LSHTM Research Online

Sharma, G; (2017) An Investigation into Quality of Care at the Time of Birth at Public and PrivateSector Maternity Facilities in Uttar Pradesh, India. PhD thesis, London School of Hygiene & TropicalMedicine. DOI: https://doi.org/10.17037/PUBS.04646087

Downloaded from: http://researchonline.lshtm.ac.uk/4646087/

DOI: https://doi.org/10.17037/PUBS.04646087

Usage Guidelines:

Please refer to usage guidelines at https://researchonline.lshtm.ac.uk/policies.html or alternativelycontact researchonline@lshtm.ac.uk.

Available under license: http://creativecommons.org/licenses/by-nc-nd/2.5/

https://researchonline.lshtm.ac.uk

ANINVESTIGATIONINTOQUALITYOFCAREATTHETIMEOFBIRTHATPUBLIC

ANDPRIVATESECTORMATERNITYFACILITIESINUTTARPRADESH,INDIA

GAURAVSHARMA

Thesissubmittedinaccordancewiththerequirementsforthedegreeof

DoctorofPhilosophyoftheUniversityofLondon

JULY2017

DepartmentofInfectiousDiseaseEpidemiology

FacultyofEpidemiologyandPopulationHealth

LONDONSCHOOLOFHYGIENE&TROPICALMEDICINE

Researchgroupaffiliation(s):CentreforMaternal,ReproductiveandChildHealth(MARCH)

Funding:MSDforMothers

Page2of248

Declarationbycandidate

I, Gaurav Sharma, confirm that the work presented in this thesis is my own. Where

informationhasbeenderivedfromothersources, Iconfirmthishasbeen indicated inthe

thesis

Signed:………………………………..

Date:7December2017

Page3of248

Abstract

Background:Ensuringhighqualitycareduringlabourandchildbirthisimportanttoeliminate

preventable maternal deaths, neonatal deaths and intrapartum stillbirths. My PhD

investigatesqualityofcare(QoC)duringnormallabourandchildbirth,andexamineswhether

QoCisinfluencedbymanagementpracticesat26publicandprivatesectormaternityfacilities

inUttarPradesh,India.

Methods:First,Iconductedclinicalobservationsoflabourandchildbirth.Iuseddescriptive

statisticsandmultivariateanalysistechniquestodescribeandcomparedifferencesinoverall

QoC,andqualityforobstetricandneonatalcare.Second,Iusedquantitativeandqualitative

methods to describe existing patterns of mistreatment encountered by women. Third, I

describedexistingmanagementpracticesusinga separate surveydatasetand linkedboth

QoCandmanagementdatasetstoexaminetherelationshipbetweenmanagementpractices

andQoC.

Results:QoCwasfoundtobepooratbothpublicandprivatesectorfacilities.Theprivate

sectoroutperformedpublicsectorfacilitiesforoverallessentialcareatbirth,andforboth

obstetricandnewborncare.Allwomenencounteredatleastoneindicatorofmistreatment.

TherewerenosignificantdifferencesbetweenqualifiedandunqualifiedpersonnelforQoC

andmistreatmentlevels.Qualitativeresultssuggestthatinformalpaymentsarewidespread,

maternitycarepathwaysarenon-functional,andtherearepoorhygienestandards.Lastly,I

foundthatmaternityfacilitiesscoredpoorlyonmanagementbestpractices.Overall,Ifound

noassociationbetweentotalmanagementscoresandQoC.

Conclusions:TheresultsofmyPhDstudyindicatethatin2015,inmaternityfacilitiesofUttar

Pradesh,unqualifiedpersonnelprovidedthebulkofmaternitycare,adherencetoevidence-

basedobstetricandneonatalcareprotocolswasgenerallypoorandallwomenencountered

at leastonepracticeofmistreatment. Theseresultssuggesttheneedtocomprehensively

measureandurgentlyimproveQoCatthetimeofbirthinUttarPradesh,India.

Acknowledgements

IamgratefulforalltheguidanceandsupportthatIhavereceivedandthepeoplethatIhave

metinthisjourney.ForapassionatepublichealthdoctorfromNepal,itwasalife-longdream

tocometotheLSHTMandlearnadvancedresearchskills.

Thisworkwouldnothavebeenpossiblewithoutthementorshipofmyexcellentsupervisors.

I am deeply grateful to Professor Veronique Filippi, my primary supervisor, who always

providedmewith timely and insightful guidance, and treatedmewith immense kindness

during the entire process. I am also indebted to Dr. Timothy Powell-Jackson, my co-

supervisor,who provided superbmentorship and guidance throughout this process. I am

gratefultoMs.LovedayPenn-Kekanaforheradviceonthequalitativeaspectsofmyworkand

for many interesting discussions. Dr. John Bradley has been an integral member of my

supervisorypanelandprovidedtimelyguidanceonquantitativeaspectsofmywork.

Iwouldalsoliketoacknowledgemanyotherinspiringacademicswithinthematernalhealth

groupandtheMARCHcentreattheLSHTM,whoseworkhasandwillcontinuetoinspireme.

Inparticular,IhavebenefittedfrommanyinformaldiscussionswithProfessorOonaCampbell

and Iamgrateful toher forgenerouslygivingmetimeandsharingherwisdom. Iamalso

thankfulformyfriends,SchadracAgblaandAnowerHossainforsharingthisjourneywithme,

their encouragement and for always helping me with any statistical problems that I

encountered.

MyPhDworkhasalsobeendeeplyinformedbymypastprofessionalexperiences,andIam

grateful to my previous supervisors and mentors for their encouragement and ongoing

support.MystayinLucknowwouldhavebeenmiserableifitwasnotforMr.BirenThapar

andIamgratefulforhisfriendshipandkindness.

ThisPhDthesisisdedicatedtomyparentsandtomywifeJune.Thankyouforallowingmeto

undertakethis;and,foryourunconditionallove,supportandsacrificesduringthischallenging

process.

Page5of248

TableofContents

Chapter1.Introduction........................................................................................................................13

1.1. Motivationforthethesis..........................................................................................................13

1.1.1:Thehighandinequitableburdenofmaternalandneonataldeaths..................................13

1.1.2:Relevanceoftheresearchtoongoingglobalefforts.........................................................13

1.1.3:Importanceofcareatthetimeofbirth.............................................................................15

1.1.4:EvidenceonQoCformaternityservicesintheprivatesectorislimited............................16

1.1.5:Managementpracticesatmaternityfacilitiesisanunder-researchedarea......................17

1.2. Purposeofthethesis............................................................................................................19

1.3. Outlineofthesis...................................................................................................................20

Chapter2:Literaturereview................................................................................................................22

2.1:Introduction..............................................................................................................................22

2.2:TheimportanceofqualityatmaternityfacilityinLMICsettings..............................................23

2.3:Qualityofessentialcareatthetimeofbirth............................................................................24

2.3.1:Background........................................................................................................................24

2.3.2:Conceptualisinganddefininghighqualitymaternitycarepathwaysatfacilities..............26

2.3.3:Skilledbirthattendance.....................................................................................................28

2.3.4:Interventionsrecommendedforcareatthetimeofbirth................................................30

2.3.5:Interventionsnotrecommendedforuseduringthetimeofbirth...................................33

2.3.6:Theimportanceofrespectfulmaternitycareduringlabourandchildbirth......................34

2.4:Frameworksofqualityinhealthanddefinitions.......................................................................37

2.4.1:Definitionsofqualityofcareinhealthservices.................................................................37

2.4.2:Elementsofqualityofcareinhealthservices...................................................................38

2.5:FrameworksanddefinitionsofQoCspecifictomaternalandnewbornhealth........................39

2.6:MeasurementofQoCformaternalandnewbornhealthinLMICsettings...............................41

2.6.1:Measuringstructureelementsofqualityofcare...............................................................41

2.6.2:Measuringprocesselementsofqualityofcare................................................................42

2.6.3:Measuringhealthoutcomemeasuresofqualityofcare...................................................47

2.6.4:Summaryofmeasuringqualityofcareinmaternalandnewbornhealth.........................49

2.7:EmpiricalevidenceondeficienciesinQoCduringlabourandchildbirthinIndia.....................49

2.8:Managementpracticesatmaternityfacilities...........................................................................56

2.8.1:Theoreticalconceptsonmanagementpractices...............................................................56

2.8.2:Empiricalevidenceonhospitalmanagementpracticesandquality..................................60

Chapter3:Researchsettingandthecontextforthedoctoralresearch..........................................63

3.1Studysetting...........................................................................................................................63

Page6of248

3.2:HealthcaresysteminIndia...................................................................................................65

3.3:Maternalandnewbornhealthservicesprovidedatpublicsectorfacilities.........................67

3.4:Maternalhealthprogrammesandpolicies...........................................................................68

3.5:ContextofthePhDresearchwithintheMatrikaproject......................................................70

Chapter4:Roleofthecandidate,fundingandresearchtimeline.......................................................72

4.1:Theroleofthecandidate.........................................................................................................72

4.2:Funding......................................................................................................................................73

4.3:Researchtimeline.....................................................................................................................74

Chapter5:Conceptualframework,aims,objectivesandstudydesign...............................................76

5.1:ConceptualframeworkformyPhD...........................................................................................76

5.2:Aimsofthedoctoralresearch...................................................................................................77

5.3:SpecificObjectives.....................................................................................................................77

5.4:Studydesign..............................................................................................................................78

5.5:Datacollectioninstruments......................................................................................................78

5.5.1:Qualityofcareassessments...............................................................................................78

5.5.2:Surveyonmanagementpracticesatmaternityfacilities...................................................79

5.6:Samplesizecalculations............................................................................................................80

5.6.1:Forclinicalpracticeobservations.......................................................................................80

5.6.2:Theassessmentofmanagementpractices........................................................................81

5.7:Samplingstrategy......................................................................................................................81

5.7.1:Clinicalpracticeobservations.............................................................................................81

5.7.2:Managementsurvey..........................................................................................................82

5.8:Datacollection..........................................................................................................................82

5.8.1:Clinicalpracticeobservations.............................................................................................82

5.8.2:Assessmentofmanagementpractices...............................................................................83

5.9:Studyparticipants.....................................................................................................................83

5.10:Dataanalysis............................................................................................................................83

5.11:Researchethics.......................................................................................................................86

5.12:Datamanagement...................................................................................................................87

Chapter6:Qualityofessentialcareatthetimeofbirth:Findingsfromclinicalobservationsofspontaneouslabourandchildbirthat26publicandprivatesectorfacilitiesinUttarPradesh,India.88

6.1:Introduction..............................................................................................................................90

6.2:Methods....................................................................................................................................91

6.3:Analysis......................................................................................................................................96

6.4:Results.......................................................................................................................................96

6.5:Discussion................................................................................................................................102

Page7of248

Chapter7:AninvestigationintomistreatmentofwomenduringlabourandchildbirthinmaternitycarefacilitiesinUttarPradesh,India:amixedmethodsstudy..........................................................106

7.1:Introduction............................................................................................................................108

7.2:Methods..................................................................................................................................110

7.3:Analysis....................................................................................................................................113

7.4:Results.....................................................................................................................................114

7.5:Discussion................................................................................................................................126

7.6:Limitations...............................................................................................................................130

7.7:Conclusions.............................................................................................................................131

CHAPTER8:Managementisnotassociatedwithqualityofcareduringlabourandchildbirth:evidencefromacross-sectionalstudyofmaternityfacilitiesinUttarPradesh,India.......................132

8.1:Introduction............................................................................................................................134

8.2:Methods..................................................................................................................................136

8.3:Results.....................................................................................................................................141

8.4:Discussion................................................................................................................................148

8.5:Limitations...............................................................................................................................150

8.6:Conclusions.............................................................................................................................151

Chapter9:Discussionoftheresultsoftheoveralldoctoralresearch...............................................153

9.1:Summaryofkeyfindings.........................................................................................................153

9.1.1:QualityofcarewasgenerallypooracrossthesampledpublicandprivatesectormaternityfacilitiesinUttarPradeshin2015..............................................................................................154

9.1.2:MistreatmentofwomenfrequentlyoccurredatmaternityfacilitiesinUttarPradeshin2015............................................................................................................................................160

9.1.3:OverallmanagementscorewasnotassociatedwithQoCatmaternityfacilitiesinUttarPradeshin2015..........................................................................................................................165

9.2:Plansfordissemination...........................................................................................................167

9.3:Reflections,strengthsandlimitations.....................................................................................168

9.3.1:Forobjective1:QoCduringlabourandchildbirthatmaternityfacilitiesinUP..............168

9.3.2:Forobjective2:MistreatmentofwomenatmaternityfacilitiesinUP...........................173

9.3.3:Forobjective3:ManagementanditsrelationshipwithQoC..........................................176

9.4:Implicationsofthedoctoralstudy..........................................................................................180

9.4.1:Recommendationsforthefutureresearchagenda.........................................................181

9.4.2:Recommendationforprogrammes..................................................................................186

9.4.3:Recommendationsforpolicy...........................................................................................187

Chapter10:Conclusions.....................................................................................................................189

11.Listofreferences..........................................................................................................................190

12.ListofAppendices........................................................................................................................210

Page8of248

Appendix1:QoCassessmenttoolfornormallabourandchildbirthinUttarPradeshin2015.210

Appendix2:ToolforassessmentofmanagementpracticesinmaternityfacilitiesinUttarPradeshin2015..........................................................................................................................220

Appendix3:Informationsheetsandconsentforms..................................................................230

Appendix4:Ethicalapprovallettersandpermissions...............................................................235

Appendix5:PublishedmanuscriptforChapter6.......................................................................240

Appendix6:Tableshowingfrequencyofmistreatmentbysector............................................248

ListofTables

Table1:Interventionsrecommendedforuseduringintrapartumandpostpartumperiod................31

Table2:Interventionsforintrapartumandpostpartumcarethatdonothaverecommendationsforuse................................................................................................................................................33

Table3:DemographicandhealthindicatorsinUttarPradeshandstudydistricts..............................64

Table4:MaternalandnewborncarestandardsatIndianpublicsectorfacilities...............................67

Table5:Summaryoftheevolutionofqualityinmaternalhealth.......................................................68

Table6:TimelineforthePhD...............................................................................................................74

Table7:IndicesforQualityofCare......................................................................................................95

Table8:Samplecharacteristics............................................................................................................97

Table9:VariationsinessentialcareatbirthacrosspublicandprivatesectorsinUttarPradesh,India....................................................................................................................................................100

Table10:Resultsfromthemultilevelmixedeffectslinearregression..............................................101

Table11:Socio-demographiccharacteristicsofthesamplebytwooveralllevelsofmistreatment.114

Table12:Bivariateanalysisofthesignificancebysocio-demographicfactorsandtheprevalenceofobservedindicatorsofmistreatment.........................................................................................118

Table13:Themesandtheircomposition-clinicalobservationsoflabourandchildbirthatmaternityfacilities......................................................................................................................................120

Table14:Correlationresultsbetweenindependentassessorsratingsformanagementdimensions....................................................................................................................................................142

Table15:Maternityfacilitysamplecharacteristicscategorisedbytheirmanagementscores.........143

Table16:Relationshipbetweenfacilitysamplecharacteristicsandmanagementscores................144

Table17:MixedeffectslinearregressionexaminingtherelationshipbetweenoverallQoCatbirthandZscoreindexfortotalmanagementscoreat26maternityfacilities..................................146

Table18:Mixedeffectslinearregressionexaminingtherelationshipbetweenqualityofcareandzscoresindexesformanagementsub-categoriesfor26maternityfacilities..............................147

Listsoffigures

Figure1:Schematicdiagramofmaternitycarepathwaysfordelivery................................................27

Figure2:WHOQualityofCareFrameworkformaternalandnewbornhealth...................................40

Figure3:Dimensionsofmanagementpracticesatmaternityfacilities...............................................59

Figure4:SchematicrepresentationofthepublichealthsysteminIndia............................................65

Figure5:ConceptualframeworkformyPhD.......................................................................................77

Figure6:StudyflowdiagramfortheassessmentofQoCduringlabourandchildbirth......................93

Figure7:Qualityofcareitemsforobstetricandnewborncarebysectorusingweighteddata........99

Figure8:EstimatedHawthorneeffectacrosssampledobservationsin26hospitalsofUttarPradesh....................................................................................................................................................102

Figure9:Quantitativeresultsshowingtheprevalenceofindicatorsofmistreatmentinpublicandprivatesectormaternityfacilities..............................................................................................116

Figure10:Overallstudyflowdiagram-investigatingtherelationshipbetweenmanagementpracticesandqualityofcareduringlabourandchildbirth........................................................................139

Figure11:Histogramshowingtotalmanagementscoresacrosssampledfacilities(n=33)...............142

Figure12:Graphshowingscoresfortotalandindividualmanagementdomainsatpublicandprivatesectorfacilities...........................................................................................................................142

Figure13:Weightedestimatesofqualityofcareatmaternityfacilitiescategorisedbytheirmanagementscores...................................................................................................................145

Listofabbreviations

AMDD AvertingMaternalDeathandDisabilityProgram

AMTSL ActiveManagementofthethirdstageoflabour

ANC Antenatalcare

ANM AuxiliaryNursemidwife

ARR AnnualRateofReduction

ASHA AccreditedSocialHealthActivists

ASME AdvancedStatisticalMethodsinEpidemiology

BEmOC BasicEmergencyObstetricCare

CEmOC ComprehensiveEmergencyObstetricCare

CHC CommunityHealthCentres

EmOC EmergencyObstetricCare

EmONC EmergencyObstetricandNewborncare

ENAP EveryNewbornActionPlan

EPMM Endingpreventablematernalmortality

IMPAC IntegratedManagementofPregnancy&Childbirth

FIGO TheInternationalFederationofGynaecologyandObstetrics

FRU FirstReferralUnits

JSY JananiSurakshaYojana

LMICs LowandMiddle-IncomeCountries

LSHTM LondonSchoolofHygiene&TropicalMedicine

MARCH CentreforMaternal,Adolescent,ReproductiveandChildHealth

MBA MastersofBusinessAdministration

MET MaternalhealthcaremarketsEvaluationTeam

MCHIP MaternalChildHealthIntegratedProgram

MDG MillenniumDevelopmentGoals

MMR MaternalMortalityRatio

MNCH Maternal,NewbornandChildhealth

MNH MaternalandNewbornhealth

Page12of248

MOHFW MinistryofHealthandFamilyWelfare

MSD MerckSharpandDohmeCorp.

MICS MultipleIndicatorclustersurvey

NICE NationalInstituteforHealthandClinicalExcellence

NGOs Non-GovernmentalOrganizations

NRHM NationalRuralHealthMission

NMR NeonatalMortalityRate

PHS PublicHealthcareSociety

PhD DoctorofPhilosophy

PDSA PlanDoStudyAct

PCACL-R PerceptionsofCareAdjectiveChecklist

QoC QualityofCare

QPP-I Intrapartum-specificQualityfromthePatientsPerspectivequestionnaire

RMNH Reproductive,MaternalandNewbornHealth

RHFA RapidHealthFacilityAssessments

SARA ServiceAvailabilityandReadinessAssessment

SBA SkilledBirthAttendant

SDGs SustainableDevelopmentGoals

SME StatisticalMethodsinEpidemiology

SPA ServiceProvisionAssessment

SSQ SixSimpleQuestions

UP UttarPradesh

UK UnitedKingdomofGreatBritainandNorthernIreland

UHC universalhealthcoverage

USA UnitedSatesofAmerica

USAID UnitedStatesAgencyforInternationalDevelopment

UN UnitedNations

WHO WorldHealthOrganization

Page13of248

Chapter1.Introduction

1.1. Motivationforthethesis

Althoughtherehasbeenconsiderableprogressinmaternalandnewbornhealthoverthepast

twodecades,provisionofhigh-qualitycareforwomenoncetheyreachhealthcarefacilities

hasemergedasanimportantchallenge.1Poorqualityofcareatthetimeofbirthhampers

healthoutcomesforwomen,childrenandcommunities;andresearcheffortsshouldidentify

waystoimprovethecurrentstateofaffairs.

1.1.1:Thehighandinequitableburdenofmaternalandneonataldeaths

Maternal andnewbornhealthare important issues for sustainabledevelopment.Withan

estimatedannual210millionpregnanciesand140millionlivebirthsglobally,ensuringthat

everywoman and every newborn across the globe has the right to high quality care is a

formidablechallenge.2Theeraof theMillenniumDevelopmentGoals (MDGs) led togood

progress andmaternal deaths declined by nearly half (44%). However, this progresswas

inconsistentacrossmanypartsoftheworld,andmanycountriescouldnotachievetheMDG

5atargetofa75%reductioninthematernalmortalityratio(MMR).2

In2015,theMMRinhigh-incomecountries(12per100 000livebirths)wasfoundtobe46

timeslowerthanthehighestMMRinsub-SaharanAfrica(546per100 000).2Similarly,the

lifetime risk formaternaldeaths in2015wasmore than100 timeshigher in sub-Saharan

Africa: one in 36 compared to one in 4900 in high-income countries.2 During this time,

inequalitiesalsoworsened.Forexample,in1990thepooledMMRfor10countrieswiththe

highest levels ofmaternalmortalitywas 100 times greater than the pooledMMR for 10

countrieswiththelowestMMRlevels.However,by2013,thisgaphaddoubledto200times

greater.2Thesedatasuggestthatimprovingmaternalhealthisstillanunfinishedagenda.

1.1.2:Relevanceoftheresearchtoongoingglobalefforts

In2016,worldleaderswelcomedtheSustainableDevelopmentGoals(SDGs),whichunlike

theMDGs,haveabroaderdevelopment focus.3Goal threeof theSDGs isconcernedwith

ensuring healthy lives for all, and has five health targets including a specific target for

maternalmortality. 3 The targets for reducingpreventablematernalmortality are thatby

2030,allcountriesshouldreduceMMRbytwothirdsandnocountryshouldhaveanMMR

Page14of248

above140.4TheWorldHealthOrganization(WHO)andpartnershavecalledforintensified

action,particularly incountrieswithMMRgreater than420per100,000 livebirths.4 It is

hopedthatwithcollectiveefforts,theglobaltargetofMMRoflessthan70per100,000live

birthsby2030canbeachieved4.Atthenationallevel,twocountries,Nigeriaat19%(58,000

deaths)andIndiaat15%(45,000deaths)contributeduptoonethirdoftheglobalburdenof

maternal deaths in 2015.5 Therefore, both thesepopulous countries have tomake rapid

reductions in maternal mortality if the Global Strategy for Women’s, Children’s and

Adolescents’Health’stargetsaretobemetby2030.6

ThemainstrategyusedtoachievematernalandnewbornhealthtargetsduringtheMDGs

wastoexpandcoverageofsimpleandeffectiveinterventionsproventoworkagainstthemain

causesofdeaths.Successwasprimarilymeasuredthroughincreasedpopulationcoverageof

indicatorssuchasinstitutionalbirths,deliveriesbyskilledattendantsorantenatalcare.7There

wasgoodprogressbetween1990to2013;theproportionofbirthsoccurringwithskilledbirth

attendants(SBA)increasedfrom57%to74%;theproportionofwomenreceivingoneormore

antenatalcare(ANC)visitsincreasedfrom65%to83%;andfourormoreANCvisitsrosefrom

37%to64%.8,9However,increasingcoveragealonewithoutaspecificfocusonQoCmaynot

beoptimalforreasonsoutlinedbelow.7

First,thereisnowincreasingresearchevidencesuggestingthat,despiteincreasedcoverage

of institutional births, associated declines in perinatal and neonatal mortality have been

modestasshownbystudiesinIndia10,11andRwanda.12Inarecentcross-sectionalstudyfrom

Malawi,researchersalsofoundthatpoorhealthfacilityqualitywasassociatedwithhigher

riskofneonatalmortality.13

Second, global monitoring efforts are primarily designed to support global, regional and

nationalcomparisonsofcoverageindicatorssuchasANCvisits,institutionalbirthsandSBA

presenceatdelivery.Theseindicatorstrackuseofhealthcareratherthancontentofcare;

therefore,aqualitygapmayarisedespitetheincreasedpopulationcoverage.14Furthermore,

features beyond SBA and ANC coverage are likely to be important. For example, a high

populationdensityandlongtraveltimesmaycausedelaysinaccesstoemergencyobstetric

Page15of248

care (EmOC), andwomen's underlying health conditions, nutritional status and other life

circumstancesalsoinfluencebirthoutcomes.15

Third,themaincausesofmaternaldeathsin2015werefoundtohaveshiftedawayfromthe

morepreventabledirectcausestoindirectcausessuchasnon-communicablediseasesand

other intractable direct causes such as ectopic pregnancies, embolism and gestational

diabetes. 2 Therefore, there is growing recognition that improvingmaternal andnewborn

healthoutcomesintheSDGerawillrequireanadditionalemphasisonqualityofcareonce

womenreachhealthfacilities.7Iwillelaborateontheseissuesfurtherintheliteraturereview

sectioninchapter2.

1.1.3:Importanceofcareatthetimeofbirth

Despite the focus on promotion of institutional deliveries, the quality of routine care for

normal labour and childbirth has not received enough research and programmatic

attention.15,16Thetimearoundchildbirthhasalwaysbeentheriskiest forwomen inmany

partsoftheworld.16-18Recentestimatessuggestthatclosureofthequalitygapthroughthe

provisionofeffectiveandwoman-centredcareforallwomenandnewbornbabiesdelivered

infacilitiescouldpreventanestimated113,000maternaldeaths,531,000stillbirths,and1·32

millionneonataldeathsannuallyby2020.19

Consensusexistsonaminimumcarepackageof interventions requiredduringpregnancy,

labour and childbirth20. High quality, routine care during labour and childbirth has the

potentialtopreventmanymaternalandneonataldeaths,eitherthroughthepreventionof

complications or by timely intervention prior to the development of complications.21 For

example,oneoftheelementsofroutinecareincludestheuseofapartograph,whichifused

correctly,canalertustothestartofprolongedorobstructedlabour.Similarly,theprovision

ofactivemanagementofthethirdstageoflabour(AMTSL)canreducetheriskofpost-partum

haemorrhage.

Inadditiontothispackageforroutinecare,somewomenandbabiesrequirehigher-levelcare

for complications. Facilities that provide such emergency obstetric and neonatal care are

classifiedasBasicEmergencyObstetricCare(BEmOC)facilitiesorComprehensiveEmergency

Page16of248

ObstetricCare(CEmOC)facilitiesbasedontheprovisionofspecifiedsignalfunctions.22Signal

functions includes clinical capabilities like providing injectable antibiotics, magnesium

sulphate, oxytocics and procedures like assisted vaginal delivery, blood transfusion,

caesareanoperationsandothers.22

However, there are widespread examples in the literature which indicate that a high

proportionofbirthsoccurinfacilitiesthatarenotfullycapableofprovidingtheappropriate

signalfunctionsforobstetriccare.15,23,24,25Forexample,providingassistedvaginaldeliveries,

injectable oxytocics or blood transfusion services are challenging in many resource-

constrainedsettings.16,23

Existingroutinehealthinformationsystemsdonotcaptureinformationonspecificelements

ofcareduringnormallabourandchildbirthfromwomengivingbirthinLMIChospitals.These

individualleveldataonqualityofroutinecareareessentialforimprovementpurposes,but

areonlyavailablethroughdedicatedstudiesandhence,thereislimitedinformationonthis

topic.However,therearecurrentlyongoingeffortsatthegloballeveltodefinemetricsfor

qualityofcareat thetimeofbirth26andonelementsofskilledattendanceatbirth (SAB),

whichmakethisPhDrelevanttotheseongoingglobalefforts.

1.1.4:EvidenceonQoCformaternityservicesintheprivatesectorislimited

The private sector provides a range of health services including maternity care in LMIC

settingsandasIwillshowintheliteraturereviewsection(chapter2),evidenceonqualityof

healthservicesprovidedbytheprivatesector is limited.Theprivatesectorcanvaryfrom

smalltolargefor-profitcompanies,orprivatepracticesformedbyagroupofhealthworkers

orclinicsrunbynationalandinternationalnon-governmentalorganizationsandclinicsrunby

individualhealthworkersandpharmacies.27

Thereareargumentsforandagainsttheroleoftheprivatesectorinprovidingessentialhealth

services.Proponentsarguethattheprivatesectorisalreadyanestablishedproviderofhealth

servicesinmanysettingsandcanmakesignificantcontributionstoexpandefficientandhigh-

qualityhealthservicestounderservedpopulations.28,29Scepticsarguethatsincetheprivate

sectorprioritisesprofitsoverpublichealthimpacts,theyareunlikelytoprovidehighquality

servicesatlowcostsparticularlyinunderservedpopulations.30

Page17of248

Despite these ideological arguments, the size and themarket share of the private sector

acrossLMICsettingsappearstobeincreasing.31,32Although,thepublicsectorstillprovides

themajority of services globally, across the continuumof care, in terms of reproductive,

maternalandnewbornhealth,theprivatesector’scontributionissubstantialandestimates

indicate that 19% of maternity care, 32% of antenatal care, and 22% of family planning

servicesgloballyareprovidedintheprivatesector.33

Therapidgrowthoftheprivatesectorhasdrawnattentiontomanyproblemsthatitoften

shareswiththepublicsector,whichincludeslowstandardsofcare,poorinfrastructure,lack

ofqualifiedstaff,inadequateorpoorequipmentandmedicalmalpractice.31Inaddition,the

abilitytoregulatetheprivatesectorhasalsonotkeptpacewithitsgrowth.Somechallenges

forregulationhaveincludedlackofgovernmentinstitutionalcapacity,thelargesizeofthe

privatesector,lackofresourcesandoften-corruptrelationshipsbetweenstateandprivate

sectoractors.31,34

TheheterogeneityandcomplexityoftheprivatesectorinLMICssuchasIndiaalsomeansthat

high-qualityresearchevidenceonQoCintheprivatesectorislimited.Manypublishedstudies

havefocussedontheincreasingmedicalisationofchildbirthintheprivatesector,especially

givenhighratesofcaesareansectionsamongwomenseekingcareintheprivatesector.24,35-

39However,detailedevidenceonqualityof routinecare fornormalbirths inLMICprivate

sectorfacilitiesislimited.Therefore,furtherresearchtoinvestigatetheQoCfornormallabour

and childbirth in the private sector is important, especially in places like India, where

estimatesindicatethat22%ofalldeliveriesoccurintheprivatesector.40

1.1.5:Managementpracticesatmaternityfacilitiesisanunder-researchedarea

Strongmanagementcompetenciesarethoughttobeessentialtoensurethathealthsystems

canrespondtopopulationneeds.41While,thesecompetenciesareimportantinallsettings,

they seem particularly indispensable in LMIC settings, which are characterised by high

burdensofmaternalandneonatalmortalityandhospitalshereoperateinanenvironmentof

resource-scarcity.41Further,sincehospitalsarethemostexpensive,resource-intensiveand

politicallysensitivecomponentsofhealthsystems,managementpracticesathealthfacilities

seemparticularlyimportant.Inbothpublicandprivatesectorfacilities,goodmanagement

Page18of248

practices seem essential tomaintain effective hospital operations, hospital performance,

hospitaltargetsandtoensuregoodhuman-resourcemanagement.42-45

AsIwillexplainingreaterdetailintheliteraturereviewsection(chapter2)empiricalevidence

onmanagementpracticesatmaternityfacilitiesinLMICsettingsislimited.Perhapsbecause

management practices are hard to measure quantitatively and because methodological

advancesinmeasurementhavebeenrecent,mostresearchonthistopicoriginatesfromhigh-

income settings.42-45 Consequently, there is limited informationonmanagement practices

anditsrelationshipwithQoCatmaternityfacilitiesinLMICsettings.

Inhigh-incomesettings, large-scaledatacollectioneffortssuchastheWorldManagement

survey(http://worldmanagementsurvey.org/),whichcollectsdatafromover2,000hospitals

inninecountriesexist.Thesemanagementdatacanoftenbelinkedtoroutinelycollected

clinicaldataavailablefromelectronicmedicalrecordsofhospitalsinhigh-incomesettingsand

therelationshipbetweenmanagementandQoCexamined.However,suchdataisgenerally

notavailableinhospitalsinLMICsettings.

Given that management practices have the potential to influence all elements of the

maternitycarepathwayatfacilities,therelationshipbetweenmanagementandQoCneeds

detailedinvestigation.Examiningwhethermanagementpracticeshavethepotentialtodrive

gains in quality in LMIC settings is an innovative and interesting area of research with

significant evidence gaps. In addition, given recent methodological advances, a

comprehensiveassessmenttool,42,43,46 isavailablethatcanbeadaptedandusedtoassess

managementpracticesatmaternityfacilitiesinIndia.

Page19of248

1.2. Purposeofthethesis

ThecentralpurposeofmythesisistodevelopadetailedunderstandingofQoCduringlabour

andchildbirthat26publicandprivatesectormaternityfacilitiesinUttarPradesh,India.Uttar

Pradesh(UP)isapopulousstatewithlowratesofinstitutionaldeliveriesandfutureprogress

inthisIndianstatehasimportantimplicationsformaternalandnewbornsurvivalinIndia.

To fulfil my research aim, I conducted clinical practice observations and assessed QoC

providedduringlabourandchildbirthfor275mother-babypairsat26maternityfacilities.I

assessedanddescribedoverallqualityofcareatthetimeofbirthandspecificallyqualityof

obstetricandneonatalcareatthesematernityfacilities.IalsoinvestigatedwhetherQoCis

associated with characteristics of the women, characteristics of health workers and

characteristicsofmaternityfacilitiesinthreedistrictsofUttarPradesh,India.

Thereafter, I identified practices that constitute mistreatment of women, assessed and

describedthenature,patternsanddeterminantsofmistreatmentencounteredbywomen

during labour and childbirth at these maternity facilities. I also investigated whether

mistreatmentisassociatedwithsocio-demographiccharacteristicsofwomen,characteristics

ofhealthworkersandcharacteristicsofmaternityfacilities.

AnotherinnovativecomponentofmyPhDistheinvestigationintomanagementpracticesat

maternityfacilities,whichIassessedthroughaseparatesurveywithhealthfacilitymanagers

in Uttar Pradesh. I described existing management practices at maternity facilities and

examinedwhether there is a relationship between quality andmanagement practices at

maternityfacilities

InvestigatingQoCfornormallabourandchildbirthscomprehensivelyincludinganyobserved

mistreatmentisanimportantareaforresearchparticularlysincetherearemanyinformation

gaps related to quality, especially in private sector. Moreover, investigating whether

managementpracticeshavethepotentialtoinfluencequalityofcareisanunder-researched

area.Generatingevidenceontheseimportantquestionscouldsupportqualityimprovement

effortsinmaternalandnewbornhealthinlow-resourcesettings.

Page20of248

1.3. Outlineofthesis

This is a “research-paper-style” thesis with three prepared manuscripts presented as

chapters.Chaptertwopresentsareviewoftheliteraturerelevanttotheresearchquestions

addressed by my thesis. In this chapter, I provide an overview of why quality of care is

importantatmaternity facilities; introduceconceptsofessentialcareat thetimeofbirth;

maternity care pathways; skilled birth attendance and outline interventions that are

recommendedandnotrecommendedforprovisionofcareatthetimeofbirth.Ialsodiscuss

theconceptsofrespectfulmaternitycare.Thereafter,Idiscussconceptsofqualityofcare,

frameworks,definitionsandmeasurementofQoCinmaternalandnewbornhealthusingthe

frameworkofstructure,processandoutcomes.Thereafter,Ioutlinetheempiricalevidence

ondeficienciesinQoCatthetimeofbirthbasedonmyreviewoftheliteraturefromIndia

using thequalityof care framework. Finally, in the last sectionof the literature review, I

summarisethetheoreticalconceptsandempiricalevidenceonmanagementpracticesand

qualityofcare.

Inchapter3,Idescribethestudysetting,provideanoverviewofthehealthsystem,maternal

andnewbornhealthservicesprovidedatpublicsectorfacilitiesanddiscusstheevolutionof

qualityinmaternalhealthprogrammesinIndia.Ithendiscussmydoctoralresearchwithin

thelargerevaluationoftheMatrikaproject.

In chapter 4, I outline my role in conducting this doctoral research, funding and overall

timelineforthisresearch.Inchapter5,IpresentaconceptualframeworkformyPhD,and

discuss the aim, objectives and design of the studies described in this PhD. Thereafter, I

provideanoverviewofthemethodsusedinthedifferentresearchstudiespresentedinthis

thesis. However,detailedmethods foreachstudyarealsopresented in individual results

chapters(chapters6-8).

Chaptersixpresentsthefirstresearchpaperentitled“Qualityofessentialcareatthetimeof

birth:Findingsfromclinicalobservationsofspontaneouslabourandchildbirthat26public

andprivate sector facilities inUttarPradesh, India.” Addressingobjectiveone, thispaper

describestheoverallqualityofcare,andqualityforobstetricandneonatalcareduringnormal

labourandchildbirthat26maternity facilities inUttarPradesh.Thismanuscripthasbeen

Page21of248

peer-reviewedandaslightlyshortenedandeditedversionwaspublishedbytheBulletinof

theWHOinaspecialseriesonqualityofcare(publishedinJune2017).47

Chapter seven presents the second research paper entitled “An investigation into

mistreatment of women during labour and childbirth inmaternity care facilities in Uttar

Pradesh, India:amixedmethodsstudy”.Addressingobjective two, thispaper investigates

anddescribethenatureandpatternsofmistreatmentobservedduringlabourandchildbirth

atmaternityfacilities.IaimtosubmitthismanuscripttoReproductiveHealth.

Chaptereightpresentsthethirdresearchpaperentitled“Managementisnotassociatedwith

qualityofcareduringlabourandchildbirth:evidencefromcross-sectionalstudyofmaternity

facilities in Uttar Pradesh, India”. Addressing objective three, this paper describes

managementpracticesatmaternityfacilitiesinUttarPradesh,Indiaanddemonstratesthat

overallmanagementpracticesarenotassociatedwithQoCduring labourandchildbirth in

maternity facilities inUttarPradesh. IaimtosubmitthismanuscripttoHealthAffairsora

similarjournal.

Chapterninesynthesizesthemainfindingsfromthesepapers,discussesmyreflectionsonthe

different studies described in this thesis, and the strengths and limitations of individual

studies. I thendiscuss the implicationsofmyPhD findings for research, programmesand

policy,andproviderecommendations.InChapter10,IpresenttheconclusionsofmyPhD.

Page22of248

Chapter2:Literaturereview

2.1:IntroductionThischapteraimstoimproveourunderstandingoftheliteratureonqualityofcareatthetime

ofbirthandmanagementpracticesathospitals.Iconductedacomprehensivereviewofthe

literature on quality of care during labour and childbirth and management practices at

hospitalswithafocusonLMICsettings.Asmyresearchobjectivescoveredabroadrangeof

topicsandIwasinterestedinresearchfromarangeofdisciplinessuchaseconomics,hospital

management, and health care administration, itwas not possible undertake a systematic

literaturereview.However,Ihavetriedtoensurethatmyliteraturereviewiscomprehensive

andcoversallkeyaspectsofmanagementpracticesathealthfacilitiesandqualityofcarefor

maternalandnewbornhealth.

