothers (70%). The proportion ofwomen who knew of ORS packets increased witheducation and increasing wealth index. Knowledge ofORS packets was lowest among mothers belonging tothe lowest wealth quintile (59%). Knowledge of ORSpackets was lowest among mothers who were notregularly exposed to any mass media. There was a clear



dichotomy between knowledge and practice. Despitethree-fourth of women knowing about ORS, only one-fourth used it when their child suffered from diarrhea. Thisdifference can not be explained only on the basis of use ofother home available fluids as only 43% mother usedeither ORS or increased fluids (including home availablefluids) during diarrhea.

The trends related to management practices related todiarrhea in the communities are not very encouraging.Table IV shows that ORS use rates have not improvedsignificantly in the seven years from NFHS-2 (1998-99) toNFHS-3 (2005-06) despite improvement in knowledge.The figures of ORS use rates in NFHS-1 are not strictlycomparable as these were related to ‘ever use of ORS’ incomparison to ‘ORS use during current episode ofdiarrhea’ in NFHS-2 and NFHS-3. UNICEF surveysreport a better ORS use rates (38%-43%) in comparison toNFHS surveys.

Health providers’ practices for management ofdiarrhea was assessed in UNICEF ten-district survey [13].85-100% of doctors practicing modern medicine claimedthat they prescribed ORS. However, as earlier stated,mothers’ reports suggested low level of ORS prescription.In the MoHWF-UNICEF rural India survey [83], mostpractitioners were aware of ORS; only few actuallyprescribed it. Only 26% stocked ORS with them, out ofwhich almost half had open packets, and were probably re-dispensing it as medicine in smaller pouches. In othersmall-scale prescription surveys, ORS use rates was alsounsatisfactory.

The results underscore the need for informationalprograms for mothers and supplemental training forhealth-care providers that emphasizes the importance ofORS, ORT, and increased fluid intake during episodes ofdiarrhea.

C. Feeding during diarrhea

Continued feeding during a diarrheal episode helps in

faster recovery and reduces the chances of gettingmalnourished. In the NFHS-3 survey, only 37% ofchildren were given the same as usual to eat whenrecently suffering from diarrhea [3]. Two percentchildren were given more to eat, 31% were given‘somewhat less than the usual’ amount of food, 11% weregiven much less than the usual amount of food, and 4%were not given any food. Rural mothers were more likelyto reduce feeding during diarrhea. Behavior contrary torecommendations for proper management of diarrheasuggests the need for public education program on properfeeding practices during diarrhea.

The practice of giving semi-solids to children duringdiarrhea showed a marginal improvement from 15% inNFHS-2 to 20% in NFHS-3. However, this could berelated to more number of older children in NFHS-3 ratherthan an improvement in diarrhea management practice asthe NFHS-3 catered to mothers having children less thanfive years in comparison to NFHS-2, which surveyedmothers having children less than three years of age.

D. Zinc

As zinc has been introduced in the National Program onlyrecently, the data evaluating use of zinc in diarrhea arescarce. NFHS-3 attempted to obtain the data on zinctreatment given during diarrhea [3]. Only 0.3% of themothers reported the use of zinc during the precedingepisode of diarrhea. However, 30% of mothers reporteduse of unknown drugs which might have included zinc. Inthe UNICEF ten-district survey [13], around 1% ofmothers had knowledge of zinc and a similar proportionutilized zinc in the management of their child sufferingfrom diarrhea. The awareness regarding zinc is presentlynegligible in the community. However, government andprivate practitioners of modern medicine have begun toprescribe zinc. In UNICEF ten-district survey, 30% ofgovernment and 36% of private practitioners of modernmedicine claimed to prescribe zinc to a child sufferingfrom diarrhea. In a prescription audit from a tertiary care

TABLE IV TEMPORAL TRENDS IN ORAL REHYDRATION THERAPY

Indicator Survey (Year)

MoHFW, IMRB, NFHS-1 NFHS-2 NFHS-3 UNICEF 10- UNICEF coverageUNICEF Rural (1992-93) (1998-99) (2005-06) district survey evaluation surveyIndia survey (1990) (2009) (2009-10)

Knowledge of ORS 37% 43% 62% 74% 70% NA

ORS use rate 6% 26%* 27% 26% 38% 43%

Increasing fluids NA 10% 22% 10% 10% 10%

*Ever use of ORS in NFHS-1; use during most recent episode in other surveys; ORS: Oral rehydration solution; NFHS: National Family HealthSurveys; MoHFW: Ministry of Health and Family Welfare; Data from references [83,3,11,12].



center in Chennai [87], 65% of children hospitalized withdiarrhea received zinc therapy.

E. Other medications/ antibiotics/ antidiarrheals

In NFHS-3 survey, significant proportion of childrenwere treated with drugs (mostly unnecessary), including30 percent who were treated with ‘unknown’ drugs and 16percent who were treated with antibiotics [3]. Irrationaluse of antibiotics was particularly common for childrenof more educated mothers and for children in householdsbelonging to the higher wealth quintiles. Use ofantimotility drug was fortunately rare with only 1.5%reporting their use. However, the validity of this data islimited because of ignorance of mothers in a significantproportion about the type of drugs given to their children.NFHS-3 did not provide information about receipt ofinjections during diarrheal episode. However, in NFHS-2and NFHS-1, 14-15% of mothers reported that injectionswere given to their child to treat diarrhea [11,12]. In therecent UNICEF ten-district survey [13], the proportionreceiving injections during a diarrheal episode was 23%.In the MoHFW-UNICEF rural India survey [83], 80% ofprivate health providers believed antibiotics to be themost useful drugs in management of diarrhea, and 40% ofchildren received injections during an episode ofdiarrhea.

Prescription audits of doctors have consistently shownvery high usage rates of antibiotics in diarrhea (WebTableVI). More than two-thirds of prescriptions given tochildren suffering from diarrhea had antibiotics. Fixeddrug combination (antibacterials + antiprotozoals) are alsovery commonly used in treatment of diarrhea.

Overall, the practices of healthcare providers inmanagement of diarrhea are far from satisfactory. Use ofORS is low, and antibiotics and other drugs are usedirrationally and indiscriminately.

Conclusions: BOX 6

F. Barriers to appropriate health practices

Information on why ORS is not prescribed or why it is notused commonly by mothers is very scarce. There ishardly any published information on this aspect. It isapparent from the prescription audits that the prescriptionrate of ORS is poor amongst health providers whose careis sought most commonly by the communities. As almostthree-fourth of the mothers seek care from healthproviders for their child during diarrhea, and in majorityORS was not prescribed, the ORS usage rate is bound tobe low.