To examine the literature on QoC at the time of birth at maternity facilities in LMICs, I

searchedpublishedpapersand the latestWHO,UnitedNations (UN) resourcesandother

reportsexaminingQoCinhealthservicesandspecificallyQoCinmaternalandnewbornhealth

inLMICsettings.SearcheswerecarriedoutinMedlineandgooglescholarbycombiningthe

relevantfree-textandMedicalSubjectHeadings(MeSH)forterms,suchas‘qualityofcare,’

with those for the field of interest (‘maternal health,’ ‘safemotherhood,’ or ‘obstetrics;’

‘newborn’or‘neonatal;’or‘childbirth’or‘intrapartum’or‘intra-partum’or‘hospital’‘health

facility’ ‘maternity facility’; ‘postpartum’ or ‘post-partum’; ‘puerperal’ or ‘puerperium’ or

‘pregnancy’or‘delivery’).InMedline,Iappliedsearchlimitsandrestrictedresultstostudies

fromLMICsettings, involvinghumansubjects,articles inEnglish,publishedduring1980to

2016.Additionalarticlesandreportswereidentifiedthroughwebsearchesoforganisations

workinginmaternalandneonatalhealth,conferencesormeetingreports,andfromexperts

in the field. Additional references were also identified from the reference lists of peer-

reviewedjournalarticlesandpublishedreports.

Forthestudyonmanagementpracticesathospitals,thesearchstrategyinvolvedthreekey

free-text search terms:management, quality of care and hospital setting. Searches were

conductedusingthesefree-textsearchtermsinMedlineandgooglescholar.Inadditionto

thesekeyterms,MedicalSubjectHeadings(MeSH)termswereusedinMedline,whichwere

Page23of248

‘exploded’toincludeallMeSHsubheadings.Limitswereappliedtorestrictarticlespublished

inEnglish,involvinghumansubjects,between1980and2016.Inaddition,Ialsoconducted

extensive web searches to identify reports produced by management consulting firms,

workingpapersineconomics,hospitalmanagementproducedleadingexpertsinthefieldof

managementandeconomics.Additionalreferenceswerealsoextractedfromthereference

listsofrelevantpublishedmanuscripts,monographsandreports.

In thesubsequentsectionsofchapter2, Iwillpresent thekey findings frommy literature

review.

2.2:TheimportanceofqualityatmaternityfacilityinLMICsettingsDespite the impressive improvements inmaternal and child health during the era of the

MillenniumDevelopmentGoals,approximately5.6millionwomenandbabiesdiedin2015

duringpregnancy, labour,childbirthandtheneonatalperiod.5,48,49 Inordertoachievethe

newmortality targets setout in the SustainableDevelopmentGoals, thereneeds tobea

renewed emphasis in research, programmes and policies that aim to reduce preventable

maternaldeaths,neonataldeathsandstillbirths.50

Formaternalmortality,asindicatedintheintroduction,theglobaltargetfor2030isanMMR

oflessthan70/100,000livebirthswithdifferentsub-targetsforspecificcontexts.4Countries,

dependingontheirbaselinelevelsin2015,shouldeitherreducetheirMMRbyatleasttwo-

thirdsoftheirbaseline,nothaveanMMRgreaterthan140/100,000livebirthsby2030,or

achievereductionsininequalitiesinMMRatasubnationallevel.Thesesub-targetsrequirean

annualrateofreduction(ARR)ofmortalitygreaterthan5.5%inthecountrieswiththehighest

MMRs(MMR>420/100,000).4,50

Forneonataldeaths,theEveryNewbornActionPlansetanabsolutetargetof12orfewer

neonataldeathsper1000livebirthsineverycountryby2030.AnARRof4.3%willbeneeded

toachievetheglobalNMRtarget,butthisvariesconsiderablybetweencountries,with29

countriesneedingtoatleastdoubletheirARR.18,51Forstillbirths,theENAPsetanabsolute

targetof12orfewerstillbirthsper1000totalbirthsinallcountriesby2030.Toachievethis,

aglobalARRof4.2%isneededand56countrieswillneedtodoubletheirARR.52

Page24of248

Researchers have also demonstrated that it is essential to go beyond ensuring increased

coverageandutilisationofservicestoreducematernalandneonataldeaths.53Asnotedin

theintroduction,studiesfromIndia,MalawiandRwandahaveshownthatgreateraccessto

institutionaldeliverieswasnotassociatedwith reductions inneonatalmortality; a finding

theyattributetopoorqualityofcareathealthfacilities.10,12,13Inamulti-countrystudy,higher

than expected maternal mortality was also found in hospitals in high-mortality, LMIC

countries, despite the availability of essential medicines. This suggest gaps in clinical

managementortreatmentdelaysforhospitalisedwomenwhohadlife-threateningobstetric

complications(maternalnear-miss).53

Recentglobaltrendshavebeenencouragingasupto74%ofdeliveriesarenowconductedby

skilledbirthattendantsandupto63%ofdeliveriesoccurininstitutions.8Withthisincrease

ininstitutionalbirths,higherproportionsofavoidablematernalandperinatalmortalityand

morbidityhavealsomovedintohealthfacilities.53Inaddition,arisingproportionofmaternal

deathsarenowduetoindirectcauses(27.5%),whilethemajorityofmaternaldeaths(over

70%)stilloccurbecauseofcomplicationsthatrequirefacility-basedcare,suchaspost-partum

haemorrhage,hypertensivedisorders,sepsisandcomplicationsrelatedtoabortions.54

Similarly, up to 85% of neonatal deaths are due to complications of preterm birth, birth

asphyxia, intrapartum-relatedneonataldeathsandsevereneonatal infectionsthatrequire

facility-basedcare.55 In2015,estimatesalso indicatedthatoverhalfofthestillbirthsthat

occur,dosoafterthestartoflabouratmaternityfacilities.56

PoorQoCatthetimeofbirthalsocausessignificantphysicalandpsychologicalmorbidities

forwomenwithnegativeconsequencesforthehealthandsurvivalofinfantsandaffectsthe

futurefinancialsecurityoffamilies.57,58Therefore,improvingthequalityoffacility-basedcare

at the time of birth offers tremendous opportunities to reduce maternal and perinatal

deaths.23

2.3:Qualityofessentialcareatthetimeofbirth

2.3.1:Background

Although expanding coverage rather than quality has been the focus of maternal health

programmeshistoricallyinLMICsettings,expertshavearguedthateffortstoimproveQoCat

Page25of248

institutions have lagged behind efforts to increase demand for institutional maternity

services.7,15,59

Manyfacilitiesinhigh-burdenLMICsettingsareill-equippedtoprovideemergencyobstetric

careparticularlylowerlevelsfacilities.15Acomprehensiveassessmentof86healthfacilities

in sevendistricts inGhana found that thequalityof routineandemergencyobstetricand

newborncarewasgenerallypoorandtherewasalargequalitygapatfacilities.60Similarly,a

cross-sectionalstudyfromNigeriafoundthatonly40%ofprimaryhealthcarefacilitiescould

provideemergencyobstetriccareandthatmostEmOCsignalfunctionswerenotprovided

regularly.61

InIndia,cross-sectionalresearchevidencefrommultiplestateshasfoundthatmostmaternity

facilities have poor EmOC capability and are not able to provide all the basic signal

functions.62-64Theknowledge,skillsandcompetenceofSBAsprovidingmaternityservicesin

institutionswasalsofoundtobedeficientinastudyfromMadhyaPradeshstateinIndia.65

In addition, researchers have also highlighted systemic problems such as bed shortages,

inadequatesupplies,shortagesofskilledstaff,whichisnotconducivetotheprovisionofhigh-

qualityandrespectfulcareatthetimeofbirth.65-67Moreover,manyfacilitiesinLMICsoften

lackbasicrequirementssuchasregularelectricityandcleanwatersupply.15,61,68

Although, skilled birth attendants working within an enabling environment has been

promotedasanessentialstrategytoprovidehigh-qualityintrapartumcare69,manywomen

deliveringat facilities inLMICreportdoingsowithoutskilledbirthattendants. InSenegal,

datafrom2009–14indicatesthat28%ofbirthsinlower-levelfacilitiesand8%inhospitals

occurredwithout skilledbirth attendants.15 In India, studies inRajasthanhave found that

unqualified providers are frequently involved in maternity care provision in institutions,

includinginuptohalfofallobservedcaseswithsignificantdeficienciesinQoCatthetimeof

birth.70,71

Otherstudieshavealsofoundthatskilledbirthattendantsoftendonothavetherequired

skills72 and that numbers of SBAs deployed are frequently not enough which further

exacerbatespoorfacilityEmOCcapability.73Inastudyatninesub-SaharanAfricancountries,

researchersfoundthatskilledbirthattendants lackedadequateknowledgeandskillssince

theirtrainingcurriculadidnotincludetrainingsonmanualremovaloftheplacenta.74Some

Page26of248

governmentsalsodesignatecadresasskilledbirthattendants,despitethemlackingrequisite

midwiferycompetencies.75,76

AvailableresearchevidencefromfacilitiesinLMICshighlightsmanydeficienciesinessential

careatthetimeofbirth,suchasnon-adherencetorecommendedprotocolsforcare,71,77,78

mistreatmentofwomen79-81andearlydischargefromfacilitieswithoutadequatepostpartum

monitoring.82The2016Lancetmaternalhealthseriesarticulatedthesedeficienciesincareas

“TooLittleTooLate”whichreferstoabsent,delayedorinadequatecareandas“TooMuch

TooSoon”,referringtoover-medicalizationthatresultsinovertreatment.83

Thereasonsbehindpoorqualityofcareatfacilitiesaremulti-facetedandcouldarisedueto

many different reasons such as: lack ofmaterial resources, limited knowledge and skills,

inappropriateapplicationsoftechnology84,inabilityoforganizationstochange85,failureto

align health worker’s incentives and quality improvement efforts to improved health

outcomes.86Giventhemulti-facetednatureofqualityashighlightedabove,ensuringQoCat

thetimeofbirthhasprovedtobechallenging.

Thebulkoftheavailableresearchevidenceonqualityofessentialcareatthetimeofbirth-

mostlyfrompublicsectorLMICfacilities-highlightstheneedtocarefullyexamineexisting

deficiencies inQoC at the time of birth andwork towards improvingQoC in institutions.

ResearchevidenceshowsthatitispossibletoimproveQoCbutinordertodosoitisessential

todefine,measureandthendevelopappropriatestrategiesforqualityimprovement.87,88In

thenextsection,Iwillelaborateonconceptualisinganddefininghigh-qualitymaternitycare

atfacilities.

2.3.2:Conceptualisinganddefininghighqualitymaternitycarepathwaysatfacilities

Thereisconsensusthatinordertoreduceavoidablematernalandneonatalmortality,every

pregnantwomanandnewbornbabywillneedskilledcareatthetimeofbirthwithevidence-

based clinical and non-clinical interventions delivered in a compassionate and enabling

environmentwhich ensures that respect, dignity and equity of care aremaintained.89 In

figure1,Ihaveconceptualisedamaternitycarepathwaythatoutlinesthedifferentwaysin

whichapregnantwomancouldarriveatahospital,eitherattheonsetoflabourorfullyin

labour; her care pathway within the hospital until her discharge from the hospital after

Page27of248

childbirth.Thisis,insomeways,issimilartowhatothershavereferredtoasthe“gate-to-

gate”approach.90

Figure1:Schematicdiagramofmaternitycarepathwaysfordelivery

A pregnant woman may directly come to the hospital once labour begins or may be

transferredtotheexaminationorlabourroomsfromanotherplacewithinthehospitalsuch

astheoutpatientclinicortheemergencyroom.Uponarrival,thefirststepwillbedetermined

bywhether the labour has actually started.Anobstetric examination to assess change in

uterinecontractionsanduterinecervix (effacementanddilation)willhelptoestablishthe

stageoflabour.Dependinguponthestageoflabour,shemaybetransferredtodifferentareas

ofthehospitalasoutlineinthefigure1.

To implement this maternity care pathway, it is essential that other fundamental

requirementsforprovisionofhigh-qualityservicesareavailable.Forexample,teamsofskilled

andauxiliaryhealthworkersshouldbeavailableatthehospitalround-theclock.Staffshould

adheretorelevantclinicalprotocolsforobstetricandnewborncare.Infectionpreventionand

control measures should be implemented rigorously. Equipment must be accessible and

functional,andsubjecttochecksduringeverydutyshift.Drugsandconsumablesshouldbe

availableround-the-clock.Daily roundsshouldbeconductedbymanagers to identifygaps

andbottlenecks,andthesemustbecorrectedonanurgentbasis.Thetimetakenfromarrival

ofwomanatthehospitaltotheactualreceiptofservicesshouldbeminimisedtotacklethe

Page28of248

thirddelay.91Specialistback-upwithinthehospitalorreferraltoanotherhigherlevelfacility,

ifneeded,shouldalsobeapartofthematernitycarepathway.15

Itistheoreticallypossiblethatprovisionofsuchamaternitycarepathwayalongwithother

essentialrequirements(staff,equipment,drugs,electricity,waterandothers)andefficient

transferofwomenincaseofcomplications,couldleadtoprovisionofhigh-qualitymaternity

careathospitals.

2.3.3:Skilledbirthattendance

As noted earlier, an important strategy employed to prevent maternal and neonatal

mortality, has been to ensure that skilled birth attendants (SBA), working in enabling

environments,areabletoattendeverychildbirth.69SBAsaredefinedas“anaccreditedhealth

professional such as a midwife, doctor or nurse who has been educated and trained to

proficiency in theskillsneededtomanagenormal (uncomplicated)pregnancies,childbirth

and the immediate postnatal period, and in identification, management and referral of

complicationsinwomenandnewborns.”69

AlthoughSBAsarewelldefined,theenablingenvironmentislesswell-definedbutconsidered

toincludetheavailabilityofequipment,drugsandafunctionalreferralpathway.92Several

studieshaveshownacorrelationbetweenanincreasedproportionofbirthsattendedbySBA

anda reducedmaternalmortality ratio.93-96Modellingsuggests thatacritical thresholdof

40%ofpopulationcoverageofbirthsattendedbyaSBA isessential forany reductions in

maternalmortalityandstillbirths.97

ThecorecompetenciesidentifiedforSBAsincludetheabilitytocommunicateinacaringand

respectful manner and provide holistic “women-centred” care, with the appropriate

knowledge and skills to provide evidence-based obstetric and neonatal care in a timely

manner.80,81,98Unfortunately,researchevidenceindicatesthat,womeninmanysettingsdo

notreceiveappropriateinterpersonalcareandthatSBAsmayoftenhavelimitedskillsand

confidence.72,74,99-101Researchershavealsoreportedthatsomecountriesmayalsodesignate

cadresasskilledattendants,despitethemlackingtherequisitemidwiferyskills.75,76

The indicator- the percentage of births delivered by skilled attendant, assesses progress

towards “skilled attendance at birth”. This indicator was used for the Millennium

DevelopmentGoal(MDG)reports9andtheCountdownto2015report.102Thisindicatorhas

Page29of248

also been proposed as a core coverage monitoring indicator by the Ending Preventable

MaternalMortality (EPMM) initiative4, the Every Newborn Action Plan (ENAP)103 and the

SDGs6.Reportingofthisindicatoratthepopulationlevelreliesheavilyonnationalhousehold

surveys such as theMultiple Indicator Cluster Surveys (MICS)104 or the Demographic and

Health Surveys (DHS).105 In reality, most population-based surveys only measure births

attendedbyskilledattendantsratherthanthequalityofcaretheyprovideortheenvironment

inwhichtheseSBAswork.Thesequestionsonqualityandtheenablingenvironmentaremuch

hardertoanswerthroughpopulation-basedsurveysandalsovarydependingonthenational

context.

Further,therearemanyissueswiththeSBAindicatoratthecountrylevelsincethereislack

ofclarityintermsofwhichcadreisconsideredaskilledbirthattendantinaparticularcountry.

Forexample,manycountriesdonothaveaformalmidwiferycadreinsteadtheyhaveother

multipurpose workers such as auxiliary nurse midwives that do not undergo specialised

midwifery training. There is also a problem in terms of standardization of names and

responsibilities of different cadres, and task- shifting to less trained providers which

complicates measurement efforts.74 As a result, researchers have found that in many

countriestherearelargegapsbetweenthedefinedstandardsandcompetenciesofSBAand

their ability to manage normal labour and childbirth and other obstetric and neonatal

complications.100

Availability of adequate numbers of SBAs at the national and subnational levels is also

important.The2014updateoftheGlobalHealthWorkforcestatisticsindicatesthatamongst

132countriesforwhichdatawasavailable,64countriesdidnotmeettheminimumcritical

thresholdof23midwives,nurses,anddoctorsper10,000populationneededtoimplement

primaryhealthprogrammes includingmaternity care services.106 In addition, shortagesof

specialistssuchasobstetricians,anaesthetistsandneonatalnursesisalsofrequentinLMIC

settings.16,107

EvenwhenSBAsareavailable,theymaybepoorlydistributedwithinurbanandruralareasor

withinthepublicandprivatesectors.15,107,108Thisisparticularlychallenginginremoteand

ruralareaswherereasonssuchaspoor infrastructure, limitedcareeropportunities, family

reasonslikeschoolingforchildrenandothers,becomesachallengeforSBArecruitmentand

deployment.107,109Asaresultofthesefactors,womenmaynotbeabletoreceivetimelycare

Page30of248

andendupeitherdeliveringaloneorwithoutappropriatelyqualifiedorskilledattendants,

despitegoingtoinstitutionsformaternitycareservices.15

2.3.4:Interventionsrecommendedforcareatthetimeofbirth

In theirvision forqualityofcare, theWHOandother internationaldevelopmentpartners

envision a future where “Every mother and newborn receives quality care throughout

pregnancy,labour,childbirthandpostnatalperiod”.4,89,110 Recentincreasesininstitutional

birthsacrosstheworld,offerauniqueopportunitytorealisethisvision.However,toachieve

this vision, health workers must apply evidence-based interventions consistently while

providingcare.Adherencetobest-practiceguidelinesforessentialcareatthetimeofbirth,

togetherwitheffectiveimplementationstrategies,111,112havethepotentialtosupporthealth

workersinmakingcorrectdecisionsattherighttimeanduseeffectiveinterventionswhile

providingcare.83

Arecentsystematicreview83publishedasapartoftheLancet2016maternalhealthseries

reviewedallavailableclinicalpracticeguidelinesfortheprovisionofroutineintrapartumcare

and postnatal care and provided up-to date guidance on recommended interventions

identifiedusingarigorousreviewmethodology.83Researchersretained51guidelinesoutof

163 guidelines reviewed, fifteen of them focussed specifically on intrapartum care and

nineteencoveredpostnatalcare.MostoftheretainedguidelineswereissuedbytheWHO,

the International Federation of Gynaecology and Obstetrics (FIGO), and the national

obstetricsandgynaecologysocietiesoftheUSA,Canada,UK,andGermanyandtheremaining

werefromNon-GovernmentalOrganizations(NGOs)inlow-incomesettings.Unfortunately,

mostof the governmental guidelines from low-income countriesdidnotmeet their strict

criteria(researchersusedtheAGREE–IIinstrumentandonlyretainedguidelinesthatreceived

ascoreof6ormore).83Table1belowsummarisestheinterventionsrecommendedforuse

duringintrapartumandpostpartumperiods.

Page31of248

Table1:Interventionsrecommendedforuseduringintrapartumandpostpartumperiod

Recommendedinterventionsfortheintrapartumperiod1. Respectfulcare,communicationandbirthcompanions

• Offerwomenthepossibilityofbeingcaredforbyamidwife;provideone-to-onecontinuoussupportivecare

• Allowandencouragewomentohaveabirthcompanionoftheirchoice• Treateverywomanwithrespect,provideherwithall informationaboutwhatshemight

expect,askheraboutherexpectations,andinvolveherinthedecisionsabouthercare2. Assessmentsandmonitoringoflabourprogress,andmaternalandfoetalhealth• Performvaginalexaminationevery4hours• Routinelyassessthefrequencyofuterinecontractionsevery30min• Routinely assessmaternal pulse every hour,maternal blood pressure and temperature

every4h,andfrequentlyassesspassingofurine• Considerthepsychologicalandemotionalneedsofthewoman• Offerintermittentauscultationofthefoetalheartratetowomeninestablishedfirststage

oflabourinallbirthsettings(recommendationsincludefrequency,timing,andrecording)• Considerusingapartograph;usea4-houraction line tomonitor theprogressof labour

duringsecondstage• Document the presence or absence of substantial meconium-stained fluid when

membranesrupture(watersbreak)3. Painrelief• Assess the labouring woman's pain level and her desire for non-pharmacological and

pharmacologicalapproachestopainrelief• Encouragewomentoadoptanyuprightpositiontheyfindcomfortablethroughoutlabour• Advisewomenthatbreathingexercises, immersion inwater,andmassagemightreduce

painduringfirststageoflabour,andthatbreathingexercisesandmassagemightreducepainduringsecondstageoflabour

• Ensuretheavailabilityofopioids(e.g.,pethidine,diamorphine)inallbirthsettings;informwomenabouttheirside-effects;ifopioidsareusedforpainrelief,provideanti-emeticsincaseofnauseaorvomiting

• Ensuretheavailabilityofnitrousoxide(1:1mixturewithoxygen)forpainreliefinallbirthsettings;informwomenaboutitsside-effects

• Inobstetricunits,ensuretheavailabilityofregionalanalgesia;informwomenaboutrisksand benefits and potential implications of epidural analgesia during labour; provideregionalanalgesiaforwomenwhorequestit(includingrecommendationsfordrugs,dosing,maintenance, co-interventions, and precautions); ensure intravenous access beforeinitiationofanalgesia

4. Careduringfirst-stageandsecond-stagelabour• Routinehygienemeasurestakenbystaffcaringforwomenin labour, includingstandard

hand hygiene and single-use sterile gloves are recommended to reduce cross-contaminationbetweenwomen,babies,andhealth-careprofessionals

• Allowandencouragewomentodrinkwater,juiceorisotonicdrinks,andeatlightmealsorsnacksduringlabour

• Encourageandhelpwomentomoveandadoptanypositiontheyfindmostcomfortablethroughoutlabourandchildbirth,exceptsupineorsemi-supine

• Informwomenthatinthesecondstagetheyshouldbeguidedbytheirownurgetopush5. Careduringthird-stageandfourth-stagelabour• Informwomenthatactivemanagementofthirdstagepreventspost-partumhaemorrhage

Page32of248

• Oxytocin(10IU,intravenouslyorintramuscular)istherecommendeddrugforpreventionofpost-partumhaemorrhage

• Ergometrineor600μgoforalmisoprostolcanbeusedasanalternativeifoxytocinisnotavailable

• Delayed cord clamping (done 1–3min after birth) is recommended for all births whileinitiatingessentialnewborncare

• Early cord clamping (<1 min after birth) is not recommended unless the neonate isasphyxiatedandneedstobemovedimmediatelyforresuscitation

• Controlledcordtractionandpalpationshouldbeusedaftercordclampinginsettingswithskilledbirthattendants.

• Encouragewomentohaveskin-to-skincontactwiththeirbabiesassoonaspossibleafterbirth

• Avoidmother–babyseparationbeforethefirsthourfollowingbirth,unlessatthemother'srequest; delay postnatal routine procedures (e.g. weighing, bathing, and measuring);monitortheneonate'sconditionduringskin-to-skincontact.

• Encourageandsupportbreastfeedinginitiationwithinfirsthour.RecommendedinterventionsforthePostnatalperiod

6. Woman-centredrespectfulmaternitycare• Provideindividualised,culturallyandcontextuallyappropriatecare,responsivetochanging

needs,andbasedonindividualcareplan7. Duringpostnatalfacilitystay• Followinganuncomplicatedvaginaldelivery,womenareadvisedtostayatleast24hinthe

facility• Evaluatepost-partumbleeding,maternalbloodpressure,anddocumenturinevoid• Evaluate perineal healing and look for signs of infection to identify and treat puerperal

infectionorsepsis(referwhennecessary)• Providepainrelief• Askwomenaboutheadache,assessbowelmovements,andpromoteearlymobilisationto

preventthrombosis• Facilitate rooming-in (mother and baby should stay in the same room 24 h a day) and

promoteparentparticipationineducationalactivitiesrelatedtonewbornbabies'health• Anti-D immunoglobulin should be offered within 72 h to every non-sensitised Rh-D-

negativewomanfollowingmiscarriageorbirthofapositivebaby• Evaluaterubellaimmunisationandofferimmunisation8. Atdischargefromhealthfacility• Attimeofdischargefromhealthfacility,provide informationaboutdangersignsforthe

motherandbaby,andcounselwomenonadequatenutrition,hygiene,handwashing,andsafesex

• Provideironandfolicacidsupplementsfor3months• Promoteexcusivebreastfeedingfrombirthuntil6monthsofage;observebreastfeeding

techniquebeforehospitaldischarge• Inmalariaendemicareas,advisemothertosleeptogetherwiththebabyunderinsecticide-

impregnatedbednets.9. Organisationandcontentofpostnatalcareafterdischarge• Recommendtwotothreepost-partumvisitsafterfacilitydischarge• Ateachpost-partumvisit,provideinformationaboutdangersignsforthemotherandbaby,

andcounselwomenonadequatenutrition,hygiene,handwashing,andsafesex• Askaboutdyspareuniaandresumptionofsexualintercourse,andrecommendpelvicfloor

exercises• Assessmentalhealthandwellbeingorpost-partumdepressionusingscreeningquestions

Page33of248

• Exploresocialsupportandassessforsignsofdomesticabuse• Promote excusive breastfeeding from birth until 6 months of age; mothers should be

counselledandprovidedwithsupportforexclusivebreastfeedingateachpostnatalcontact• Inmalariaendemicareas,advisemothertocontinuetosleeptogetherwiththebabyunder

insecticide-impregnatedbednetsSource:Millerat.al(2016)83

2.3.5:Interventionsnotrecommendedforuseduringthetimeofbirth

Havingidentifiedinterventionsrecommendedfortheprovisionofroutineintrapartumand

postpartumcare,Table2belowoutlinestheinterventionsthatdonothaverecommendations

for use, but still continue to be used frequently during provision of intrapartum and

postpartumcare,particularlyinLMICsettings.83,113-116Lackofup-to-dateknowledge,attrition

of skills, low levels of motivation, restrictive institutional policies and health system

bottlenecks canperpetuate theuseof these interventions thatarenot recommended for

providing care during labour and childbirth. 16,117-120 As shown in Table 2,many of these

interventions such as routine use of enemas, prophylactic insertion of intravenous fluids,

administration of oxytocics before delivery, routine episiotomy and others, do not have

evidence of effectiveness.114 Adoption of these ineffective practices into routine care is

harmfulespecially inLMICsettingswithweakhealthsystems,whereservicequality isnot

routinelymonitoredandwherewomenmaynotregularlycometofacilities.

Table2:Interventionsforintrapartumandpostpartumcarethatdonothaverecommendationsfor

use

1. Duringlabourandtheintrapartumperiod• Donotcarryoutaspeculumexaminationifmembraneshavecertainlyruptured.• Donotperformcardiotocographyonadmissionforlow-riskwomeninsuspectedorestablished

labourinanybirthsettingaspartoftheinitialassessment.• Donotperformroutinefetalpulseoximetry.• Donotmakeanydecisionaboutawoman'scare in labouronthebasisof cardiotocography

findingsalone.• Restrictionoffoodandfluidsduringlabour.• Routineintravenousinfusioninlabour.• Repeatedorfrequentvaginalexaminations,especiallybymorethanonecaregiverForpainrelief:• Donotoffertranscutaneouselectricalnervestimulationtowomeninestablishedlabour• Donotofferlidocainespraytoreducepaininthesecondstageoflabour• Donotofferoradvisearomatherapy,yoga,acupressure,acupuncture,orhypnosis,orwater

papulesforpainrelief• DonotoffereitherH2-receptorantagonistsorantacidsroutinelytolow-riskwomen2. Recommendedagainstinterventionsforcareduringfirstandsecondstageoflabour.

Page34of248

• Donotofferoradviseclinicalinterventioniflabourisprogressingnormallyandthewomanandbabyarewell(includingamniotomyandoxytocinaugmentation,eveninwomenwithepiduralanalgesia).

• Discouragethewomanfromlyingsupineorsemi-supineinthesecondstageoflabour.• Donotperformroutineperinealshavingorenemas.• Donotperformperinealmassageinthesecondstageoflabour.• Donotcarryoutaroutineepisiotomyduringspontaneousvaginalbirth.• DonotperformKristellermaneuver.• Administrationofoxytocinatanytimebeforedeliveryinsuchawaythattheeffectcannotbe

controlled.• Sustained,directedbearingdowneffortsduringthesecondstageoflabour.• Massagingandstretchingtheperineumduringthesecondstageoflabour.• Donotperformfundalpressureduringlabour.• Asaroutine,donotmovelaboringwomantoadifferentroomatonsetofsecondstage.• Donotencouragewomantopushwhenfulldilationornearlyfulldilationofcervixhasbeen

diagnosed,beforewomanfeelsurgetobeardown3. Recommendedagainstinterventionsforcareduringthepostnatalperiod• Palpationormeasurementofuterusinabsenceofabnormalbleedingisnotrecommended.• Donotperformmanualexplorationoftheuterusafterdelivery.• Donotperformlavageoftheuterusafterdelivery.• Do not use antibiotics routinely in low-riskwomenwith a vaginal delivery for endometritis

prophylaxis• Aspirinforthromboprophylaxisisnotrecommended.• VitaminAsupplementationforthepreventionofmaternalandinfantmorbidityandmortality

isnotrecommendedSources:WHO1999114,Easonetal.2000115,Nielson1998121;LudkaandRoberts1993116,Milleretal.2016.83

2.3.6:Theimportanceofrespectfulmaternitycareduringlabourandchildbirth

Inrecentyears,researchersandorganizationssuchasFIGO,WHOandothershavehighlighted

theimportanceofprovidinghighqualitycareduringlabourandchildbirthbyusingevidence-

basedinterventionsanddeliveringtheminahumaneanddignifiedmannerwithrespectfor

women’s human rights.80,81,122,123 The Lancet 2014 midwifery series also identified that

womenvalue relevant, timely informationand support, so that theyareable tomaintain

dignityandcontrolduringthebirthingprocess.124

There is often a complex interplay of factors such as those at the individual level (socio-

economicstatus,educational levels,caste), institutional levels (policiesoncompanionship,

clinical guidelines, lack of resources for example: inadequate privacy screens) and at the

healthworkerlevel(deficienciesinknowledge,skillsandincreasedworkloads)thatmayresult

inmistreatmentofwomenatmaternityfacilities.79,80ArecentWHOstatement(2014)on

preventingandeliminatingdisrespectandabusehighlightedtheurgencyofaddressingthis

Page35of248

issueand recommended that thereneeds tobeagreater researchandaction inorder to

improverespectfulcareduringlabourandchildbirth.123

This growing importanceofensuring respectfulmaternity carehasevolved from research

evidencethathaspreviouslyconceptualisedthisissueasdisrespectandabuse125,obstetric

violence126anddehumanisedcare.127There isnowincreasingresearchevidence indicating

that that this phenomenon occurs in both high80,128-131 and low income settings132-134,

indicatingthatthisisauniversalissue,andnotjustlimitedtoresource-constrainedsettings.

Women’s experiences of maternity care are negatively influenced by factors such as

unhygienic conditions at facilities, any disrespect and abuse they encounter, limited

informationorexplanationsprovidedpriortoconductinginvasiveprocedures,discrimination

and inequitable care provision.81,135 Low cost of treatment, convenience, kindness,

interpersonalqualitiesofthehealthworker,attentionandtimereceived,technicalcapability,

communicationandintegrityareallvaluedqualitiesbywomenwhenitcomestochoosing

healthworkers.136-140

Inalandscapeanalysisfrom2010,BowserandHilldescribedsevencategoriesofdisrespectful

and abusive care during childbirth: physical abuse, non-consented clinical care, non-

confidentialcare,non-dignifiedcare,discrimination,abandonment,anddetentioninhealth

facilities.125Sincethen,researchershaveadvancedthisconceptandproposedadefinitionto

articulate the criteria for determining when an interaction with a health worker or

circumstancesatmaternityfacilitiesthatshouldbeconsideredabusiveanddisrespectful.81,99

Freedmanetal.(2014)proposedthatacomprehensivedefinitionofmistreatmentneedsto

capture the health, human rights and socio-cultural dimensions of mistreatment; while

measurementeffortsneedtocapturewhere,howandwhymistreatmentoccurs.81Further,

measurementeffortsshouldalsobeabletocapturewhethermistreatmentwasintentional

ornot,andtheroleoflocalsocietalnorms(forexample,women’sstatus,patient-provider

dynamics) that influences women’s perceptions of mistreatment in different contexts.81

Giventhisbackground,Freedmanetal.(2014)defineddisrespectandabuseduringchildbirth

as “interactions or facility conditions that local consensus deem to be humiliating or

undignified,andthoseinteractionsorconditionsthatareexperiencedasorintendedtobe

humiliatingorundignified.”99

Page36of248

In2015,theWHOconductedasystematicreviewandtriedtoestablishtheevidence-basefor

mistreatmentglobally.80Theauthorsreviewed65studies(53qualitativeand12quantitative)

from34 countries and found thatmost studiesuseddifferentoperational definitions and

measurementapproaches.80Amongstthequantitativestudies,onlythreestudiesreporteda

prevalenceofmistreatmentatmaternityfacilities,whichvariedfrom15to98%.80Thisreview

also proposed a typology of items consideredmistreatment and identified the following:

physical, verbal or sexual abuse, stigma and discrimination, failure to meet professional

standardsofcare,poorrapportbetweenwomenandprovidersandhealthsystemconditions

andconstraints.80

However,mistreatmentandpoorqualityofclinicalcarearecloselyinterlinked.83Asthe2016

Lancet maternal health series noted, there are two extremes of maternal health care

provisioninagrowingnumberofLMICs.83Thefirstextremeisassociatedwithover-treatment

ortheroutineover-medicalisationofnormallabourandbirths,whichtheyreferredas“Too

MuchTooSoon”. The secondextreme isunder-treatmentorunderuseof evidence-based

practicessignifiedbytheterminology“TooLittle,TooLate”whichistheunderlyingcauseof

high maternal mortality and considerable morbidity.83 Both over-medicalisation such as

increased use of unnecessary procedures like episiotomies without indications or under-

treatmentsuchasabsenthygienicstandardsatmaternityfacilitiesarealsoagainsttherights

ofchildbearingwomen.141

Therefore, mistreatment of women during labour and childbirth can occur because of

inappropriate care practices, which may include those related to disrespect and abuse

(intentional harmor degradation), over-treatment, or under-treatment. Regardless of the

terminologyused,mistreatmentofwomenfallsunderpoorqualityofcare.Aswomenwho

aremistreatedare less likelytocomebacktofacilitiesforfuturepregnancies,142this isan

importantissuethatneedstobeaddressedurgently.

Inthenextsection,Iwillpresentfindingsfrommyliteraturereviewonqualityofcareinhealth

systemsandqualityasitrelatestomaternalandnewbornhealth.Thereafter,Iwilloutline

approachestomeasurevariouselementsofQoCinmaternalandnewbornhealth.QoCfor

healthsystems includesbroader issuesthanQoC inhealth facilitiesandtheybothrequire

different interventions for improvement. However, since both of these issues are closely

interlinked,itisusefultounderstandthemeaningoftheseconceptsandunderstandtheways

Page37of248

thattheyoverlaporinteract.Inthesubsequentsection,Iwillthensummarisefindingsfrom

myliteraturereviewontheempiricalevidenceonQoCduringlabourandchildbirthinIndia

usingthequalityofcareframework.

2.4:Frameworksofqualityinhealthanddefinitions

Over the past two decades, numerous frameworks on quality have been conceptualised

basedondifferingnotionsofquality.Sincequalityismulti-dimensional,itiswidelyaccepted

thatthereisnosingleconceptorframeworkofqualityofcare.

Previous frameworks of quality of care for health services have included the perspective

model143focussedonclientandprovider’sperceptionsofquality,thecharacteristicmodel144

whichfocussedonspecificcareelements(safety,efficacy,timeliness,patientcenteredness

etc.)andthesystemsmodels145whichconsideredqualityasaby-productofgoodstructures

andprocessesresultingtogoodoutcomes.

2.4.1:Definitionsofqualityofcareinhealthservices

Historically,manydefinitionshaveexistedforQoCinhealthservices.Earlierdefinitionsseem

to have favoured biomedical outcomes alone. For example, Donabedian (1980) defined

qualityas“theapplicationofmedicalscienceandtechnologyinamannerthatmaximisesits

benefit to health without correspondingly increasing the risk. The degree of quality is,

therefore,theextenttowhichthecareprovidedisexpectedtoachievethemostfavourable

balanceofrisksandbenefits.”146

In1988,RoemerandMontoya-Aguilarmadeadistinctionbetweenassessmentofqualityand

assuranceofqualitybasedonpre-definedsetofstandards.147Theywrote,“Qualityofhealth

careconsistsoftheproperperformance(accordingtostandards)of interventionsthatare

knowntobesafe,thatareaffordabletothesocietyinquestion,andthathavetheabilityto

produceanimpactonmortality,morbidity,disability,andmalnutrition.”147

Anotherdefinitionwhichstressedondecreasingthegapbetweendesiredandactualhealth

outcomeswastheInstituteofMedicinedefinition(1990)whichdefinedqualityofcareas,

“thedegreetowhichhealthservicesforindividualsandpopulationsincreasethelikelihood

ofdesiredoutcomesandareconsistentwithcurrentprofessionalknowledge.”148

Page38of248

Otherdefinitionshavebeenmorecomprehensive.WilsonandGoldsmith(1998)described

QoC as “the sum of four components: technical quality, resource consumption, patient

satisfactionandvalues”.149Perhaps,thesimplestandmostwell-accepteddefinitionofquality

isGodlee’s(2009),whodefinedqualityas“clinicaleffectiveness,safety,andgoodexperience

forthepatient.”150

2.4.2:Elementsofqualityofcareinhealthservices

Elements of quality of care in health services are generally assessed using Donabedian’s

classicframeworkofstructure,processandoutcomes.146,151AsexplainedbyPeabodyetal.

(2006) in describing elements of quality for health systems, “structure refers to physical

infrastructure,supplies,commodities,resources,financingofhealthservicesandothers.”86

Process refer to “healthworker and client interactionswhich occur during consultations,

examinationorprocedures.” 86Outcomes refer to indicators “thatmeasurehealth status,

mortalityanddisabilityadjustedlifeyearsofthepopulation”.86

Morerecently,theInstituteofMedicine’sreport“CrossingtheQualitychasm:anewhealth

systemforthe21stcentury”broadenedtheconceptofqualitybyexpandingoncontextual

elements of quality to illustrate how improved processes can actually lead to improved

quality152.Accordingtotheirframework,qualityofhealthcaremeansprovisionofservices

thatare:

1. Effective: delivering health care that is adherent to an evidence-base and results in

improvedhealthoutcomesforindividualsandcommunities,basedonneed;

2. Efficient:deliveringhealthcare inamannerwhichmaximizes resourceuseandavoids

waste,deliveringhealthcarethatistimely,geographicallyreasonable,andprovidedina

settingwhereskillsandresourcesareappropriatetomedicalneed;

3. Acceptable/patient-centred:deliveringhealthcarewhichconsidersthepreferencesand

aspirationsofindividualserviceusersandtheculturesoftheircommunities;

4. Equitable: delivering health care which does not vary in quality because of personal

characteristicssuchasgender, race,ethnicity,geographical location,orsocioeconomic

statusand

5. Safe:Deliveringhealthcare,whichminimizesrisksandharmtoserviceusers.

Page39of248

2.5:FrameworksanddefinitionsofQoCspecifictomaternalandnewbornhealth

Similartoframeworksonqualityofhealthservices,therehavealsobeenmanyframeworks

toassessqualityinmaternalandnewbornhealth.153,154Someoftheseincludesframeworks

thathaveassessedQoC fromclientsperspectives,138 rights–basedapproaches,155provider

needs156andmodelstoovercomedelays.157

Perhaps, themostwidely used frameworkwas developed byHultonet al.153 (2000)who

adaptedtheIOMdefinitionofqualitywhileincorporatingtheconceptsofeffectiveandtimely

accessandofreproductiverights.153Theydefinequalityofmaternalhealthas“thedegreeto

whichmaternal health services for individuals and populations increase the likelihood of

timelyandappropriate treatment for thepurposeofachievingdesiredoutcomes thatare

both consistent with current professional knowledge and uphold basic reproductive

rights.”153

More recently, in 2015, theWHOpublished its vision forQoC formaternal andnewborn

health89.TheWHOvisionwasinformedbyHulton’sframework153andtheIOMdefinition,and

defines QoC as ‘the extent to which health services provided to individuals and patient

populationsimprovedesiredhealthoutcomes.Inordertoachievethis,healthcareneedsto

besafe,effective,timely,efficient,equitable,andpeople-centred.’89

SimilartotheHultonframework153,theWHOframeworkforQoCinmaternalandnewborn

health (Figure2)alsoconceptualisesqualityasbothprovisionofevidence-basedcareand

positiveexperiencesforwomenseekingcare.89Satisfactionofwomenwithmaternitycareis

closelylinkedtowomen’sexperiencesofcareintheQoCframeworkandisassociatedwith

allelementsofstructure,processandoutcomes.Forexample,alackofadequatesuppliesor

skilledpersonalleadstopoorsatisfactionamongstwomen.Similarlyprocessofcareelements

suchasgoodinterpersonalbehaviours,emotionalsupport,andtreatmentwithrespectand

dignityareessentialtoensuresatisfactionwithmaternitycare.89,158Outcomeindicatorsfor

qualityincludethoserelatedtowomen’ssatisfactionandlabourandchildbirthoutcomes.89

However,thereisacomplexrelationshipbetweensatisfactionandQoC.159Itispossiblethat

care receivedbypatients is of high technical quality but inadequate in termsof patient’s

satisfaction.Inaddition,women’ssatisfactionisalsoassociated,atleast,inpart,withlabour

andchildbirthoutcomes.Researchsuggeststhatwomenwhoaredissatisfiedwithmaternity

Page40of248

servicesalso tendtohavepoorpregnancyoutcomessuchaspoorpostnatalpsychological

outcomes,apreferenceforcaesareansections,negativefeelingstowardstheneonateand

problemswithbreastfeeding.160-162

TheWHOframeworkrecognisesthatqualityisanormativeconcept,therefore,standardsfor

care are needed for assessment and improvement purposes.89 WHO guidelines for both

routineandemergencyobstetricandnewborncarearewell-defined.163TheWHOframework

alsorecognisestheimportanceofrobustinformationsystemstocapturedataonQoC,and

theneedforeffectivereferralsystemsincaseofemergencies.89

TheQoCframeworkis linkedtothesixWHOhealthsystembuildingblocks164of1)service

delivery; 2) health workforce; 3) information systems; 4) medical products, vaccines and

technologies;5)healthfinancingand;6)leadership/governance.Ittherebycreateslinkages

sothatanalyticalworkandimprovementprojectstoimproveQoCcanbetakenusingahealth

systemsapproach.Finally,theframeworkrecognisesthathealthsystemsareplatformsthat

enableaccess tohighQoCandallowsprocesses tooccur, along two importantand inter-

linkeddimensionsofserviceprovisionandexperienceofcareleadingtoimprovedindividual

andfacility-leveloutcomes.89

Figure2:WHOQualityofCareFrameworkformaternalandnewbornhealth

Page41of248

2.6:MeasurementofQoCformaternalandnewbornhealthinLMICsettings.

GiventherecentemphasisonemphasisonaccountabilityinMNH,165nationalgovernments

anddevelopmentpartnersneedtogenerateorhaveaccesstohigh-quality,representative

dataonQoCtoinformtheirpolicyandprogrammedecisions.166However,measurementof

QoCisoftendifficultgiventhewide-rangingissuesthatqualityencompassesandweaknesses

ininformationsystemsinmanyLMICsettings.

Donabedian’sapproachtomeasuringqualitybyassessingelementsofstructure,processand

outcomesisalsowidelyusedformeasuringQoCinmaternalandnewbornhealth.167These

elementsaredescribedinthesectionsbelowwithaspecificfocusonmeasurementofquality

ofMNHservices.

2.6.1:Measuringstructureelementsofqualityofcare

Dataonstructuralelementsofqualityareperhapstheeasiesttoobtain.Routinemonitoring

dataiscollectedbynationalhealthsystemsormonitoringsystemsofimplementingagencies

andareoftenanobvioussourceofinformationonstructuralelementsofQoC.Forexample,

facility inventoriesofdrugsandsuppliesareoftenavailablethroughlogisticsmanagement

information systems. Serviceutilisationdataon indicators suchasnumberof institutional

births,deliveriesbyskilledbirthattendantsandothersisavailablethroughthenationalhealth

informationsystems.

Potential advantagesof routinedata for structuralmeasures include theiravailabilityata

relativelylow-cost,onacontinuousbasis,dataareoftendisaggregateduptothefacilityor

district level. In addition, routine data provide more detailed information on service

availability and utilisation compared to household surveys.166 However, there are also

limitations with using routine data, for example: many elements of MNH care are not

collected through routine systems,denominators are limited to those in contactwith the

healthsystem,datamayalsobeofpoorquality,incompleteorupdatedinfrequently.166

Giventheseissues,dataonstructuralelementsofQoCtendstobecollectedthroughspecial

surveys or censuses. Readiness which often refers to the availability of necessary drugs,

commoditiesandtrainedhealthworkers,isoftenusedasaproxyforstructuralquality.Some

large-scale facility-based surveys regularly measure structural elements. Some of these

include: the World Health Organization’s Service Availability and Readiness Assessment

Page42of248

(SARA)168, the DHS Program’s Service Provision Assessment (SPA) surveys,169 and the

MEASUREEvaluation’sRapidHealthFacilityAssessments (RHFA).170Thesesurveyscapture

information on training, supervision, availability of services, tools and guidelines,

infrastructure conditions, availability and storage conditions ofmedications, supplies and

equipment.60,171However,noneofthesemethodsassesscompetencyofhealthworkers.166

SpecifictoMNH,materialssuchasEmOCneedsassessmenttoolkit22,UnitedStatesAgency

for International Development’s (USAID)- Maternal and Child Health Integrated Project’s

(MCHIP)QoCsurveys172haveseparatemodulesonfacilityinventoryassessmentthatcapture

information on infrastructure, availability and conditions of commodities, supplies, and

equipmentrequiredforprovisionofmaternityservices.

However, structural improvements by themselvesmay not improve health outcomes. 151

Therefore,inmaternalandnewbornhealth,measurementofinputsalone,suchasreadiness,

either of facilities (throughmeasurement of signal functions) or of the provider (through

measurement of knowledge and skills) does not provide a comprehensive picture and

therefore,measurementofprocessofcareisimportant.

2.6.2:Measuringprocesselementsofqualityofcare

Theoretically, processes of care can be measured during every health care encounter.

However,insomecases,theprivatenatureofhealthworker-clientinteraction,absenceof

appropriatemeasurementscalesorinstrumentslimitsmeasurementefforts.173Overthepast

decade, therehavebeenmanymethodological advances inmeasurementofprocessesof

care for MNH. There is also robust research evidence, which suggests that measuring

processes of care, as a part of quality improvement efforts can lead to improved health

outcomes.88,111,174ThismakesprocessmeasurementapreferredapproachtoassessQoCfor

maternalandnewbornhealth.

Below Idiscussnineapproaches tomeasureprocessesof care formaternalandnewborn

health such as standardizedpatients, clinical vignettes, reviewofmedical records, audits,

simulations or clinical skills and drills, direct clinical observations, video filming and

satisfactionsurveys.Allmethodshavetheirownadvantagesanddisadvantages.86,175

Standardized patients are a popular method to assess processes of care and have been

employedbyanumberofstudiesinAsiaandAfricatomeasureQoCforchildhoodillnesses

Page43of248

such as diarrhoea, acute respiratory infections,176 and sexually transmitted infections177.

Standardized patients are trained actors, often from local communities, who make

unannouncedvisitstoahospitalandpresentsymptomsofasimulatedcondition.178These

patientscompleteanassessmentchecklistonprovidersclinicalactionsafterthevisit.175Since

this methodology employs cases that are standardised and predetermined, it allows for

qualitycomparisonsacrossdifferenttypesofprovidersandcontexts.179

Someproponentsofstandardizedpatientsarguethatthatsincehealthworkersdonotknow

the true identity of standardized patients, their behaviours approximates that of “real-

patients”andhence,healthworkersarelesspronetoHawthorneeffect.180Hawthorneeffect

isaphenomenonwherebyhealthworkersbecomeawarethattheyarebeingobserved,and

thereafter,exertadditionaleffortwhich isachange intheiractualbehaviour.181However,

predictinghealthworker’sbehavioursinreallifeiscomplex.Forexample,healthworkersmay

providebettercaretosomeonetheyknowpersonallyorprovidediscriminatorycaretoother

patients.Moreover,thesesimulatedpatientsarenotsuitableforassessingQoCforinvasive

proceduresorconditionslikechildbirththatcannotbesimulatedbyactors.

Clinicalvignettesweredevelopedformeasuringqualitywithinagroupofproviders86,179,182

and theyhavebeenused to studyQoC fora rangeof conditions, including formeasuring

EmOC capability65 and intrapartum decision-making of midwives.183 Vignettes can be

administeredeitheronpaper,bycomputer,orovertheInternet.86Whenclinicalvignettes

areusedtoassessmanyproviders,eachproviderisgiventhesamecaseorthesamesetof

cases.86Healthworkersfollowthatparticularwrittenclinical-case,respondtoquestionsthat

replicatecertaincomponentsofapatient’svisit, forexample-history-taking,examination,

orderingofinvestigationsorprescribingatreatmentplan.86Thequestionsareopen-ended

and include interactive responses that simulate a patient’s visit and evaluate the health

workersknowledge.Healthworkers’performanceisassessedagainstacriteriaformanaging

theparticularcondition.184

Vignetteshave severaladvantages, suchasallowingcomparisonbetweenhealthworkers,

andcomparisonbeforeandafterimplementationofanewpolicy.86Theyarealsocheap,easy

toadministerandeasytoanalysewhichmakesthemuseful.86However,researchershave

argued that health worker’s behaviours during an actual consultation is not accurately

captured by vignettes, and that knowledge does not always translate into actual clinical

Page44of248

practice.185Therefore,althoughvignettesareausefulqualityassessmenttool,theyareoften

incompletewhenusedin isolation,andshouldbeusedwithothermethodssuchasdirect

observationsofclinicalpractice.185186

Recordreviewsareoneofthemostfrequentlyusedmethodstoevaluateclinicalqualitysuch

asforemergencycaesareansections.187Theirmainadvantagesarethatmedicalrecordsare

availableaftereveryhealthcareencounterand theyareeasilyobtained. However,often

whenmedicalrecordsarehandwritten,theymaynotbelegibleormayhavebeenwrittenfor

otherpurposes likeobtainingpayments,ormedico-legalreasonsratherthantodocument

details of procedures.86 Their utility is perhaps greater in high-income settings where

electronicmedicalrecordsareroutinelyused.Incontrast,suchsystemsdonotexistinmost

low-resourcesettingsandthereisofteninconsistency1andpoorclinicaldocumentationfor

indicatorsof interestsuchaspartographuse, timingofoxytocin,orbloodtransfusionand

others.

Auditssuchasnear-missaudits,maternalandperinataldeathreviewshavealsobeenused

extensivelytoidentifyandaddressdeficienciesinprocessesofMNHcare.90,188-193Auditshave

beendefinedas:‘thesystematicandcriticalanalysisofthequalityofmedicalcare,including

the procedures used for diagnosis and treatment, the use of resources and the resulting

outcome and quality of life for the patient.194 193 Audits often combine information from

different sources, whichmakes them superior to othermethods such as record reviews.

However, it is important toensure that thepurposeofconducting theauditasa learning

exercise aimed to improve clinical practices is communicated effectively for them to be

acceptedathospitals.193

AvarietyofstudieshaveusedauditstomeasureandimprovequalityinMNHandevidence

indicates that under certain contextual conditions audits can be feasible, effective and

acceptable.90,192,193195However,likerecordreviews,auditsareretrospectiveandrequirea

trainedhealthworker toundertakedetailedabstractionof records fromdifferent sources

whichmakeitatimeconsumingendeavour.175

Clinicalskillsanddrillsapproachesliketheobstetricemergencyskillsanddrillsmethodshave

been used extensively to maintain health workers’ competence in managing obstetric

emergenciesthathealthworkersmaynotalwaysencountersuchaseclampsiaorpost-partum

Page45of248

haemorrhage.196,197 In these skills and drills approaches, participants are given clinical

scenarios,andareinstructedtodemonstrateclinicalskillsonmannequinsorothersimulators.

Simulation-based-trainingisconsideredtobeaproactiveapproachtoreduceerrorsandrisk

inobstetricsandaimstoprovideparticipantsarangeoftransferrableskillstoimprovetheir

actual clinical performance. 196 However, these methods have mostly been used for

educationalpurposes rather than formeasuringquality, 198 and simulators canalsobea

costlyinvestment,particularlyforuseinLMICsettings.

Clinicalpracticeobservationsaredirectobservationsofcareprocessesastheyhappenand

areanestablishedmethodforevaluationofQoC.1,86Theygenerallyutiliseexternalobservers

andareseparatetoongoingsupervisionandmentorshipduringregularclinicalpracticewhich

mayinvolveobservations.Fromanethicsstandpoint,itisessentialthatbothhealthworkers

andpatientsareinformedpriortothestartofclinicalobservations.Thismayoftenintroduce

a bias referred to as Hawthorne effect.181 Clinical practice observations and standardized

patients are thought to be gold-standardmethods to assessQoC 1,86,179 but they are not

suitableforoutcomesthatareinfrequentorconditionsthatcannotbesimulatedbyactors,

forexample:neonatalresuscitationormaternalcomplicationsofpregnancy.Theyarealso

resource-intensiveandthereforemaynotbesuitableforfrequentorroutinemonitoringof

quality.

Clinical practice observations have been utilised by various studies to examine quality of

obstetricandneonatalcareinmanyLMICsettings.70,71,78,199-201TheAvertingMaternalDeath

and Disability (AMDD) programme of the Columbia University, which initiated the needs

assessment of emergency obstetric and newborn care22, USAID/ MCHIP QoC surveys172,

HelpingBabiesBreatheprogrammeforneonatalresuscitation202andassessmenttoolsfrom

the Gaala study203 have specific sections onmeasuring processes of care during routine

labourandchildbirth.Theyalsohavespecificsectionsonintrapartumandimmediatepost-

partum care including aspects of woman-centered respectful maternity care. These

instruments been used in multiple countries and are based on globally recognised best

practicessuchastheWHO’scareinnormalbirthandIntegratedManagementofPregnancy

&Childbirthmanuals.114,204

SomeexcitingrecentadvancesinmeasuringQoChaveincludedtheuseofvideo-filming205

which is suitable for rarer outcomes, events that unfold over a shorter period of timeor

Page46of248

involve a series of steps such as neonatal resuscitation or observation of oxytocin use.

However,thecostsofclosed-circuitcameras,ethicalandsensitiveissuesaroundtheuseof

videofilming,consentprocedures,dataanonymizationanddatamanagementneedscareful

consideredpriortousingsuchvideofilmingmethodsinLMICsettings.

Clients’experiencesincludingsatisfactionwithcareisgenerallyassessedusingcross-sectional

surveys.Donabedian(1980)definedusersatisfactionas“patient’sjudgmentonthequality

and goodness of care”146. Linder-Pelz and Struening (1985) have argued that satisfaction

comprisesof“multipleevaluationsofdistinctaspectsofhealthcarewhicharedetermined(in

some way) by the individual’s perceptions, attitudes and comparison processes.”206 As

highlightedbythesedefinitions,theconceptofsatisfaction ismultidimensional207andany

evaluation of satisfaction is likely to be influenced by individual women’s personal

preferences,theirexpectations,theculturalandsocialcontextandactualcarereceivedby

them.208

Althoughsatisfactionisconsideredtobeimportantforfutureutilizationandchoiceofhealth

facility,209 further research is needed to fully understand the mechanism through which

womenperceivesatisfactionwithmaternityservices.210-212Surveystomeasuresatisfaction

have been criticised for limitations such as measurement errors and inability to assess

changesovertime.Forexample,surveysmayoftenuseasingleitemtoassesssatisfaction

withcareignoringthemulti-dimensionalnatureofsatisfaction.208,210,213Researchindicates

multipledeterminantsthatinfluencewomen’ssatisfactionsuchasstaff-womaninteraction,

informationexchange,involvementindecisionmaking,controlduringthebirthingprocess,

painrelief,andbirthenvironment.214-216,210,217Detailedinformationonthesedeterminantsis

notalwayscollectedinsatisfactionsurveys.

Someresearchershavealsoarguedthatsurveysonsatisfactionwithmaternitycarearenot

groundedinconceptsandtheory.218,207Othershavealsonotedthathighlevelsofsatisfaction

are frequently reported in surveys which questions the reliability and validity of existing

measurement tools.Oftendata fromsatisfactionsurveysshowsa lackofvariabilitywhich

questionstheabilityofsurveystodiscriminate.211Lastly,measuresofsatisfactionreported

inresearchstudiesoftendonotalwaysdifferentiatebetweentheactualexperienceoflabour

andchildbirth(suchaspainormistreatment)andtheoverallexperienceofcareduringthe

hospitalstay.208,219,220

Page47of248

Arecentsystematicreviewthataimedtoidentifyexistinginstrumentstomeasuresatisfaction

with labour and childbirth found that there were only a small number of validated

instruments.208 Based on a detailed review of the literature, researchers identified all

availablemulti-itemscalesofsatisfactionofcareduringlabourandchildbirthandassessed

whether psychometric information (such as information on questionnaire construction,

reliabilityandvalidity)wasavailable.Basedontheirfindings,researchersrecommendedthat

for a detailed investigation of satisfaction with maternity care, the Intrapartum- specific

Quality from thePatientsPerspectivequestionnaire (QPP-I)was themostappropriate. 218

Othershorterinstrumentsfoundtohavegoodreliabilityandvaliditywerereportedtobethe

SixSimpleQuestions(SSQ)161andthePerceptionsofCareAdjectiveChecklist(PCACL-R).221,222

However, research evidence examining the extent towhich these instruments have been

usedinLMICsettingsishardtoobtain.

Although, measuring satisfaction with maternity services has been discussed under

measuringprocessesofcare,Iwillnotmeasurewomen’ssatisfactionwithmaternitycareas

apartofmyPhD.

2.6.3:Measuringhealthoutcomemeasuresofqualityofcare

Outcomemeasuresareindicatorsofthehealthstatusofthegroupofpatientsusingfacilities

andofbroaderpopulations,however,measuringhealthoutcomesalone,isnotnecessarily

idealformeasuringqualityofcareforthreereasons.

First,apatientmayreceivepoorqualitycarebutmayrecovercompletelyormayreceivehigh

quality carebut stillmaynot recover.Second,adversehealthoutcomessuchasmaternal

deathsand/ormaternalcomplicationstendtoberare.86Third,inhealthfacilities,casefatality

and complication rates are influenced by the case mix of patients using facilities which

complicatesanalysisandinterpretationefforts.

Atthepopulation levelandwiththeaimofassessinghealthsystemquality,therearefive

opportunities to collectdataonoutcomes suchasmaternalmortality.These includedata

fromdeathregistrationsystems,routinedatafromhealthfacilities,censuses(onceeveryten

years), specialised surveys and surveillance efforts.223 Other composite methods such as

ReproductiveAgeMortalityStudies(RAMOS)alsoexist,whichdrawuponacombinationof

Page48of248

these methods. 73 Essentially, in these studies researchers aim to identify all deaths of

reproductive age woman and then ascertain cases of maternal deaths and identify the

circumstancesbehindthesematernaldeath.95,223 However,it isgenerallyacceptedthatin

LMICcountrieslackingcompletevitalregistration,noapproachisguaranteedtogiveprecise

population-basedestimates, inparticularasmaternaldeath isstatisticallya relatively rare

event.

Somecommonproblemsassociatedwithreportingofmaternalmortalityaremisclassification

andunderreportingofmaternaldeaths.Misclassificationmayoccur incaseswheredeaths

areassociatedwithinducedabortion(especiallywhereitisillegal);earlypregnancydeaths

(resultingfromectopicormolarpregnancy),wherethepregnancymayhavebeenunknown

to the woman or her family; indirect maternal causes (malaria, anaemia, tuberculosis,

hepatitis,orcardiovasculardisease).Itmayalsohappenifdeathsoccursometimeafterthe

endofchildbirth,especiallyincaseswherethedeathoccursinanon-obstetrichospitalwards,

forexample, inan intensivecareorotherspecializedunits.224Underreportingofmaternal

deathsinLMICsettingsisalsothoughttooccurbecauseoflimitedincentivestoreportvital

events,differentialunder-reportingofdeathsbysexandinaccurateclassificationofmaternal

deathsashighlightedearlier.223

Measuringmaternalhealthoutcomeswithcertaintyatthepopulation level, thereforecan

requireresearchstudiesconductedonavery largescale,whichmakesthemanexpensive

endeavour.

However, depending on the research question, studies frequently measure outcomes to

assesstheeffectivenessofclinical interventionsorprogrammaticapproaches in improving

maternalhealthatthehealthfacilitylevel.Forexample,arecentlarge-scaletrialknownas

theWOMANtrialenrolledover20,000womenacross21countries,examinedtheeffectof

Tranexamicacidonriskofmortalityfrompost-partumhaemorrhage(outcome)andfound

that Tranexamic acid reduced death due to bleeding in women with post-partum

haemorrhage with no adverse effects.225 Similarly, Dumont et al. (2013) conducted a

pragmatic cluster randomised trial and investigated the impact of a multi-faceted

intervention(trainings,auditcycles,maternaldeathreviews,refreshertrainings,certification

andothers)onreducinghospital-basedmaternaldeath(outcomemeasure)in46hospitalsin

Senegal and Mali. 226 Their results showed that that this multi-faceted intervention was

Page49of248

successfulinreducingmaternaldeaths(oddsratio[OR]0.85;95%CI0·73to0·98,p=0.029)

inprimaryreferralhospitals.226

However,itisimportanttonotethatwhilemeasuringoutcomesatthehealthfacilitylevel

suchasmaternaldeathsorcomplications,researchershavetobecarefulininterpretingtheir

findingsbecauseof issuessuchasmisclassificationandunderreporting (describedearlier),

but also larger health system factors that influence maternal and perinatal outcomes in

facilities.

2.6.4:Summaryofmeasuringqualityofcareinmaternalandnewbornhealth

Measurementofstructurealone,suchasreadiness,eitheroffacilities(throughmeasurement

ofsignalfunctions)oroftheprovider(throughmeasurementofknowledgeandskills)does

not provide a comprehensive picture. Similarly, a focus on clinical outcomes alone is not

enough, as most pregnancies are uneventful, complications may occur, and negative

outcomesmayalsooccurinthepresenceofgoodclinicalcare.Therefore,measurementof

QoCinobstetricsneedstofocusontheprocessesofcareandshouldincludebothtechnical

quality as well as experiences of care that women receive while seeking institutional

maternitycare.

2.7:EmpiricalevidenceondeficienciesinQoCduringlabourandchildbirthinIndia

This section will present the findings of my literature review on QoC during labour and

childbirthatmaternityfacilitiesincludingrelevantliteratureonhealthsystemsissuesinIndia.

Ihaveusedtheframeworkofstructure,processandoutcomestosummarisethefindingsof

myliteraturereviewrelatedtoQoCduringlabourandchildbirthinIndia.Additionaldetails

ontheIndianhealthsystemareprovidedinthestudysettingsection.

2.7.1:ForstructuralelementsofQoCinIndia

This sectionwill discussdeficiencies in structural elementsofquality, bothat the levelof

healthsystemsandathealthfacilitiessincetheyarebothintegraltotheprovisionofhigh-

qualitymaternitycareatfacilities.

AlthoughthereseemstobeashortageofclinicalworkforceacrosseveryIndianstate,this

situationisparticularlyacuteinstateswithpooresthealthindicatorssuchasUttarPradesh,

which also has the lowest density of healthworkers.227,228 Data from the IndianNational

SampleSurvey(2011-2012)estimatedthatthedensityofdoctors,nursesandmidwivesof6.4

Page50of248

per 10,000 population was significantly below theWHO benchmark of 22.8 workers per

10,000population.227

Overall,theIndianpublichealthsystemisknowntosuffersfromstaffshortages,imbalances,

mal-distribution, poor work environments, low personnel productivity, numerous vacant

posts, high staff turnover, loss of qualified personnel to private sector, andmigration of

workerstourbanareas229,whichmakehumanresourcesshortagesasignificantconcern.The

distributionofqualifiedhealthworkersinthecountryalsoseemstobeskewedinfavourof

urbanareas;77.4%ofthequalifiedworkforcelivesinurbanareas,whereas31%oftheIndian

population is urban.227 Moreover, there aremany challenges to recruiting and retaining

qualifiedstaffinthepublicsectorespeciallyinruralareas.228

India’shealthworkforcealsoincludesdoctorstrainedinIndiansystemsofmedicinesuchas

Ayurveda, Yoga, Unani, Sidha and Homeopathy which are collectively known as AYUSH

providersandtheyofferhealthcarethroughbothpublicandprivatesectorfacilities.230

ApartfromAYUSHpersonnel,therearealsoalargenumberofinformalmedicalpractitioners

commonlyreferredasregisteredmedicalpractitioners(RMPs).TheseRMPsareoftenthefirst

point of contact, particularly for the rural population and the urban poor. Although they

practiceallopathicmedicine,RMP’softendonothavetherequiredformalqualificationsor

licensetodoso.230DetailedinformationonRMP’squalificationsandskillsarehardtoobtain,

however,onestudyestimatedthatanaverage25%ofRMP’sclassifiedasallopathicdoctors

reportednomedicaltraining(42%inruraland15%inurbansettings).231 AnotherstudyinUdaipurdistrict,inthestateofRajasthanfoundthat41%ofprivatepractitionerswhocalled

themselvesdoctorshadnomedicaldegree,18%hadnomedicaltrainingatalland17%had

notevengraduatedfromhighschool.232

Theseunregisteredprivate“doctors”areconsideredtoprovideasubstantialproportionof

maternalandnewborncare,althoughasmentionedearlier,disaggregateddataontheirshare

of themarket ishard toobtain.However,available research suggests that these informal

providersoftenhavestrongprofessionalnetworkswithqualifiedprivatesectordoctors(or

theprivatepracticesofpublicsectordoctors),pathologylaboratories,andprivatefor-profit

hospitalsandtheymakereferralstotheseplacesinreturnforcommissionsonprocedures,

diagnostics,medicinesandconsultations.233

Page51of248

Furthermore,Indiaalsohasotherpractitionersoftraditionalmedicinesuchasherbalistsand

faithhealers, traditional birth attendants (dais), andotherswhoare also involved in care

provision.Asaresult,availableresearchevidencesuggeststhatthereisahighrelianceon

unqualifiedpersonnel,particularlyinruralareas.230

Specifictomaternitycare,thereisnoformalmidwiferycadreinIndia.234Aneedsassessment

reportby JohnsHopkinsProgramforInternationalEducationinGynaecologyandObstetrics

(Jhpiego)Indiain2015estimatedthattherewasadeficitofovertwomillionnurses,withover

18%posts of staff nurses and auxiliary nursemidwives at primary and community health

centresreportedtobevacant.235 Jhpiego(2015)alsoreportsthatthetrainingcurriculaof

nursesinIndiadidnotmeettheinternationallydefinedcompetenciesforSBA.235Inaddition,

61% nursing institutions were reported as unsuitable for conducting competency-based

trainings.234

ResearchershavealsoreportedthatinsomestatesofIndiathereisalackofnationallyagreed

minimum standards for drugs, supplies and equipment that results in procurement of

resourcesofvariablequality.16Inaddition,poorhospitalinfrastructureandstrictinstitutional

policies (for example, not allowing nurses to give injectable/s or not allowing birth

companionsinlabourrooms)alsohampertheprovisionofhighqualitycareatthetimeof

birth.

ThehealthfacilityenvironmentinIndiaalsoappearstohavemanystructuralconstraints.For

example, a facility survey fromUttar Pradesh (2009) reported that cleanwaterwas only

availablein57%andessentialdrugsandsupplieswereonlyavailablein29%ofprimaryhealth

centresandregularelectricitysupplyisamajorchallenge236.Agapanalysisoffirstreferral

units(FRU)inUttarPradeshconductedbytheUPgovernmentandpartners(November2013)

foundthatonefifthofhigherlevelfacilitiessuchasdistricthospitalsandmedicalcollegesdo

nothaveadequatespaceallottedforlabourrooms.237Thisstudyreportedthatcomparedto

higher-levelfacilities,greaterproportionsoflowerlevelfacilitiesperformedpoorlyformany

structural indicators. For example, just 35% of Community Health Centres (CHC) had

appropriate handwashing areas with elbow-operated taps, 16% had functional and clean

toiletsattachedtothelabourroom,31%hadtheadequateavailabilityofessentialequipment

andsuppliesand31%ofCHCshadsufficientnumberofbeds.237Thesedataindicateindicates

significantstructuraldeficienciesatmaternityfacilitiesinUttarPradesh.

Page52of248

Othercross-sectionalevidence from Indianstudies suchas inMadhyaPradesh,whichhas

similar indicatorstoUttarPradesh,hasfoundthat86%ofdeliveriesoccur inpublicsector

facilitiesthatareunabletoprovidetherecommendedBEmOCsignalfunctions.62Inthisstudy,

researchers reported that amongst 29 facilities that could perform caesarean operations,

nonecouldperformalltheBEmOCfunctions.62Capacitytoprovidesignalfunctionssuchas

parenteralanticonvulsants,manualexplorationoftheuterus,removalofretainedproducts

of conception and assisted vaginal deliverieswere particularly problematic.62 In addition,

researchers reported that CEmOC services were more likely to be available in a greater

proportion of private-sector facilities compared to public sector facilities, just one in six

public-sectorfacilitiescouldprovideallCEmOCservices.62

OtherstructuralconstraintsdocumentedinIndianmaternityfacilitieshaveincludedlimited

triaging mechanisms, limited availability of round-the clock services, weak referral and

transportationservices,limitedonsitebloodtransfusionandanaesthesiaservices;allofwhich

indicatethattherearesignificantstructuralchallengesfortheprovisionofhigh-qualitycare

atthetimeofbirthinIndia.16,238-240

2.7.2:ForprocesselementsofQoCinIndia

MostoftheavailableresearchevidenceonprocessmeasuresofQoCduringnormallabour

andchildbirthinIndiaemergesfromcross-sectionalstudiesinthepublicsector.Information

onQoCfromtheprivatesectorisscarce.Availableresearchevidenceindicatespoorquality

of maternity care as shown by high rates of labour augmentation, routine conduct of

episiotomies,non-adherencetoactivemanagementofthirdstageoflabour,limiteduseof

partograph or foetal heart rate monitoring, early discharge from the hospital, limited

preparedness for neonatal resuscitation, poor initiation of breastfeeding, and inadequate

thermalcareofneonates.241-244

Evidencefromaquantitativestudyin2007inRajasthanfoundthatupto85%ofalldeliveries

wereaugmented,67%ofwomenweresubjectedtostrongfundalpressureandmorethan

halfofpostpartumwomenweredischargedbefore24hours(nationalguidelinesrecommend

48hours).243Similarly,anotherobservationalstudyfoundthatfoetalheartratemonitoring

was not performed regularly, preparedness for neonatal resuscitation was minimal,

episiotomy;perineal-shavingandenemawerecommon.244Theauthorsalsoreportthatthe

presenceofSBAsduringchildbirthatfacilitiesdidnotguaranteethereceiptofskilledcareby

Page53of248

themother and her newborn, and that unqualified attendants are frequently involved in

maternitycareprovisioninfacilities.244

Someresearchershavearguedthatlimitedknowledgeofmaternitycarepersonnelisbehind

thepoorQoCatfacilitiesinIndia.65Forexample,anIndianstudyutilisingclinicalvignettes,

foundthataslittleas20percentofnursesworkingasSBAsappearedcompetentinmanaging

eclampsiaandhaemorrhage,andonly10%seemedcompetentinperformingacorrectinitial

assessmentofwomenwithpregnancycomplications.65Twocross-sectionalstudieshavealso

reportedthatANMs,nursesandmedicalofficers(doctors)didnothavetherequiredskillsand

werenotconfidentinprovidingbasicEmOCservicesincludingstabilisationpriortoreferral.245,246

Researchers have also found overuse of prophylactic antibiotics during labour in India

irrespectiveofthetypeofdelivery.247Thisoveruseofantibioticswithoutproperindications

isthoughttobeduetohealthworkers’beliefsregardingpoorhygieneandinfectioncontrol

standards at maternity facilities and their own assumptions of poor personal hygiene of

womenthatcomefordeliveries.247

Stanton et al. (2014) conducted an observational study in two Indian states, and found

widespread non-adherence to existing protocols on uterotonic drug use at public sector

facilities.242Theyfoundthatthatlabouraugmentationratesrangedfrom78.6%(Karnataka)

and99.1%(UttarPradesh),correctuseofoxytocicsforpostpartumhaemorrhagevariedfrom

6%–8.8%inUttarPradeshand41.2%–76.4%inKarnataka.Activemanagementofthethird

stageoflabourwasfoundtobeperformedcorrectlyinlessthan10%ofdeliveriesinboth

districtsandthatstorageofuterotonicsatroomtemperaturewascommon.242

OtherqualitativeresearchevidencefromIndia66,239hasdescribedsituationswherelabouring

women were mistreated (shouted at or slapped), cases where women were not given

adequateinformationabouttheproceduresbeingdone,birthsoccurringinhospitalswithout

a health professional in attendance, and cases where post-partum women were not

monitoredorsupportedafterchildbirth.66,239

Evidence from various Indian states has also revealed poor routines in care, such as

inappropriate monitoring during labour and childbirth, use of harmful and unnecessary

practices,limitedpreparednessandwidespreadstaffshortagesathealthfacilities.66,67,70,71,78

Page54of248

Thereisalsosomeresearchinformationonwhousesprivatesectorfacilities.Cross-sectional

research evidence indicates that socio-demographic factors are a key determinant for

choosingprivatesectorformaternitycare.40Theeffectofeducationseemstobeimportant,

with one study finding a positive effect in India.248 Other factors such as ethnicity and

caste/tribestatuswerefoundtobenegativelyassociatedwiththeuseofprivatefacilitiesin

India.248Cognitionwhichmeansprovider–clientinformationexchangewasidentifiedasthe

mostimportantdeterminantforservice-utilisationinsouthAsia.239Womenwhoattendeda

greaternumberofANCvisitsweremorelikelytousetheprivatesectorduringchildbirth.248

However, there is also contradictory evidence on whether obstetric complications could

promptwomen to seek care inaprivate sector. 40,249Ahigher socio-economic statusand

urbanresidencewasassociatedwithgreateruseofprivatesectorfacilitiesformaternitycare

inIndia.248

Most published studies from India (and specifically from Uttar Pradesh ) have not

comprehensively measured QoC during labour and childbirth, most have employed

qualitative methodologies, were conducted in the public sector, examined home-based

childbirth practices70 or focus on specific issues such as PPH management77, labour

augmentationwithoxytocin,250,251neonatalcordcare,breastfeedingorthermalcare.252Asa

result, there is limited descriptive information from robust studies that comprehensively

measureQoCduringlabourandchildbirth.ThisisespeciallytruefortheprivatesectorinIndia

whichprovidesapproximatelyaquarterofmaternitycareservicesinIndia.253,254However,

available evidence from the private sector does indicate that there is increasing

medicalisationofchildbirthdrivenprimarilybycaesareansinprivatesectorhospitalsinsouth

AsiaincludinginIndia.25,39

2.7.3:HealthoutcomemeasuresofqualityofcareatthetimeofbirthinIndia

Giventheincreasingglobalattentiontowardsimprovingmaternalandnewbornhealth(MNH)

in India and effective advocacy by grassroots organizations, there appears to be lot of

momentumaroundimprovingMNHinIndia.TheGovernmentofIndia,throughtheNational

RuralHealthMission’seffortsandtheJananiSurakshaYojanaprogrammehasbeensuccessful

inincreasinginstitutionalbirthsfrom41%in2004to73%in2012.255

Page55of248

However,thereislimitedavailabilityofinformationonfacility-basedoutcomemeasuressuch

ashospitalmortalityrates,obstetriccase-fatalityrates,near-misseventsornever-eventsin

UttarPradesh.Thisinformationcouldnotbeobtainedfromexistingroutinedatasourcesat

healthfacilitiesandthesearescarceinthepeer-reviewedliterature.

Available population-based outcome measures illustrate the high burden of maternal,

neonatalandperinatalmortalityinIndia.TheMDG5atargetforIndiawastoreducetheMMR

to 109 maternal deaths per 100,000 live births by 2015.256 However, despite impressive

progress, with declines inMMR from 437 to 178 per 100,000 live births (a 59% decline)

between1990and2012,IndiacouldnotachievetheMDG5atargets.Furthermore,national

estimateshidestrikingdisparitiesbetween Indianstates.Forexample, theMMRinAssam

was found tobe328per100,000compared toUttarPradesh,where theMMRwas240,

considerablyhigher than states likeKeralawhere theMMRwas just 66per 100, 000 live

births.253Availableresearchevidencealsoindicatesthatthemajorcausesofmaternaldeaths

inIndiawereduetodirectobstetriccausessuchas-haemorrhage(38%),sepsis(11%),unsafe

abortion(8%),hypertensivedisorders(5%)andobstructedlabour(5%)257,althoughdataon

causesofdeathafter2003isnotavailable.

For neonatal mortality, in 2013, India had the highest burden globally with 0.75 million

neonatal deaths. 258 Currently, theNMR stands at 28 per 1000 live births.259 The annual

burdenofneonataldeathsreducedfrom1.35millionin1990to0.75millionin2013258with

rapidaccelerationofNMRdeclines(33%)between2000-2013comparedto17%declines

between1990and2000.260InIndia,themaindirectcausesofneonataldeathin2015were

prematurity(43.8%),birthasphyxia/trauma(18.9%)andsepsis(13.6%)261whichsuggeststhat

careatthetimeofbirthisansignificantconcern.