As most users and providers have the knowledge ofORS, but do not make efforts to use it, it reflects low

salience of the product in the mind of target group. TheMoHFW-UNICEF survey addressed some of these issuesin the 1990 survey [83]. The mothers mostly did not useORS as it was not prescribed by the doctor. Those whoreduced the fluid intake during episode of diarrhea did sobecause either the child rejected fluids (30%), or theybelieved that the cooling effect of fluids can exacerbatediarrhea (21%) and water content in the fluid canaggravate diarrhea (17%). Fortunately, there were nocultural barriers to the adoption of ORS and ORT, and thelevel of satisfaction was high amongst the mothers whoused it. Rural doctors perceived the medication(capsules/tablets) to be the most important element indiarrhea management, and ORS only as an adjunctivetreatment. Sixty-five percent did not rank ORS among the3 most important elements of treatment in the early stagesof diarrhea. In the qualitative survey, the reasons for notprescribing ORS were mentioned as (i) it does not stopdiarrhea, (ii) it is indicated only in dehydration, and (iii)presence of perceived contraindications such asvomiting, cough/cold and fever. The reasons for highantibiotic usage were (i) strong belief on its benefit indiarrhea, (ii) expectation of patient of receiving amedicine from health practitioner, and (iii) regularinteraction with chemists about new available drugs.Reasons for prescribing antibiotics in UNICEF ten-district survey [13] were presence of dehydration, veryfrequent loose motions and vomiting. In some other smallscale studies, presence of fever prompted the doctor to

BOX 6

• Three-fourth seek care during diarrhea; a largemajority from private providers.

• Knowledge of ORS/ORT amongst mothers ofunder-five children is good (~70-75%) but thereis a big gap between knowledge and practice asreflected in poor ORS usage rates (25-40%).

• The knowledge regarding ORS has shown apositive trend but the use rates have beenconsistently poor.

• Provider’s knowledge of ORS is universal but thisagain does not translate commonly into practice.

• Knowledge and use of zinc is negligible, thoughappears to be improving.

• Practice of advising increased fluids andcontinued feeding during diarrhea needsimprovement.

• Irrational use of antibiotics, other drugs andinjections to treat diarrhea is common.

• Use of antisecretory agents is rare.



prescribe antibiotics in cases of diarrhea.

Regarding stocking ORS, the majority (79-90%) ofservice providers working in the government sector(practitioners of modern medicine, ANMs/AWWs/ASHAs) and private chemists stocked supplies of ORS[13]. In contrast, most private practitioners of modernmedicine, unregistered medical practitioners and private/government AYUSH practitioners kept no stock of ORSpackets. Stock-outs of ORS were reported in 24-42% ofgovernment-sector community health functionaries(ANMs, AWWs and ASHAs). In the MoHFW-UNICEFrural India survey [83], stocking of ORS was reported byonly 26% of rural doctors; almost half of them wereprobably opening the packets and re-dispensing asmedicine in smaller paper-pouches (pudiyas).

Lack of adequate training in ORT has also been citedas one of the reasons for poor ORS usage in the UNICEFten-district survey [13]. Although 72% of ANMs/AWWs/ASHAs reported that they had received some training onmanagement of childhood diarrhea, most (88%) reportedhaving received training for less than eight hours.

An operational study for implementation of ORT wasconducted in West Bengal through the existing healthservices facilities [92,93]. All the grassroot level healthworkers, including their supervisors at various levels weretrained regarding the management of patients of diarrheaby ORT. Training was done for one working day (inbatches of 30-40 health workers) using lectures withslides. After five months, the training was repeated for oneday, in batches, using modern methods which includedmodule-based approach, discussion, problem-solvingexercises, and demonstrations. Another block in the samedistrict with similar demographic features was notprovided with this intervention, and served as control.After 22 months of observation, it was evident that despiteadequate training, the performance of workers was notencouraging as there was low utilization of both homeavailable fluids and oral rehydration solution in the studyarea [93]. Diarrhea associated mortality was also similar(2.8/1000 under-five children) between the study andcontrol group. Though evaluation of training of workersusing modern modular methods revealed a strikingimprovement in their knowledge regarding signs ofdehydration, preparation and use of ORS etc., this did nottranslate into practice. The knowledge and level of skillsalso went down to a considerable extent after 12 months.In spite of providing necessary forms to maintain records,no CHGs or AWWs actually did so. There was nosupervision at all at the PHC level. ORS supply was alsogrossly inadequate as it was not replenished. In spite of theelaborate training imparted to the community health

guides and anganwadi workers, the overall usage rate inthe study area was only 11% in comparison to 12% incontrol areas. Home available fluid (HAF) usage rateswere also low in both the areas (27% in study vs 20% incontrol). Further, it was revealed that two-thirds of themothers did not even know their village level worker in thestudy area. In spite of repeated training of health workersto educate mothers to use ORT earliest in an episode ofdiarrhea and in adequate amonts, the grassroot level healthworkers failed to do so even amongst the smallporoportion of mothers (12.4%) educated by them. Thereasons for failure cited in this study were: (i) logisticfailure, (ii) lack of motivation of health workers, and (iii)lack of proper supervision at all levels, and (iv) absence ofcontinued supply of ORS [93]. Box 7 presents a compiledlist of barriers to ORS/ORT use, and for preventingantibiotic use [83, 92-98].

A recent review has tried to address the problem ofpoor performance by community health workers [99].Most schemes for these workers assume that there is asufficient pool of volunteers to participate in the socialservice in rural areas and urban settlements. Mostprograms pay their community workers a salary orhonorarium and there is no system of sustained communityfinancing. Other financial incentives range from a smallsalary from the state to payments for attendance at trainingsessions. However, the costs entailed by lost economicopportunities may be too high even if they are working onpart-time basis. A high attrition rate contributes to poorstability of the program and increases training costsbecause of the need for continuous replacement. Paymentsand commissions related to drug dispensing and sales mayencourage inappropriate treatment at the expense ofprevention and overuse of medications. Non-financialapproaches to improving performance may have lesspotential to distort care than fee-for-service payments orthose associated with drug sales. The review concludedthat policymakers should consider using a mix of financialand non-financial incentives tailored to localcircumstances to improve health workers’ performance,training programs should be tailored to the literacy level ofthe community health workers, and well-organisedsupervisory systems should be developed to improvemotivation and provide professional development [99].

G. Social Marketing of ORS

NFHS surveys reported that knowledge of ORS wasconsistently better in mothers who are exposed to any kindof mass media [3,11,12]. Attempts at social marketing ofORT through mass media have been done in past.Television has been effectively used impart knowledge inthe communities regarding use of ORS and ORT.



A small-scale, questionnaire-based study from theslums of Delhi, evaluated the effectiveness of the Ministryof Health’s mass media campaign to promote ORT useduring diarrheal episodes [100]. The knowledge of 59mothers who watched the television (TV) advertisementwith celebrities delivering simple and clear messages werecompared to 90 mothers who had received ORT messagesfrom other sources such as health workers. Mothers in thefirst group were considerably more likely to know how tocorrectly prepare ORS than those who learned about ORTfrom other sources (62.7% vs 37.7%). However, no

significant difference in use of ORT at home between the 2groups existed (69.49% and 53.33%, respectively). TVadvertisements were more likely to teach educatedmothers how to correctly prepare and to use OR at home,than health staff (81.5% vs 35.5% and 81.5% vs 41.9%,respectively). These results showed that social marketingof ORS packets via TV was successful in increasing ORTacceptability, knowledge and use, and especially amongeducated mothers. Similar studies in Bangladesh [101]showed that education incites changes in attitude andbehavior of mothers, which makes them more receptive ofnew knowledge and modern medicine. Another possibilityfor the education difference may be that TV was able tointerest educated mothers better than health staff. Thesefindings indicate a need to strengthen education programsin this area using effective media such as television forboth mothers and health care providers [102-104].

Knowledge gaps and directions for future research

There is an urgent need to find out the reasons for lowORS use by the users as well as health providers throughlarge scale representative qualitative studies in thecommunity. The system-related barriers need to beurgently addressed by strengthening and overhauling thehealth system. Acceptability of different preparations ofORS such as flavoured and ready-made ones need to becompared to the standard packaging in the community.National Rural Health Mission is a welcome step in thisdirection, and its impact on care-seeking behavior andpractices needs to be continuously monitored. The role ofmedia in improving the practices needs to be evaluated inoperational research programs. With the revolution intelecommunications, the role of mobile phones incommunity education need to be explored. Urgentoperational research is needed to find out the ways toimprove the community level workers’ performance.Their capability in handling various programs (integratedvs vertical) also need to be explored.

6. Operational Research on Zinc Use in Diarrhea

The efficacy of zinc treatment during diarrhea has beenproved by many randomized, placebo-controlled trials,and systematic reviews of randomized controlled trials.However, the effectiveness under real-life conditions isinfluenced by the community’s knowledge, attitudes andperceptions, and the quality of training component andfinancing system. Data regarding scaling up of zinc usefrom effectiveness studies in communities need to beexamined to evaluate its impact on diarrhea management(Table V).

Bhandari, et al. [104] conducted a cluster-randomizedtrial in Haryana utilizing government providers (doctors,

BOX 7

Barriers in ORS Use

A. System side barriers

Lack of motivated staff at most peripheral level.

Lack of supervision

Lack of confidence of people in public healthsystem

Lack of awareness of people about public healthsystem

Health care providers are not convinced ofbenefits of ORS, and do not prescribe it despiteknowing about it.

- Perception that it does not stop diarrhea

- Perception that it is only indicated indehydration

- Preference of intravenous fluids in case ofdehydration

ORS-stock outs common with govt. healthfunctionaries

Lack of IEC material laying emphasis on ORT

Inadequate training and inadequate retention ofthe knowledge and skills about ORT

B. Demand-side barriers

Parents, caregivers and providers are notconvinced of benefits of ORS.

(have knowledge of ORS but don’t use it)

Perception that it is not a medicine (does not stopdiarrhea)

Social beliefs (cooling effect of fluids, chances ofexacerbation of illness)

C. Barriers to Preventing Antibiotic Use

Fever, dehydration and young age promptscaregiver to prescribe antibiotics

Health workers not convinced of absence of anybenefit.

Demand for antibiotics by patients/parents is notcommon but expectation of a product/ medicineby parents is common.



TA

BL

E V

OPE

RA

TIO

NA

L R

ESE

AR

CH

ON

USE

OF

ZIN

C D

UR

ING

DIA

RR

HE

A

Aut

hor

Year

Pla

ceSa

mpl

e si

ze a

ndP

rofi

le o

fTy

pe o

f tra

inin

gR

esul

tsC

omm

ents

[Ref

.]ch

arac

teri

stic

she

alth

wor

ker

Bha

ndar

i,P

rim

ary

Car

eC

lust

er ra

ndom

ized

Gov

ernm

ent a

ndSt

anda

rd tr

aini

ngA

fter

3 m

onth

s of

Ext

ensi

ve IE

C c

ampa

igns

et a

l. 20

08se

ttin

g, H

arya

na,

tria

l Six

clu

ster

s of

priv

ate

prov

ider

sin

dia

rrhe

a m

anag

e-in

terv

enti

on:

(pos

ters

, ann

ounc

emen

ts) u

sed

[104

] In

dia

30,0

00 p

eopl

e, e

ach

and

vill

age

heal

thm

ent a

nd re

ferr

al,

zinc

use

dur

ing

diar

rhea

l epi

sode

:in

inte

rven

tion

are

as.

rand

omly

ass

igne

d to

wor

kers

incl

udin

g ap

prop

riat

e36

.5%

OR

S u

se:

34.

8% v

s. 7

.8%

Con

tinu

ous

free

sup

plie

s of

inte

rven

tion

and

use

of z

inc.

Dur

atio

n:C

ompl

ianc

e w

ith

14 d

ay z

inc:

70%

zinc

and

OR

S a

scer

tain

ed.

cont

rol s

ites

.ha

lf a

day

for p

hysi

cian

sD

iarr

hea

in la

st tw

o w

eeks

: 13%

Reg

ular

sup

ervi

sion

.In

terv

enti

on: Z

inc

and

a fu

ll d

ay fo

r hea

lth

vs.

18%

Ant

ibio

tic

use:

4%

vs.

16%

Mot

ivat

ion

mea

sure

s fo

ran

d O

RS

use

pro

mot

edw

orke

rs. E

duca

tion

Dia

rrhe

a ho

spit

aliz

atio

ns:

heal

th p

rovi

ders

(if a

ny) n

otth

roug

h tr

aini

ngs

cam

paig

ns p

rom

oted

zin

cA

fter

6 m

onth

s of

inte

rven

tion

:cl

ear.

Inje

ctio

n us

age

and

othe

rC

ontr

ol: O

nly

OR

Sas

'ton

ic' f

or p

reve

ntin

gzi

nc u

se d

urin

g di

arrh

eal e

piso

de:

drug

usa

ge re

mai

ned

high

inpr

omot

ion

futu

re e

piso

des.

59.8

% O

RS

use

: 59.

2% v

s. 9

.8%

both

are

as.

Com

plia

nce

wit

h 14

day

zin

c: 6

2%A

ctua

l com

plia

nce

wit

h fu

ll 1

4D

iarr

hea

in la

st tw

o w

eeks

: 14%

vs.

day

cour

se w

as 2

2-36

%.

23%

Ant

ibio

tic

use:

2%

vs.

15%

Bha

ndar

iP

rim

ary

heal

thS

urve

y of

236

4 ho

use-

gove

rnm

ent

Two

day

trai

ning

inT

here

was

a s

hift

in c

are-

seek

ing

Pil

ot s

tudy

; for

med

the

basi

s of

et a

l. 20

05ce

nter

of T

igao

nho

lds

havi

ng u

nder

-fiv

epr

ovid

ers

diar

rhea

man

agem

ent,

from

pri

vate

pro

vide

rs to

com

mun

ity

the

late

r clu

ster

-ran

dom

ized

[105

]in

Far

idab

adch

ildr

en s

erve

d by

(phy

sici

ans

zinc

dis

trib

utio

n an

dhe

alth

wor

kers

. The

pre

scri

ptio

n an

dtr

ial.

dist

rict

, Har

yana

,pr

imar

y he

alth

cen

ter.

and

aux

ilia

ryus

e, a

nd re

ferr

al c

rite

ria.

use

rate

s of

OR

S d

urin

g di

arrh

eaIn

dia.

nurs

e m

idw

ives

incr

ease

d m

arke

dly

from

7%

at

at th

e pr

imar

yba

seli

ne to

ove

r 40%

at 3

to 6

heal

th c

ente

r),

mon

ths.

Zin

c w

as p

resc

ribe

d in

mor

epr

ivat

eth

an h

alf o

f epi

sode

s an

d 72

-74%

prac

titi

oner

sre

port

ed g

ivin

g th

e co

mpl

ete

14-d

ayan

d co

mm

unity

cou

rse.

wor

kers

und

erth

e IC

DS

sch

eme.