Forstillbirths,in2015,Indiahadthehighestratesintheworldwithapproximately592,100

stillbirths,contributingupto22.6%oftheglobalburden.52Ahospital-basedstudyhasfound

thatpregnancy-inducedhypertension,eclampsia,abruptio-placenta,birthasphyxia,andpre-

term labour are the underlying causes contributing to stillbirths in India.262 Researchers

suggestthatpoorqualityofantenatalandintrapartumcarearetheleadingcausesformost

preventablestillbirthsinIndia.262,263

Page56of248

2.7.4:SummaryoftheliteraturereviewonQoCinIndia.

Theprevioussectiondiscussedthefindingsofmyliteraturereviewandestablishedacasefor

investigatingtheQoCatthetimeofbirthinIndia.

Insummary,thefindingsofmy literaturereviewhighlightedthatthebulkoftheavailable

evidenceonqualityofessentialcareduringlabourandchildbirthinIndiaemergesfromthe

public sector. Most identified studies did not examine care at the time of birth in a

comprehensive manner. The literature review on structural aspects of QoC identified

deficiencies relatedto inadequatehumanresources, limited functioningofEmOCfacilities

andinadequateprovisionofrecommendedsignalfunctions.Ialsofoundproblemsrelatedto

limitedsuppliesofessentialdrugsandcommodities,gapsinknowledgeofhealthworkersand

widerinfrastructuralconstraints.

The literature review on process aspects of QoC identified deficiencies related to non-

adherencetoevidence-basedprotocolsformaternalandneonatalcareandmistreatmentof

womenatmaternityfacilitiesinIndia.Inparticular,researchevidenceonqualityofmaternity

careprovidedintheprivatesectorandresearchstudiesthathavecomprehensivelyaddressed

careatthetimeofbirthwerefoundtobelimited.

Outcomedeficienciesidentifiedwererelatedtoalackofinformationfromhealthfacilitieson

outcomeindicatorsofQoCsuchasinformationonobstetriccasefatalityrates,near-misses

ornever-events.Overall,populationbasedoutcomemeasuresshowhighratesofmaternal

mortality,neonatalmortalityandstillbirthsinIndia.AllofthesefindingssuggestthatQoCat

thetimeofbirthisanextremelyimportantconcernforIndia.

2.8:Managementpracticesatmaternityfacilities

2.8.1:Theoreticalconceptsonmanagementpractices

Management capacity has often been identified as a critical bottleneck for poor health

indicators in LMICs264 but their potential in improving QoC at hospitals has not been

extensively studied. TheWorld Health Organization (2005) defines goodmanagement as

“providingdirectionto,andgainingcommitmentfrompartnersandstaff,facilitatingchange

andachievingbetterhealthservicesthroughefficient,creativeandresponsibledeployment

ofpeopleandotherresources.”264

Page57of248

In hospitals, management competencies are needed to identify and prioritise problems,

developappropriateplans,effectivelyutiliseavailableresourcesandtrackprogresstowards

achievement of institutional goals.264 Theoretically, it seems likely that if resources are

available,amanagerwhounderstandstherequirementsnecessarytoprovidehigh-quality

maternitycarepathways(figure1)willputinplaceappropriatesystemsandproceduresto

supporthigh-qualityclinicalcareandrespectfulmaternityservicesatinstitutions.

Itappearsthatmostofthetheoreticalconceptsaroundmanagementcomefromthebusiness

sector,whichhasbeenprogressivetotestinnovativemanagementstrategiesandquantify

the impact ofmanagement practices inmonetary terms rather than gains in quality. The

notionthathealthserviceswillbemoreeffectiveifstaffwithmanagerialcompetenciesare

employedataseniorleveliswellestablished,andappearstohavebeeninfluencedbymany

factors.265First,itisthoughtthateffectivecost-containmentathospitalswillnotbeachieved

withoutdrawinghealthworkersintoaframeworkofaccountability.Second,thereisabelief

thatmodernmanagementpracticessuchasthoseemployedinthebusinesssectorcouldbe

appliedtohospitalstoincreasetheirproductivity.Third,funders(whetherpublicorprivate)

requireaccountabilityforthelargesumsofmoneytheyinvestintohealthinstitutions.Fourth,

theprocessesandtransactionscreatedbyhospitalssuchascontracting,costingandbilling

arethoughttobetterdealtbyprofessionallytrainedmanagersandfifth,thereisabeliefthat

goodmanagementpractices could lead tobetter returnson investment through financial

earningsandcostsavings.265

However,inhospitalsunlikeotherorganizations,managersneedtounderstandnotjustthe

operational,human, institutionalandstructural factorsbutalso issuesaroundclinicalcare

provision,patientsafetyandmedicalerrors.266,267Thesefactorsmaketheroleofahospital

managerparticularlychallenging.Moreover,atpublic-sectorhospitalsinLMICsettings,there

aremorefundamentalbarrierstoprovidinghighQoCthatareoftenbeyondthecontrolof

individual managers. Some of these include limited availability of essential drugs,

commodities,irregularelectricitysupply,infrastructuraldeficienciesandlackofskilledhuman

resourcesasdescribedinprevioussections.16

Basedonmyreviewoftheliteratureonmanagement,qualityofcareandhospitalsetting,

managementpracticesathospitalshavebeenconceptualisedasoperationsmanagement,

Page58of248

performance management, target management and people management.42-45 These

practicesareillustratedinfigure3andfurtherdetailsareoutlinedbelow.

1. Operationsmanagement:Operationsmanagement refers to thedesign,management,

andimprovementofsystemsthataffectahospital’sperformance.Areviewofoperations

managementliteratureidentifiesthefollowingbestpracticecategories:“leansystems”

to eliminate waste and non-value-added activities, planning and control systems and

qualitymanagement systems. 268-273 Goodhospitaloperationsare supported ifhealth

workersadheretoclinicalguidelinesandcarepathways274thatfacilitateefficientpatient-

flowthroughthehospital275andbyimprovingqualityofclinicaldocumentation.276Bloom

etal.andDorganetal.havealsoproposedthatefficientlayoutofpatientflow,focuson

continuous improvementandoptimalutilizationofresourcesarealsooperationalbest

practices. 277,43 In the Indian context, although government guidelines on operational

standardsexist,thereislimitedresearchexaminingwhethertheseoperationalstandards

arefollowed.278,279

2. Targetsmanagement:Performancetargetsaretoolsdesignedtoimproveaccountability,

transparency and performance of health facilities.280,281 Effective target management

referstosettingrealistic,well-defined,time-boundandspecifictargetsformaternitycare

servicesatfacilities.282Bloometal.andDorganetal.suggestthatbestpracticesintarget

managementrequirestargetstoincludeoperationalandfinancialdimensionsandhave

short and long-term timeframes. 43,277 The introduction of targets and performance

contractsisthoughttoencouragebetterhospitalperformance283andthatincentivestend

tobemoreeffectivewhenlinkedtoinstitutionaltargets.284

3. Peoplemanagement:Humanresourcesatmaternityfacilitiesarecomprisedofteamsof

medical,paramedicalandauxiliarystaffresponsibleforvariousindividualfunctionswith

anaimtoprovidehigh-qualitycontinuityofcaretopregnantwomanfromthetimeof

admission to discharge from the facility.16 To a large extent hospital performance,

dependsontheknowledge,skillsandmotivationsofindividualsresponsibleforproviding

services.285Astudyof61hospitalsinEnglandfoundapositiveassociationbetweengood

humanresourcepractices(specifically,performanceappraisal)andpatientmortality.286

Evidencefromhigh-incomecountriessuggeststhatincentivesandteam-basedworking

increasesjobsatisfaction287,employeemotivation,retention288andcauseslessstress289.

Page59of248

Systemsof training, supervision and career development are also thought to improve

retentionofhealthworkers.290,291Bloometal.277andDorganetal43suggestthatgood

practices forpeoplemanagement includea structuredapproach to recruit, retainand

managehealthworkers.Theyalsosuggestthateffectiveincentivesystemsarelinkedto

performanceappraisalsandshouldbalancebothfinancialandnon-financial incentives.

Merit-based promotions rather than by tenure or seniority are also considered best

practices.277However,inLMICsettings,humanresourcesystemsarenotasdeveloped,

andpeoplemanagementisoftenchallenging.Forexample,inthepublicsector,decision-

making on recruitment, positing and transfers is usually centralized and adequate

supervisionisanon-goingchallenge.292

Figure3:Dimensionsofmanagementpracticesatmaternityfacilities

4. Performance management: Performance management allows managers to identify

deficiencies in serviceprovision293 and if doneeffectively, is thought to improve care-

processes and clinical outcomes.294,295,296Researchers have argued that multiple

performanceindicatorsarerequiredtomeasurehospitalperformanceaccurately.297-299

According to Scott et al., an effective performance monitoring system is based on

evidence-basedclinicaldecision-making(throughguidelines,protocolsandpathways);it

incorporates systems for process evaluation (audits, feedback, clinical indicators and

processmeasures);andsupportsindefining,implementingandmonitoringappropriate

Page60of248

indicatorstomeasurequalityimprovement.274ThereisrobustevidencefromaCochrane

systematic reviewthat supports theeffectivenessofauditsand feedback in improving

professionalpracticeandhealthoutcomes.195

2.8.2:Empiricalevidenceonhospitalmanagementpracticesandquality

Although it is generally accepted thatmanagement practices influence quality of care at

hospitals, empirical evidence examining this relationship is limited.300 Perhaps, because

managementishardtomeasureusingquantitativemethodsandmethodologicaladvancesin

measurement have been recent, there is limited evidence to support these claims,

particularlyasthereisadearthofstudiesfromLMICsettings.

The bulk of the peer-reviewed literature consists of non-empirical articles such as case

studies, opinion-pieces, editorials but these often lack empirical data examining the

relationshipbetweenmanagementandqualityofcare.301,302303Manyresearchershavenoted

thisasanimportantevidencegap.300,304-306Theonlysystematicreviewwhichexaminedthe

role of hospital managers in quality and patient safety found limited and inconsistent

evidence on this relationship.300 The modest evidence that exists does suggest that

managers’ time spent, engagement andwork specifically on quality assurance influences

indicatorsofclinicalqualityandpatient-safetypositively.300Managerialactivitiesthoughtto

improvequalityincludeactivitiessuchasestablishinggoalsandstrategiestoimproveQoC,

setting thequality agenda,promotingaquality improvement cultureandprocurementof

institutionalresourcestoensurequalityofcare.300

Thepastdecadehasseena rise in the innovativemeasurementefforts thathave tried to

quantify therelationshipbetweenmanagementandQoCoutcomes.Mostof this research

stems from the field of health economics and are primarily from studies in high-income

countries.42,46,300Notableamongstthese,isthepioneeringworkbyBloometal.(2010)who

initiallystudiedmanagementpracticesacrossmanufacturingfirmsinnumerouscountries.307

ThisworkhassincebeenreplicatedinthehealthsectorandthetoolsdevelopedbyBloomet

al.(2010)havenowbeenusedformeasuringmanagementpracticesindiversehealthsystem

contexts suchas inhigh-income (Australia,Canada, France,Germany, Sweden,UK,USA),

upper-middleincome(Brazil)andlower-middleincomecountries(India).43,282,308,309

Page61of248

These research efforts have employed a telephone-based interview methodology and

assessedmanagementpracticesunderfourkeydimensionsdescribedearlier:measuresof

hospitaloperations,measuresofhospitalperformance,measuresoftargetsmanagementat

hospitalsandmeasuresofpeoplemanagementathospitals.44,310,311

Thesemeasurementeffortshaveuncoveredsomeinterestingresults.Forexample,inacross-

sectionalstudyatcardiacunitsinUSA,managementpracticesweresignificantlyassociated

withmortalityaswellasprocessofcaremeasures.312Inanotherstudyatsubstanceabuse

clinics in the USA, researchers have found a strong association between management

practicesandclientdaystotreatmentandincreasedrevenuegeneratedattheseclinics.313

Similarly,inUKhospitals,managementpracticeswerefoundtobestronglyassociatedwith

both health outcomes (improved survival rates after acute myocardial infarction) and

financialindicators.310

OtherstudiesthatconductedsecondaryanalysisofdataconductedasapartoftheWorld

ManagementSurveyefforts(http://worldmanagementsurvey.org/),whichcollectsdatafrom

over 2,000 hospitals in nine countries have found that hospitals with more effective

management practices provide higher quality care.46,42,311 One of these studies which

restrictedanalysistodatafromhospitalsintheUSAandEnglandfoundthatwhenhospital

boardspaidmoreattentiontoclinicalquality,managersweremorelikelytopayattention

to clinical quality and that hospital boards which used clinical quality measures more

effectivelyhadhigherscoresontargetmanagementandoperationsmanagement.46

Bloomet.al’stoolhasalsobeenusedinIndia,whereitwasadministeredtomanagersof

3,892privatesectorhospitalsasapartofadescriptivestudy.44InIndia,theresearchersfound

that the average total management scores in Indian hospitals were poorer compared to

hospitals in other high income countries but this study did not examine the relationship

betweenmanagementscoresobtainedbyhospitalsandQoCoutcomes.44

However, it is important to note that the majority of research studies examining the

relationshipbetweenmanagementpracticesandQoCweredoneinhigh-incomecountries,

mostly in the private sector and none of them specifically focused on examining the

relationship between management practices and quality of maternity care provision.

Page62of248

Therefore,examiningwhether there isa relationshipbetweenmanagementpracticesand

QoCofferedduringlabourandchildbirthinmaternityfacilitiesisakeyevidencegap.

Given thewide application of the Bloomet al.’s study instrument, the standard research

methodologyusedacrossmultiplecountrieswhichsupportscomparabilityandthe limited

time-frameofaPhD,thistoolappearspromisingforadaptationanduseatmaternityfacilities

inUttarPradesh,India.

Chapter3:Researchsettingandthecontextforthedoctoralresearch

3.1Studysetting

Uttar Pradesh (UP) is India’smost populous statewith approximately 200million people

(about17%ofthepopulationof India) livingacross its18divisionsand75districts.314The

population ispredominantlyrural (77%).Eightcities inUttarPradeshcontainmorethan1

millionpeople(Kanpur,Lucknow,Ghaziabad,Agra,Varanasi,Meerut,AllahabadandBareilly).

Seventy-fivepercentof thepopulation lives in rural areasandabout33% livesbelow the

povertyline.Theper-capitaincomeofUPwasUS$522comparedtoIndia’saverageofabout

US$1,097in2010-11.314UPhasconsistentlybeenrankedwithinthebottomthirdamongst

allIndianstatesontheHumanPovertyIndexsince1981,andhaspoorhumandevelopment

indicators compared to other Indian states.314 The overall literacy rate is 70%,with 60%

femaleliteracycomparedtoanationalaverageof74%and65%,respectively.315

The religiousandcastecharacteristicsofUttarPradeshshowthestrongpresenceofboth

Hindu and Muslim populations, and of Scheduled Caste groups (marginalised groups).3

Approximately80%ofthepopulationofUttarPradeshisHindu,withtheremaining20%being

Muslim.Itisestimatedthatabout21%ofthepopulationbelongtosocalledScheduledCaste

communities.253,316

In2010-2011,UttarPradesh’smaternalmortalityratio(359per100,000livebirths)wasthe

secondhighest in the country.317 Neonatalmortality (45per1,000 livebirths) and infant

mortalityrates(63per1000livebirths)arethehighestinthecountry.318Thetotalfertility

rateof3.8isthehighestinIndia,althoughcontraceptivecoverageisincreasing.319Anaemia

(85%inchildrenand51%inwomen)andmalnutritionaresignificantconcernswithahigh

percentageofchildrenunderweight(42%),wasted(20%)andstunted(52%).320

MydoctoralstudywasconductedinthreedistrictsofUttarPradesh:KanpurNagar,Kanpur

DehatandKannauj.WithKanpurasitscapital,KanpurNagardistrictisthemostpopulatedof

the studydistrictswith4.6millionpeople.KanpurDehatandKannaujdistrictsarealmost

exclusivelyruralwith90%and83%ofthepopulationresidinginruralareasandapopulation

of1.8millionand1.7millionrespectively.TheproportionofMuslimpopulationislowerin

KanpurDehat(10%)thaninKannauj(17%)andKanpurNagar(16%).KanpurDehathasthe

Page64of248

highestproportionofScheduledCastecommunities(26%)whileKannujhas19%andKanpur

Nagarhas18%.253

Despitegovernmentschemestoimproveratesofinstitutionalbirths,54%ofdeliveriesoccur

athomeinUttarPradesh(57%inKannauj,40%inKanpurNagarand52%inKanpurDehat).

Ofthehomedeliveries,11%,53%and29%wereconductedbyskilledhealthpersonnelin

Kannauj,KanpurNagarandKanpurDehat,respectively.253

Table3highlightsimportantReproductive,MaternalandNewbornHealth(RMNH)indicators

inKannauj,KanpurNagarandKanpurDehatascomparedtotheUP-stateaverage.Kanpur

Nagarispredominantlyurban,withhigherliteracyandlowermortalityratesthanthestate

average.Bycontrast,KannaujandKanpurDehataremoreruralwithlowerlevelsofliteracy

andhighermortalityratesclosertothestateaverage.Specifically,Kannaujdistrictfaresthe

worstacrossmostoftheseRMNHindicatorscomparedtotheothertwostudydistricts.315

Table3:DemographicandhealthindicatorsinUttarPradeshandstudydistricts

Indicator UttarPradesh

Kannauj KanpurNagar

KanpurDehat

Population(inmillions) 199.8 1.7 4.6 1.8

Ruralpopulation(%) 78 83 34 90

Literacy(%) 57 61 71 65

Fertility(lifetime) 3.3 3.7 2.6 3.2

Maternalmortalityratio(per100,000) 345 267 267 267

Underfivemortality(per1,000) 94 99 52 97

Infantmortalityrate(per1,000) 71 78 36 65

Neonatalmortalityrate(per1,000) 50 55 24 43

CurrentuseofmodernFPmethodsamongcurrentlymarried(women(%)

31.8 23.2 39.7 38.6

Unmetneedforfamilyplanningamongcurrentlymarriedwomen(%)

29.7 43.2 23.7 25.0

ANC3+visit(%) 29.6 14.5 51.0 32.3

Institutionalbirthrate(%) 45.6 42.4 59.7 47.7

DeliveryathomeconductedbySBA(%) 21.8 11.2 53.2 28.6

Motherreceivedpost-natalcheck-upwithin48hours(%)

68.4 48.8 66.5 72.7

Newbornwascheckedwithin24hoursofbirth(%) 68.2 49.9 71.7 74.4

Page65of248

Newbornbreastfedwithinonehourofbirth(%) 32.9 27.0 41.0 47.0

Sources:Census2011315andAnnualHealthSurvey2011316

Note:TheMMRestimatesaregroupedbystatesduetosamplesizelimitationsandthethreedistrictshavethesameMMRgrouping.

3.2:HealthcaresysteminIndia

ThehealthsysteminIndiaismixedandconsistsofthepublicsector,theprivate-for-profitand

theprivatenot-for-profitsector.Thepublichealthsystemisthree-tieredincludingprimary,

secondary and tertiary level facilities (Figure 4)321. The private-for-profit sector is

heterogeneousandvariesinsizeandcapacity,dependingoncontextofthedistrictwhereas

theprivatenot-forprofitsectorisrelativelysmall.

Inthepublicsector,thesub-centreistheprimaryunitinruralareasandoftenthefirstpoint

ofcontactwherewomengotoreceiveantenatalcareservices.Insomeinstances,maternity

servicesarealsoavailableat sub-centres, if auxiliarynursemidwife (ANM), femalehealth

worker (cadre) or staff nurses are available.At thenext level areprimaryhealth centres,

which although, envisioned as round-the-clock BEmOC sites, may not always provide

maternity services. Community health centres (CHCs) are sites where obstetricians and

paediatriciansareavailableandtheymayfunctionattheBEmOCorCEmOClevel.Firstreferral

units (FRUs) are upgraded CHCs, sub-district hospitals, district hospitals and specialist

hospitalsthatshould,inprincipal,provideCEmOCcare.322

Figure4:SchematicrepresentationofthepublichealthsysteminIndia

Source:https://doi.org/10.1371/journal.pone.0159793

Page66of248

Inurbanareas,therearedispensaries,urbanhealthcentresandurbanhealthposts,which

provideantenatalcareandreferralservices.Someurbanhealthcentreshavebeenupgraded

toprovidematernityservices.Atthenextlevelarematernityhospitalsandothersecondary

and tertiary level hospitals that provide both BEmOC and CEmOC services.323 There are

medical colleges inmost districts that provide specialised tertiary care services including

CEmOCservices.

Healthworkersinthepublicsectorarepaidaregularmonthlysalarydependingontheirgrade

andlevelofexpertise.Thesesalariestendtobelowerthanearningsintheprivatesectorand

hencethesejobsarenotaslucrative.Inrecenttimes,therehavemanyinnovativeschemes

toattracthealthworkerstothepublicsectorsuchasadditionalmonetaryincentivesifthey

work in rural areas, educational incentives, promotion and career enhancement

opportunitiesandinnovativepublic-privatepartnershipschemes.However,thedistinction

betweenprivateandpublicsectorprovidersisnotstraightforwardashealthcareworkersin

publicsectorsmayalsohavetheirownprivatepracticeswheretheyworkaftertheirregular-

workinghoursinthepublicsector.

Qualifiedhealthworkers providingmaternity care services at institutions includedoctors,

nurses, auxiliary nursemidwiveswho receive 5 years, 3 years and 2 years of pre-service

training respectively. Thesequalifiedhealthworkers are regulated, and legally allowed to

provideinstitutionalservices.However,inIndia,asdescribedpreviously,availableevidence

indicatesthatthedistributionofqualifiedhealthworkersisinequitable:77.4%ofthequalified

healthworkforcelivesinurbanareas,although,just31%oftheIndianpopulationisurban.227

Also,researchevidencesuggeststhatthereisahighrelianceonunqualifiedpersonnelinrural

areasincludingathospitals.70,230Inaddition,thereisnoseparatemidwiferycadreinIndia

and it is thought that a serious human resource shortages exist throughout the health

system.228

Intheprivatesector,thereisextremeheterogeneityoffacilitiesinIndia.Anestimated75%

ofprivatehealth facilitiesaremicro-enterprisesandtherestaremediumto largemedical

establishments.324 In my study, private sector maternity facilities were either private-for

Page67of248

profitorNGOownedfacilitieswithbasicemergencyobstetriccare(orhigher)capability.The

majorityofprivatesectorhospitalsandbedsarelocatedinurbanareasandareoperatedby

qualifiedandregisteredprivatesectordoctorsornurses.325Generally,largerprivatesector

facilitiestendtoberegisteredwithpublicauthoritiesandtherefore,intheoryhavesomelevel

ofregulationandmonitoring.Dependingonthetypeoffacility,healthworkersintheprivate

sectorearnafixedmonthlysalaryorreceivefee-for-services.

3.3:Maternalandnewbornhealthservicesprovidedatpublicsectorfacilities

UndertheumbrellaoftheNationalRuralHealthMission,Indianpublichealthstandardshave

beendefinedthatprescribeuniformstandardsforMNHservicesabovethePHClevel.These

standardsareshowninTable4.TheyemphasizeSBA,EmONCcapabilityandefficientreferral

capacityatfacilities.326Privatesectorfacilitiesarealsoencouragedtoprovidesimilarstandards

ofMNHservicesbutdetailedinformationonadherencetothesestandardsbyprivatesectoris

notavailable.

Table4:MaternalandnewborncarestandardsatIndianpublicsectorfacilities.

Level2:Institutional(BasicLevel) Level3Institutional(ComprehensiveLevel)PHC-Basic Obstetric and Neonatal Care (round-the

clockservicesatPHCs,CHCsotherthanFRUs)FRU-Comprehensive Obstetric and Neonatal Care

(DH,SDH,RH,CEmOC,selectedCHCs)

StandardsforintrapartumcareAllinLevel1(deliverybySBAorhomedeliveryoratCHCs,PHCsnotfunctioninground-theclockplus:Availabilityoffollowingservicesroundtheclock• Episiotomyandsuturingcervicaltear• Assisted vaginal deliveries like outlet forceps,

vacuum• Stabilisation of patients with obstetric

emergencies,e.g.eclampsia,PPH,sepsis,shock• Referrallinkageswithhigherfacilities

All in Level 2 plus availability of following servicesroundtheclock:• Managementofobstructedlabour• SurgicalinterventionslikeCaesareansection• Comprehensive management of all obstetric

emergencies, e.g. Eclampsia, Sepsis, PPH,retainedplacenta,shock.

• In-housebloodbank/bloodstoragecentre• Referral linkageswithhigher facilities including

medicalcolleges

EssentialnewborncareasinLevel1plus• AntenatalCorticosteroidstothemotherincase

of pre-term babies to prevent RespiratoryDistressSyndrome

• ImmediatecareofLBWnewborns(>1800grams)

EssentialnewborncareasinLevel2plus• CareofLBWnewborns<1800gm.

PostnatalandNewbornCareAllinLevel1plus

• 48 hours stay post-delivery and all postnatalservices for zero and third day tomother andbaby.

• Timely referral of women with postnatalcomplications.

AllinLevel2plus• Clinical management of all maternal

emergencies such as PPH, Puerperal Sepsis,Eclampsia, Breast Abscess, post-surgicalcomplication, shock and any other postnatalcomplicationssuchasRHincompatibilityetc.

• Newborn Care as in Level 2 plus in districthospitalsthroughSickNewbornCareUnit

Page68of248

• Stabilisation of mother with postnatalemergencies, e.g. PPH, sepsis, shock, retainedplacenta

• ReferrallinkageswithhigherfacilitiesNewbornCareasinLevel1plus:• Stabilisationofcomplicationsandreferral• CareofLBWnewborns>1800gm.• Referral services for newborns that are <1800

gm.andothernewborncomplications• Managementofsepsis

• Managementofcomplications• CareofLBWnewborns<1800grams.• Establishreferrallinkageswithhigherfacilities

3.4:Maternalhealthprogrammesandpolicies

Maternalandnewbornhealth in India fallsunder the remitof theMinistryofHealthand

FamilyWelfare(MoHFW).SinceIndiaobtainedits independencein1947,therehavebeen

significant shifts with regards to programmes and policies on reproductive andmaternal

health. The evolution of quality in maternal health and major programmatic efforts for

maternal newborn health are highlighted in Table 5.327 Although, previous policy and

programmeeffortswerefocussedonexpandingcoverage,since2000,thereappearstohave

beenaconsiderableemphasisonQoCforMNH.

Asmentionedearlier,despiteimpressiveprogress,withdeclinesinMMRfrom437to178per

100,000livebirths(a59%decline)between1990-91and2010-12, Indiacouldnotachieve

MDG5atargets.256Since2000,theGovernmentadoptedthreemajorpolicies-theNational

PopulationPolicy(2000)328,theNationalPolicyforEmpowermentofWomen(2001)329and

theNationalHealthPolicy(2002)330,allofwhichhaveaspecificfocusonqualityinMNH.

Table5:Summaryoftheevolutionofqualityinmaternalhealth

Timeperiods Keyevents1947-60 • Focusonexpansionofservicesinunderservedareas

• Limitedhealthsectorfunding• Launchofverticaldiseaseeradicationprogrammeswithfirstfive-yearplan• Maternalandchildhealthpriorityareawithexpandedprogramminginfirst

five-yearplan• Noevidenceofeffortonqualityassurance-focusrestrictedtoequityand

humanresources1960-80 • Adoptionoftarget-basedfamilyplanningapproach;pressureformeeting

targetsdamagescommunitymaternalandchildhealthservices• ForcedsterilizationsduringEmergency(1975-77)leadtoneglectofmaternal

healthcareservices• Qualityconcernsrestrictedtoequityandqualityofhumanresources

Page69of248

• AlmaAtadeclaration(1978)renewsfocusonprimaryhealthandinspiresconcernforqualityinhealthcare

1980-90 • Verticalprogrammesonimmunizationandmaternalhealthlaunched• Qualityscopelimitedtoequity,humanandphysicalresourcesand

effectiveness;noactionstrategies1990-2000 • StructuralAdjustmentProgrammeleadstoriseinprivatesectorhealth

investmentinIndia• ReproductiveandChildHealthprogrammeintroducesintegratedmaternal

andchildhealth,familyplanningandreproductivehealthservices• EffortstoensureessentialobstetriccareandEmOCthroughstrengthening

healthfacilitiesandtransportfacilities,improvingfirstreferralunitsandbloodtransfusionservices

• Qualityconcernsvoicedincreasinglybutnoactionstrategiesformulated2000-present • NationalPopulationPolicy(2000)outlinesReproductiveandChildHealth

strategyandsetsspecificmaternalandinfantmortalityreductiongoals• Qualityfocusintenthandeleventhplanswithstrategiesforquality

assuranceandappraisal• NationalRuralHealthMissionlaunched,leadingtoexpandedfundingand

decentralizedprogrammeimplementationo Emphasisonfacilitybirthso Focusonskilledbirthattendance(SBA)o InfrastructurestrengtheningforbasicandcomprehensiveEmOC

throughreformsundertheNRHMo CapacitybuildingforSBA-trainingofnurse-midwivesforSBA,task

shifting–generalphysicianstrainedforanaesthesiaforEmOCandforCaesareansection

o Raisingdemandforfacilitybirths-theJSYcashtransferprogramofferingincentivestowomenandtoASHAs

• Qualityfocusandactionstrategiesinbothprogrammesalongwithregularmonitoringandfeedbackmechanisms

• QualityinitiativesincludeIndianPublicHealthStandardsforqualityassuranceinprimarycare;QualityAssuranceCommitteesatdistrict/StatelevelandassistancetostatesbyNationalAccreditationBoardforHospitalsandHealthcareProviders(NABH)forqualitycertification

Source: Srivastavaetal.(2013),Chaturvedietal.(2015)

In2006,theMoHFWinitiatedaconditionalcashtransferprogrammeknownastheJanani

SurakshaYojana(JSY)thatpaysacashincentivetowomenattendinginstitutionsforbirth.

TheJSYhasbeenaleadingprogrammeoftheNationalRuralHealthMission(NRHM)ofthe

GovernmentofIndia.331Themonetaryincentivesforwomendiffersbasedonthecontextof

individualstates.Inlowperformingstates,theJSYprogramprovidesacashincentiveofINR

1400andINR1000(equivalent£12-17)towomengivingbirthinapublicoraccreditedprivate

healthfacility.Inhigh-performingstates,thecashincentiveisabouthalfofthatamountand

Page70of248

is restricted to women living below the poverty line and those from marginalised

communities.332

TheNRHMhasalsoappointedcommunityhealthworkersknownasAccreditedSocialHealth

Activists(ASHAs)ineveryIndianvillage.333Motivatingpregnantwomen,accompanyingthem

toinstitutionsforchildbirthandarrangingsuitabletransportationtohospitalsfallsunderthe

responsibilitiesoftheASHAs,whoreceiveINR600(equivalent£7)forthesetasks.

Intheinitialyears,JSYbenefitswererestrictedtowomenabove19yearsofageandtowomen

withparityofuptotwo,butduetooppositionfromadvocacygroups,theserestrictionswere

laterremoved.Therewerealsootherconditionsthatwomenhadtofulfillsuchascompleting

allantenatalcarevisitswhichwaslaterremovedsinceitwasnotfeasibletomonitorthis.A

mandatary48-hourpostpartumstayathospitalswasalsoaconditionforobtainingpayments

butthishasalsonotprovedpracticalsincemanywomenprefertobedischargedearly332and

vacatingbedsforotherclientsisanimportantpriorityespeciallyinhigh-volumefacilities.The

JSYisoneofthelargestconditionalcashtransferprogrammesintheworld,withanestimated

80millionbeneficiaries.Despitecontributingtoremarkableincreasesininstitutionalbirths,

resultsfrommanyevaluationshavenotfoundassociateddeclinesinmortality.10,11,64

3.5:ContextofthePhDresearchwithintheMatrikaproject

MyPhDresearchbenefittedfromfundingbyMSDforMothersobtainedbyLSHTMacademics

leading the external evaluation of Matrika project funded by MSD for mothers and

implemented by two NGOs – Pathfinder International (lead) and World Health Partners

(partner)inUttarPradesh.

The aim of the Matrika project was to increase access to, and use of, basic emergency

obstetriccareandfamilyplanningservices.ItoperatedinKanpurNagar,KannaujandKanpur

DehatdistrictsofUttarPradesh.TheprojectworkedtowardsthreeobjectivesbetweenMarch

2013toMay2016;(1)Establishasocialfranchisenetworkofprivateprovidersandfunctional

referral centres offering affordable antenatal care, emergency obstetric care, and family

planningservices;(2)Strengthencapacityofandlinkagesbetweenruralprivateandpublic

sectorhealthproviderstoofferhighqualityservices;and(3)Improvecommunityawareness,

demandandlinkageswithmaternalhealthservicesamongruralpopulations.

Page71of248

TheoverallimpactevaluationofMatrikausedquantitativeandqualitativeresearchmethods

toassesstheimpactoftheprojectonawiderangeofstudyoutcomes,coveringhealthcare

utilisation,qualityofcare,patientexperience,healthybehaviours,healthstatusandfinancial

strain.Thefindingsfromtheimpactevaluationshowedthatthemulti-facetedprojectdidnot

haveameasurableimpactonthevastmajorityofoutcomes,withtheexceptionofasmall

effectonrecommendeddeliverycarepractices.Notably,Matrikawasfoundtohavenoeffect

on antenatal care (ANC) utilisation, ANC content of care, or ANC knowledge and

preparedness.

TheMatrikaevaluationwasledbyDr.TimothyPowell-Jackson,Ms.LovedayPennKekanaand

Dr.AndreiaSantosatLSHTMandwasdoneincollaborationwithanIndianresearchagency

calledSambodhiResearchandCommunications(KultarSingh,PareshKumarandDr.Kaveri

Halder)based inNewDelhi, India.MyPhD researchworkwasdonealongside theoverall

impact evaluation and contributed some important information to the impact evaluation.

Apartfromreceivingregularguidancefrommyco-supervisorDr.TimothyPowell-Jackson,

andguidancefromMs.LovedayPenn-Kekanaontheanalysisofqualitativedata,noneofthe

other academics involvedwith the largerMatrika impact evaluation provided substantial

inputsonmyPhD.Iconceivedandexecutedallaspectsofthethreeseparateresearchstudies

describedinchapterssix,sevenandeightofthisthesis.

Chapter4:Roleofthecandidate,fundingandresearchtimeline

4.1:Theroleofthecandidate

IamamedicaldoctorfromNepalandgraduatedfromtheKathmanduUniversityin2004.I

thenreceivedfurthertraininginpublichealthfromtheUniversityofAberdeengraduatingin

2007. After that, I went back to Nepal and worked in a variety of positions with non-

governmentalorganizations,bilateraldonorsandUNagenciesinNepalandothercountries

in southAsia and south-eastAsia. The focus ofmyprofessional career has been towards

improvingmaternalandnewbornhealthinresource-constrainedsettings.

InSeptember2013,IjoinedtheLSHTManddevelopedmyPhDresearchideaonQoCduring

labourandchildbirth,afterdiscussionwithseveralacademicsattheLSHTM,myfuturecareer

interests and the scope to conduct this work within the Matrika evaluation. Prior to

finalisation of the protocol, I undertook a preparatory field-visit, designed all the data

collectiontools,finalisedtheresearchmethodologyandsubmittedanethicsapplicationfor

thestudy.IwrotemydoctoralresearchprotocolthatwasapprovedbyLSHTMexaminersas

apartofmyupgradingdocument.

IlivedinLucknow,UttarPradesh,Indiaforapproximatelysevenmonthsleadingallresearch

activitiesrelatedtothePhD.Duringthattime,I initiatedcontactwithgovernmentofficials

from National Rural Health Mission and obtained the necessary permissions prior to

approaching hospitals. I coordinated and managed relationships with local partners at

Sambodhi,PathfinderandWorldHealthPartners. I finalisedall thestudy instrumentsand

providedoversighttothe:I)translationofthetoolsintoHindi;ii)pretestingofthetools;and

iii)developmentofthedatacollectionmechanisms.

Iconductedapilotstudytofinalisetheoveralllogisticsforthestudy.Icarriedoutsampling

asdescribedintheprotocolandmadesiteselectionvisitstomosthospitalsinthethreestudy

districts.Iprovidedmanagerialandtechnicaloversighttotheprimarydatacollectionefforts

andkeptallpartnersinformedoftheprogress.Idevelopedatrainingmanualusingclinical

training skills methodology and conducted the trainings myself to ensure that field-

researchersweretrainedtocompetencyovera5-daytrainingperiod.Iensuredthatthedata

werecheckedregularlyforqualityandconsistencyduringtheentiredatacollectionphase.I

travelledtoallthetwenty-sixhospitalsnumeroustimesduringdatacollectionandmanaged

Page73of248

all logisticalandadministrative issues. Iwasresponsible fordeploying14-fieldresearchers

andthreefield-supervisorsduringthistime.

For themanagement survey, I adapted a pre-existing tool so that it was relevant to the

contextofUttarPradeshandledathree-dayorientationprogrammetofield-researcherson

management.Ipilotedthemanagementsurveyinstrumentatonehospitaloveraday,made

required changes to the tool after piloting and then finalised the survey instrument. I

conductedall the interviewswithmanagersat thirty-threehospitalsmyself,prepared the

transcripts,enteredthedata,ensureddataqualityandconsistency.

Duringmy time in Uttar Pradesh, I also provided regular updates tomy supervisors and

incorporated their feedback into the on-going work. I was responsible for all the data

cleaning,dataanalysisandinterpretationofalltheworkinthisthesis.Iwrotethefirsttwo

draftsofall thethreepapersthathavebeen includedasresultschaptersandIhavebeen

managingtheprocessofcollaboratingwithallmyco-authorsandsupervisors.

IwashiredasaresearchassistantbyMETfordoingthisworkinUttarPradeshandpaida

monthlysalaryforthedurationofthefieldwork.Inaddition,theQoCtoolsthatIdeveloped

were used for subsequent studies in Uganda and in Rajasthan, where I conducted the

trainings.AsapartofmycontractwithMET,IalsosubmittedapreliminaryreportonQoCand

managementpracticesatmaternityfacilitiesinUttarPradeshtoMSDformothers.Although,

my research provided important information to the largerMatrika evaluation, I was not

involvedinotheraspectsofthelargerimpactevaluationstudy.

4.2:Funding

Funding for this research was obtained from Merck Sharp & Dohme Corp. (“MSD”), a

subsidiaryofMerck&Co.,Inc.,Kenilworth,NJ,USA,throughitsMSDforMothersprogramme.

Fundingwasusedforgeneral financialsupport,salaries, travelandoverheadcostsandall

datacollectionactivities.Non-financialsupportwasreceivedfromSambodhiResearchand

Communications,NewDelhi,Indiawhoprovidedtwo-researchassistants’pro-bonothatwere

involvedinthemanagementsurvey.MSDhadaroleinthedesign,collection,analysisand

interpretationofdata,inwritingofthethesisorthemanuscriptscontainedinthisthesisorin

thedecisiontosubmitthemanuscriptsforpublication.