Aw

asth

i,O

utpa

tien

t hea

lth

2,00

2 ch

ildr

en a

ged

No

deta

ils

onN

o de

tail

s on

trai

ning

In fi

ve o

f six

sit

es, O

RS

use

in c

ases

Qua

si-r

ando

miz

ed tr

ial;

uni

tet

al.

2006

faci

liti

es in

six

2 to

59

mon

ths

prof

ile

of h

ealt

h-of

hea

lthw

orke

rs.

wit

h co

ntin

ued

diar

rhea

on

days

3 to

5of

rand

omiz

atio

n be

ing

the

[106

]ce

nter

s in

five

Inte

rven

tion

: zin

cw

orke

rs; p

ro-

Loc

ally

dev

elop

edw

as th

e sa

me

in th

e tw

o gr

oups

or

day

coun

trie

s (B

razi

l,(2

0 m

g on

ce d

aily

babl

y do

ctor

s as

cult

ural

ly s

peci

fic

high

er in

zin

c gr

oup,

exc

ept i

n B

razi

lE

thio

pia,

Egy

pt,

for 1

4 da

ys) w

ith

the

stud

y is

don

em

essa

ges

emph

asiz

ing

whe

re it

was

low

er in

zin

c gr

oup.

Indi

a an

dO

RS

in o

utpa

tien

tth

erap

euti

c an

d pr

e-O

vera

ll a

dher

ence

to z

inc

supp

le-

Phi

lipp

ines

Con

trol

: OR

S a

lone

heal

th fa

cili

ties

vent

ive

bene

fits

of z

inc.

men

tati

on w

as 8

3.8%

(95%

CI 8

1-86

).L

ocal

ly d

evel

oped

Ant

ibio

tic/a

ntid

iarr

heal

use

was

less

ercu

ltur

ally

spe

cifi

c in

the

zinc

gro

up.

mes

sage

s fo

r pro

-m

otin

g zi

nc.

Gup

ta, e

t al.

Rur

al a

rea

Ran

dom

ized

dou

ble-

Vil

lage

sur

-M

odul

ar tr

aini

ng in

the

Low

er in

cide

nce

of d

iarr

hea

in th

eE

valu

ated

zin

cco

ntd.

..



ANMs and Aanganwadi workers) and privatepractitioners (including unlicenced) as channels ofdelivery of intervention (provision of zinc and ORS alongwith caregiver education on their use) after a brief trainingin diarrhea management and zinc use. The earlier pilotwork [105] by this group had demonstrated the feasibilityto train various government and community channels topromote zinc as a treatment for acute diarrhea through theprimary healthcare system. Information education andcommunication (IEC) activities involved posters andcampaigns to promote zinc as a treatment for diarrhea, aswell as a “tonic” that prevents diarrhea over the ensuingmonths, to ensure compliance with the full 14-day course.Uninterrupted supplies of ORS packets and zinc wereensured. The intervention resulted in reduction in theprevalence of diarrhea and hospitalizations. ORS useincreased significantly in the intervention areas, and theintervention also resulted in reduction in the use ofunwarranted oral and injectable drugs during diarrhea.

Operational feasibility of introducing zinc in thetreatment of diarrhea has also been demonstrated from amulticentric study in outpatient health facilities of sixcenters in five countries [106] where zinc supplementationwas taken by 84% of participants, and it resulted inreduction in antibiotic/antidiarrheal use. This studyhowever, did not demonstrate a significant increase inORS use with zinc supplementation. Operationalfeasibility of preventive zinc therapy has also beendemonstrated from rural area of West Bengal [107].Evidence from a communities of Bangladesh [108] alsodocumented a significantly higher use of ORS and otherhome fluids in the zinc intervention areas than those in thecomparison areas. The probability of use of anantimicrobial during diarrhea was only about one-third inthe intervention children compared to that in thecomparison children.

The improvement in ORS utilization and reduction inantibiotic use in the populations receiving zincdemonstrates that the benefits of zinc supplementationextend well beyond reducing the diarrheal morbidity andmortality. The reasons for reduction in antibiotic and otherunknown drug usage could be a reflection of fulfilling thecaregivers’ expectation of medicine by dispensing zinc,and a shift in careseeking behavior from private providerto government functionary. Improvement in ORS use ratesare likely to be the effect of enthusiastic IEC campaignsincorporating zinc messages along with ORS. Overall, thestudies have demonstrated that an intervention to improvediarrhea management with ORS and zinc is feasible,acceptable and effective in rural Indian communities.Further scaling-up of this intervention should be givenpriority in India.A

utho

r Ye

arP

lace

[R

ef.]

Sam

ple

size

and

Pro

file

of

Type

of t

rain

ing

Res

ults

Com

men

ts

2007

[10

7]co

mpr

isin

gbl

ind

plac

ebo

cont

-ve

illa

nce

offi

cers

loca

l lan

guag

e, in

clud

ing

supp

lem

ente

d gr

oup

(RR

0.7

4, 9

5%su

pple

men

tatio

n gi

ven

of 1

1 vi

llag

es in

roll

ed c

omm

unit

y(a

t lea

st s

econ

dary

diar

rhea

man

agem

ent,

CI 0

.64-

0.87

) 96%

of m

othe

rs a

dmin

is-r

outi

nely

and

not

dur

ing

Wes

t Ben

gal

base

d s

tudy

enr

olli

ngsc

hool

edu

cate

d)pr

epar

atio

n of

OR

S a

ndte

red

syru

p w

eekl

y to

thei

r chi

ldre

n.di

arrh

eal e

piso

des.

1712

chi

ldre

n ag

edse

lect

ed in

con

sul-

refe

rral

.be

twee

n 6

and

48ta

tion

wit

h vi

llag

em

onth

s in

a ra

ndo-

Pan

chay

at.

miz

ed d

oubl

e bl

ind

stud

y

Baq

ui,e

t al.

Ban

glad

esh

Com

mun

ity b

ased

Com

mun

itySt

anda

rd tr

aini

ng to

OR

S u

se 7

5% v

s. 5

0% A

ntib

ioti

cC

onti

nuou

s fr

ee s

uppl

ies

of20

04 [

108]

clus

ter r

ando

miz

edhe

alth

com

mun

ity

heal

thpr

escr

ipti

on 1

3% v

s. 3

4%zi

nc a

nd O

RS

asc

erta

ined

.tr

ial 3

,974

chi

ldre

n in

wor

kers

and

wor

kers

Oth

er m

edic

ines

15%

vs

45%

Reg

ular

sup

ervi

sion

.th

e in

terv

enti

on c

lust

ers

com

mun

ityM

otiv

atio

n m

easu

res

for

and

4,09

6 in

con

trol

volu

ntee

rshe

alth

pro

vide

rs (i

f any

) not

clus

ters

, Int

erve

ntio

n:cl

ear.

zinc

and

OR

S p

rom

otio

nan

d di

stri

buti

on; C

ontr

ol:

OR

S d

istr

ibut

ion

and

emph

asis

on

feed

ing



reduce antibiotic prescription and improves ORSutilization. Zinc therapy can be upscaled through existinginfrastructure, introducing the training component andinformation, education and communication activities.