Page74of248

4.3:Researchtimeline

ThissectionprovidesatimelineofactivitiesinvolvedinconductingmyPhDresearch,which

startedwhenIenrolledintoLSHTMinlateSeptember2013.Asmentionedearlier,overthe

courseofmyPhDwork,IspentapproximatelysevenmonthsinUttarPradesh.Furtherdetails

onthetimelineofthePhDarepresentedintable6below.

Table6:TimelineforthePhD

Year Month Activities

2013-2014

September2013-Dec2014

• Preparatoryreadingandformativeresearch• Finalisedresearchtopic• Conceptnotefinalisation• ExploratoryvisittoIndia• Tookrelevantmodules:Extendedepidemiology,Statistical

MethodsinEpidemiology,AdvancedStatisticalMethodsinEpidemiology,Datamanagement

• Researchinstrumentdevelopment• Submittedforlocalethics• Upgradingdocumentfinalisedandupgradingseminar

2015

Jan–April2015

• Preparatoryworkforfieldworkbegins.• TranslationofdatacollectiontoolsintoHindiandback-

translation• Conductedfurtherpretestingandformativeresearchto

finalizetools.• SubmittedtoLSHTMethics• Resubmissiontoethics

May–July2015

• TravelledforfieldworktoUttarPradesh.• MeetingwithlocalpartnersatPathfinderandWHP• MeetingwithMoHandhospitalauthorities• Conductedpretestingofthetoolsandtrainingmanual• ConductedapilotstudyforQoCassessments:3daysat

publicsectorand4daysatprivatesectorusing2observersforprivatesectorand4forpublicfacility

• Amendmentsoftoolsbasedonresultsofthepilotstudy• PilottestedthedataentrysoftwareinCS-pro• Regularcommunicationwithsupervisors.• DatacollectionbeginswithclinicalobservationsandQoC

studyends.

2015

Aug-Oct2015 • SiteSelectionforthemanagementsurvey• Trainingtofieldresearchers• Pilotstudyatonepublicsectorhospitalx1day• Amendmentofthedatacollectioninstrument• Datacollectionanddataentrybegins.

Oct-Dec2015 • Datacleaningandpreliminaryanalysis

Page75of248

• SubmittedreporttoMSDformothers

2016 Jan-Dec2016 • Dataanalysisandwritingup• Finalizedpaper1• Finalisedpaper3

2017 Jan-July2017 • Dataanalysisandwritingup• Resubmissionofpaper1-acceptedatBWHO• LearntqualitativeanalysisusingNVIVO• Finalizedpaper2-submittedatRHjournal• Finalisedfirstdraftofpaper3• Finalisedthesisandsubmitted

Chapter5:Conceptualframework,aims,objectivesandstudydesignThis chapterprovidesanoverviewof theconceptual framework,aims,objectivesand the

studydesignofmyPhDstudy.ThefirstsectionoutlinestheconceptualframeworkofmyPhD

andtheaimandobjectives.Ithendiscussthestudydesignwithsectionsonthedevelopment

of the data collection instruments, sample size calculations, sampling strategy, study

participants,dataanalysis,ethicalissues,anddatamanagement.

5.1:ConceptualframeworkformyPhD

A conceptual framework is the composition of various concepts developed from the

theoretical underpinnings to guide and better explain the proposed research work. The

conceptualframeworkformyPhD(Figure5)wasdevelopedtoassessQoCfornormallabour

andchildbirthatpublicandprivatesectormaternityfacilities(BEmOCorhigher)inUP,India

andcombinestheDonabedianQoCcausalchainmodel(structure,processandoutcome)145

withtheHulton239andWHOframework89.

Atthelevelofmaternityfacility,QoCcomprisesofclientsexperiencesofcareandclinicalcare

provision239.Experiencesofcaredependuponinterpersonalaspectsofcarereceivedduring

thelabourandbirthingprocesssuchasensuringprivacy,allowingabirthcompanion,freedom

tochoosebirthingposition,righttoinformation,respectforchoiceandpreferences,freedom

fromdiscrimination and others.334 Clinical care provision or adherence to evidence-based

practices depends upon numerous factors such as organizational factors, financing,

infrastructure,healthworkforce,existingqualityimprovementmechanismssuchauditsand

feedbackmechanisms.174ThiscorrespondstostructureinDonabedian’sframework;i.e.the

contextinwhichcareisprovided.145

ThesecondboxintheframeworkrelatestomyoperationaldefinitionofQoCwhichisthe

applicationofevidence-basedguidelinesandrespectfulmaternitycarepracticesbymaternity

care personnel. This corresponds to theprocess element in Donabedian’s framework.145

Finally,thelastboxcorrespondstooutcomessuchasimprovementsinclinicaloutcomes335

(reduction of maternal, neonatal deaths, disability and complications) and positive client

experiences.157Although,acomprehensiveconceptualframeworkhasbeenpresented,my

Page77of248

doctoral research focuses on assessment of selected structure and process measures.

OutcomemeasurementwasbeyondthescopeofthisPhDstudy.

Figure5:ConceptualframeworkformyPhD

5.2:AimsofthedoctoralresearchUltimately, my doctoral research aims to provide policymakers, public health managers,

academics and clinicians with novel information about QoC during normal labour and

childbirthinUttarPradesh.IwillalsoexaminewhethermanagementpracticesinfluenceQoC

at26publicandprivatesectormaternityfacilitiesinU.P,India.

5.3:SpecificObjectives1. TodescribeQoCfornormallabourandchildbirthat26maternityfacilitiesandtoexamine

whetherQoC isassociatedwithcharacteristicsof thewomen,characteristicsofhealth

workersandcharacteristicsofmaternityfacilitiesinthreedistrictsofUttarPradesh,India.

2. To investigate and describe patterns of mistreatment encountered by women during

labour and childbirth at 26 public and private sectormaternity facilities and examine

whethermistreatment isassociatedwithsocio-demographiccharacteristicsofwomen,

characteristicsofhealthworkersandcharacteristicsofmaternityfacilities.

3. To assess and describe existing management practices at 33 maternity facilities and

examinewhetherthereisarelationshipbetweenQoCandmanagementpracticesat26

maternityfacilitieswhereclinicalobservationshadtakenplace.

Page78of248

Inadditiontoadvancingtheevidence-baseonthesetopics,Ihopethatthedatacollection

instrumentsdevelopedinthisstudywillbeusefulformeasuringandimprovingQoCinlabour

andchildbirthinhigh-burdensettingsofsouthAsiaandsub-SaharanAfrica.

5.4:Studydesign

Thisthesisusescross-sectionaldatasetsfromtwoseparateresearchstudies.Overaseven-

monthperiod, I led twoprimarydatacollectionefforts in threedistrictsofUttarPradesh,

India.Toaddressobjectiveoneand two, I conductedclinicalpracticeobservationsof275

mother-newbornpairsat26publicandprivatesectormaternityfacilitiesutilisingastructured

tool designed to assess QoC during normal labour and childbirth including aspects of

mistreatmentofwomenatmaternityfacilities.

To address objective three, I conducted a separate cross-sectional survey by interviewing

hospitalmanagersat33maternityfacilitiesbyusingapreviouslytestedsurveyinstrument

thatwasadaptedtothecontextof ruralUttarPradesh. These33maternity facilitiesalso

includedthe26facilitieswhereclinicalobservationshadtakenplace.

5.5:Datacollectioninstruments

5.5.1:Qualityofcareassessments

Fordevelopingtheclinicalpracticeobservationtools,Ireviewedavailableguidanceonbest

practicesinmanagementofnormallabourandchildbirth.TheseincludedtheWHOguidelines

for care during normal labour and childbirth,204,336NICE guidelines for intrapartum care, 5

AMDDEmONCneedsassessmenttools22andresearchinstrumentsfromtheGaalastudy337.

Ialsoconductedexploratoryvisitstothestudysitesinordertounderstandthesocio-cultural

factors, maternity-facility context, health worker characteristics, facility caseloads and

existingmaternitycarepathwaysathealthfacilities.Afterreviewingtheavailableliterature,

andlearningfromthefieldvisits,Idevelopedanearlyversionofthestudyinstrumentsthat

underwentpeer-reviewsbyIndianandLSHTMresearchers.

Ialsopre-testedthesetoolswithfield-researchersinLucknow,UPandmaderelevantchanges

afterpre-testing.Apilotstudywasalsoconductedtotestthefeasibilityofmethodsanddata

collectionproceduresover3daysinasampleofpublicsectorfacilitiesand4daysinasample

ofprivatesectorfacilitiespriortofinalizingtheclinicalpracticeobservationtools.

Page79of248

Overall, the QoC assessment tool has three sections. The first section is a screening

questionnairethatcapturedmedicalandobstetrichistoryfromclientcaserecordstoensure

participants fulfil case definitions for normal labour and childbirth. Normal labour and

childbirthwasdefinedaslaboursthatarespontaneousinonset,low-riskatthestartoflabour

withasingletonpregnancy,inavertexpresentation,withagestationalagebetween37to42

completedweeksofpregnancy.Thesecondsectioncapturedinformationondemographic,

socio-economicandeducationalstatuswhichwasadaptedfromtheNationalFamilyHealth

Survey questionnaire (2014-2015).338 The third section included modules that captured

provisionoftechnicalinterventionsandrespectfulmaternitycareprovisionfromthetimeof

admission of women up to one-hour post-partum. This section was developed based on

reviewofWHOIntegratedManagementofPregnancyandChildbirth(IMPAC)guidelinesand

NICEguidelinesforcareduringnormallabourandchildbirth.204,339TheentireQoCassessment

toolisavailablefromappendix1.

5.5.2:Surveyonmanagementpracticesatmaternityfacilities

Iadaptedandusedamanagementsurveytoolthathaspreviouslybeenusedformeasuring

managementpracticesindiversehospitalsettingsincludinginIndia282,308,309andtailoreditto

be applicable to the context of health facilities in rural Uttar Pradesh. Essentially, this

interview-based tool assessed management practices at hospitals through four key

managementdomainsasdescribedpreviously.Questionswerestructuredbutopen-ended.

Ascoringgrid(between1to5)wasusedbyinterviewerstogivescoresforresponsestoall

questionsdependingonhow closely answersmatcheddescriptors for eachquestion. The

entiremanagementassessmentinstrumentisavailablefromappendix2.Briefly,operations

managementandperformancemanagementsectionsofthetoolassessedhowwellmodern

managementtechniqueswereappliedatmaternityfacilities,whethersystemsforcontinuous

improvementexisted; andwhether facilityperformancewas regularly trackedwithuseful

indicators.Targetmanagementsectionassessedwhetherappropriatetargetshadbeenset,

whethertheypushedmaternity facilitiesto improvetheirperformance,andwhetherthey

were communicated, effectively throughout the hospital. People management section

assessedwhethertherewasanemphasisongoodhumanresourcepractices.42,277,307,310

Page80of248

5.6:Samplesizecalculations

5.6.1:ForclinicalpracticeobservationsTheprimary focusonmystudy is toexamine thequalityofnormal labourandchildbirth,

whichisthecommonesteventatanymaternityfacility.Clinicalobservations,therefore,were

themostappropriatemethodtoassessprocessesofcare. Samplesizecalculationswere

doneinthecontextofadefinedevaluationwithearmarkedfundsfordatacollection.Sample

sizeestimationwasdictatedbylogisticalfeasibilityofobtainingnecessarycasesforclinical

observationsandresponseratesat individualpublicandprivatesectorfacilities.Obtaining

goodresponseratesandadequatenumberof‘normalvaginalbirth’caseswasfoundtobe

particularlychallengingintheprivatesector.ToensurethatIfollowedascientificapproach

forestimatingtherequirednumberofclinicalobservationsatmaternityfacilities,Iconducted

powercalculationsforimportantindicatorsofinterestsuchasActiveManagementofThird

StageofLabour(AMTSL),partographandoxytocinuse.

For calculating the required numbers of observations, I used themethodology of cluster-

randomisedtrials,withclustersdivided into1)Publicsectorfacilitiesand2)Privatesector

facilities.Thetotalnumberofclustersrequired,denotedby!,iscalculatedusingthefollowingequation340which ismultipliedbytheconstant(4/3)toaccountfortheapproximately3:1

ratioofpublic(n=18)andprivatefacilities(n=7).341

! = 1 + (&' ( + &))( ,- 1 − ,- / + ,0 1 − ,0 / + 1((,-( + ,0()

(,- − ,0)(×2× 43…… . (1)

Where,&' (isthelevelofsignificanceand&) ispower,,0and,-aretheproportionsofuseofevidencebasedpracticesatfacilitiesinbothsectors,/isthenumberofclinicalpractice

observationsateachhealthfacilityand1isthecoefficientofvariation,ameasureofvariation

betweenheathfacilities.

ForActiveManagementofThirdstageoflabour:TherecentStantonet.alstudyfoundthat

AMTSLwaspracticed in10%of all births inpublic sector facilities in similarneighbouring

districtsofUP.242Theaboveformulashowsthat,assumingacoefficientofvariationof0.25,

thenif10observationseachareconductedat18publicsectorfacilitiesand7privatesector

Page81of248

facilities,thestudywillhave80%powertodetectdifferencebetween10%useofAMTSLin

publicsectorfacilitiesand28%useofAMTSLinprivatesectorfacilities.

Partographuse:Arecentcross-sectionalstudydonein44publicsectorfacilitiesinasimilar

neighbouringstateofIndiafoundthatpartographwasusedin11%ofallbirths.243Usingthe

aboveformula,assumingacoefficientofvariationof0.25,thenif10observationseachare

conductedat18publicsectorfacilitiesand7privatesectorfacilities,thestudywillhave80%

powertodetectdifferencebetween11%partographuseinpublicsectorfacilitiesand30%

partographuseinprivatesectorfacilities.

Useofoxytocics: Iyengaret.al’s study inaneighbouringstateofUttarPradeshfoundthat

oxytocin was given in 57% of all childbirths243. Using the above formula and assuming a

coefficientofvariationof0.25,thenif10observationseachareconductedat18publicsector

facilitiesand7privatesectorfacilities,thestudywillhave80%powertodetectadifference

betweentheuseofoxytocinin57%ofpublicsectorfacilitiesand88%oxytocinuseafterthe

birthofthebabyinprivatesectorfacilities.

5.6.2:Theassessmentofmanagementpractices

The assessment ofmanagement practices atmaternity facilities was purposive. Separate

samplesizecalculationswerenotdoneforthemanagementsurvey.Instead,I interviewed

managersatallthefacilitieswhereclinicalpracticeobservationshadtakenplace.

5.7:Samplingstrategy

5.7.1:Clinicalpracticeobservations

Iusedamultistagesamplingmethod.Theinitialsamplingframeincluded59facilitiesinUttar

Pradesh thatprovidedmaternity services:all29of the largerpublic facilities listedby the

IndianDepartmentofHealthi.e.facilitiesthatreportedatleast200deliveriespermonth342

and in theorywere round-the clockBEmOC sites. In addition, I also identified30private

facilities that, in theory, provided continuousmaternity care. The private facilities were

identifiedbykeyinformantsfromSambodhiResearchandCommunications(Lucknow,India),

anorganizationthathasworkedinhealthresearchinthestudydistrictsforseveralyears.47

In thesecondstageofsampling, Iattemptedtoselectsixpublic facilitiesperdistrict.This

includedarandomselectionoffourcommunityhealthcenters,onemedicalcollegeandone

districthospitalperdistrict.SinceKanpurDehatdidnothaveamedicalcollege,Ihadtoselect

Page82of248

anadditionaldistricthospital.Weinvitedthe18selectedpublicfacilitiesandall30private

facilitiestoparticipateinourstudy;however,13facilities–allprivate–refusedtoparticipate.

Among the nine private facilities that agreed to participate, no deliveries occurredwhile

observerswerepresent.TheobservationaldatathatIanalyzedthereforecamefrom18public

and eight private facilities. Further details on the sampling strategy and the study flow

diagramareprovidedinchapter6andfigure6andpublishedelsewhere47.

5.7.2:ManagementsurveyA purposive sampling technique was utilized and all maternity facilities where clinical

observationshadtakenplacewereselectedforthemanagementsurvey.Allselectedfacilities

had complex organizational structures, defined as facilities with separate administrative,

information,therapeutic,diagnosticandsupportservicesandgreaterthanfivepostnatalcare

beds.Ireceivedabetterresponserateandwasabletointerviewmanagersat33facilities

whereasclinicalobservationscouldonlybeobtainedin26maternityfacilities.Thesampling

strategyforthemanagementsurveyisillustratedinFigure10inchapter8.

5.8:Datacollection

5.8.1:Clinicalpracticeobservations

At health facilities, female observers with nursing or midwifery backgrounds visited

admissions, emergency, labour roomand postnatalwards to identify pregnantwomen in

latentphaseoflabour(regularuterinecontractionswithcervicaldilatationlessthan4cm),

whoarelikelytoundergonormalvaginaldeliveries.Theyprovidedinformationsheetsand

consentformstothesewomenandobtainedaninformedconsentfollowingethicalconsent

procedures. After obtaining informed consent, they collected background information on

women’smedicalandobstetrichistoryfromtheircaserecordstoensurethatshewaseligible

forparticipationinthestudy.Theyprospectivelyobservedcareprovidedtothesepregnant

womenduringtheentiredurationoflabourandchildbirthuptoonehourpostpartum,using

a structured, paper- based, clinical observation toolwithout interfering in any aspects of

clinicalcareprovision.Accompanyingfamilymembersorcompanionswerealsoapproached,

consent takenanddetailed informationondemographic, socio-economicandeducational

characteristic of women was collected from them, to minimise distress to the labouring

woman.

Page83of248

5.8.2:Assessmentofmanagementpractices

Iestablishedtelephonecontactwithfacilitymanagersearlyonandsetupappointmentsto

ensureahighresponserate.Themanagementsurveywasconductedasafollow-upactivity

to the QoC assessments. Interviews were presented as confidential conversations about

management experiences and challenges and did not cover sensitive issues, for example

financialearningsofthehospital.Theparticipantswerenotawarethattheywerebeingrated

fortheirresponsestothemanagementquestionnaire.Alltheinterviewsweredouble-scored;

whileIconductedalltheinterviews,anotherresearcheralsoscoredthemindependently.

5.9:Studyparticipants

Clinicalpracticeobservationsoflabourandchildbirth:Studyparticipantsincludedpregnant

womenwithspontaneous,uncomplicatedlabouroperationallydefinedaswomenwithlow-

risk, gestational agebetween37and42 (+0)weekswith singletonpregnancywith vertex

presentation admitted to facilities who consented to participate in the study and their

newborns.

Assessment of management practices at maternity facilities: Study participants for the

managementsurveyincludedadministratorsorclinicalleadersat33maternityfacilities(10

privateand23publicsector).

5.10:Dataanalysis

Inthissection,Ihavesummarisedthedataanalysisplanformydoctoralstudy.However,the

individualresultschapters(chapters6to8)describethemethodsandtheanalysisplanfor

eachobjectiveingreaterdetail.

The data obtained from clinical practice observations (for objective 1 and 2) and the

management survey (objective 3) were coded either as binary, continuous or categorical

variables. Both QoC and management datasets were double entered. Frequencies were

calculatedforallvariables,andoutliersorerrorsinthedatasetwereidentified.Incasesthere

wereinconsistencies,Iwentbacktothepaper–basedquestionnairesandverifiedtheentered

data. After crosschecking for data accuracy and completeness, I conducted appropriate

statisticalteststoanswerthestudyobjectivesusingSTATA14(StataCorp.LP,CollegeStation,

UnitedStatesofAmerica).

Page84of248

Forobjective1:

Foreveryclinicalobservation,Iassessed42clinicalitemsrelatedtoQoCduringlabourand

childbirth. Each item was coded as 1 if completed, and 0 otherwise. I finalised a

comprehensiveframeworktoassessQoCduringlabourandchildbirth,bymappingthese42

clinical items into17overall essential carepractices.Nineof these clinical practiceswere

related to obstetric care and eight clinical practiceswere related to neonatal care. Some

practiceswerebasedonasingleitem(e.g.,earlyinitiationofbreastfeeding).Otherpractices

werebasedonmultipleitems(e.g.ActiveManagementofThirdStageofLabour). Further

detailsonthedevelopmentoftheQoCindicesusingduringanalysisareprovidedinchapter

six.

Iappliedprincipalcomponentanalysistodataonownershipofacommonsetofassets,and

thereby,generatedquintilesofwealthstatusforindividualwomen.343Ialsoappliedweights

using data on facility caseload of normal deliveries, the idea being to correct

underrepresentationoffacilitieswithfewercases.

DescriptiveanalyseswasconductedatthelevelofindividualwomenusingSvycommandin

STATA to account for clustering of patients within facilities. Prevalence, proportions,

frequencies, andmeanswere calculated for covariates disaggregatedby public or private

sector.Summaryscoresforobstetriccare(ninepractices),newborncare(eightpractices)and

anoverallessentialcareatbirth index(17practices)werecalculatedasthepercentageof

practicescompletedperwoman(i.e.thenumberofpracticesdonedividedbythenumberof

practicesmeasuredwithintheQoCdomain).

For investigating whether QoC was associated with characteristics of the women, health

workersandhealth facilities, Iusedatwo-level linearmixedeffectsmodelwitharandom

effectat the facility level toaccount forclustering.344Theexposurevariablewaspublicor

private sector and the explanatory variables were women’s characteristics (parity, age,

referral status, caste, wealth, time and day of admission), health worker characteristics

(deliverybyqualifiedpersonnel,dutyhours)andfacilitycharacteristics(volume).Ialsoadded

a dummy variable for each observer in the regression model to mitigate biases across

observers.

Page85of248

Forobjective2:

I analysed quantitative data on 15 potentially harmful interventions obtained from

quantitativechecklistusedduringclinicalobservationsof275normallabourandchildbirthin

maternity facilities. I also used qualitative data obtained from open-ended observers’

commentsrecordedattheendofeveryclinicalobservation.

Forthequantitativeanalysis,eachitemofmistreatmentwascodedas1ifobserved,and0

otherwise. An aggregate score for mistreatment was calculated for every woman, which

rangedfrom0-15.Descriptiveanalyseswerecarriedoutatthelevelofindividualwomento

describe patterns ofmistreatment that occurs atmaternity facilities. I then conducted a

bivariateanalysistoexaminetherelationshipbetweenindicatorsofmistreatmentandsocio-

demographic characteristics of women. Means, proportions and a summary total

mistreatmentscorewerecalculatedforallcovariates.Chisquaretestswereusedtoassess

whether there was a significant difference amongst the use of practices considered

mistreatment and the relevant co-variates. Since, this paper was conceptualised as a

descriptivepaperwrittentodocumentandexplainthecontextandreasonsformistreatment,

I did not conduct any advanced regression analysis. Instead, I used qualitative insights

obtainedfromobserver’scommentstofurtherexplainquantitativedataonmistreatment.

For analysing the qualitative data obtained fromobservers’ comments, I used a thematic

approach to data analysis using Nvivo 11 software (QSR International). Comments that

summarisedsimilarfindingsacrossobservationswereusedasexamplestodescribedifferent

themesofmistreatment.

Forobjective3:

Two separate analyses were done to address objective three. First, to analyse the

determinantsofmanagementpracticesat33maternityfacilities,Icalculatedtotalscoresfor

overall management, operations management, performance management, target

managementandpeoplemanagementatmaternityfacilities.Ithencategorisedthesampled

maternityfacilitiesbasedontheirmanagementscoresandconductedadescriptiveanalysis

ofthedeterminantsofmanagementpracticesatmaternityfacilities.

Second,toinvestigatetherelationshipbetweenmanagementandQoC,ImergedtheQoCand

managementdatasets,whichmeantthatdatawereanalysedatthelevelof275individual

Page86of248

women. I then calculated Z scores for overall management and specific management

dimensionstostandardizethesemanagementscores.Aswiththeearlieranalysis,Iapplied

samplingweightssothatequalweightsweregivenforobservationsateachfacility,thereby,

correctingforunderrepresentationoffacilitieswithfewercases.Therelationshipbetween

QoC (outcome) and total management and specific management Z scores (explanatory

variables)werethenanalysedusingmultilevel,mixed-effectslinearregressionmodels.The

modelsincludedrobuststandarderrors,accountedforclusteringattheleveloffacilitiesand

usedsamplingweights.Ialsoincludedadummyvariableforobserverratingsandcontrolled

forrandomeffectsatthelevelofindividualfacilitiesandhealthworkers.

5.11:Researchethics

EthicalapprovalforthestudywasobtainedfromthePublicHealthcareSociety(PHS)Ethics

Review Board in India and the London School of Hygiene & TropicalMedicine in the UK

(LSHTMEthicsRef:8858)whichincludedspecificdetailsontheQoCassessmentsandthefinal

study instruments. The study protocol also received clearance from the National Health

MissioninUttarPradesh.

ThefocusofthisPhDresearchwasontheobservationsoflabourandchildbirthatmaternity

facilities. It did not involve clinical interventions or other controversial issues such as

collectionofbiologicalsamplesorconductofclinicalexaminations.Alltheinvestigators,field

researchersandstafffromthelocalresearchpartnerwereexternalanddidnothavearolein

project implementation or provision of services at maternity facilities. All the observers

employedforclinicalpracticeobservationswerefemales.Effortsweretakentoensurethat

respect,dignity,privacyandculturalsensitivityweremaintainedasmuchaspossible.

Theobserverswereprovidedafive-daytrainingincludingfourhoursspecificallydevotedto

research ethics and informed consent as a part of their training. Unless, therewere life-

threateningemergencies, theobserverswere instructednot todirectly intervene in cases

wheretheyobservedmedicalmistreatmentorininstanceswheresubstandardcarewasbeing

delivered.Inthetrainingmanuals,Ihadalsodesignedvariouscase-studiesoutliningdifferent

scenariosthatobserverscouldcomeacrossandthesuggestedprocessofdealingwithsuch

instances.Theobserverswereinstructedtoreportsuchinstancestomeattheendofevery

Page87of248

dayandIdiscussedtheseissueswithfieldsupervisorsandinformedthefacilityin-charges,as

appropriate.

Informed written consent was taken prior to the start of data collection from all health

workers,allthelabouringwomenandmanagersofhealthfacilitiesthatparticipatedinthe

study.Aninformationsheetwithdetailsontheobjectivesoftheresearch,confidentialityof

datacollected,thevoluntarynatureofparticipation,thepossibilityofrefusaltoparticipate

atanytimewithoutprovidingaspecificreason,andmycontactdetailswereprovidedtoall

theparticipantspriortothestartofdatacollection.Allparticipantsinthestudyprovideda

written informed consent. The copies of the ethical approval letters and samples of the

consentandinformationformareprovidedinappendix3and4.

5.12:Datamanagement

Dataconfidentialitywasmaintainedthroughsecurestorageofpaper-basedquestionnaires,

anonymizationofdataonceenteredandsecurestoragewithpasswordprotection.Allefforts

weremadetoensurethattheriskofconfidentialitybreacheswasminimum.Datacollected

fromQoCassessmentandmanagementsurveywereonly linkedatthetimeofanalysisby

me. Data were not shared with anyone apart from my supervisors. All completed

questionnaireshavebeenstoredinalockedcupboardandwillbedestroyedaftermyPhdis

completed.

Chapter6:Qualityofessentialcareatthetimeofbirth:Findingsfromclinicalobservationsofspontaneouslabourandchildbirthat26publicandprivatesectorfacilitiesinUttarPradesh,India.PrefaceThischapterpresentsthefirstofthreeresearchpapers,whichformtheresultssectioninmy

PhDthesis.TheobjectiveofthispaperwastodescribevariationsinQoCatthetimeofbirth

inpublicandprivate sectormaternity facilitiesandexaminewhetherquality isassociated

withcharacteristicsofwomen,healthworkersorfacilities.Thispaperutilisedprimarydata

collectedfromclinicalobservationsof275mother-babypairsat26hospitalswhichwerethen

weighted to obtain population-based estimates for the study districts. I also developed

innovativeframeworksforthemeasurementofQoCatthetimeofbirth,bydevelopingthe

overallessentialQoCindex,anindexforqualityofobstetriccareandanindexforqualityof

neonatal care. I conducted a descriptive analysis using the Svy command to incorporate

weightsandaccountforclusteringatthefacilitylevelandusedamulti-levelmixedeffects

linearregressiontechniqueusingthemixedcommandinStatatoinvestigatetheassociation

ofQoCwithcharacteristicsofwomen,healthworkersandhealth facilities. Mixedeffects

regressionsarearobustmethodthatoffersapracticalwaytoanalyseclustereddatasuchas

data from different hospitals and these techniques account for random effects and fixed

effectsinthelinearregressionmodel.

Thisisthepre-copyedited,finalauthorapprovedversionofthearticlesubmittedafterpeer

reviewtothepublishers.Sincethepublishedversionofthearticlewascopy-editedfurther

for language and style and could not accommodate many of the interesting figures and

importantdiscussionpoints,inchapter6,Ihavesubmittedthefinalauthor-approvedversion.

Ashortercopy-editedversionofthispaperwaspublishedbytheBulletinoftheWorldHealth

Organizationinaspecialthemeissueonmeasuringqualityofcare(publishedJune2017).It

is available online at http://www.who.int/bulletin/volumes/95/6/16-179291.pdf. The

publishedversionofthefinalmanuscriptisalsoavailablefromappendix5.

RESEARCH PAPER COVER SHEET PLEASE NOTE THAT A COVER SHEET MUST BE COMPLETED FOR EACH RESEARCH PAPER

INCLUDED IN A THESIS.

SECTION A – Student Details

Student Gaurav Sharma

Principal Supervisor Véronique Filippi

Thesis Title An investigation into quality of care at the time of birth at public and private sector maternity facilities in Uttar Pradesh, India

If the Research Paper has previously been published please complete Section B, if not please move

to Section C

SECTION B – Paper already published

Where was the work published? Bulletin of the WHO

When was the work published? June 2017

If the work was published prior to registration for your research degree, give a brief rationale for its inclusion

No

Have you retained the copyright for the work?* Yes Was the work subject to

academic peer review? Yes

*If yes, please attach evidence of retention. If no, or if the work is being included in its published format, please attach evidence of permission from the copyright holder (publisher or other author) to include this work. SECTION C – Prepared for publication, but not yet published

Where is the work intended to be published? The Bulletin of the World Health Organization

Please list the paper’s authors in the intended authorship order:

Gaurav Sharma, Timothy Powell-Jackson, Kaveri Haldar, John Bradley, Véronique Filippi

Stage of publication In press SECTION D – Multi-authored work

Page90of248

6.1:IntroductionThequalityofcare(QoC)offeredatmaternityfacilitiesaffectspregnantwomen,physically

and emotionally, but also impacts the survival and long-term health of mothers and

newborns.16,21 An increased focus on care during childbirth has multiple returns on

investmentthroughthereductionofmaternalandneonataldeaths,preventionofstillbirths

andfuturedisability.16,18

Manycountrieshaveadoptedpoliciestoencouragebirthsinhealthfacilitiesandglobally72%

ofalldeliveries,including69%ofdeliveriesinSouthAsiaarenowatinstitutions.345Failuresin

processes of care result in badobstetric andneonatal outcomes346,347 andpoor quality is

associatedwithlowdemandformaternalhealthservices.348,349Inaddition,aschildbirthisa

normalphysiologicalprocess,somecareprovidedcanbeineffectiveorevenharmful.114

Despite substantial efforts to promote evidence-based obstetrics, the uptake of

recommended interventions into clinical practice has been limited worldwide.119,120,337

Clinical practices are influenced bymultiple factors such as health-worker characteristics,

patientcharacteristics,task-complexity,theinstitutionalenvironment,andthesocio-cultural

environment,350,351makingpracticesdifficulttochange.

Indiaisthesecondhighestcontributortomaternaldeathsglobally(45000deathsin2015)352

andachievingthe“Survive”targetsformothersandnewbornsasapartoftheglobalstrategy

for women’s, children’s and adolescents’ health (2016-2030)6 is dependent upon future

progressinIndia.MaternityservicesinIndiaareavailableinbothpublicandprivatesectors,

fromanenormousrangeofhealthproviders.Maternitycareinthepublicsectorisprovided

throughanetworkoflevel1,2and3facilities,whichinprincipleprovideroutinecare,Basic

EmergencyObstetricCareandComprehensiveEmergencyObstetricCarerespectively.353The

privatesectorisheterogeneousandrangesfromsmallmaternityhomestolarge(100ormore

beds)multispecialtytertiaryhospitalsandmedicalcolleges.

Increasingly,theprivate-sectorhasemergedasanimportantproviderofmaternityservices.

ArecentanalysisofDemographicandHealthSurveysfor57countries(2000–2013)foundthat

theprivate-sectorshareamongappropriatedeliverieswas9–56%acrossworldregionsandit

Page91of248

is often less equitable than the public sector.33 India has a mixed health system with a

dominantprivatesectorandextremeheterogeneityoffacilities.Anestimated75%ofprivate

health facilities are micro-enterprises and the rest are medium to large medical

establishments.324Recentestimatesindicatethatupto22%ofalldeliveriesinIndiaoccurin

theprivatesector.40Womenwithpreviouslynegativepregnancyoutcomestendtochoose

privatesector.40Highersocio-economicstatusandaccessibilityareassociatedwithincreased

privatesectoruse.40Scheduledcasteandtribestatusarenegativelyassociatedwithuseof

privatefacilities.248Theprivatesectorismoreexpensivethanthepublicsectorandthereisa

perceptionthattheyprovidebetteramenitiesandahigherstandardofcare.248

Although,there isconsiderable literatureonthequalityofemergencyobstetriccare,354,355

thereislimiteddescriptiveinformationonQoCforuncomplicatedspontaneousvaginalbirths

inIndia,particularly,fromprivatesectorfacilities.Mostoftheavailableevidenceisfromthe

publicsectorandfromqualitativestudies.ThesestudieshavefoundpoorQoCwithhighrates

of labour augmentation, routine episiotomies, no choice of position, non-adherence to

protocols,limitedmonitoring,earlydischargefromthehospitalandpoorneonatalcare.71,77,78

It iswell-established that theprivate sector is adriverof caesarean section rates inmost

world-regions.36,39,356,357Inaddition,a2011studyusingmultivariateanalysisofover11000

deliveryrecordsinThailandfoundthatwomenwhodeliveredintheprivatesectorwere9.4

timesmore likely tohavehadacaesarean than thosewhodelivered inapublic sector.358

Usingprimarydataobtainedfromclinicalobservations,wesoughttodescribeQoCforlabour

andchildbirthinpublicandprivatefacilitiesandexaminewhetherqualityisassociatedwith

characteristicsofthewomen,healthworkersandfacilitiesinUttarPradesh,India.

6.2:Methods

6.2.1:Studysetting

ThisstudywasnestedwithinalargerresearchprojectinthreedistrictsofUttarPradesh(UP):

Kannauj,KanpurNagar,andKanpurDehat.359AmongstIndianstates,UPisthemost-populous

anditsmaternalmortality(258per100,000livebirths)wasthesecondhighestandneonatal

mortality (49per1,000 livebirths)wasthethirdhighest in2012-13.253Neonatalmortality

ratesinthethreestudydistrictswerehigh(Kannauj-55,KanpurDehat-41andKanpurNagar

-24per1000livebirths),aswasthematernalmortalityratio(240per100000livebirths).253

Upto39%ofdeliveriesinUPoccuratpublicsectorfacilities(43%inKannauj,46%inKanpur

Page92of248

Dehatand34%inKanpurNagar).253Theprivatesectordeliveryshareisestimatedtobe18%

inUP(15%inKannauj,34%inKanpurNagar,and10%inKanpurDehat).253Therearealso

widespread inequities across the continuum of care for all reproductive, maternal and

newbornhealthindicatorsinthethreestudydistricts.253Primarydataonestimatedcaseloads

atourstudysitesshowthatthemedianannualnumberofdeliverieswas2,216(range:1,433-

5,126)andthecaesareansectionratewas6%(range:0%-34%)inoursampleofpublicsector

facilities.Forprivatesectorfacilities,themedianannualnumberofdeliverieswas697(range:

234-2,392)andthecaesareansectionratewas32%(range:2%-59%).

6.2.2:Sampling

We used a multistage stratified sampling methodology. The sampling frame included 29

public sector facilities (obtained from the Department of Health) and 30 private facilities

(identifiedbykeyinformants).Thepublicsectorfacilitieswereeligibleiftheyhad200ormore

deliveriesmonthlybasedonHMISdata342andwereround-theclockBEmOCsites.Therewas

nosamplingframeavailablefortheprivatesectorandacensusofallprivatesectorfacilities

wasnotfeasible.Wereliedonlocalknowledgeofourcollaboratingorganisation(whichhas

workedinhealthresearchinthestudydistrictsforyears)todrawupalistofprivatesector

facilities providing 24/7maternity care in the study districts and selected all facilities for

inclusion in the study. In the second stage, 18 public facilities were randomly selected

stratifiedby typeof facilityandallagreedtoparticipate.Amongst the30private facilities

invitedtoparticipate,13facilitiesrefused.Therewerenocasesatanadditionalnineprivate

facilities during the oneweek that researcherswere stationed there. Figure 6 shows the

overallstudyflowchart.Powercalculationsweredonetoestimatetherequirednumberof

observations at each facility (discussed in chapter 5, section 5.6).We expected up to 10

observationsintwodaysperpublicsectorfacilityand10observationsperweekinprivate

sectorfacilitiesandultimately,couldobserveanaverageof12and8observationsinpublic

andprivatesectorfacilities,respectively.

6.2.3:Studyparticipants

Study participants included pregnant women with spontaneous, uncomplicated labour

operationallydefinedaswomenwithlow-risk,gestationalagebetween37and42(+0)weeks

withsingletonpregnancywithvertexpresentationadmittedtofacilitieswhoconsentedto

participateinthestudyandtheirnewborns.

Page93of248

Figure6:StudyflowdiagramfortheassessmentofQoCduringlabourandchildbirth

• 59maternityfacilitiesinUttarPradeshassessedforeligibility:

• 29fromthepublicsector• 30fromtheprivatesector

•

13privatefacilities–declinedtoparticipate

• 218deliveriesobserved

Uncomplicatedvaginaldeliveriesobservedinall18publicfacilities

• 18publicfacilitiesenrolled:• 12communityhealthcenters• 4districthospitals• 2medicalcollegeteachinghospitals

Uncomplicatedvaginaldeliveriesobservedinonly8oftheprivatefacilities

17privatefacilitiesenrolled:• 8maternityhomes• 7multispecialtyhospitals• 2medicalcolleges

• 64deliveriesobserved(noreferralsordeaths)

Randomsampleof18publicfacilitiesandall30privatefacilitiesinvitedtoparticipate

• Analysisofobservationaldataon211deliveriesinpublicfacilities

• Analysisofobservationaldataon64deliveriesinprivatefacilities

• 7deliveriesexcludedfromanalysis:

• 5neonataldeaths• 1referredforspecialist

careelsewhere• 1referredforcaesarean

section

Page94of248

6.2.4:Datacollection

WedevelopedaQoCassessment toolbasedona critical assessmentofpreviously tested

instruments337,360andWHOguidelinesforcareduringpregnancyandchildbirth.361Questions

capturing educational, demographic and socio-economic status were adapted from the

NationalFamilyHealthSurveyquestionnaire.338TheQoCassessmenttool isavailablefrom

Appendix 1. At maternity facilities, 14 trained enumerators with maternal and newborn

healthbackgroundsvisitedtheadmissions,emergency,labourroomandpostnatalwardsto

identify pregnant womenwhowere likely to undergo uncomplicated vaginal births. Two

enumeratorswere stationedat each facility andobserved round-the-clock careprovision.

Datawerecollectedafterobtainingwomen’sinformedwrittenconsentbetween26thofMay

to8thofJuly2015.EthicalapprovalwasobtainedfromthePublicHealthcareSocietyEthics

ReviewBoardandtheIndianCouncilforMedicalResearchinIndia,andtheLondonSchoolof

Hygiene&TropicalMedicineintheUK.

6.2.5:Measures:

Learning from previous quality measurement efforts,151,153 we operationalized QoC as

encompassingclinicalcareprovisionandclients‘experiencesofcare.Clinicalcareprovision

meansapplicationofevidence-basedprocesses includingprinciplesofdoingnoharmand

experiencesof care relate towoman-centred respectful carepracticesduring thebirthing

process.141Wecollecteddataon42itemsforeveryobservation.Eachitemwascodedas1if

completed,and0otherwise.Wethenaggregateditemsintoclinicalpractices–nineobstetric

carepractices,eightnewborncarepracticesand17practicesoverallforessentialcareatbirth

(Table7).Somepracticeswerebasedonasingleitem(e.g.earlyinitiationofbreastfeeding

wascoded1 if themotherwasobservedto initiatebreastfeedingwithinonehour).Other

practiceswerebasedonmultipleitems(e.g.ActiveManagementofThirdStageofLabourwas

codedas1 ifuterotonicwithin1min,cordclampingandcontrolledcordtractionwereall

done). Finally, summary scores for obstetric care (nine practices), newborn care (eight

practices)andanoverallessentialcareatbirth index(17practices)werecalculatedasthe

percentageofpracticescompletedperwoman(i.e.thenumberofpracticesdonedividedby

thenumberofpracticesmeasuredwithinthequalityofcaredomain).

The exposure variable was public or private sector and the explanatory variables were

women’s characteristics (parity, age, referral status, caste, wealth, time and day of

Page95of248

admission),healthworkercharacteristics (deliverybyqualifiedpersonnel,dutyhours)and

facility characteristics (volume). Principal component analysis was applied to data on

ownershipofacommonsetofassets,andquintilesofwealthstatusweregenerated.343

Table7:IndicesforQualityofCare

Timing Obstetric Foetal/Neonatal

Onadmissionandfirststageoflabour

Regularmonitoringoflabourusingapartograph(1item:labourmonitoredregularlywithpartograph)

Checksfundalheightandfoetalpresentation(2items:fundalheightchecked;foetalpresentationchecked)

Maternalinfectionpreventionmeasuresduringadmission(2items:hand-washingpriortoexam;sterilegloveswornpriortovaginalexam)

Foetalheartratemonitoredatregularintervals(1item:foetalheartratemonitoredatregularintervals)

Preeclampsia/eclampsiascreening(2items:BPmonitoredandurinetestedforproteins)

Duringsecondstageoflabourtocompletionofchildbirth

MaternalInfectionpreventionmeasures(2items:healthworkerwearssterilegloves,cleansthevulvaandperineumwithanantiseptic)

Healthworkerspreparedforresuscitationifrequired(2items:ventilationbagavailable;newbornmaskavailableandlaidout)

Activemanagementofthethirdstageoflabour(3items:uterotonicwithin1min;cordclamping;andcontrolledcordtraction)

Neonatalsterilecordcare(1item:sterilecordcutting)

Maternalbloodlossassessment(3items:completenessoftheplacentaandmembranes;assessingforvaginaltears;andlacerationsandmonitoringbleeding.)

Appropriatenewbornthermalcare(3items:babydried;skintoskincontact;babycoveredwithadrytowel)

Womencentredrespectfulcarepractices(5items:processoflabourexplainedtothemotherorsupportpersonatleastonce;companionallowedtobewiththemotherduringlabour;womeninformedpriortovaginalexamination;visualprivacyensured;motheraskedaboutchoiceofposition)

Apgarscore1minand5minutes(2items:Apgarscoreassessedat1minafterbirth;Apgarscoreassessedat5minafterbirth)

Initiatesearlybreastfeedingwithin1hour(1item:motherinitiatesbreastfeedingwithin1hourofbirth)

Noharmfulorunnecessaryinterventionsdoneformotherduringthelabourandchildbirthperiod(6items:noenema;nopubicshaving;noapplyfundalpressuretohastendeliveryofbabyorplacenta;nouterinelavageafterdelivery;nomanualexplorationoftheuterusafterdelivery;nouseofepisiotomywithoutindication)

Noharmfulorunnecessarypracticesforthenewbornduringtheearlyneonatalperiod(3items:noroutineaspirationofthenose;noslapthenewborn;noholdingthenewbornupsidedown)

Page96of248

Noharmfulorunnecessarybehavioursdonetothemotherduringthelabourandchildbirthperiod(3items:norestrictfoodandfluidduringlabour;noshout,insultorthreatenthewomanduringlabourandchildbirth;noslap,hitorpinchthewomanduringlabourandchildbirth)

6.3:Analysis

DescriptiveanalyseswerecarriedoutatthelevelofindividualwomenusingtheSvycommand

inSTATA14(http://www.stata.com/)toincorporateweightsandaccountforclusteringatthe

facilitylevel.Post-samplingweightswereappliedtoobtainpopulation-basedestimatesusing

dataonfacilitycaseloadofnormaldeliveries,theideabeingtogivegreaterweighttotheQoC

providedbyfacilitieswithmorepatients.Prevalence,proportions,frequencies,andmeans

werecalculatedforcovariatesdisaggregatedbysector.Atwo-levellinearmixedeffectsmodel

wasusedwitharandomeffectatthefacilityleveltoaccountforclustering.344Theregression

includedtheexplanatoryvariablespreviouslydescribedaswellasadummyvariableforeach

enumerator to mitigate biases across observers. Estimation was by restricted maximum

likelihood.WeusedaWaldtesttogenerateanoverallp-valueforeachcategoricalvariable

(e.g. agegroup) toassesswhether therewasanassociationbetweenagivenexplanatory

variableandthequalityofcareoutcome.

6.4:Results6.4.1:SamplecharacteristicsMostobservationswereconductedinthepublicsector(n=211,77%)andmostwomencame

directlytofacilities(92%)(Table8).Mostwomenwerebetween20-35yearsofage(90%),

multi-parous(56%)andbelongedtotheotherbackwardcastecategory(51%).Womenofthis

castewere inhigherproportionatprivatematernities than thepublic sector (p=0.002).A

higherproportionofprivatesectorclientswerefromthehighestquintileandthirdquintile

thanpublicsectorpatients(p=0.07).Agreaterproportionofdeliveriesintheprivatesector

(73%) compared to public sector (27%) were performed by qualified personnel (doctors,

nurses,andmidwives)(p=0.01).Agreaterproportionofcaseswereadmittedtotheprivate

sector (99.5%) during daytime work-hours compared to public (93%) maternity facilities

(p=0.003).

Page97of248

Table8:Samplecharacteristics

Unweighted Weighted

Total

(n=275)

Public

(n=211)

Private

(n=64)

Total

(n=52047)

Public

n=41512

Private

(n=10535)

Pvalue

Women’sage 0.85a. <20years 16/275(6%) 12/211(5.6%) 4/64(6.2%) 5.5% 6% 4.4%b. 20-35years 247/275(90%) 191/211(90.5%) 56/64(87.5%) 90.4% 90% 90.5%c. 35yearsormore 12/275(4%) 8/211(4%) 4/64(6.2%) 4.1% 4% 5.1%Parity 0.7a. Primipara 119(43% 90/211(43%) 29/64(45.3%) 44% 41.6% 53.4% b. Multipara 156(57%) 121/211(57%) 35/64(54.7%) 56% 58.4% 46.6%Referralstatus a. Patientdirectlytothisfacility 243/275(88.4%) 197/211(93.4%) 46/64(72%) 91.5% 96% 74% 0.003b. Patientreferredfromanotherfacility 32/275(11.6%) 14/211(6.6%) 18/64(28%) 8.5% 4% 26%Castea. “Scheduledcaste” 59/275(21.4%) 53/211(25.1%) 6/64(9.4%) 24.2% 29% 6.4% 0.002b. “Scheduledtribe” 2/275(0.7%) 0/211(0%) 2/64(3.1%) 0.3% 0% 1.4%c. “Otherbackwardcaste” 153/275(55.6%) 111/211(52.6%) 42/64(65.6%) 51.4% 49% 61.1%d. “Generalcaste” 61/275(22.2%) 47/211(22.3%) 14/64(22%) 24.1% 22.3% 31%Socio-economicstatusa. 1stquintile(lowest) 56/275(20.4%) 49/211(23.2%) 7/64(11%) 22.5% 24.2% 16% 0.07b. 2ndquintile 54/275(19.6%) 46/211(22%) 8/64(12.5%) 18% 19.5% 11%c. 3rdquintile 55/275(20%) 36/211(17%) 19/64(30%) 18% 18% 18%d. 4thquintile 55/275(20%) 46/211(22%) 9/64(14%) 19.5% 22% 10%e. 5thquintile(highest) 55/275(20%) 34/211(16.1%) 21/64(33%) 22.5% 17% 45.2%Typeofbirthattendanta. Qualifiedbirthattendant 113/275(41%) 75/211(35.5%) 38/64(59.4%) 36% 27% 73% 0.01b. UnqualifiedSBA 162/275(59%) 136/211(64.5%) 26/64(40.6%) 64% 73% 27%Admissionduringworkhours?a. Withinworkhours(9:00AM-17:00PM) 254/275(92.3%) 191/211(90.5%) 63/64(98.4%) 94.4% 93% 99.5% 0.003b. Outofhours(17:01PMto8:59am) 21/275(7.6%) 20/211(9.5%) 1/64(1.5%) 5.5% 7% 0.5%Admissionduringweekends?a. Admissionduringweekdays 211/275(77%) 158/211(75%) 53/64(83%) 77% 76% 82% 0.58b. Admissionduringweekends. 64/275(23%) 53/211(25%) 11/64(17%) 23% 24% 18%

Page98of248

6.4.2:VariationsinessentialcareatbirthacrosspublicandprivatesectorsfacilitiesTable 9 below shows the QoC by sector for each of the clinical practicesmeasured. For

obstetric care provision, monitoring of labour using partograph (2%), screening for pre-

eclampsia/eclampsia(2%),woman-centredcare(4%),noharmful/unnecessaryinterventions

(4%) and AMTSL (24%) were particularly low in both sectors. Facilities in both sectors

performed relatively better formaternal infection preventionmeasures during admission

(76%)andnoharmfulhealthworkerbehaviours(74%).However,partographuse(p=<0.001),

maternalinfectionpreventionmeasuresduringchildbirth(p=0.05)andmaternalbloodloss

assessment(p=0.01)weresignificantlybetter intheprivatesectorcomparedtothepublic

sector.Wedidnotfindanysignificantdifferencesbetweensectorsinuseofnoharmfulor

unnecessary maternal care interventions (p=0.2) or in harmful health worker behaviours

towardsmothers(p=0.45).

Forfoetal/neonatalcare,foetalheartratemonitoringatregularintervals(20%),assessment

of foetalpresentationand fundalheight (1%),andassessmentofApgar scoresat1and5

minutes(1%)wereespeciallypooracrossbothsectors.Facilitiesinbothsectorsperformed

relativelybetterforresuscitationpreparedness(68%),sterilecordcare(95%)andsupportfor

earlyinitiationofbreastfeeding(70%).However,significantdifferenceswereseenbetween

publicsector(7%)andprivatesector(73%)forfoetalheartratemonitoring(p=<0.001).Figure

7showsdatadisaggregatedfurtherbyeachofthe42itemsobserved.

Qualityofessentialcareduringlabourandchildbirthwasfoundtobedeficient(36%)across

theentiresampleofmaternityfacilities(Table9).Onaverage,45%ofclinicalpracticeswere

completed amongstwomen giving birth in the private sector compared to 33% in public

sector facilities (p=0.01). For obstetric care, private sector clients received 40% of the

recommended obstetric care practices compared to 28% in the public sector (p=0.01).

Neonatalcarewasalsobetterintheprivatesector(p=0.02)whereclientsreceived51%of

recommendedpracticescomparedto39%inthepublicsector.

Page99of248

Figure7:Qualityofcareitemsforobstetricandnewborncarebysectorusingweighteddata.

TheresultsfromthemultivariateanalysisrevealthatoverallQoCwas6percentagepoints

higher (p=0.03) inprivate sector facilities thanpublic sector facilities, after controlling for

confounders(Table10).Wefoundnoassociationbetweenuseofqualifiedpersonnel,facility

caseloadorclientcharacteristicsandoverallQoCatthetimeofbirth.Specifically,therewere

no statistically significant differences in quality of carewith respect to thewoman’s age,

parity, referral status, caste, or socio-economic status. However, admission during the

weekendswasassociatedwitha3-percentagepointpoorerstandardofcare(p=0.03).

We examined adjusted variances between healthworkers and health facilities and found

greater variation within health workers than between health workers for QoC (SD

within=0.004,SDbetween=0.002,intraclasscorrelationof0.33).Similarly,therewasgreater

variationwithinhealthfacilitiesthanbetweenhealthfacilities(SDwithin=0.005,SDbetween

=0.002,intraclasscorrelationof0.27).WefoundthatQoCdidnotchangesignificantlybythe

orderofobservation,suggestingthathealthworkerswerenotexertingmoreeffortsimply

becausetheywerebeingobserved.GraphshowinglimitedHawthorneeffectispresentedin

Figure8.

Page100of248

Table9:VariationsinessentialcareatbirthacrosspublicandprivatesectorsinUttarPradesh,India

Measures Unweightedestimates(n,%) Weightedestimates(%)

Total

(n = 275)

Public

(n = 211)

Public

sector

95%CI

Private

(n=64)

Private

sector95%CI

pvalue

Total

(n = 52 047)

Public

(n = 41 5

12)

Public

sector95%

CI

Private

(n = 10 53

5)

Private

sector95%

CI

pvalue

Clinicalpracticesforobstetriccare

Regularmonitoringoflabourusingpartograph 3(1.1) 1(0.5) 0.1to3.3 2(3.1) 0.7to12 0.07 1.6 0.2 0.1to1.9 7.2 1.7to26 <0.001

MaternalInfectionpreventionmeasuresduringadmission

212(77) 159(75.4) 69to81 53(83) 71.4to90 0.21 76.4 73.4 65to80 88.2 77to94 0.1

ScreeningforPreeclampsia/Eclampsia 3(1.1) 2(0.9) 0.2to3.7 1(1.5) 0.2to10.5 0.67 2.3 2.22 0.5to9.3 2.5 0.3to16 0.9

MaternalInfectionpreventionmeasuresduringchildbirth

115(42) 76(36) 30to43 39(61) 48.4to72.2 <0.001 45.5 38.3 31to46% 74.1 59to85 0.05

Activemanagementofthethirdstageoflabour 73(26.5) 58(27.4) 22to34 15(23.4) 14.6to35.5 0.52 24.5 25.4 19.3to32.5 21 11to36 0.7

Maternalbloodlossassessment 124(45.1) 81(38.4) 32to45 43(67.2) 54.7to77.6 <0.001 43 34.5 27.4to42.4 75.7 61to86 0.01

Womencentredrespectfulcarepractices 12(4.4) 9(4.3) 2.2to8 3(4.7) 1.5to14 0.88 3.4 3 1to6 5.6 1.1to24 0.5

Noharmfulinterventionsdonetothemother 15(5.4) 14(6.6) 4to11 1(1.5) 0.2to10.5 0.12 4.3 5 3to9 1.5 0.2to10 0.2

Noharmfulhealthworkerbehaviourstowardsthemother

215(78.2) 162(77) 70.5to82 53(83) 71.4to90.3 0.306 74 72.4 64to79 81 57to93 0.45

Clinicalpracticesfornewborncare

Checksfundalheightandfoetalpresentation 4(1.4) 1(0.5) 0.1to3 3(4.7) 1.5to13 0.014 1.1 0.5 0.1to3.6 3.4 0.7to14 0.08

Foetalheartratemonitoredatregularintervals 61(22.2) 20(9.5) 6.2to14 41(64) 51to75 <0.001 20 6.6 45to10.5 73.3 58to84 <0.001

Healthworkerspreparedforresuscitation,ifrequired 179(65.1) 132(62.6) 56to69 47(73.4) 61.2to83 0.11 68 67.2 60to74 71.5 51to86 0.8

Neonatalsterilecordcare 265(96.4) 202(96) 92to98 63(98.4) 89.5to99.8 0.3 95.2 94.6 89to97.5 97.5 84to99 0.5

Appropriatenewbornthermalcare 84(30.5) 62(29.4) 23to36 22(34.4) 23.7to47 0.4 38 36.5 29to45 42.4 26to62 0.7

Apgarscore1minand5min 1(0.36) 0(0) 0to0 1(1.5) 0.2to10.5 0.07 0.9 0 0to0 4.7 0.6to27 0.08

Initiateearlybreastfeeding 191(69.4) 148(70) 64to76 43(67.2) 55to77 0.6 70 71 62to78 65.6 49to79 0.6

Noharmfulorunnecessarypracticesforthenewborn 95(34.5) 70(33.2) 27to40 25(39) 28to52 0.3 38 35.3 28to43.5 49 31to67 0.3

Aggregateindicesofqualityofcareattimeofbirth

ObstetriccareIndex 31.2 29.6 28to31 36.5 33to39.5 0.03 30.5 28.2 26to30.5 40 35to44 0.01

Neonatalcareindex 40 37.6 36to39 48 44to51.6 0.02 41.3 38.9 37.2to41 51 45to57 0.02

Essentialcareatbirthindex 35.3 33.4 32to35 42 38to45 0.01 35.6 33.3 31.6to35 45 40to49 0.01

Page101of248

Table10:Resultsfromthemultilevelmixedeffectslinearregression

Outcome:Essentialcareatthetimeofbirthindex Coef. pvalue 95%Conf.intervalExplanatoryvariablesBysector• Publicsector Base • Privatesector 0.06 0.03 0.01-0.11Wastheadmissiononaweekend? • Weekdayadmission Base • Weekendadmission -0.03 0.03 -0.06-0.003Numberofdeliveriesatmaternityfacilitylastyear • lowvolume<2000deliveries/year Base • averagevolume(2000-2999deliveries/year) 0.01 0.77 -0.05-0.06• Highvolume(>3000deliveries/year) -0.02 -0.08-0.05Woman'sage • Lessthan20years Base • 20-34years 0.01 0.91 -0.04-0.05• 35andgreater 0.01 -0.05-0.08Parity • Primipara Base • Multipara 0.01 0.22 -0.01-0.03Referraltothehospital? • Patientdirectlytothisfacility Base • Patientreferredfromanotherfacility 0.00 0.84 -0.04-0.03Caste • Scheduledcasteandscheduledtribe Base • Otherbackwardcaste 0.02 0.15 -0.01-0.04• Generalcaste 0.03 0.00-0.06Socio-economicstatus • 1stquintile(lowest) Base • 2ndquintile(lower) 0.00

0.08

-0.03-0.03• 3rdquintile(average) 0.00 -0.03-0.03• 4thquintile(higher) 0.00 -0.03-0.03• 5thquintile(highest) 0.04 0.0- 0.07Admissionduringworkhours? • Withinworkhours(9:00AM-17:00PM) Base • Outofhours(17:01PMto8:59am) -0.01 0.62 -0.05-0.03Whoconductedthedelivery? • Non-qualifiedbirthattendant Base • Qualifiedbirthattendant 0.01 0.61 -0.02-0.04

Page102of248

Figure8:EstimatedHawthorneeffectacrosssampledobservationsin26hospitalsofUttarPradesh

6.5:DiscussionUsingdatafromclinicalobservations inUttarPradesh,wefoundthatessentialcareatthe

timeofbirthprovidedtowomenandtheirnewbornswaspoorquality.Thereweresignificant

differencesamongstsectors,withprivatefacilitiesoutperformingpublicsectorfacilitiesfor

overallcareatbirth,obstetricandneonatalcare.Theprivatesectoralsoperformedbetterfor

specificproceduressuchasmaternalbloodlossassessment,monitoringofprogressoflabour

andmonitoringoffoetalheartrate.Preventivemeasuresagainstmajorcausesofmaternal

mortalitysuchashaemorrhage,sepsisandhypertensivedisorderswerefrequentlynotdone

atfacilitiesinbothsectors.

Our study advances the descriptive evidence base on QoC at the time of birth in India,

particularlyfortheprivatesectorwhichhasanincreasingmarketshareformaternitycare.33

We used direct observations of clinical practices that offer many advantages over other

qualityassessmentmethods.Wefoundnoevidencethatobservinghealthworkersgenerated

aHawthorneeffect.Inaddition,weprovidedacomprehensivemeasureofQoCthatincludes

adherencetoevidence-basedguidelines,useofharmfulandunnecessaryinterventionsand

behaviours, and respectful care practices. The essential care at birth, obstetric care and

neonatalcareindicescouldbeusedformonitoringQoCinothersettings.

The findings from themultivariate analysis confirmed that the private sector provided a

higher standard of care compared to the public sector andQoCwas not associatedwith

Page103of248

characteristics of thepatient, facility, ormidwiferypersonnel.However, admissionduring

weekendswasassociatedwithpoorerqualityofcare.Ourfindingsaresimilartootherstudies

thathavefoundaweekend-effectwithpoorerobstetricandneonatalcareoutcomesduring

weekends.362,363

Althoughcarewaslesslikelytobeprovidedbyqualifiedstaffinthepublicsector,qualified

personnelattendingbirthswasnotassociatedwithbetterqualityofcare.Previousstudies

haveshownthatevenwhenaqualifiedbirthattendantispresenttheymaynotbeadequately

skilled.78,364AstudyusingstandardizedpatientsinIndiaalsofoundminordifferencesbetween

trainedanduntrainedprovidersandQoC,although,thisstudydidnotfocusonmaternaland

newborncare.178

WedidnotfindanyrelationshipbetweenfacilitysizeandQoCatbirth.Thiscouldperhapsbe

explainedbythefactthatourobservationswerelimitedtouncomplicatedvaginalbirthsand

QoC in this settingwasdeficientacrossall sampled facilities.Previous studieshave found

betterQoCathigher level facilities,potentiallyexplainingwhypatientsbypass lower level

facilities.348 Although, we do not have robust evidence on factors influencing quality of

obstetricandneonatal careat facilities in India, there isevidence fromother low income

countrieswhichshowsthatprovidereffortmaybeakeydeterminantforQoC365andthatthe

private sector provides better QoC because it has superior operational andmanagement

systemsincludingbetterincentiveschemestoattractbetterqualifiedandmotivatedstaff.178

Weintendtoexploresomeoftheseissuesinsubsequentanalyses.

Our findingsaresimilar tootherstudies from India thathave foundpartographuse tobe

especially weak and that monitoring often consists of repeated unhygienic vaginal

examinationswithinadequateattentiontoeitherfoetalormaternalwell-being.71Wefound

slightlyhigherratesofAMTSLcomparedtoarecentstudyinneighbouringdistrictsofUP.77

Respectfulcarewaspoorinbothsector:only4%womanreceivedrights-basedcare.141Verbal

(13%) and physical abuse (8%)was endured by somewomen. Our informal observations

during data collection were consistent with other studies, in Madhya Pradesh66 and

Rajasthan78,thatfoundlabourroomenvironmentswerechaoticandhealthworkerscanbe

dominant,abusiveand threateningonoccasions.66Someresearchershavesuggested that

inadequateknowledgeandskills,staffingshortages,poorqualityin-servicetrainings,lackof

enablingenvironmentsandlimitedsupportivesupervisioncouldbeunderlyingcausesofpoor

Page104of248

quality care in India.66,71 We note that the Government of India and its partners are

implementing a range of schemes to improve the quality of intrapartum and immediate

postpartumcare.366Givenimmenseshortagesofskilledhumanresourcesformaternitycare,

focusedeffortstoestablishaprofessionalcadreofmidwivescouldbebeneficial.Wefound

greater variance in QoCwithin individual health workers than between them. This could

indicatethathealthworkersdonotfollowstandardprotocolsorprovidepreferentialcare.

Wenoteseverallimitationsofthestudy.First,theremayhavebeenobserverbiasduetothe

generalperceptioninthecommunitythattheprivatesectorissuperiorbecauseithasbetter

infrastructureandbettertrainedpersonnel,leadingtoanover-estimationofqualityinprivate

facilities. Second, there were challenges to sampling the private sector. Not only did 13

privatefacilitiesrefusetoparticipate,wehadnoofficialsamplingframefromwhichtoselect

thefacilities.ItispossiblethattheQoCoftheparticipatingprivatefacilitieswasdifferentfrom

thosethatwereeithernotsampledorrefusedtoparticipate.Third,aggregatingnumerous

indicators masks variations between individual indicators but was essential to report

comprehensivelyonQoC.Indevelopingaggregatemeasuresofquality,wegaveequalweight

toeachindicatorastherewasnoscientificbasisforapplyingintervention-specificweights.

Thevalidationoftheindexwasbeyondthescopeofthepresentstudy.Ontheotherhand,

therewerenorefusalsbywomentorecruitmentandastrictcase-definitionwasfollowed

whichminimisesselectionbiasatthelevelofparticipants.Researcherswerewell-trainedand

astructuredinstrumentwasusedwhichlimitssubjectivity.

Although,thegovernmenthashadsuccessinencouragingwomentodeliverinfacilities,we

foundlimitedevidence-basedcarepracticedatpublicandprivatesectormaternityfacilities

inUP.Ourfindingssuggestthreekey implications. First,there isaneedforauthoritiesto

introduceasystematicefforttomeasureandidentifyexistingqualitygapsduringlabourand

childbirth especially in high-burden states. These efforts should include private-sector

facilitiesastheyprovideasubstantialproportionofmaternitycareinIndia.Second,reasons

for high rates of untrained personnel providing maternity care and widespread non-

adherencetorecommendedprotocolsshouldbeinvestigatedfurther.Thepracticeofrelying

heavilyonpersonnel,notformally-trained,toprovidematernitycareisaworryingmodelof

service provision in the 21st century, which makes improving QoC particularly difficult

because such personnel are invisible within the health system. Third, tailored quality

Page105of248

improvementinitiatives88mustbedesignedforfacilitiesinbothsectorswithregularauditing

ofactualcare-processeslinkedtofunctionalaccountabilitymechanisms.

Page106of248

Chapter7:AninvestigationintomistreatmentofwomenduringlabourandchildbirthinmaternitycarefacilitiesinUttarPradesh,India:amixedmethodsstudy

Preface:

Chapter6presentedresultsonoverallqualityofcareatthetimeofbirthinpublicandprivate

sector facilities. Since there is limited research evidence on mistreatment of women in

maternityfacilitiesinUttarPradesh,Idecidedtoinvestigatemistreatmentindetail.

In this chapter, I report on amixed-methods study employing structured clinical practice

observations and analysis of open-ended observer comments to describe the nature and

contextofmistreatmentofwomenatpublic andprivate sector facilities inUttarPradesh

India.Forthequantitativedata,Iusedabivariatedescriptiveanalysistechniqueandforthe

qualitativedata,Iusedathematicapproachtoanalyseopen-endedobservercommentsand

describepatternsofmistreatmentinpublicandprivatesectormaternityfacilities.

Theresultsofthestudyshowamixedpictureofcareduringlabourandchildbirthatpublic

andprivatesectormaternityfacilitieswithahighprevalenceofcertainharmfulpractices.I

demonstratethatmistreatmentofwomenfrequentlyoccursinmaternityfacilitiesinUttar

Pradeshasaresultofcomplexfactorsrelatedtopolicy,infrastructureandresources,ethics,

cultureandpoorstandardsatmaternityfacilities.

Page107of248

RESEARCH PAPER COVER SHEET PLEASE NOTE THAT A COVER SHEET MUST BE COMPLETED FOR EACH RESEARCH PAPER INCLUDED

IN A THESIS.

SECTION A – Student Details

Student Gaurav Sharma

Principal Supervisor Véronique Filippi

Thesis Title An investigation into quality of care at the time of birth at public and private sector maternity facilities in Uttar Pradesh, India

If the Research Paper has previously been published please complete Section B, if not please move to

Section C

SECTION B – Paper already published

Where was the work published?

When was the work published?

If the work was published prior to registration for your research degree, give a brief rationale for its inclusion

Have you retained the copyright for the work?* Choose an item. Was the work subject to

academic peer review? Choose an item.

*If yes, please attach evidence of retention. If no, or if the work is being included in its published format, please attach evidence of permission from the copyright holder (publisher or other author) to include this work. SECTION C – Prepared for publication, but not yet published

Where is the work intended to be published? Reproductive Health

Please list the paper’s authors in the intended authorship order:

Gaurav Sharma, Loveday Penn-Kekana, Kaveri Halder, Véronique Filippi

Stage of publication Not yet submitted SECTION D – Multi-authored work

For multi-authored work, give full details of your role in the research included in the paper and in the preparation of the paper. (Attach a further sheet if necessary)

As first author on this paper, I developed the idea for the paper, undertook the analysis, wrote the first two draft of the manuscript and incorporated co-author comments

Student Signature: Date:

Page108of248

7.1:IntroductionThenumberofmaternaldeathsremainshighinIndiawith45,000estimateddeathsin2013.5

Since2006,theGovernmentof Indiahaspromotedskilledattendanceatbirthandrapidly

expandedtheJananiSurakshaYojana(JSY)programmethatnowbenefitsapproximately40%

of India’sbirth cohort.367 The JSY is a cash transferprogramme thatprovides amonetary

incentivestowomendeliveringinhealthfacilities.332

However,recentevidencefromJSYhasbeencautionaryandhighlightstheneedtoimprove

QualityofCare(QoC),concomitantlywitheffortstoincreaseinstitutionalbirths.10Ensuring

highQoCatthetimeofbirthencompassestheapplicationofevidence-basedobstetricand

neonatal care and efforts to ensure positive birth experiences for pregnant woman. 89

Respect, dignity and emotional support, although, integral to ensuring positive birth

experienceshavebeenoverlookedinresearch,policy,programmesandpractice.368,15

There is now increasing research evidenceonmistreatment ofwomenduring labour and

childbirthfrombothhigh,80,128-131and lower incomesettings132-134.Mistreatmenthasbeen

previously described as disrespect and abuse,125 obstetric violence126 and dehumanised

care.127 However, conceptualising what constitutes mistreatment, and therefore how to

measuremistreatmentarebothcomplex.Acomprehensivedefinitionofmistreatmentneeds

tocapturethehealth,humanrightsandsocio-culturaldimensionsofmistreatment,while,

measurement efforts need to capturewhat,where, howandwhymistreatmentoccurs.81

Freedmanetal.havehighlightedthatmeasurementeffortsshouldalsobeabletocapture

whether mistreatment was intentional or not, and the role of local societal norms (for

example-women’sstatus,patient-providerdynamics)thatinfluenceswomen’sperceptions

ofmistreatmentindifferentcontexts.81

Giventhesechallenges,arecentWHOsystematicreviewtriedtoestablishtheevidence-base

formistreatmentglobally.80They reviewed65 studies (53qualitativeand12quantitative)

from34countriesandfoundthatmoststudieshaveuseddifferentoperationaldefinitions

and measurement approaches.80 Amongst the quantitative studies, only three studies

reportedaprevalenceofmistreatmentatmaternityfacilities,whichvariedfrom15to98%.80

Thisreviewalsoproposedatypologyofitemsconsideredmistreatment,andidentifiedthe

following: physical, verbal or sexual abuse, stigma and discrimination, lack of informed

Page109of248

consent,breachesofconfidentiality,neglectandabandonment,refusaltoprovidepainrelief,

lack of supportive care, detainment in facilities, bribery and extortion.80 The review

incorporated elements from the work by Bowser and Hill (2010), who proposed seven

categoriesofdisrespectandabuse,namely:1)physicalabuse(beating,slapping,punching),

2) non-consented care (prior to vaginal examination or caesarean operation), 3) non-

confidentialcare(lackofprivacy),4)non-dignifiedcare(shouting,scolding,anddemeaning

care),5)discrimination(basedonage,wealthstatus,castegroup),6)abandonmentduring

care(beingleftaloneafterchildbirth),and7)detentioninfacilities(ifclientscannotpayuser

fees).125

However,aphenomenonoftenoverlookedinthedisrespectandabusediscourserelatesto

theoveruseofinappropriateorunnecessaryinterventionsforcareatnormalbirth.Thereare

examples of health workers in both high and low-income settings underusing simple,

inexpensiveinterventions(forexample,birthcompanionshiporcounsellingonbreastfeeding)

and overusing ineffective interventions that are more technical, lucrative or convenient

despite potential for harm (for example: labour augmentation without indications or

caesareansections).369-371

For this study, we operationalised mistreatment as those related to the following: 1)

disrespectandabuse(noprivacy,nobirthingpositionchoice,notinformingwomenpriorto

a vaginal examination, not allowing birth companions, not explaining reasons for

augmentation of labour, restricting food and water and informal payments); 2)

Overtreatment(routineuseofenema,routineuseofperinealshaving,applicationofextreme

fundalpressure,routineuterinelavage,routinemanualexplorationoftheuterusandroutine

episiotomy);andlastly,3)Under-treatment(deficienciesininfectionpreventionbyindividual

health workers, deficiencies in hospital environmental hygiene and use of unqualified

attendants). Research and programme efforts to improveQoC at the time of birth have

largelyneglectedtoexamineandaddressmistreatment insuchacomprehensivemanner.

Further,itisalsopossibleforbothundertreatmentandovertreatmenttooccurwithinthe

samepatientandwithinthesamefacility370whichmakesinterpretingdatadifficultbutthis

shouldbeconsideredbyresearchersworkingtoimproveQoC.

Uttar Pradesh (UP) is India’s most populous and deprived state.253 In related work, we

previouslydescribedoverallpoorqualityofcareatthetimeofbirth47butdidnotspecifically

Page110of248

examinemistreatmentofwomenatmaternityfacilities.Therearelimitednumberofstudies

thathavedescribedpatternsandthecontextofsuchcareatmaternityfacilitiesespeciallyin

theprivatesectorwhichhasanestimated18%ofthemarketshareformaternitycareinUP.253 This information is essential for understanding the context of care provision and in

developingeffectiveinterventions,policyandadvocacyapproachesforimprovementofQoC

at the time of birth. Available research evidence indicates that women with previously

negativepregnancyoutcomestendtochooseprivatesector.40Highersocio-economicstatus

andaccessibilityareassociatedwithincreasedprivatesectoruse.40Scheduledcasteandtribe

statusarenegativelyassociatedwithuseofprivatefacilities.248Theprivatesectoristhought

tobemoreexpensivethanthepublicsectorandthereisageneralperceptionamongstIndian

womenthattheprivatesectorprovidesbetteramenitiesandahigherstandardofcare.248

Qualitative studies in India have describedmany challenges to ensuring high QoC during

childbirth such as overcrowding of labour rooms, chaotic work environments, poor

coordinationbetweenhealthworkers, limited skills and competenceof healthworkers in

performing routine care procedures.65,66,372 They have also described situations where

labouring women have been left unsupported, were shouted at or slapped, not given

informationaboutwhatprocedureswerebeingdoneandwhytheywerereceivingit.66,239

In this paper,we report on thenature and context ofmistreatment recordedduring 275

clinicalobservationsoflabourandchildbirthin26maternityfacilitiesinUttarPradesh.This

richobservationaldatahelpsusindescribingthecontextofcare-provisioninalow-resource

settingincludingwhat,howandwhymistreatmentofwomenduringlabourandchildbirth

occursatmaternityfacilities.

7.2:Methods

7.2.1:Studysetting

ThestudywasconductedinthedistrictsofKannauj,KanpurNagarandKanpurDehatofUttar

Pradesh.In2012-2013,thematernalmortalityacrossUttarPradeshwas240per100,000live

births.253Atthistime,theneonatalmortalityratewere55per1000livebirthsinKannauj,41

inKanpurNagarand24inKanpurDehat.253Despitegovernmentschemestoimproverates

ofinstitutionalbirthsinpublicsectorfacilities,approximately39%ofdeliveriesinUP(43%in

Kannauj,46%inKanpurDehatand34%inKanpurNagar)occurredatpublicsectorfacilities

Page111of248

in2012-2013.253Theprivatesectordeliverysharewasestimatedtobe18% inUP(15% in

Kannauj,34%inKanpurNagar,and10%inKanpurDehat)duringthattime.253TheNational

RuralHealthMission has also appointed community healthworkers known asAccredited

Social Health Activists (ASHAs) in every Indian village.333 Motivating pregnant women,

accompanying them to institutions for childbirth and arranging suitable transportation to

hospitalsatthestartoflabouralsofallsundertheresponsibilitiesofASHAswhoarepaida

smallmonetaryincentive(INR600-equivalent£7)forthesetasks.

7.2.2:SamplingOursamplingframeincludedallhigh-volumepublicsectorfacilities(>200monthlydeliveries

based onHMIS data342) and established private sector facilities providing round-the-clock

basic emergency obstetric care identified by Sambodhi Research and Communications

(Lucknow,UttarPradesh)thathasextensiveexperienceofworkinginhealthresearchinthe

studydistricts.Aftermappingoffacilities,weselectedsixpublicsectorfacilitiesperdistrict

byconductingarandomselectionoffourcommunityhealthcentres,onemedicalcollegeand

onedistricthospitalandweinvitedallidentifiedprivatesectorfacilitiestoparticipate.Since

Kanpur Dehat did not have amedical college,we selected an additional district hospital.

Amongst the selected facilities, all public-sector facilities agreed to participate while 17

privatefacilities(outof30)agreedtoparticipate.Atnineoftheprivatefacilitiesthatagreed

to participate, there were no deliveries while observers were present. Therefore, the

observational data that we analysed came from 18 public facilities and 8 private sector

facilities. Further details on the samplingmethods are described elsewhere.47Theoverall

studyflowdiagramwaspresentedinFigure6.

7.2.3:StudyparticipantsStudy participants included pregnant women with spontaneous, uncomplicated labour

(definedaswomenwithlow-riskpregnancy,ofgestationalagebetween37and42weeksand

singleton vertex presentation, admitted to facilities who consented to participate in the

study)andtheirnewborns.

7.2.4:DatacollectionWecollecteddataon15potentiallyharmfulinterventionsasoutlinedpreviously.Eachitem

was codedas1 if observedand0otherwise.Anaggregatemeasureofmistreatmentwas

developedwhichwasthemeanofobserveditemsofmistreatmentforeverywoman(Range:

0-15). Potential covariates included women’s age, parity, referral status, caste, socio-

Page112of248

economic status, deliverybyqualifiedpersonnel, admissionduringwork-hours, admission

duringweekendsandpublicorprivate sector. For socio-economic status,wealthquintiles

weregeneratedusingprincipalcomponentanalysisusingdataonownershipofhousehold

assets.343

Weconceptualisedmistreatmentofwomenduringlabourandchildbirthasdisrespectand

abuse,overtreatmentandundertreatmentduringthetimeofbirthasdescribedpreviously.