Results of this review are in consonance with theexisting diarrhea prevention and control strategies. Recentanalysis using Lives Saved Tool (LiST) estimated thatdiarrheal deaths can be drastically reduced (by 78%-92%)if key interventions (ORS, zinc, antibiotics for dysentery,rotavirus vaccine, vitamin A supplementation, basic water,sanitation, hygiene, and breastfeeding) were scaled-up inthe 68 high child mortality countries [109].

Previous reviews tended to have methodologicallimitations such as incorporation of outdated data; orselective inclusion (or omission) of evidence supporting aparticular viewpoint. Another strength of this review is thatwe could access current data relevant to India frommultiple sources including Health Ministry documents,NFHS series, UNICEF surveys etc. Therefore, thissystematic review can be regarded as current,comprehensive and oriented to facilitating informeddecision making, especially at program level.

Nevertheless, some limitations of this review must alsobe recognized. We did not undertake detailed qualityassessment of the included publications. Therefore wehave not presented insights into the applicability,transferability or appropriateness of cited evidence withspecific reference to the Indian context. Secondaryanalysis of the data presented in the included publicationswas also not undertaken. Therefore, we are unable topresent a weighted average for numerical data or othermeta-analyses. We have reported data as presented in theoriginal publications, without filtering or treating them tofit a common reporting format. This can make it slightlydifficult to compare outcomes presented in variablemanner. Although we have accorded highest priority torecent systematic reviews, some conclusions presented insystematic reviews could stem from a limited number oftrials (in some cases, even one RCT) and participants. Itmust also be noted that we have not undertaken literaturesearch for some issues like efficacy and effectiveness ofzinc and ORS in management of diarrhea as these are wellestablished interventions, and the need of a re-appraisal onthese was not identified in the finalization of questions forthis systematic review.

Competing interests: None stated; Funding: UNICEF.

Disclaimer: The views expressed by authors are their own anddo not reflect the Institutional policies, to which they belong.

REFERENCES

1. Sample Registration System: Special Survey of Deaths.

Conclusions: Box 8

Knowledge gaps and directions for future research: Zinchas now been introduced in the National diarrheatreatment guidelines. Efforts are on to promote its usageby involving organizations such as Indian Academy ofPediatrics. The providers’ practices need to becontinuously monitored to document any effect of thesestrategies. The efficacy of zinc and its effect on health-seeking behavior, and ORS and antibiotic use rates needto be monitored in real program based settings.Compliance with the use of zinc for 14 days needs to beevaluated in the communities. The preventive effects ofusing zinc in the program need to be monitored byconducting follow-up incidence and prevalence studies.The capability of the existing health system to deliverzinc along with other strategies in management ofdiarrhea need to be assessed by continuous monitoringand supervision.

DISCUSSION

This systematic review methodologically collected andcollated evidence from a variety of sources (including butnot restricted to peer reviewed publications), to guide theinitiation and/or scaling-up of advocacy and actions fortackling the burden of childhood diarrhea in India. Thisreview is a critical first-step in today’s era of evidence-informed decision-making. The review concluded thatdiarrhea is widely prevalent at any point among under-five children in India, and accounts for 14% of the totaldeaths in this age group. Exclusive breastfeeding,handwashing and point-of-use water treatment wereidentified to be the most effective strategies forprevention of all-cause diarrhea. Rotavirus vaccines,though efficacious in preventing rotavirus specificdiarrhea, need further evaluation for their efficacy andeffectiveness in India or similar settings. The current userates for ORS and zinc are unacceptably low.Incorporation of zinc therapy in the health programs can

BOX 8

• Introduction of zinc in treatment of diarrheareduces antibiotic prescription and increases ORSusage provided all health functionaries (includingprivate providers) are involved in the program anduninterrupted supplies are maintained.

• Compliance with zinc is acceptable after IECactivities.

• Intervention of zinc treatment given duringdiarrhea can be upscaled through existinginfrastructure, introducing the training componentand IEC activities.



Report on Causes of Death: 2001-03, Office of RegistrarGeneral of India, Ministry of Home Affairs, New Delhi.2009.

2. United Nations Children’s Fund and Ministry of Healthand Family Welfare, Government of India. 2009 CoverageEvaluation Survey. All India Report. New Delhi: UNICEF,2010. Available from: URL: http://www.unicef.org/india/health_6679.htm. Accessed February 24, 2012.

3. International Institute for Population Sciences (IIPS) andMacro International. 2007. National Family Health Survey(NFHS-3), 2005–06: India: Volume I. Mumbai: IIPS.

4. Mathew JL, Shah D, Gera T, Gogia S, Mohan P, Panda R,et al. UNICEF-PHFI Newborn and Child Health Series –India. Systematic Reviews on Child Health Priorities forAdvocacy and Action: Methodology. Indian Pediatr.2011;48:183-9.

5. Kosek M, Bern C, Guerrant RL. The global burden ofdiarrhoeal disease, as estimated from studies publishedbetween 1992 and 2000. Bull WHO. 2003;81:197-204.

6. Bern C, Martines J, de Zoysa I, Glass RI. The magnitude ofthe global problem of diarrhoeal disease: a ten-year update.Bull WHO.1992;70:705-14.

7. Snyder JD, Merson MH. The magnitude of the globalproblem of acute diarrhoeal disease: a review of activesurveillance data. Bull WHO.1982;60:604-13.

8. Jones G, Steketee RW, Black RE, Bhutta ZA, Morris SS,and the Bellagio Child Survival Study Group. How manychild deaths can we prevent this year? Lancet. 2003; 362:65-71.

9. Black RE, Cousens S, Johnson HL, Lawn JE, Rudan I,Bassani DG, et al., for the Child Health EpidemiologyReference Group of WHO and UNICEF. Global, regional,and national causes of child mortality in 2008: a systematicanalysis. Lancet.2010;375:1969-87.

10. Million Death Study Collaborators, Bassani DG, Kumar R,Awasthi S, Morris SK, Paul VK, et al. Causes of neonataland child mortality in India: a nationally representativemortality survey. Lancet. 2010;376:1853-60.

11. International Institute for Population Sciences (IIPS) andORC Macro. 2000. National Family Health Survey(NFHS-2), 1998–99: India: Volume I. Mumbai: IIPS.

12. International Institute for Population Sciences (IIPS) andMacro International 1995. National Family Health Survey(NFHS-1), 1992–93: India: Volume I. Mumbai: IIPS.

13. UNICEF. Management Practices for Childhood Diarrheain India: Survey of 10 Districts. New Delhi: UNICEF;2009.

14. Anvikar AR, Dolla C, Dutta S, Rao VG, Gadge VS, ShuklaGP, et al. Role of Escherichia coli in acute diarrhoea intribal preschool children of central India. Paediatr PerinatEpidemiol. 2008 ;22:40-6.

15. Anand K, Sundaram KR, Lobo J, Kapoor SK. Are diarrhealincidence and malnutrition related in under five children?A longitudinal study in an area of poor sanitary conditions.Indian Pediatr. 1994;31:943-8.

16. Gupta N, Jain SK, Ratnesh, Chawla U, Hossain S,Venkatesh S. An evaluation of diarrheal diseases and acuterespiratory infections control programmes in a Delhi slum.Indian J Pediatr. 2007;74:471-6.

17. Awasthi S, Pande VK. Cause-specific mortality in underfives in the urban slums of Lucknow, north India. J TropPediatr. 1998;44:358-61.