Specifically, our questionnaire captured information on ensuring adequate privacy,

explainingtheprocessoflabour,restrictingfoodandfluids,informingwomenpriortovaginal

examinationandpriortolabouraugmentation,performinganenema,perinealshaving,not

allowing a birth companion, not offering choice of birthing position, routine episiotomy,

physical abuse (slapping or hitting), verbal abuse (insult, threaten and shout), routine

applicationoffundalpressure,routineuterinelavageandroutinemanualexplorationofthe

uterusafterchildbirth.

Questionscapturingeducational,demographicandsocio-economicstatuswereadaptedfrom

the National Family Health Survey questionnaire.338 At the end of every case, clinical

observerswithmaternal and child health backgroundswere encouraged to record open-

endedcommentsabouttheQoCtheyobserved,particularly,anythingtheyfeltwasimportant

toexplainthecontextandthingsthatwereparticularlystrikingtothem.Observershadbeen

orientedtotheprinciplesofrespectfulmaternitycareduringfield-leveltrainings.141Ateam

of14clinicalobserversworkinginpairsateachfacilityobservedcareroundtheclock.They

visited theadmissions,emergency, labour roomandpostnatalwards to identifypregnant

womenwhowerelikelytoundergouncomplicatedvaginalbirthsandobservedcareprovided

from admission to one hour postpartum. Data were collected after obtaining women’s

informedwrittenconsentbetween26thofMayto8thofJuly2015.

7.2.5:EthicsEthicalapprovalwasobtainedfromthePublicHealthcareSociety(PHS)EthicsReviewBoard

in India and the London School of Hygiene and Tropical Medicine in the UK

(LSHTMEthicsRef:8858).ThestudyalsoreceivedgovernmentclearancefromtheNational

HealthMissioninUttarPradesh.

Page113of248

7.3:Analysis7.3.1:MeasurementWecollecteddataon15potentiallyharmfulinterventionsasoutlinedpreviously.Eachitem

was codedas1 if observedand0otherwise.Anaggregatemeasureofmistreatmentwas

developedwhichwasthemeanofobserveditemsofmistreatmentforeverywoman(Range:

0-15). Potential covariates included women’s age, parity, referral status, caste, socio-

economic status, deliverybyqualifiedpersonnel, admissionduringwork-hours, admission

duringweekendsandpublicorprivate sector. For socio-economic status,wealthquintiles

weregeneratedusingprincipalcomponentanalysisusingdataonownershipofhousehold

assets.343

7.3.2:QuantitativeanalysisDescriptiveanalyseswerecarriedoutatthelevelofindividualwomenusingSTATA14(Stata

Corp.LP,CollegeStation,UnitedStatesofAmerica).Sincepreliminaryanalysisshowedthat

allwomenencounteredatleastoneitemofmistreatment(Appendix6),wecategorisedthe

sampleintotwogroupsbasedonthemediannumberofitemsofmistreatmentobserved,as

shown in Table 11. We then conducted a bivariate analysis to examine the relationship

between indicators of mistreatment and socio-demographic characteristics. Means,

proportionsandatotalmistreatmentscorewerecalculatedforallcovariates.Chisquaretests

wereusedtoassesswhethertherewasasignificantdifferenceamongsttheuseofpractices

consideredmistreatmentandtherelevantco-variates.

7.3.3:QualitativeanalysisTheopen-endedcommentsweretranscribedinHindiandtranslatedtoEnglishandanalysed

usingNvivo11software(QSRInternational).Athematicanalysisapproachwasutilised.Two

researchers(GS,LPK)independentlyreviewedcommentsline-by-lineandthenagreedona

setofcodesbroadlycategorisedintocodesrelatedtothequantitativechecklistandcodesfor

otheremergingissues.Bothresearchersthenjointlycodedalltheopen-endedcomments.In

caseswheredisagreementsarosebetweenresearchers, furtherdiscussiontookplaceuntil

consensuswasachieved.Throughouttheanalysisprocess,researchersreflectedonhowtheir

background,trainingandworldviewmightinfluencetheirinterpretationofresultsandefforts

were taken to minimise them. We triangulated the quantitative data with qualitative

comments.Commentsthatsummarisecommonfindingsacrossobservationsarereported.

Page114of248

7.4:ResultsWe first report onwomen’s socio-demographic characteristics categorised by two overall

mistreatment levels. Next, we present bivariate analysis of the prevalence of specific

indicators of mistreatment for which quantitative data are available and examine their

relationship with socio-demographic characteristics of the sample. Finally, we report our

qualitative findings, which provide additional information, and triangulate these to the

quantitativeresults,wherepossible,tofurtherexplainthenatureandthecontextinwhich

mistreatmentoccurs.

1. DemographiccharacteristicsThemajorityofobservationswere conducted in thepublic sector (n=211,77%)andmost

women came directly to facilities (88%) (Table 11). Amongst our sample, themajority of

participantswerebetween20-35yearsofage(90%),multi-parous(53%),belongedtotheso-

called“otherbackwardclass”category(55%)andwerefromthelowestwealthquintile(20%).

Mostdeliverieswereperformedbyunqualifiedpersonnel(59%)duringregularwork-hours

(92%)onweekdays(77%).Theonlyvariablesignificantlydifferentwastimingofadmission

andagreaterproportionofmistreatmentwasobservedincasesadmittedduringworkhours

comparedtoobservationsdonebeyondregularworkinghours(p=0.02).

Table11:Socio-demographiccharacteristicsofthesamplebytwooveralllevelsofmistreatment

Total(n=275)N,(%)

LessthanorequaltomediannumberofmistreatmentitemsN,(%)

Greaterthanmediannumberofmistreatmentitems

N,(%)

Pavalue

1. Women’sage

a. <20years 16(5.8) 14(7.5) 2(2.3)0.23b. 20-35years 247(89.8) 165(88.2) 82(93.2)

c. 35yearsormore 12(4.4) 8(4.3) 4(4.6)2. Parity

a. Primipara 119(43.3) 76(40.6) 43(48.9)0.32b. Multipara 145(52.7) 102(54.6) 43(48.9)

c. Grandmultipara 11(4.0) 9(4.8) 2(2.3)3. Referralstatus

a. Patientcomesdirectlytothisfacility 243(88.4) 164(87.7) 79(89.8) 0.62b. Patientreferredfromanotherfacility 32(11.6) 23(12.3) 9(10.2)

4. Castegroupb

a. “Scheduledcasteandtribe” 61(22.2) 38(20.3) 23(26.1)0.40b. “Otherbackwardcaste” 153(55.6) 109(58.3) 44(50.0)

c. “Generalcaste” 61(22.2) 40(21.4) 21(23.9)5. Socio-economicstatus

Page115of248

a. 1stquintile(poorest) 56(20.4) 41(21.9) 15(17.1)

0.56

b. 2ndquintile 54(19.6) 35(18.7) 19(21.6)c. 3rdquintile 55(20.0) 39(20.9) 16(18.2)d. 4thquintile 55(20.0) 39(20.9) 16(18.2)e. 5thquintile(wealthiest) 55(20.0) 33(17.7) 22(25.0)

6. Deliverybyqualifiedattendants

a. Qualifiedattendantsc 113(41.1) 78(41.7) 35(39.8)0.76

b. Unqualifiedattendantsd 162(58.9) 109(58.3) 53(60.2)7. Timingofadmission

a. Withinworkhours(9:00AM-17:00PM) 254(92.4) 168(89.8) 86(97.7)0.02

b. Outofhours(17:01PMto8:59am) 21(7.6) 19(10.2) 2(2.3)8. Admissionday

a. Admissionduringweekdays 211(76.7) 141(75.4) 70(79.6) 0.45b. Admissionduringweekends 64(23.3) 46(24.6) 18(20.5)

9. Sector

a. Public 211(76.7%) 138(73.8) 73(82.9) 0.09b. Private 64(23.2%) 49(26.2) 15(17.1)

aForthecomparisonoftheproportionsforlessthanorequaltomediannumberofitemsofmistreatmentobservedandgreaterthanmediannumberofitemsofmistreatmentthatwereobserved.bThecastesysteminIndiaisasystemofsocialstratificationthatplacespeopleinoccupationalgroups.Membersofscheduledcastesarethelowestcastesinsocietyandprotectedbythegovernmentthroughspecialconcessions.373Forcaste,wehaveusedtheexactlanguageofthevariousethniccategoriesgiveninIndiannationalfamilyhealthsurveyquestionnaires.cDoctors,nursesornurse-midwives–withatleast5,4and2yearsofpre-servicetraining,respectively–whoarelicensed,regulatedandendorsedbythegovernmenttoprovidematernitycareathealthfacilities.dAccreditedsocialhealthactivists,cleaners,hospitalporters,othercommunityhealthworkers,traditionalbirthattendantsandotherswhoarenotlegallyallowedbythegovernmenttoprovidematernitycareathealthfacilities.

2. Patternsofmistreatmentbysocio-demographiccharacteristicsFigure 9 below shows that amongst mistreatment practices, birthing position choice not

offeredtothelabouringwoman(92%),manualexplorationoftheuterusafterdelivery(80%)

andreasonforaugmentationnotexplained(46%)wereparticularlyhighatfacilitiesinboth

sectors.

Page116of248

Figure9:Quantitativeresultsshowingtheprevalenceofindicatorsofmistreatmentinpublicandprivatesectormaternityfacilities

Tabletwelvebelowillustratesthatamongstallsocio-demographiccharacteristics,thehighest

mistreatmentscores(mean)forwomen,werefoundinwomenabove35yearsofage(5.1);

primiparous women (5.2); those that were referred from another facility (5.0); amongst

womenbelonging to“scheduledcasteand tribes” (5.0); those in the fifth (richest)wealth

quintile(5.1),andamongstcasesadmittedduringwork-hours(5.0)onweekdays(5.0)inthe

public sector (4.9). However, the timing of admission (during weekdays or weekends)

influencedagreaternumberof indicatorsofmistreatmentcompared toadmissionduring

regularwork-hours,despitetotalmistreatmentscoresbeingsimilaracrossbothco-variates.

More women admitted during weekdays underwent episiotomies (p=0.04) and enemas

(p=0.01) whereas, more women admitted during weekends were not informed prior to

vaginal examination (p=0.03) and did not have the process of labour explained to them

(p=0.04).Wefoundthatmorewomenadmittedduringregularwork-hoursdeliveredwithout

adequate privacy (p=0.01), underwent enemas (p=0.03) and extreme fundal pressure

(p=0.01)morefrequently.

0 10 20 30 40 50 60 70 80 90 100

UterinelavageafterdeliveryperformedRestrictfoodandwaterintakeduringlabour

PhysicalabusebyhealthworkerCompanionnotallowed

PublicshavingdoneVerbalabusebyhealthworker

EpisiotomyperformedWomannotinformedpriortovaginalexam

ApplicationforfundalpressureVisualprivacynotensured

EnemagivenpriortochildbirthProcessoflabournotexplained

HWdoesnotexplainreasonforaugmentationManualexplorationoftheuterusperformed

Positionchoicenotoffered

Indicatorsofm

istreatmen

tduringlabo

urand

childbirth

Total Public Private

Page117of248

Tabletwelveshowsthatthepublicsectorperformedworsethantheprivatesectorfornot

ensuringprivacyofthelabouringwomen(p=<0.001),notinformingwomenpriortoavaginal

examination (p=0.01)andforphysicalviolence (shout,hitorpinch) towardsthe labouring

woman (p=0.04).On theotherhand, theprivate sectorperformedworse than thepublic

sectorfornotallowingbirthcompanionstoaccompanythelabouringwoman(p=0.02)and

forperinealshaving(p=<0.001).

Page118of248

Table12:Bivariateanalysisofthesignificancebysocio-demographicfactorsandtheprevalenceofobservedindicatorsofmistreatment

Noprivacy

%

NoPositionchoice%

Womannotinformedpriorto

vaginalexam%

Companion

notallowed

%

Processoflabour

notexplained

%

Reasonforaugmentationnotexplained

%

Restrictfoodandwater%

Enema%

Publcshaving

%

Fundalpressure

%

Uterinelavage%

Manualuterus

exploration%

Episiotomy%

Physicalabuse%

Verbalabuse%

Totalscore(mean)

TotalNreportingmistreatment(N=275) 82 252 74 23 99 40 21 84 27 79 10 221 65 21 37 Range

1-15Women’sage<20years 18.8% 81.3% 25.0% 0.0% 18.8% 12.5% 0.0% 62.5% 6.3% 18.8% 0.0% 68.8% 43.8% 0.0% 0.0% 4.4

20-35years 30.4% 92.3% 27.1% 8.9% 36.0% 15.0% 8.5% 28.7% 10.5% 28.7% 4.0% 81.4% 23.1% 7.3%14.2%

4.9

35yearsormore 33.3% 91.7% 25.0% 8.3% 58.3% 8.3% 0.0% 25.0% 0.0% 41.7% 0.0% 75.0% 8.3% 25.0%16.7%

5.1

Chisquare 0.59 0.30 0.97 0.46 0.10 0.79 0.28 0.02 0.43 0.42 0.56 0.42 0.08 0.04 0.26 ParityPrimipara

26.1% 91.6% 24.4% 9.2% 31.9% 20.2% 6.7% 36.1% 16.% 34.5% 5.0% 80.7% 45.4% 7.6%16.0% 5.2

Multipara33.1% 91.0% 30.3% 8.3% 41.4% 10.3% 7.6% 24.1% 4.8% 25.5% 2.8% 78.6% 7.6% 8.3%

11.7% 4.7

Grandmultipara27.3%

100.0% 9.1% 0.0% 9.1% 9.1% 18.% 54.5% 0.0% 9.1% 0.0% 100.0% 0.0% 0.0% 9.1% 4.3

Chisquare 0.45 0.59 0.22 0.57 0.05 0.07 0.39 0.02 0.003 0.10 0.50 0.23 <0.001 0.61 0.55 Referralstatus

Patientcomesdirectlytothisfacility 29.6% 91.8% 27.2% 7.4% 36.6% 13.2% 7.4% 30.0% 9.9% 30.0% 2.9% 79.8% 21.8% 7.4%

12.3% 4.9

Patientreferredfromanotherfacility 31.3% 90.6% 25.0% 15.6% 31.3% 25.0% 9.4% 34.4% 9.4% 18.8% 9.4% 84.4% 37.5% 9.4%

21.9% 5.0

Chisquare 0.85 0.83 0.80 0.11 0.55 0.07 0.69 0.62 0.93 0.19 0.07 0.54 0.05 0.69 0.14 Caste“Scheduledcasteandtribe” 32.8% 93.4% 36.1% 8.2% 39.3% 13.1% 9.8% 27.9% 6.6% 34.4% 1.6% 78.7% 19.7% 11.5%

13.1% 5.0

“Otherbackwardcaste”28.1% 92.2% 24.2% 6.5% 35.3% 13.1% 8.5% 30.1% 10.5% 24.2% 3.9% 82.4% 20.3% 6.5%

15.0% 4.8

“Generalcaste” 31.1% 88.5% 24.6% 13.1% 34.4% 19.7% 3.3% 34.4% 11.5% 34.4% 4.9% 77.0% 36.1% 6.6% 9.8% 4.9Chisquare 0.77 0.58 0.19 0.11 0.82 0.44 0.33 0.72 0.61 0.18 0.60 0.63 0.04 0.44 0.60 Socio-economicstatus1stquintile(lowest) 41.1% 89.3% 42.9% 7.1% 46.4% 17.9% 5.4% 25.0% 8.9% 30.4% 0.0% 83.9% 10.7% 3.6%

12.5%

4.9

2ndquintile 27.8% 90.7% 37.0% 3.7% 33.3% 11.1% 7.4% 29.6% 3.7% 27.8% 5.6% 74.1% 16.7% 14.8%20.4%

4.8

3rdquintile 23.6% 96.4% 18.2% 5.5% 43.6% 12.7% 12.7% 38.2% 5.5% 20.0% 9.1% 74.5% 25.5% 3.6% 7.3% 4.7

4thquintile 32.7% 92.7% 21.8% 5.5% 32.7% 12.7% 7.3% 20.0% 5.5% 30.9% 3.6% 83.6% 21.8% 10.9%16.4%

4.8

5thquintile(highest) 23.6% 89.1% 14.5% 20.0% 23.6% 18.2% 5.5% 40.0% 25.5% 34.5% 0.0% 85.5% 43.6% 5.5%10.9%

5.1

Chisquare 0.22 0.62 0.002 0.01 0.09 0.76 0.59 0.11 0.001 0.53 0.05 0.37 0.001 0.10 0.31 Deliverybyqualifiedattendants*

Page119of248

Unqualifiedattendants 30.2% 93.2% 32.7% 4.9% 36.4% 15.4% 9.3% 28.4% 6.2% 29.0% 1.9% 78.4% 17.3% 9.9%16.0%

4.8

Qualifiedattendants 29.2% 89.4% 18.6% 13.3% 35.4% 13.3% 5.3% 33.6% 15.0% 28.3% 6.2% 83.2% 32.7% 4.4% 9.7% 4.9Chisquare 0.85 0.26 0.01 0.01 0.86 0.62 0.23 0.35 0.02 0.90 0.06 0.33 0.003 0.09 0.13 Admissionduringworkhours#Withinworkhours 31.9% 90.9% 28.0% 9.1% 36.2% 15.0% 7.1% 32.3% 10.6% 30.7% 3.9% 80.7% 24.8% 7.9%

13.8%

5.0

Outofhours 4.8%100.0%

14.3% 0.0% 33.3% 9.5% 14.3% 9.5% 0.0% 4.8% 0.0% 76.2% 9.5% 4.8% 9.5% 3.7

Chisquare 0.01 0.15 0.18 0.15 0.79 0.50 0.23 0.03 0.12 0.01 0.35 0.62 0.11 0.61 0.58 Admissionduringweekends?Admissionduringweekdays

30.8% 90.0% 23.7% 10.0% 32.7% 14.2% 7.1% 34.6% 11.% 29.4% 4.7% 82.0% 26.5% 8.5%14.7%

5.0

Admissionduringweekends.

26.6% 96.9% 37.5% 3.1% 46.9% 15.6% 9.4% 17.2% 4.7% 26.6% 0.0% 75.0% 14.1% 4.7% 9.4% 4.6

Chisquare 0.52 0.08 0.03 0.08 0.04 0.78 0.55 0.01 0.12 0.66 0.08 0.22 0.04 0.31 0.28 SectorPublicsector 35.5% 91.0% 30.8% 6.2% 38.9% 14.7% 7.1% 28.9% 6.2% 31.3% 2.8% 78.2% 21.8% 9.5%

15.2%

4.9

Privatesector 10.9% 93.8% 14.1% 15.6% 26.6% 14.1% 9.4% 35.9% 21.% 20.3% 6.3% 87.5% 29.7% 1.6% 7.8% 4.7Chisquare <0.01 0.49 0.01 0.02 0.07 0.90 0.55 0.29 <0.01 0.09 0.20 0.10 0.19 0.04 0.13

Page120of248

3. Specificpatternsofmistreatmentthatoccuratmaternityfacilities

Thesectionbelowsummarisesqualitativeinformationobtainedfromobservers’open-ended

comments on mistreatment. It provides contextual insights into the quantitative data

presented earlier, as well as additional information on categories and themes of

mistreatmentsuchasdeficienciesininfectionprevention,lackofanalgesiaforepisiotomy,

informalpaymentsandpoorhealthfacilityenvironmentalhygienewhichwerenotcaptured

bythequantitativechecklist(Table13).

Table13:Themesandtheircomposition-clinicalobservationsoflabourandchildbirthatmaternityfacilities

Categories Themes Composition1. Over-

treatmenta) Extremefundal

pressureOccursfrequentlyandhelpoftensoughtfromotherspresent

b) Routineepisiotomy

Occursfrequentlyandoftenconductedwithoutanyanalgesia.

2. Under-treatment

c) DeficienciesinInfectionpreventionbyindividualhealthworkers

Usingdirtyclothestocleantheperinealandvaginalareas,unhygieniccareprocedures,conductingunnecessarymanualexplorationofuterusoruterinelavageandusingunsterileglovesandequipment.

d) Unqualifiedbirthattendants

Chronicstaffshortagesmeanthatunqualifiedhealthworkersareofteninvolvedprovidingmaternitycareservices.

e) Healthfacilityenvironmentalhygiene

Limitedadherencetoinfectionmanagementprotocols,nofacilitiesforhandwashing,nouseofantiseptics,non-availabilityofprotectivegear,inadequatesterilisationofequipments,apronsorfacemasks,nowastedisposalsystemsandstrayanimalssuchasdogsandcowsinpremises.

3. Disrespectandabuse

f) Physicalviolenceandverbalabuse

Healthworkersareoftenanxiousandsometimesusephysicalviolenceandverbalabuse.Physicalabuserangedfromslappingthepregnantwoman,tohittingandpinchingherthighsorrestrainingforcefully.Verbalabuserangedfromtalkingdowntothepregnantwoman,usingfoullanguageandthreateningwomenwithcaesareansections,iftheydidnotstopshoutingorcrying.

g) Informalpayments

Frequentinbothpublicandprivatesectormaternityfacilities.TheserangefromRupees200–2000,equivalent£2.4to£24

1. Overtreatmentbyhealthworkersa) FundalPressure:Ourquantitative results (Figure9) show that theprevalenceof fundal pressurewas29%;

similar across both sectors (p=0.09) but donemore frequently during regularwork-hours

Page121of248

(p=0.01) compared to outside regular work hours. The descriptions of fundal pressure

recordedbyobserversinopen-endedcommentsrangedfromapplicationoflightpressureto

extremepressureontheupperabdomendirecteddownwardstothebirthcanal. Ina few

instances,observersnotedthatmaternitycarepersonnelclimbedontopofthebedanduse

bothhandstopushdownforcefullyontheabdomen.Oftenphysicalviolencewasalsoused

whileperformingfundalpressure.Although,fundalpressurewasmostlydonebypersonnel

attendingtothedelivery,helpwasalsosoughtfromotherspresentinthelabourroomsuch

asmother-in lawsandayahs. Thecircumstances leadingtothedecisiontoapplyextreme

fundal pressure included to expedite the delivery process, when the woman could not

tolerate labourpainsorcouldnotbeardownorpushproperly.Thequotebelowillustrate

someexamplesofhowfundalpressurewasdescribedinthefieldnotes.

‘Thelabourroomofthedistricthospitalconductsdeliveriesinamiserablestate.Theygive

fundalpressureontheabdomenthewaypeopleusepumpsforfillingair incycletyres.

Theywerepressingtheirabdomenwiththeirelbowsduringdeliveryandalsoslappedthe

ladybadly’.(Clinicalobservationof35-year-old,primiparousatdistricthospital.)

b) Episiotomy:

Quantitative results indicate that episiotomy was done in 24% of cases and that the

prevalence was similar across both sectors (p=0.19). However, amongst cases where

episiotomywasgiven,noanalgesiawasgiven in25%ofcases, similaracrossbothsectors

(p=0.09).Commentsrecordedbyobserverscorroboratethatanalgesicswereoftennotgiven

duringepisiotomiesdespitewomencryingandshoutinginpain.Anecdotalevidencecollected

duringfieldworksuggeststhathealthworkersseemtobelievethatwomendonotrequire

analgesia during episiotomy as they are already in so much pain and will not feel any

additionalpain.Thequotesbelow illustrate twoexamplesofepisiotomy recorded in field

notes.

“Episiotomy was conducted without analgesia because of which the patient was

constantly shouting. The nurse consoled her saying it was only a few stitches, but no

analgesiawasgivenandinsteadthenursescoldedherbeforegivingherstitches”(Clinical

observationatadistricthospitalina34-yearmultigravidawoman.)

Page122of248

“Familymemberswerenotallowedtoenter inthelabourroomwhenpatientcamefor

delivery.Thiswasherfirstlabourandshelookedscared.Shewasstoppingthenursefrom

doingPVexamination.“Thepregnantwomensaidthatshehasbeeninpainfora long

timebutnooneispayingattention.Insidelabourroom,whenwomenaskedforwater,

nursesaidnottogivewaterassheisjustdoingdrama.Veryhighpressurewasappliedon

abdomenandepisiotomywasdoneduringlabour.Stitchesweregiven2-3hourslater.The

womanwascryingandsaidthatshewon’tevercomeagaintopublichospitalasnurse

insultedherbadly.”(Clinicalobservationatadistricthospitalina22-year-oldprimiparous

woman.)

2. Undertreatment:

c) Deficienciesininfectionprevention:

Deficienciesininfectionpreventionbyindividualhealthworkerswasalsoanimportanttheme

in theobservers’comments. Thesedeficienciesby individualhealthworkers rangedfrom

using dirty clothes to clean the perineal and vaginal areas, pouring oil over the vagina/

perineum,conductingunnecessarymanualexplorationofuterus,andusingunsterilegloves

andequipment.Althoughquantitativedataisnotavailableforallofthesepractices,available

quantitativeresultscorroborateahighprevalence(80%)ofmanualexplorationoftheuterus

whichwassimilarinbothsectors(p=0.10).Enemaswerealsoobservedin30%ofcases,risking

possiblefaecalcontamination.Itisencouragingtonotethatmosthealthworkersusedsterile

gloves;useofunsterileglovestoconductvaginalexaminationswaslowandhappenedinjust

2.2%ofallcases,allinthepublicsector(3%).Uterinelavageafterdeliverywasalsoinfrequent

inbothpublic(3%)andprivatesectors(6.3%)cases.Observer’scommentsalsoindicatethat

insomefacilities,instrumentsweresterilisedonceadayandoftenjustdippedinwarmwater

andchlorhexidinesolutionandreusedmultipletimes.Vaginalexaminationswereobserved

tobeconductedmultipletimesbydifferenthealthworkers. Inafewinstances,observers’

commentsmentionthatusedsyringeswereleftdiscardedonthefloor,whichisapotential

hazardforneedle-stickinjuries.

The quote below illustrates some examples of deficiencies in infection prevention by

individualhealthworkers:

“Whilesuturingtheepisiotomy,ayahacceptedaphonecall,alsotouchedthebedwith

herglovedhandsand thencontinuedwith the suturing.Manualexplorationof the

Page123of248

placenta was also done to check whether anything was left inside” (Clinical

observationinacommunityhealthcentreofa28-year-oldmultiparouswoman)

“Here,glovesaretakenoutfromthepowder.Idon’tknowiftheyuseautoclaves.They

did not inform me. They just wash instruments with water only. Mostly they dip

instruments inwarmwater,butthebloodstainsarestill there.Cheatle forcepsare

availablebuttheydonotkeepitproperly.”(Clinicalobservationatadistricthospital

ina30-year-oldgrandmultiparouswoman).

d) Healthfacilityenvironmentalhygiene:

Thewiderfacilityenvironmentandhospitalinfectionpreventionandcontrolmeasureswere

alsonotedasaseriousconcerninmanyoftheobservers’comments.Thisthemecaptures

issuesbeyondthecontroloftheindividualhealthworkers,suchasthoseattheinstitutional

level,andhasbeenconceptualisedasunder-treatment,whichconstitutesmistreatmentof

women since it is unethical to allow women to deliver in such unhygienic conditions.

Observer’s comments frequently describe limited adherence to infection management

protocolsatfacilities,nofacilitiesforhandwashing,nouseofantiseptics,non-availabilityof

protectivegear, inadequate sterilisationofequipments, apronsor facemasks. Systems for

segregationofwastes (used injectionvials, sharp instrumentsorwastes suchasplacenta,

otherfluids)suchascolour-codedbinswerenon-functional.Afrequentfindingwasthatstray

animals suchasdogsandcows roamed throughout the facility compoundandoften took

shelter in thewards or labour rooms. Clean towels and sterile padswere frequently not

availableathospitals;instead,women’soldclothessuchasoldsariswereusedtowipethe

womanandnewbornafterchildbirth. Suctionmachinesandradiantwarmers,evenwhen

available,wereoftenfoundtobeunusedanddirty.Bedssheetswerenotchangedregularly

andmultiplewomenwereobservedgivingbirthinthesamebed.Thequotesbelowillustrate

someexamplesofcommentsrecordedunderthistheme.

“Instrumentshereareneitherwashedproperlynorplacedintheautoclave.Theyclean

itwithwaterandusethemagain.Doctor,nurse,ayah-noneof themtakecareof

anything.There isnowateravailable inthebathroom.Noonecleansthebedafter

deliveryfornextpatient.Anotherwomanwasaskedtolayoverthesamebedwhere

therewasblood fromthepreviousdelivery.” (ClinicalobservationataCommunity

healthcentreofa25-year-oldmultiparouswoman).

Page124of248

“Thehospitalisprivatebutitdoesn’tlooklikeone.Repeateddeliveriesareconducted

withoutevencleaningthebedproperly.Inthelabourroom,thestaffchewandspit

tobaccoandtherearestainseverywhere. There isa large focus light in the labour

roomwhichiscoveredwithdust.Therearemiceinthelabourroom.Theyneveruse

the autoclave machine although it is available.” (Clinical observation at a private

hospitalof27-year-oldmultiparouswoman.)

e) Unqualifiedbirthattendants:

Quantitativedata indicate that 59%of all birthswere attendedbyunqualifiedpersonnel,

more frequently in the public (64%) than the private (41%) sector (p=0.001). We

conceptualisedtheuseofunqualifiedpersonnelasunder-treatment.Moreover,itisunethical

forwomentoreceivedcarefromunqualifiedpersonnelatinstitutions.Ourfindingsindicate

thatgiventhechronicstaffshortages,theroleofunqualifiedpersonnelseemsimportantand

established in theprovisionof careduring labourandchildbirth. The sweeper, traditional

birthattendant (dai) and theayah (helper) tend tobe involved in supportingwork in the

labourroomsuchasbringinginstrumentsordeliverytrayswhenthedelivery is imminent.

Theyareoftenalsoinvolvedinconductingthedeliveriessincethedoctorsandnursesarenot

available or do not attend all the normal deliveries. The quotes below highlight some

examplesfromfieldnotes.

“Afterexaminingthepregnantwoman,thenurseaskedifanydaihadcheckedheras

well.Daisareroutinelyinvolvedinprovidingcareatthisfacility.Ididnotobserveany

doctorsduringmyshift”(Clinicalobservationatacommunityhealthcentreofa25-

year-oldmultiparouswoman).

“Nurses of this private hospital are not trained. They are studying now and are

workingbasedonsomeexperience.”(Clinicalobservationinaprivatehospitalofa26-

year-oldprimiparouswoman).

3. Disrespectandabuse

f) Physicalviolenceandverbalabuse

Physicalviolenceandverbalabusewereacommonthemeinobserver’scomments.Fromthe

quantitativedata,theprevalenceofphysicalabusewas7.6%;andmorefrequentinthepublic

sectorthantheprivatesector(p=0.04)andgreateramongstwomenabove35yearsofage

Page125of248

(p=0.04).Although,verbalabusewasalsomoreprevalentinthepublicsector(15%)thanin

the private sector (8%), this was not statistically significant (p=0.13). The descriptions of

physicalviolenceintheopen-endedcommentsrangedfromslappingthepregnantwomanto

hittingandpinchingherthighswhileshewasbearingdown.Slappingoftenoccurredwhile

fundalpressurewasbeingapplied.Verbalabuserangedfromtalkingdowntothepregnant

woman,usingfoullanguageandthreateningwomenwithcaesareansections,iftheydidnot

stopshoutingorcrying.Inmostinstances,field-researchersnotedthatstaffappearedanxious

atthetimeofthebirthandoftenusedphysicalviolence(suchasslapping,forcingwomanto

beardownorrestrainingthewoman)duringthebirthingprocess.Therewerenoinstances

recorded in the field notes where pregnant woman or their companions stood up to

mistreatment or abuse by health workers. The quotes below illustrate physical violence,

verbalabuseandmistreatmentofpregnantwomanencounteredduringclinicalobservations.

“The nurse said, when you are with your husbands, you don’t shout but you are

shoutingnow.Youwillcomeagainwithanotherbabysoon!”(Clinicalobservationata

districthospitalofa27-year-oldmultiparouswoman.)

“Thenursewasbadlyscoldingthepregnantwoman.Thewomenappearedrestless

and was screaming and shouting. The nurse threatened her and said that if she

continues to scream, shewouldoperateonher.” (Clinicalobservationat adistrict

hospitalina25-year-oldprimiparouswoman)

g) Informalpayments:

Thepracticeofmaternitycarepersonnelaskingforinformalpaymentswasthemostcommon

theme identified from the observers’ comments and is a form of disrespect and abuse.

However, quantitative data about this phenomenon were not captured during clinical

observations. Observers’ comments indicate that, in most instances, maternity care

personnel demandedmoney from families for doing activities that are a part of their job

descriptionsuchasdryingandwrappingthenewborn,weighingthenewborn,cleaningblood

spills on the delivery bed or labour room floor and cleaning up. Often in public sector

hospitals,maternitycarepersonneldemandedmoneyfromclientsandtheirfamiliestocover

their costs, as they were contractual staff, allegedly, without a regular monthly income

Page126of248

source.Insomeinstances,informalpaymentswerealsogiventohealthworkersasgratuity

paymentstoexpresshappinessatthebirthofnewborn.

Fieldnotesalsoindicatethatthereisanunderstandingbetweenthematernitycarepersonnel

andcommunityhealthworkersuchasASHA’swhooftenactastheintermediarybetweenthe

clientsandhealthworkers,facilitatingtheexchangeofsuchpayments.Inaddition,inmost

observations,familieswereaskedtopurchasedrugsandcommoditiessuchasgloves,baby

towels,medicines,deliverykits fromoutside,although, inprinciple these itemsshouldbe

providedfreeofcostathealthfacilitiesundertheJSYscheme.Therewerealsoafewcases

where observers documented that newbornswerewithheld from families until providers

receivedpaymentsfromfamilies.Iftheprovidersdidnotreceivemoney,womenweremore

likelytobemistreatedduringtheirhospitalstay.Theamountof informalpaymentsvaried

betweenIndianRupees200–2000,equivalentUKPoundsSterling£2.4-24.Thequotesbelow

illustratesomeexamplesofthepracticesofinformalpaymentsatmaternityfacilities.

Thejuniornursesaskformoneyinthishospital.Theysay,“GivemeRs.2000.Wehave

performedthedeliverysowell.Ifwehadnotdonethatthechildwouldhavediedinside

you.Iwilltakehalfofthemoneyandwillgivetheresttomadam.”(Clinicalobservation

atadistricthospitalofa22-year-oldmultiparouswoman).

“Nursewas fighting formoney. She conducteddeliveryonlyafter receivingmoney.

Familymembersareaskedtobringclothesforcleaningmotherandchild.Moneyfor

gloves is also taken from familymembers.” (Clinical observation at a community

healthcentreofa23-year-oldprimiparouswoman.)

7.5:DiscussionThisstudyexploredthenatureandcontextofmistreatmentamongstwomenattendingpublic

andprivatesectormaternityfacilitiesinUttarPradesh.Allwomeninthestudyencountered

atleastoneindicatorofmistreatment.Ourestimatesaresimilartoanothercross-sectional

studyfromateachinghospitalinsouth-easternNigeriawhere98%ofwomenreportedsome

kindofmistreatmentduringchildbirth.374Theprevalenceofmistreatmentreportedacross

studiesvariesdependingonhowmistreatmentisconceptualisedandmeasured.80Arecent

cross-sectional study fromUttarPradesh, India reported that 57%ofurban slum-resident

women reported some form of perceivedmistreatment during childbirth.375 In Tanzania,

Page127of248

researchers found 19%perceivedmistreatment amongst a sample ofwomenwhile using

hospital-exitinterviewsandupto28%mistreatmentamongstthesamewomenfollowed-up

athomewhichtheyattributetocourtesybiasintheexitinterviews.376However,unlikeinour

study, both of these studies measured perceived mistreatment rather than direct

observationsoflabourandchildbirth.

Wefoundthattotalmistreatmentscoreswerehigheramongstwomenabovethan35years

ofage(5.1),primiparous(5.2),thosethatwerereferredfromanotherfacility(5.0),amongst

womenbelongingtothe“scheduledcasteandtribe”(5.0),thoseinthefifth(richest)wealth

quintile(5.1),andamongstcasesadmittedduringwork-hours(5.0)onweekdays(5.0)inthe

publicsector(4.9).Thecross-sectionalstudyfromurbanslumsinUttarPradesh,mentioned

earlier also found thatwealthierwomen,migrantwomen andwomen from lower castes

reported higher levels of disrespect and abuse.375 The importance of caste is well

documented in Indiawithmany studies reporting inferior care anddiscrimination against

womenbelongingtotheseso-calledscheduledcastes.377,378Researchershavesuggestedthat

sincethesewomenarelessempowered,healthworkersaremorelikelytothinkthattheycan

getawaywithmistreatmentofthesewomen.375

Wefoundthatnotofferingwomanachoiceofbirthingposition(92%),manualexplorationof

theuterusafterdelivery(80%)andnotexplainingthereasonforaugmentation(46%)were

particularly high at facilities in both sectors. There is evidence from a systematic review

supportingthebenefitsofdeliveringinalternativepositionscomparedtosupinepositionsfor

normalbirthssuchasshorterlabourduration,fewerepisiotomiesandfewersecond-degree

tears.379Manualexplorationoftheuterusisanimportantriskfactorforpuerperalsepsisand

shock114andshouldbeavoidedunlessindicatedandconstitutesovertreatmentwhichisform

ofmistreatment.Further,itisessentialtoprovideallwomenwithadequateinformationand

obtain an informed consent prior to any invasive clinical procedures such a vaginal

examination.114

Wefoundthatthepublicsectorperformedworsethantheprivatesectorfornotensuring

privacy of the labouring women (p=<0.001), not informing women prior to a vaginal

examination (p=0.01)andforphysicalviolence (shout,hitorpinch) towardsthe labouring

woman(p=0.04). Therecouldbemanyreasonsforpoorperformanceofthepublicsector

such as inadequate infrastructure, high-workloads, poor communication skills and

Page128of248

normalisationof disrespect andabuse in actual practice.During fieldwork,wenoted that

publicsectorfacilitieswerecrowdedandthatmaternitycarepersonnelworkedinchallenging

environmentsoftenwithoutbasicamenities,limitedincentivesandtheseenvironmentswere

notconducivetopracticeevidencebasedmaternitycare.

Ontheotherhand,theprivatesectorwasfoundtoperformworsethanthepublicsectorfor

notallowingbirthcompanionstoaccompanythelabouringwoman(p=0.02)andforperineal

shaving (p=<0.001). This could perhaps be due to existing institutional polices in private

hospitallabourroomswhichdonotallowbirthcompanions.ArecentCochranereviewfound

thatthatcontinuoussupportfromachosenfamilymemberorafriendincreasedwomen’s

satisfactionwiththeirchildbearingexperience.380 Although,perinealshavingisperformed

with the belief that it reduces the risk of infection, a systematic review has found no

associatedclinicalbenefitsofshaving.381PerinealshavingisalsodiscouragedintheIndian

skilledbirth attendance trainingmaterials382,which suggests that, perhaps, private sector

healthworkersmaynothavereceivedthesetrainingsorthatqualityofsuchtrainingsispoor.