18. Grover VL, Chhabra P, Malik S, Kannan AT. Pattern ofmorbidity and mortality amongst under fives in an urbanresettlement colony of East Delhi. Indian J Prev Soc Med.2004;35:21-26.

19. Parashar UD, Burton A, Lanata C, Boschi-Pinto C,Shibuya K, Steele D, et al. Global mortality associated withrotavirus disease among children in 2004.J Infect Dis. 2009Nov 1;200 Suppl 1:S9-S15.

20. Huilan S, Zhen LG, Mathan MM, Mathew MM, Olarte J,Espejo R, et al. Etiology of acute diarrhoea among childrenin developing countries: a multicentre study in fivecountries. Bull World Health Organ. 1991;69:549-55.

21. Nair GB, Ramamurthy T, Bhattacharya MK, Krishnan T,Ganguly S, Saha DR, et al. Emerging trends in the etiologyof enteric pathogens as evidenced from an activesurveillance of hospitalized diarrhoeal patients in Kolkata,India. Gut Pathog. 2010;2:4.

22. Kuttiat VS, Lodha R, Das B, Kohli U. Prevalence ofcholera in pediatric patients with acute dehydratingdiarrhea. Indian J Pediatr. 2010;77:67-71.

23. Ajjampur SS, Sarkar R, Sankaran P, Kannan A, MenonVK, Muliyil J, et al. Symptomatic and asymptomaticCryptosporidium infections in children in a semi-urbanslum community in southern India. Am J Trop Med Hyg.2010;83:1110-5.

24. Nandy S, Mitra U, Rajendran K, Dutta P, Dutta S. Subtypeprevalence, plasmid profiles and growing fluoroquinoloneresistance in Shigella from Kolkata, India (2001-2007): ahospital-based study. Trop Med Int Health. 2010;15:1499-507.

25. Nayak MK, Chatterjee D, Nataraju SM, Pativada M, MitraU, Chatterjee MK, et al. A new variant of Norovirus GII.4/2007 and inter-genotype recombinant strains of NVGIIcausing acute watery diarrhoea among children in Kolkata,India. J Clin Virol. 2009;45:223-9.

26. Samal SK, Khuntia HK, Nanda PK, Satapathy CS, NayakSR, Sarangi AK, et al. Incidence of bacterialenteropathogens among hospitalized diarrhea patients fromOrissa, India. Jpn J Infect Dis. 2008;61:350-5.

27. Ramani S, Sowmyanarayanan TV, Gladstone BP,Bhowmick K, Asirvatham JR, Jana AK, et al. Rotavirusinfection in the neonatal nurseries of a tertiary care hospitalin India. Pediatr Infect Dis J. 2008;27:719-23.

28. Ramani S, Kang G. Burden of disease & molecularepidemiology of group A rotavirus infections in India.Indian J Med Res. 2007;125:619-32.

29. Saha DR, Rajendran K, Ramamurthy T, Nandy RK,Bhattacharya SK. Intestinal parasitism and Vibrio choleraeinfection among diarrhoeal patients in Kolkata, India.Epidemiol Infect. 2008;136:661-4.

30. Monica B, Ramani S, Banerjee I, Primrose B, Iturriza-Gomara M, Gallimore CI, et al. Human caliciviruses insymptomatic and asymptomatic infections in children inVellore, South India. J Med Virol. 2007;79:544-51.

31. Banerjee I, Ramani S, Primrose B, Moses P, Iturriza-Gomara M, Gray JJ, et al. Comparative study of the



epidemiology of rotavirus in children from a community-based birth cohort and a hospital in South India. J ClinMicrobiol. 2006;44:2468-74.

32. Bhattacharya R, Sahoo GC, Nayak MK, Ghosh S, Dutta P,Bhattacharya MK, et al. Molecular epidemiology of humanastrovirus infections in Kolkata, India. Infect Genet Evol.2006;6:425-35.

33. Bahl R, Ray P, Subodh S, Shambharkar P, Saxena M,Parashar U, et al. Incidence of severe rotavirus diarrhea inNew Delhi, India, and G and P types of the infectingrotavirus strains. J Infect Dis. 2005;192 (Suppl 1):S114-9.

34. Ballal M, Ramamurthy T. Enteroaggregative Escherichiacoli diarrhea in Manipal. Indian Pediatr. 2005;42:722-3.

35. Mishra D, Gupta VK, Yadav RB. Role of Entamoebahistolytica in acute watery diarrhea in hospitalized under-five children. Indian Pediatr. 2004;41:861-3.

36. Phukan AC, Patgiri DK, Mahanta J. Rotavirus associatedacute diarrhoea in hospitalized children in Dibrugarh,north-east India. Indian J Pathol Microbiol. 2003;46:274-8.

37. Kaur R, Rawat D, Kakkar M, Uppal B, Sharma VK.Intestinal parasites in children with diarrhea in Delhi, India.Southeast Asian J Trop Med Public Health. 2002;33:725-9.

38. Ballal M, Shivananda PG. Rotavirus and enteric pathogensin infantile diarrhoea in Manipal, South India. Indian JPediatr. 2002;69:393-6.

39. Saha MR, Saha D, Dutta P, Mitra U, Bhattacharya SK.Isolation of Salmonella enterica serotypes from childrenwith diarrhoea in Calcutta, India. J Health Popul Nutr.2001;19:301-5.

40. Jain V, Parashar UD, Glass RI, Bhan MK. Epidemiology ofrotavirus in India. Indian J Pediatr. 2001;68:855-62.

41. Government of India: Ministry of Health & FamilyWelfare; Child Health Division. Minutes of the High LevelCommittee Meeting Held on 20.3.2007 to Discuss Role ofZinc in Management of Diarrhoea and New Guidelines forTreatment of Diarrhoea in Children. Available from: URL:h t t p : / / w w w . w h o i n d i a . o r g / L i n k F i l e s /Child_Health_in_India_Child_Health_GOI_Guidelines_002.pdf.Accessed January 26, 2012.

42. Bhatnagar S, Lodha R, Choudhury P, Sachdev HPS, ShahN, Narayan S, et al. IAP Guidelines 2006 on managementof acute diarrhea. Indian Pediatr. 2007; 44:380-9.

43. Bhutta ZA, Ahmed T, Black RE, Cousens S, Dewey K,Giugliani E, et al., for the Maternal and ChildUndernutrition Study Group. What works? Interventionsfor maternal and child undernutrition and survival. Lancet2008;371:417–40.

44. Duijts L, Jaddoe VW, Hofman A, Moll HA. Prolonged andexclusive breastfeeding reduces the risk of infectiousdiseases in infancy. Pediatrics. 2010;126:e18-25.

45. Ehlayel MS, Bener A, Abdulrahman HM. Protective effectof breastfeeding on diarrhea among children in a rapidlygrowing newly developed society. Turk J Pediatr.2009;51:527-33.

46. Quigley MA, Cumberland P, Cowden JM, Rodrigues LC.How protective is breast feeding against diarrhoeal diseasein infants in 1990s England? A case-control study. Arch

Dis Child. 2006;91:245-50.47. Khadivzadeh T, Parsai S. Effect of exclusive breastfeeding

and complementary feeding on infant growth andmorbidity. East Mediterr Health J. 2004;10:289-94.

48. Saleemi MA, Zaman S, Akhtar HZ, Jalil F, Ashraf RN,Hanson LA, et al. Feeding patterns, diarrhoeal illness andlinear growth in 0-24-month-old children. J Trop Pediatr.2004;50:164-9.

49. Bhandari N, Bahl R, Mazumdar S, Martines J, Black RE,Bhan MK; Infant Feeding Study Group. Effect ofcommunity-based promotion of exclusive breastfeeding ondiarrhoeal illness and growth: a cluster randomisedcontrolled trial. Lancet. 2003;361:1418-23.

50. Arifeen SE, Hoque DM, Akter T, Rahman M, Hoque ME,Begum K, et al. Effect of the Integrated Management ofChildhood Illness strategy on childhood mortality andnutrition in a rural area in Bangladesh: a cluster randomisedtrial. Lancet. 2009;374:393-403..

51. WHO Collaborative Study Team on the Role ofBreastfeeding on the Prevention of Infant Mortality. Effectof breastfeeding on infant and child mortality due toinfectious diseases in less developed countries: a pooledanalysis. Lancet. 2000;355:451-5.

52. Ejemot RI, Ehiri JE, Meremikwu MM, Critchley JA. Handwashing for preventing diarrhoea. Cochrane Database SystRev. 2008;23:CD004265.

53. Fewtrell L, Kaufmann RB, Kay D, Enanoria W, Haller L,Colford JM Jr. Water, sanitation, and hygiene interventionsto reduce diarrhoea in less developed countries: asystematic review and meta-analysis. Lancet Infect Dis.2005;5:42-52.

54. Curtis V, Cairncross S. Effect of washing hands with soapon diarrhoea risk in the community: a systematic review.Lancet Infect Dis. 2003;3:275-81.

55. Luby SP, Agboatwalla M, Bowen A, Kenah E, Sharker Y,Hoekstra RM. Difficulties in maintaining improvedhandwashing behavior, Karachi, Pakistan. Am J Trop MedHyg. 2009;81:140-5.

56. Soares-Weiser K, MacLehose H, Ben-Aharon I, GoldbergE, Pitan F, Cunliffe N. Vaccines for preventing rotavirusdiarrhoea: vaccines in use. Cochrane Database Syst Rev2010;5:CD008521.

57. Vieira SC, Gurgel RQ, Kirby A, Barreto IP, Souza LD,Oliveira OC, et al. Acute diarrhoea in a community cohortof children who received an oral rotavirus vaccine inNortheast Brazil. Mem Inst Oswaldo Cruz. 2011 ;106:330-4.

58. Molto Y, Cortes JE, De Oliveira LH, Mike A, Solis I,Suman O, et al. Reduction of diarrhea-associatedhospitalizations among children aged < 5 Years in Panamafollowing the introduction of rotavirus vaccine. PediatrInfect Dis J. 2011;30(1 Suppl):S16-20.

59. Quintanar-Solares M, Yen C, Richardson V, Esparza-Aguilar M, Parashar UD, Patel MM. Impact of rotavirusvaccination on diarrhea-related hospitalizations amongchildren < 5 years of age in Mexico. Pediatr Infect Dis J.2011;30(1 Suppl):S11-5.

60. de Palma O, Cruz L, Ramos H, de Baires A, Villatoro N,Pastor D, et al. Effectiveness of rotavirus vaccination



against childhood diarrhoea in El Salvador: case-controlstudy. BMJ. 2010;340:c2825. doi:10.1136/bmj.c2825.

61. Richardson V, Hernandez-Pichardo J, Quintanar-SolaresM, Esparza-Aguilar M, Johnson B, Gomez-AltamiranoCM, et al. Effect of rotavirus vaccination on death fromchildhood diarrhea in Mexico. N Engl J Med.2010;;362:299-305.

62. Centers for Disease Control and Prevention (CDC).Reduction in rotavirus after vaccine introduction—UnitedStates, 2000-2009. MMWR Morb Mortal Wkly Rep.2009;58:1146-9.

63. Esparza-Aguilar M, Bautista-Márquez A, González-Andrade Mdel C, Richardson-López-Collada VL. Analysisof the mortality due to diarrhea in younger children, beforeand after the introduction of rotavirus vaccine. SaludPublica Mex. 2009;51:285-90.

64. Sinclair D, Abba K, Zaman K, Qadri F, Graves PM. Oralvaccines for preventing cholera. Cochrane Database SystRev. 2011;3:CD008603.

65. Graves PM, Deeks JJ, Demicheli V, Jefferson T. Vaccinesfor preventing cholera: killed whole cell or other subunitvaccines (injected). Cochrane Database Syst Rev.2010;8:CD000974.

66. Reddaiah VP; Kapoor SK. Does measles immunizationreduce diarrhoeal morbidity. Indian J Comm Med1993;18:115-7.

67. Narang A, Bose A, Pandit AN, Dutta P, Kang G,Bhattacharya SK, et al. Immunogenicity, reactogenicityand safety of human rotavirus vaccine (RIX4414) in Indianinfants. Hum Vaccin. 2009;5:414-9.

68. Podewils LJ, Antil L, Hummelman E, Bresee J, ParasharUD, Rheingans R. Projected cost-effectiveness of rotavirusvaccination for children in Asia. J Infect Dis.2005;192(Suppl 1):S133-45.

69. Sachdev HP, Kumar S, Singh KK, Satyanarayana L, PuriRK. Risk factors for fatal diarrhea in hospitalized childrenin India. J Pediatr Gastroenterol Nutr. 1991;12:76-81.

70. Sur D, Lopez AL, Kanungo S, Paisley A, Manna B, Ali M,et al. Efficacy and safety of a modified killed-whole-celloral cholera vaccine in India: an interim analysis of acluster-randomised, double-blind, placebo-controlled trial.Lancet. 2009;374:1694-702.

71. Gogia S, Sachdev HS. Vitamin A supplementation for theprevention of morbidity and mortality in infants six monthsof age or less. Cochrane Database Syst Rev.2011;2:CD007480.

72. Mayo-Wilson E, Imdad A, Herzer K, Yakoob MY, BhuttaZA. Vitamin A supplements for preventing mortality,illness, and blindness in children aged under 5: systematicreview and meta-analysis. BMJ. 2011;343:d5094. doi:10.1136/bmj.d5094.

73. Imdad A, Yakoob MY, Sudfeld C, Haider BA, Black RE,Bhutta ZA. Impact of vitamin A supplementation on infantand childhood mortality. BMC Public Health.2011;11(Suppl 3):S20.

74. Gogia S, Sachdev HS. Neonatal vitamin Asupplementation for prevention of mortality and morbidityin infancy: systematic review of randomised controlledtrials. BMJ. 2009;338:b919. doi: 10.1136/bmj.b919.

75. Grotto I, Mimouni M, Gdalevich M, Mimouni D. VitaminA supplementation and childhood morbidity from diarrheaand respiratory infections: a meta-analysis. J Pediatr.2003;142:297-304.

76. D’Souza RM, D’Souza R. Vitamin A for preventingsecondary infections in children with measles—asystematic review. J Trop Pediatr. 2002;48:72-7.

77. Lazzerini M, Ronfani L. Oral zinc for treating diarrhoea inchildren. Cochrane Database Syst Rev. 2008;3:CD005436.

78. Patel AB, Mamtani M, Badhoniya N, Kulkarni H. Whatzinc supplementation does and does not achieve in diarrheaprevention: a systematic review and meta-analysis. BMCInfect Dis. 2011;11:122.

79. Yakoob MY, Theodoratou E, Jabeen A, Imdad A, EiseleTP, Ferguson J, et al. Preventive zinc supplementation indeveloping countries: impact on mortality and morbiditydue to diarrhea, pneumonia and malaria. BMC PublicHealth. 2011 Apr 13;11 Suppl 3:S23.

80. Gulani A, Bhatnagar S, Sachdev HP. Neonatal zincsupplementation for prevention of mortality and morbidityin breastfed low birth weight infants: systematic review ofrandomized controlled trials. Indian Pediatr. 2011;48:111-7.

81. Aggarwal R, Sentz J, Miller MA. Role of zincadministration in prevention of childhood diarrhea andrespiratory illnesses: a meta-analysis. Pediatrics.2007;119:1120-30.

82. Bhutta ZA, Black RE, Brown KH, Gardner JM, Gore S,Hidayat A, et al. Prevention of diarrhea and pneumonia byzinc supplementation in children in developing countries:pooled analysis of randomized controlled trials. J Pediatr.1999;135:689-97.

83. Vishwanathan H, Rhode JE. Diarrhea in Rural India: ANationwide Study of Mothers & Practitioners, All IndiaSummary. New Delhi: Vision Books; 1990.

84. Kumar R, Indira K, Rizvi A, Rizvi T, Jeyaseelan L.Antibiotic prescribing practices in primary and secondaryhealth care facilities in Uttar Pradesh, India. J Clin PharmTher. 2008;33:625-34.

85. Awasthi S, Srivastava NM, Pant S. Symptom-specific care-seeking behavior for sick neonates among urban poor inLucknow, Northern India. J Perinatol. 2008;28 (Suppl2):S69-75.

86. S KI, Chandy SJ, Jeyaseelan L, Kumar R, Suresh S.Antimicrobial prescription patterns for common acuteinfections in some rural & urban health facilities of India.Indian J Med Res. 2008;128:165-71.

87. Balasubramanian S, Ganesh R. Prescribing pattern of zincand antimicrobials in acute diarrhea. Indian Pediatr.2008;45:701.

88. Chakrabarti A. Prescription of fixed dose combinationdrugs for diarrhoea. Indian J Med Ethics. 2007;;4:165-7.

89. Bharathiraja R, Sridharan S, Chelliah LR, Suresh S,Senguttuvan M. Factors affecting antibiotic prescribingpattern in pediatric practice. Indian J Pediatr. 2005;72:877-9.

90. Singh J, Bora D, Sachdeva V, Sharma RS, Verghese T.Prescribing pattern by doctors for acute diarrhoea inchildren in Delhi, India. Diarrhoeal Dis Res. 1995;



13:229-31.91. Raghu MB, Balasubramanian S, Balasubrahmanyam G,

Indumathy, Ramnath A. Drug therapy of acute diarrhoea inchildren—actual practice and recommendations. Indian JPediatr. 1995;62:433-7.

92. Gupta DN, SenGupta PG, Sircar BK, Mondal S, Sarkar S,Deb BC. Implementation of ORT: some problemsencountered in training of health workers during anoperational research programme. Indian J Public Health.1994;38:69-72.

93. Sircar BK, Deb BC, Sengupta PG, Mondal S, Gupta DN,Sarkar S, et al. An operational study on implementation oforal rehydration therapy in a rural community of WestBengal, India. Indian J Med Res. 1991;93:297-302.

94. Dua T, Bahl R, Bhan MK. Lessons learnt from DiarrhealDiseases Control Program and implications for the future.Indian J Pediatr. 1999;66:55-61.

95. Sengupta PG, Mondal SK, Ghosh S, Gupta DN, Sikder SN,Sircar BK. Review on development and communityimplementation of oral rehydration therapy. Indian J PublicHealth. 1994;38:50-7.

96. Suarez De Balcazar Y, Balcazar FE. Child survival in theThird World: a functional analysis of oral rehydrationtherapy dissemination campaigns. Behav Change.1991;8:26-34.

97. Sarkar K, Sircar BK, Roy S, Deb BC, Biswas AB, BiswasR. Global review on ORT (oral rehydration therapy)programme with special reference to Indian scene. Indian JPublic Health. 1990;34:48-53.

98. Kumar V, Kumar R, Khurana JL. Assessment of the effectof training on management of acute diarrhoea in a primaryhealth care setting. J Diarrhoeal Dis Res. 1989;;7:70-6.

99. Haines A, Sanders D, Lehmann U, Rowe AK, Lawn JE,Jan S, et al. Achieving child survival goals: potentialcontribution of community health workers. Lancet 2007;369: 2121–31.

100. Koul PB, Murali MV, Gupta P, Sharma PP. Evaluation ofsocial marketing of oral rehydration therapy. IndianPediatr. 1991;28:1013-6.

101. Green EC. Diarrhea and the social marketing of oralrehydration salts in Bangladesh. Soc Sci Med.1986;23:357-66.

102. Meyer A, Foote D, Smith W. Communication worksacross cultures: hard data on ORT. IDev Commun Rep.1985;51:3-4.

103. Rao KV, Mishra VK, Retherford RD. Mass media canhelp improve treatment of childhood diarrhoea. Natl FamHealth Surv Bull. 1998;11:1-4.

104. Bhandari N, Mazumder S, Taneja S, Dube B, AgarwalRC, Mahalanabis D, et al. Effectiveness of zincsupplementation plus oral rehydration salts comparedwith oral rehydration salts alone as a treatment for acutediarrhea in a primary care setting: a cluster randomizedtrial. Pediatrics. 2008 May;121:e1279-85.

105. Bhandari N, Mazumder S, Taneja S, Dube B, Black RE,Fontaine O, et al. A pilot test of the addition of zinc to thecurrent case management package of diarrhea in aprimary health care setting. J Pediatr Gastroenterol Nutr.2005;41:685-7.

106. Awasthi S; INCLEN Childnet Zinc Effectiveness forDiarrhea (IC-ZED) Group. Zinc supplementation in acutediarrhea is acceptable, does not interfere with oralrehydration, and reduces the use of other medications: arandomized trial in five countries. Pediatr GastroenterolNutr. 2006;42:300-5.

107. Gupta DN, Rajendran K, Mondal SK, Ghosh S,Bhattacharya SK. Operational feasibility ofimplementing community-based zinc supplementation:impact on childhood diarrheal morbidity. Pediatr InfectDis J. 2007;26:306-10.

108. Baqui AH, Black RE, El Arifeen S, et al. Zinc therapy fordiarrhoea increased the use of oral rehydration therapyand reduced the use of antibiotics in Bangladeshichildren. J Health Popul Nutr. 2004;22:440–2.

109. Fischer Walker CL, Friberg IK, Binkin N, Young M,Walker N, Fontaine O, et al. Scaling up diarrheaprevention and treatment interventions: a Lives SavedTool analysis. PLoS Med. 2011;8:e1000428.

promoting appropriate management of diarrhea: a systematic ... · objective: to identify,...

Documents