We also found some interesting associations between women’s socio-demographic

characteristics and the prevalence of specific indicators ofmistreatment. Castewas only

associatedwithepisiotomyandwomenintheso-called“generalcaste”werefoundtohave

greaterproportionsofroutineepisiotomies(p=0.04)perhapsbecausetheyusedpublicsector

facilitiesmoreoften.Womeninthefirstquintile(poorest)wereleastlikelytobeinformed

prior to a vaginal exam (p=0.002) which suggests discriminatory care based on wealth

status.378 However,women in thehighestwealthquintile (richest)weremore frequently

unaccompanied by a birth companions (p=0.01), had higher rates of perineal shaving

(p=0.001)andepisiotomy(p=0.001)whichcouldperhapsreflectgreateruseoftheprivate

sectorandconsequentovertreatmentofwomenthatattendprivatesectorfacilities.

Womenwhodeliveredwithanunqualifiedattendantweremore frequentlynot informed

priortoavaginalexamination(p=0.01)andtheseexaminationswereoftenconductedwith

unsterilegloves(p=0.04).However,deliverywithaqualifiedattendantwasassociatedwith

lower rates of birth companionship (p=0.01), and routine episiotomy (p=0.003) which

suggestseitherunfavourableinstitutionalpoliciesoroutdatedknowledgeofhealthworkers

resulting inovertreatment. Interestingly,we found that totalmistreatment scores (mean)

Page129of248

werehigherfordeliveriesconductedbyqualifiedattendants(4.9)ascomparedtounqualified

attendants(4.8)whichsupportsthenotionofovertreatmentbyqualifiedpersonnel.

Mistreatmentwasseentocoexistwithlimitedadherencetoevidence-basedpracticesinthis

setting.47 Saini et al. (2017) suggest that the primary drivers for poor care arise out of

inequalitiesofinformation,wealth,andpower.370Inthiscontext,wesuggestthatthedrivers

for mistreatment include resource constraints, shortages of health workers, limited

incentives,weakmentorshipandsupervision,restrictiveinstitutionalpolicies,lackofup-to-

date knowledge and unequal power dynamics between health workers and pregnant

women.376,383,384Someresearchershavealsoarticulatedthatlong-standingpatternsofpoor

workconditions,resourcescarcity,lowskillsoroverburdenedhealthworkersatfacilitiesand

limitedchoiceforclientsleadstopoorQoC.125Inaddition,healthworkersmayoftennotbe

awareofrights-basedapproachesorunabletoprovidehighqualitycaredespitetheirbest

intentionsduetoinherentorganizationalandwork-environmentrelatedconstraints,which

areparticularlyrelevantinthissetting.

Anotherimportantfindingofthisstudycapturedthroughobservers’commentswasinformal

payments. Upon reflection, our QoC assessment tool should have specifically captured

detailed information on informal payments. Informal payments can range from gratuity

payments from appreciative patients, payments to jump the queue, receive better or

additional care, to obtain drugs and commodities, or simply to receive any care at all.385

Informalpaymentsareconsideredtobeinequitableandconstituteinstitutionalisedbribery,

whichmay hamper the entire health system.385,386 Further, they tend to be prevalent in

settingswherehealthsystemsareunder-funded,supervisorymechanismsareweak;where

womenarenotempoweredornotawareoftheirrights,andwhereprovidersareunlikelyto

facedisciplinaryactionfortheirbehaviours.385

Insummary,theliteraturesuggeststhatmistreatmentduringlabourandchildbirthmaybe

the result of many factors such as unfavourable institutional policies, resource and

infrastructural constraints, socio-cultural factors, limited knowledge and skills of health

workers.Weargue thatnon-adherence toclinicalprotocols, includingunder treatmentor

overtreatmentalsoconstitutemistreatmentofwomenatmaternityfacilities.Animportant

questionthatemergesfromourstudyiswhetheritisethicaltoallowandencouragewomen

todeliverinconditionswherebasicstandardsofevidence-basedcare,cleanliness,hygiene,

Page130of248

dignity and equity cannot bemet.We demonstrated thatmistreatment ofwomen often

occursbecauseofover-treatmentandunder-treatmentwhichconstituteafailuretoadhere

to professional standards of care80. Over-treatment and under-treatment should be

considered intheglobaldiscourseondisrespectandabuse,astheyarealsoaviolationof

humanrightsandconstitutepoorqualityofcareatmaternity facilities. It ispossible that

somepracticessuchasthoserelatedtoindividualhealthworkers’deficienciesinknowledge

orskillsareperhapseasiertochangecomparedtolong-standingsocio-culturalfactorsthat

may give rise tomistreatment.Ultimately,mistreatment occurs, at least in part, because

governmentshavenotcommittedtoorinvestedinparticipatoryaccountabilitymechanisms

likesocialaudits,communityscorecardsandothers,whichensurethatwomen’sexperiences

and perceptions of care are addressed and that respectful maternity care standards are

followed.81Thisisoneofthekeyrecommendationsemergingfromthiswork.

7.6:LimitationsThisstudyuseddatafromanobservationalstudydesignedtocapturedescriptiveinformation

onelementsofQoCfornormallabourandchildbirth.Thestudywasnotspecificallypowered

to measure and explain mistreatment as a separate category of poor quality of care.

Fieldworkers used open-ended comments to capture information that was contextually

importantorevents thatwereparticularly striking to them.Therefore, it is likely that the

commentsperhapscapturedthemoreextremeeventsratherthanroutinecareprocesses.

Theremayalsohavebeenanobserverbias,forexample:commentsrecordedbyobservers

perhaps reflects theirownprofessional experiences, trainingandknowledgeof respectful

carepractices.Duringfieldwork,wealsonotedthatyoungerobserversweremorelikelyto

takedowndetailednotescomparedtotheolderobservers,whoweremoreexperienced,and

perhaps,moreinclinedtoacceptmistreatmentasanormaloccurrence.Oursampleofprivate

sectorfacilitieswasalso limitedbythefactthatwehadnoofficialsamplingframeforthe

privatesectorfacilitiesinthestudydistrictsandthat13privatefacilitiesrefusedtoparticipate

inthestudy.ItispossiblethattheQoCofparticipatingprivatesectorfacilitieswasdifferent

fromotherprivatefacilitiesthatwereeithernotsampledorthosethatrefusedtoparticipate.

WehavepreviouslyshownthatanyHawthorneeffectwasnegligibleinthisstudysincethe

aggregatequalityscoresforindividualobserversdidnotchangedependingontheorderof

observations.47 The mixed methods approach taken to triangulate our findings, data

Page131of248

collection round-the-clock on all seven days of theweek, and the use of clinical practice

observationsarekeystrengthsofthisstudy.

7.7:ConclusionsMistreatmentiscommoninbothprivateandpublicsectors,albeitofdifferenttypes.Efforts

toexpandinstitutionalbirthsinUttarPradeshandotherhighmaternalandperinatalmortality

settingswouldbenefit fromstrengtheningthequalityofmaternitycare inbothsectorsso

thatevidence-basedmaternitycareisprovided,andpositivebirthsexperiencesareensured.

Thereareatleastfourspecificrecommendationsemergingfromthiswork.First,thereneeds

tobeasystematicandcontext-specificefforttomeasuremistreatmentinhighburdenstates

inIndiainbothpublicandprivatesectors.Second,atraininginitiativetoorientallmaternity

carepersonneltotheprinciplesofrespectfulmaternitycarewouldbeuseful.Third,systems

topromoteaccountabilityfortheapplicationofrespectful,woman-centred,maternitycare

pathways are needed. Lastly, we note that there needs to be a long-term, sustained

investment in health systems so that supportive and enabling work-environments are

availabletofront-linehealthworkers.

Page132of248

CHAPTER8:Managementisnotassociatedwithqualityofcareduringlabourandchildbirth:evidencefromacross-sectionalstudyofmaternityfacilitiesinUttarPradesh,India.

Preface:Inchapter6,Idescribedtheresultsfromacomprehensiveassessmentofqualityofcareat

the time of birth in Uttar Pradesh. Chapter 7 described the nature and context of

mistreatmentofwomeninmaternityfacilities,whichoccursduetohealthworkeractions,

restrictiveinstitutionalandlabourroompolicies,andlackofup-todateknowledgeandskills

amonghealthworkers.

Inthischapter, Iassessanddescribemanagementpracticesatmaternityfacilities inUttar

Pradeshandexaminewhethermanagementpracticesareassociatedwithqualityofcare.I

collectedprimarydataonmanagementpracticesfrominterviewswithhospitalmanagersin

the study sites. I merged two datasets on QoC andmanagement, performed descriptive

analyses and then used multi-level mixed effects regression techniques to investigate

whethertherewasarelationshipbetweenmanagementpracticesandQoCduringlabourand

childbirth. Multi-level mixed effects regression techniques are a robust and practical

statisticalmethodtoanalyseclustereddatasuchasdatafromdifferenthospitalsandthese

techniquesaccountforrandomeffectsandfixedeffectsinthelinearregressionmodel.

MyresultsindicatethatQoCandmanagementpracticeswerebothpoorinmaternityfacilities

in Uttar Pradesh, India. In this setting, my results indicate management practices at the

institutionalleveldonotinfluenceQoCduringlabourandchildbirth.Theonlymanagement

domainthatstronglyinfluencedQoCwasperformancemanagement,whichwasassociated

withuptoaseven-percentagepointhigherqualityscore.Theseresultssupporttheroleof

performancemonitoring activities focussed on quality improvement such as auditswhich

have been found to encourage the use of evidence-based-practices, improve supportive

supervision of health workers, encourage regular monitoring, and reporting on key

performanceindicators.

Page133of248

RESEARCH PAPER COVER SHEET PLEASE NOTE THAT A COVER SHEET MUST BE COMPLETED FOR EACH RESEARCH PAPER INCLUDED

IN A THESIS.

SECTION A – Student Details

Student Gaurav Sharma

Principal Supervisor Véronique Filippi

Thesis Title An investigation into quality of care at the time of birth at public and private sector maternity facilities in Uttar Pradesh, India

If the Research Paper has previously been published please complete Section B, if not please move to

Section C

SECTION B – Paper already published

Where was the work published?

When was the work published?

If the work was published prior to registration for your research degree, give a brief rationale for its inclusion

Have you retained the copyright for the work?* Choose an item. Was the work subject to

academic peer review? Choose an item.

*If yes, please attach evidence of retention. If no, or if the work is being included in its published format, please attach evidence of permission from the copyright holder (publisher or other author) to include this work. SECTION C – Prepared for publication, but not yet published

Where is the work intended to be published? Health affairs

Please list the paper’s authors in the intended authorship order:

Gaurav Sharma, Véronique Filippi, John Bradley and Timothy Powell Jackson

Stage of publication Not yet submitted SECTION D – Multi-authored work

For multi-authored work, give full details of your role in the research included in the paper and in the preparation of the paper. (Attach a further sheet if necessary)

As first author on this paper, I developed the idea for the paper, undertook the analysis, wrote the first two draft of the manuscript and incorporated co-author comments

Student Signature: Date:

Page134of248

8.1:IntroductionManagersofmaternityfacilitiesareresponsibleforimplementingappropriatesystemsand

procedurestoensurehigh-qualitycareforlabouringwomenfromthetimeofadmissionto

theirdischargefromfacilitiesafterchildbirth.300Managerialpracticesaredefinedas“theset

of formal and informal rules and procedures for selecting, deploying, and supervising

resourcesinthemostefficientwaypossibletoachieveinstitutionalobjectives.”387

Althoughwegenerallyassumethatmanagementinfluencesqualityofcare(QoC)athospitals,

empiricalevidenceexaminingthisrelationshipislimited.300

Sincemanagementpracticesarebroadinnature,traditionallyresearchershavefoundithard

tomeasuremanagementpracticescomprehensivelythroughquantitativemethods.However

recentmethodologicaladvancesfromstudiesinhigh-incomecountriesofferaninteresting

frameworkformeasuringmanagementpracticesathospitals.42,46,300

The only systematic reviewwhich examined the role of hospitalmanagers in quality and

patientsafetyfoundlimitedandinconsistentevidencetosupporttheseclaims.300Themodest

evidence that exists does suggest that managers’ time spent, engagement and work

specifically onquality assurance influences indicators of clinical quality andpatient-safety

positively. 300 Managerial activities thought to improve quality include activities such as

establishinggoalsandstrategies to improveQoC, setting thequalityagenda,promotinga

qualityimprovementcultureandprocurementofinstitutionalresourcestoensurequalityof

care.300

Inlow-resourcesettings,although,inadequatemanagementcapacityhasbeenrecognisedas

an important bottleneck for improving maternal and newborn health, research evidence

examining this relationship is limited.388 In addition, the likely relationship between

managerial practices and QoC may also be of a lesser magnitude as more fundamental

barriers to quality such as unavailability of essential drugs, commodities, poor referral

linkages,weakinformationsystems,deficienciesinknowledge,skillsandmotivationofhealth

workersexist16thatareoftenbeyondthecontrolofhospitalmanagers.

Manyresearchersarguethattherearearangeofmanagementpracticesoriginallyappliedin

themanufacturingandserviceindustriesthatarerelevanttohealthcare.269,389-391However,

Page135of248

transferringandapplyingthesemanagementpracticestohospitalsmustconsidercomplex

healthcareissuessuchasquality,safetyandmedicalerrors.266

In the recent decade, there have been important advances in measuring management

practices fromstudies in the fieldofhealtheconomics. Forexample, inacross-sectional

study at cardiac units in USA, management practices were significantly associated with

mortalityaswellasprocessofcaremeasures.312Inanotherstudyatsubstanceabuseclinics

intheUSA,researchershavefoundastrongassociationbetweenmanagementpracticesand

clientdaystotreatmentandincreasedrevenuegeneratedattheseclinics.313Similarly,inUK

hospitals, management practices had a strong association with both health outcomes

(improvedsurvivalratesafteracutemyocardialinfarction)andfinancialindicators.310

OtherstudiesthatconductedsecondaryanalysisofdataconductedasapartoftheWorld

ManagementSurveyefforts(http://worldmanagementsurvey.org/),whichcollectsdatafrom

over 2,000 hospitals in nine countries have found that hospitals with more effective

management practices provide higher-quality care.46,42,311 One of these studies which

restrictedanalysistodatafromhospitalsintheUSAandEnglandfoundthatwhenhospital-

boardspaidmoreattentiontoclinicalquality,managersweremorelikelytopayattention

to clinical quality and that hospital boards which used clinical quality measures more

effectivelyhadhigherscoresontargetmanagementandoperationsmanagement.46

However,itisimportanttonotethatmostresearchstudiesonthistopicarefromhigh-income

countries,fromtheprivatesector,andnoneofthemhasspecificallyfocusedonexamining

the relationshipbetweenmanagementpracticesandqualityofmaternitycare.Therefore,

examining whether there is a relationship between management practices and QoC in

maternityfacilitiesisakeyevidencegap.

India has one of the highest burden of maternal and neonatal deaths352 and, available

evidencefromhigh-burdenstateslikeUttarPradeshindicatessignificantdeficienciesinQoC

atmaternityfacilitiesparticularlyaroundthetimeofchildbirth.47Giventheincreasingrates

ofinstitutionalbirths,exploringwhethermanagementcandrivegainsinqualityisimportant

toinvestigate.Wecollectedprimarydataonmanagementpracticesat33maternityfacilities

afteradaptationofapreviously testedsurvey instrument toour studysetting. Ourstudy

objectives were: 1) to assess and describe existing management practices at public and

Page136of248

privatesectormaternityfacilitatesinthreedistrictsofUttarPradesh,India;and2)toexamine

whethermanagementpracticesinfluencequalityofcareofferedduringlabourandchildbirth.

8.2:Methods8.2.1:ConceptualizationofmanagementThepastdecadehasseena rise in the innovativemeasurementefforts thathave tried to

quantify therelationshipbetweenmanagementandQoCoutcomes.Mostof this research

stems from the field of health economics and are primarily from studies in high income

countries.42,46,300Notableamongstthese,isthepioneeringworkbyBloometal.(2010)who

initiallystudiedmanagementpracticesacrossmanufacturingfirmsinnumerouscountries.307

ThisworkhassincebeenreplicatedinthehealthsectorandthetoolsdevelopedbyBloomet

al.(2010)havenowbeenusedformeasuringmanagementpracticesindiversehealthsystem

contexts suchas inhigh-income (Australia,Canada, France,Germany, Sweden,UK,USA),

upper-middleincome(Brazil)andlower-middleincomecountries(India).43,282,308,309

These research efforts have employed a telephone- based interview methodology and

assessedmanagementpracticesunderfourkeydimensions:measuresofhospitaloperations,

measures of hospital performance, measures of targets management at hospitals and

measuresofpeoplemanagementathospitals.44,310,311

Briefly,operationsmanagementandperformancemonitoringsectionsofthetoolassessed

howwellmodernmanagementtechniqueswereintroducedatmaternityfacilities;whether

systemsforcontinuousimprovementexisted;andwhetherfacilityperformancewasregularly

trackedwithuseful indicators. Targetmanagementsectionassessedwhetherappropriate

targetshadbeenset,whethertheypushedfacilitiestoimproveperformanceandhowwell

they had been communicated across the hospital. People management section assessed

whetheremphasishadbeenplacedongoodhumanresourcepractices,whethermechanism

toincentivisehighperformingstafforreprimandpoorperformingstaffexist.42

Page137of248

8.2.2:StudydesignGiventhepaucityofempiricalevidenceontherelationshipbetweenmanagementpractices

and quality of care, we conducted a cross-sectional survey to collect primary data on

management practices and conducted clinical observations of labour and childbirth at

maternityfacilitiesinUttarPradesh,India.

8.2.3:SettingsThisstudywasconductedinthreedistrictsofUttarPradesh(UP)inIndia:Kannauj,Kanpur

NagarandKanpurDehat. 359KanpurNagar ispredominanturban,withhigher literacyand

lowermortality than the stateaverage.By contrast, Kannauj andKanpurDehat aremore

typicalofthestate.Thematernalmortalityratio(240per100000livebirths)andneonatal

mortalityrates(Kannauj-55,KanpurDehat-41andKanpurNagar-24per1000livebirths)

werebothhighinthestudydistricts.253Acrossthecontinuumofcare,largediscrepanciesin

maternalandchildhealth indicatorsexistacross the studydistricts. Forexample,43%of

deliveriesinKannauj,46%inKanpurDehatand34%inKanpurNagaroccuratpublicsector

facilities.Theprivatesectordeliveryshareisestimatedtobe15%inKannauj,34%inKanpur

Nagar,and10%inKanpurDehat.253

8.2.4:DatacollectionWe conducted clinical observations of labour and childbirth care at maternity facilities

between26Mayand8July2015.Subsequently,weconductedface-to-face,interview-based

data collection onmanagement practices from 9 August to 12 of September 2015. We

establishedtelephonecontactwith facilitymanagersearlyonandsetupappointmentsto

ensureahighresponserate.Theinterviewswerepresentedasfollow-upactivitiestotheQoC

assessments and were confidential conversations about management experiences and

challenges.Wedidnotcoversensitiveissues,forexample,financialearningsofthehospital.

The participants were not aware that they were being rated for their responses to the

managementquestionnaire.All interviewsweredouble-scored;whileone researcher (GS)

conducted all the interviews, another researcher also scored them independently. The

researchershadbeenorientedonmanagementconcepts,thesurveytool,andthescoring

techniqueoverathree-daytrainingsession.

Page138of248

8.2.5:SamplingTheoverallstudyflowdiagramforthisstudyisillustratedinFigure10below.Samplingfor

the QoC assessments was described in Chapter 6 and the QoC study flow diagram was

presentedinFigure6ofchapter6.Altogether,fortheQoCassessments,wecouldobserve

careprovided to 275mother-babypairs at 18public sector facilities and8 private sector

facilities(n=26).Forthemanagementsurvey,weemployedthesamesamplingstrategyas

theQoC assessments.However,we received a better response rate for themanagement

survey (n=33) compared to the clinical observations which could only be obtained in 26

facilities.Allthesurveyedfacilitieshadcomplexorganizationalstructures-definedasfacilities

withseparateadministrative,information,therapeutic,diagnosticandsupportservicesand

greaterthanfivebedsallottedformaternitycare.

8.2.6:SurveyinstrumentWeadaptedandusedamanagementsurveytoolthathaspreviouslybeenusedformeasuring

managementpracticesindiversehospitalsettingsinhigh-income(Australia,Canada,France,

Germany, Sweden, UK, USA), upper-middle income (Brazil) and lower-middle income

countries (India)282,308,309 and tailored it to be applicable to the context ofmaternity care

provisioninruralUttarPradesh.Specifically,afterpre-testing,weremovedquestionsona

category known as target interconnection, which was not applicable in this context and

simplifiedthelanguagesothatquestionsretainedtheirmeaninginHindi.

Essentially, this interview-based tool assessesmanagement practices at hospitals through

fourkeydomains:operationsmanagement,performancemanagement,targetmanagement

and peoplemanagement as described previously. The interviewer (GS) asked a series of

structured but open-ended questions (up to four questions for every domain) so that

sufficientinsightstoscoreeachmanagementpracticewereobtained.Ascoringgrid(between

1to5)wasusedbyassessorstogivescoresforresponsestoallquestionsdependingonhow

closely answers matched descriptors for each question. The survey tool is available in

appendix2.

8.2.7:StudyparticipantsStudyparticipantsforthemanagementsurveyincludedadministratorsorclinicalleadersat

33maternityfacilities(10privateand23publicsector).ParticipantsfortheQoCassessments

Page139of248

included pregnant women and their newborns that consented to the clinical practice

observations.

Figure10:Overallstudyflowdiagram-investigatingtherelationshipbetweenmanagementpracticesandqualityofcareduringlabourandchildbirth.

Page140of248

8.2.8:EthicsEthicalapprovalwasobtainedfromthePublicHealthcareSociety(PHS)EthicsReviewBoard

in India and the London School of Hygiene & Tropical Medicine in the UK

(LSHTMEthicsRef:8610).Thestudydesignhasalsoreceivedgovernmentclearancefromthe

NationalHealthMissioninUttarPradesh.

8.2.9:Measurement8.2.9.1:MeasuresofQualityofcareTheoverallessentialcareatbirthindexwhichmeasuresQoCduringlabourandchildbirthis

theoutcomevariableinouranalysis.Detailsonthedevelopmentofthisindicatorhavebeen

providedinChapter6.

8.2.9.2:MeasuresofmanagementMeasures of management included scores for overall management and individual

managementdimensions:operations,performance,targets,andpeoplemanagement.Two

independent assessors gave individual ratings for questions asked to managers at 33

maternityfacilities.Correlationofscoresgivenbythetwoindependentassessorswashigh

(seeTable14below)soameanscorewascalculatedforeachvariable.Asisstandardpractice,

thesescores(between1to5)werethenconvertedtoZscores;whichexpresshowfaravalue

isfromthepopulationmeanandexpressesthisdifferenceintermsofstandarddeviationsby

which it differs. Z scores were calculated primarily for the purposes of the regression

analysis.277

8.2.9.3:ExplanatoryvariablesOther explanatory variables included hospital characteristics such as number of beds,

ownership,when thehospitalwasestablishedand teaching status.Adummyvariable for

individualobserver’squalityratingwasalsousedtomitigateconcernsrelatingtoobserver

bias.

8.2.10:AnalysisAnalysiswas carriedoutusingSTATA14 (StataCorp. LP,CollegeStation,UnitedStatesof

America).Total scores foroverallmanagementand individualmanagementdomainswere

calculated for every facility. Descriptive analyses were performed to examine the

determinantsofmanagementpracticesatthe33sampledmaternityfacilities.Determinants

ofmanagementincludedthenumberofbeds,ownership,teachingstatus,managers’tenure

Page141of248

in post,when thehospitalwas established andwhether therewas external development

partnersupporttothefacility.

TherelationshipbetweenmanagementpracticesandQoCwasinvestigatedaftermergingthe

managementdataset(facilityn=33)withtheQoCdataset(facilityn=26).Therefore,thedata

thatweanalysedwereatthelevelofindividualwomen(n=275)at26facilities.

Fourmultilevel,mixed-effectslinearregressionmodelswithoverallqualityindex(outcome

variable)andZscoresfortotalmanagementandZscoresforsub-categoriesofmanagement

(explanatoryvariables)weredeveloped.Thefirstmodelwastheunadjustedmodelandthe

secondmodeladjustedforfacilitycharacteristics.

All four models included robust standard errors, accounted for clustering at the level of

facilities, used sampling weights, included a dummy variable for observer ratings and

controlledforrandomeffectsatthelevelofindividualfacilitiesandhealthworkers.Sampling

weightswereappliedsothateachfacilityreceivedequalweight intheanalysis.Maximum

likelihood estimation was used. The coefficients of the multivariate regression were

interpretedtoshowtheassociationofonestandarddeviationofmanagementZscoreonthe

outcome(QoC).

8.3:Results8.3.1:MaternityfacilitysamplecharacteristicsTheaveragehospitalinoursamplewas10yearsoldandhad12bedsallocatedformaternity

services.Most sampled facilitieswerenon-teaching (88%) and in thepublic sector (70%).

Mostmanagersat surveyed facilitieshada clinicalbackground (91%);48%of themhada

postgraduateclinicalspecialisationand6%reportedtohaveaMBAdegree.

Theaveragemanagementscoreformaternityfacilitiesinoursamplewas1.6(SD+0.7)(See

figure11below). Figure12showsmanagementscoresdisaggregatedbypublicorprivate

sector.Publicsectorfacilitiesreceivedameanscoreof1.5(SD+0.4)comparedtotheprivate

sector facilities that received a mean score of 2.0 (SD+ 0.9). Private sector facilities

outperformed the public sector for allmanagement dimensions: operationsmanagement

(private:2andpublic:1.7),performancemanagement(private:1.9andpublic:1.5),targets

management(private:1.6andpublic:1.2)andpeoplemanagement(private:2.4andpublic:

1.2).

Page142of248

Table14:Correlationresultsbetweenindependentassessorsratingsformanagementdimensions

Managementscores Correlationbetweenindividualassessors

Totalmanagementscore 0.9• Operationsmanagement 0.7• Performancemanagement 0.9• Targetsmanagement 0.7• Peoplemanagement 0.9

Figure11:Histogramshowingtotalmanagementscoresacrosssampledfacilities(n=33)

Figure12:Graphshowingscoresfortotalandindividualmanagementdomainsatpublicandprivatesectorfacilities

Table15showsthedescriptiveanalysisoftherelationshipbetweenmanagementscoresat

maternity facilities and their characteristics. We did not find significant differences in

characteristicsamongstsampledfacilities.

0.5

11.

5D

ensi

ty

1 1.5 2 2.5 3 3.5 4 4.5 5Total score for management practices.

0

1

2

3

4

5

Operationsmanagementscore(mean)

Performancemanagementscore(mean)

Targetmanagementscore(mean)

Peoplemanagementscore(mean)

Totalmanagementscore(mean)

Privatesector(n=10) Publicsector(n=23)

Page143of248

Table15:Maternityfacilitysamplecharacteristicscategorisedbytheirmanagementscores

Characteristicsofmaternityfacilities

Lessthanorequaltomedianmanagementscore(facilityn=17)

Abovethemedianmanagementscore(facility

n=16)

p-value

n % n %

1. Bedcapacity

a)Lessthan15beds 10 59% 7 44%

0.38b)Morethan15beds 7 41% 9 56%

2. Ownership

a)Privatefacility 4 23.5% 6 37.5%

0.38b)Publicfacility 13 76.5% 10 62.5%

3. Teachingstatus

a)Non-teachinghospital 16 94% 13 81%

0.25b)Teachinghospital 1 6% 3 19%

4. Managerstenureinpost

a)Yearsinpost(<5years) 9 53% 9 56%

0.85b)Yearsinpost(>5years) 8 47% 7 44%

5. Hospitalestablished

a)Lessthan10years’old 11 65% 7 44%

0.23b)Morethan10years’old 6 35% 9 56%

6. DevelopmentPartnersupport

a)No 6 35% 3 19% 0.28

b)Yes 11 65% 13 81%

8.3.2:RelationshipbetweenmanagementZscoreandfacilitycharacteristicsTable 16 shows results froma linear regressionexamining the relationshipbetween total

managementZscoreandfacilitycharacteristics.Olderfacilities(established>10years)were

associatedwithhighermanagementzscores(p=0.04)andthenumberofbedswasfoundto

beborderlinesignificant(p=0.07).Othercharacteristicssuchasownership,teachingstatus,

manager’stenureinpostandsupportbydevelopmentpartnersdidnothaveastatistically

significantrelationshipwithtotalmanagementZscore.

Page144of248

Table16:Relationshipbetweenfacilitysamplecharacteristicsandmanagementscores

Totalmanagementzscore Coef. pvalue [95%Conf.Interval]

1. Numberofbeds a)Lessthan15beds Base b)Morethan15beds 0.60 0.07 -0.06 1.252. Ownership a)Privatefacility Base b)Publicfacility -0.63 0.20 -1.59 0.343. Teachingstatus a)Non-teachinghospital Base b)Teachinghospital 0.87 0.21 -0.52 2.274. Managerstenure a)Yearsinpost(<5years) Base b)Yearsinpost(>5years) -0.28 0.41 -0.95 0.405. Hospitalestablished a)Lessthan10years’old Base b)Morethan10years’old 0.67 0.04 0.04 1.296. DevelopmentPartnersupport

a) No Base b) Yes 0.30 0.41 -0.43 1.02

8.3.3:RelationshipbetweenqualityofcareduringlabourandchildbirthandmanagementpracticesFigure 13 shows variations in QoC at facilities dichotomised based on theirmanagement

scores.Facilitieswithbelowmedianmanagementscoresprovided39%ofallrecommended

interventionscomparedto34%byfacilitieswithabovemedianmanagementscoresbutthis

difference was not statistically significant (p=0.28). For maternal care, better-managed

facilities provided 30% of the recommended interventions compared to 34% for poorly

managedfacilitiesbutthisdifferenceisnotstatisticallysignificant(p=0.4).Fornewborncare,

better managed facilities provided 39% of the recommended interventions compared to

poorlymanagedfacilitiesthatprovided44%oftherecommendedneonatalcareinterventions

butthisdifferencewasalsonotstatisticallysignificant(p=0.13).

Page145of248

Figure13:Weightedestimatesofqualityofcareatmaternityfacilitiescategorisedbytheirmanagementscores

8.3.4:ResultsfromthemixedeffectslinearregressionmodelInmultivariateanalysis(seeTable17below),therewasnostatisticallysignificantrelationship

(p=0.85)betweentotalmanagementZscoreandqualityofcare intheunadjustedmodel

(model 1). This relationship remains statistically insignificant after adjusting for all

explanatoryvariables(Model2,p=0.55).

Table18showsresultsfromthemultivariateanalysisinvestigatingtheassociationbetween

the four domains of management and QoC, and we find that, amongst individual

management dimensions, performance monitoring (p= 0.02) is the only dimension

statisticallyassociatedwithQoC(outcome).One-unitincreaseinperformancemanagement

wasassociatedwithaseven-percentagepointincreaseinqualityofcare.Further,resultsfrom

bothmultivariatemodels(Table17&18)showthatdeliveringinprivatematernityfacilities

wasassociatedwith7-10%pointhigherstandardofcareforwomen.However,wefoundno

association between bed capacity, teaching status or duration since establishment of

maternityfacilitiesandqualityofcare.

34% 44.5% 39%

30% 39% 34%

0% 10% 20% 30% 40% 50% 60% 70% 80% 90%

100%

Maternalhealthindex

Newbornhealthindex

Overallessentialcareatbirth

index

Qualityofcaredu

ringlabo

urand

childbirth

Poorlymanagedfacilities Bettermanagedfacilities

Page146of248

Table17:MixedeffectslinearregressionexaminingtherelationshipbetweenoverallQoCatbirthandZscoreindexfortotalmanagementscoreat26maternityfacilities

Outcome:Essentialcareatbirth Model1 Model2Variables Coef. pvalue [95%Conf.Interval] Coef. pvalue [95%Conf.Interval]Managementscorezindex 0.00 0.85 -0.02 0.02 -0.01 0.55 -0.03 0.02Bedcapacity

• Lessthan15beds X X X X Base • Greaterthan15beds X X X X -0.01 0.76 -0.06 0.05

Ownership • Public X X X X Base • Private X X X X 0.10 0.003 0.03 0.16

Hospitalestablished • Lessthan10years X X X X Base • Morethan10years X X X X 0.00 0.92 -0.04 0.04

Teachingstatus • Non-teachinghospital X X X X Base • Teachinghospital X X X X -0.03 0.27 -0.09 0.03

Page147of248

Table18:Mixedeffectslinearregressionexaminingtherelationshipbetweenqualityofcareandzscoresindexesformanagementsub-categoriesfor26maternityfacilities

Outcome:Essentialcareatbirth Model1 Model2Variables Coef. pvalue [95%Conf.Interval] Coef. pvalue [95%Conf.Interval]Operationsscorezindex -0.04 0.03 -0.07 0 -0.03 0.12 -0.06 0.01Performancescorezindex 0.08 0.01 0.02 0.15 0.07 0.02 0.01 0.12Targetscorezindex -0.01 0.7 -0.07 0.05 -0.02 0.33 -0.06 0.02Peoplescorezindex -0.04 0.02 -0.08 -0.01 -0.03 0.09 -0.07 0.01Bedcapacity • Lessthan15beds X X X X Base • Greaterthan15beds X X X X -0.03 0.25 -0.07 0.02Ownership • Public X X X X Base • Private X X X X 0.07 0.01 0.02 0.13Hospitalestablished • Lessthan10years X X X X Base • Morethan10years X X X X 0.00 0.87 -0.04 0.04Teachingstatus • Non-teachinghospital X X X X Base • Teachinghospital X X X X -0.03 0.29 -0.08 0.03

Page148of248

8.4:DiscussionThispaperprovidesempiricalevidenceonmanagementpracticesandtheirassociationwith

quality of care at maternity facilities in Uttar Pradesh, India. Overall, we found that

managementpracticeswerepooracrossthesurveyedmaternityfacilitiesinUttarPradesh.

WedidnotfindastatisticallysignificantrelationshipbetweentotalmanagementZscoresand

QoC.However,amongstmanagementdomains,performancemonitoringwasfoundtohave

asignificantrelationshipwithQoC(adjustedpvalue=0.02).One-unitincreaseinperformance

monitoringwasassociatedwitha7-percentagepointhigherqualityscore.

Onascaleof1to5,theaveragemanagementscoreforfacilitiesinoursamplewas1.6with

facilities intheprivatesector (2.0)receivingbettermanagementscoresthanpublicsector

facilities (1.5). These findings are in line with previous studies that have found higher

managementscoresintheprivatesector.42Ourresultsarealsocompatiblewiththefindings

reportedbyalargerIndiansurveyin3,892privatesectorhospitalsthatusedthesurveytool

thatweadaptedbutwasdonethroughtelephoneinterviews.44ThisstudybyLemosetal.

(2012)reportedatotalmanagementscoreof1.9,whichiscomparabletoourprivatesector

scoreof2.0.Similarly,scoresforallmanagementdimensionsobtainedbytheprivatesector

samples inourstudywere in linewiththosereportedbythepreviouslymentionedstudy,

suchas:operationsscore(2.0to2.1),performancescore(1.9to2.0),targetmanagement

score(1.6to1.6)andpeoplemanagement(2.4to1.9).44

PreviousstudiesutilizingthesametoolhavefoundthatIndianhospitalswerepoorlymanaged

comparedtohospitalsinUS(3.1),UK(2.9),Sweden(2.7),Germany(2.6),Canada(2.5),Italy

(2.5)andFrance(2.4).44InIndia,researchersfoundawidespreadoftotalmanagementscores

acrossstates, ranging from2.2 (highest) inHaryana,1.9 inDelhi (median) to1.7 inKerala

(lowest).44HospitalsinUttarPradeshwerebelowthemedianandobtainedatotalscoreof

1.844whichis0.2pointshigherthanourtotalmanagementscore.

In our sample, most managers had a clinical background (91%) rather than a business

background(6%)and3.0%hadajointdegree(MD/MBA).InthelargerIndianstudy,30%of

managershadanMBAdegreeorsomesortofequivalentbusinesstrainingand54%hada

clinical degree.44 Thesedifferences couldperhapsbedue toour study setting,whichwas

Page149of248

predominantlyruralwithanascentprivatesectorintwodistrictsorcouldalsoindicatelimited

formalmanagement training amongst administrators. Cross-sectional evidence fromhigh-

income settings indicates that hospitals employing clinically trainedmanagers often have

bettermanagementpractices.308ResearchfromtheUnitedKingdomusingthesamesurvey

tool has also found that doctors often make better managers if they have the relevant

managementskillsandunderstandingofhospitaloperations.309

Previous research in India has found that Indian hospitalmanagers are often unaware of

modern management practices.44 Our data shows that that most public-sector hospital

managershave clinical backgrounds and tend to come into their positionsbasedon their

tenurethroughanincrementalcareerprogressionscheme.Whereas,privatesectorhospitals

weremore likely tobe familyor self-owned, for-profitenterprisesandmanagersat such

privatesectorinstitutionshadformalmanagementqualifications(6%),whichmayperhaps

explainbettermanagementscores in theprivatesector.Our impressionsduring fieldwork

wasthatmanagersofpublicsectorfacilitiesareoftenconstrainedbybureaucraticprocedures

thatlimitsfinancialautonomy,authorityforrecruitmentordismissalandabilitytoincentivise

highperformers.Thesecouldperhapsalsopartlyexplainpoorperformanceofpublic-sector

facilitiescomparedtotheprivatesector.

Wealsofoundthatolderfacilities(established>10yearsago)hadhighermanagementscores

compared tonewly established facilitieswhich could indicate that older facilities perhaps

have more standardized and established care pathways compared to newer maternity

facilities.

Although we found some variation in overall QoC between better managed and poorly

managedfacilities,thisdifferencewasnotstatisticallysignificant(p=0.28).Resultsfromthe

mixed effects model confirmed that there was no statistical association between total

managementZscoreandQoCinboththeunadjusted(Model1;p=0.85)andadjustedmodels

(Model2;p=0.55).Thisfindingisnotconsistentwithpreviousresearchevidencefromhigh-

incomesettings,43,45,308however,noneofthesestudiesweredoneinlow-incomesettingsor

specificallyfocussedonqualityofmaternitycare.

Page150of248

Amongstallmanagementdimensions,performancemonitoringwastheonlydimensionthat

hadasignificant relationshipwithQoC(adjustedpvalue=0.02)withone-unit increase in

performancemonitoring associated with a 7-percentage point higher quality score. Our

findings are in line with previous studies that have shown performance monitoring may

encourage the use of evidence-based-practices, improve supportive supervision of health

workers, encourage regular monitoring, and reporting on performance indicators.293 The

effectivenessofauditsandfeedbackwasevaluatedinaCochranereview,whichfoundthat

audits and feedback interventionshave thepotential for amodest improvement (median

+4.3%)inhealthworkercompliancewithdesiredpractice.195 Inaddition,thereviewfound

thataudits and feedbackareparticularlyeffectivewhenbaselineperformance is low, the

sourceisasupervisororacolleague,itisdonemultipletimes,deliveredinbothverbaland

writtenformatsandincludesexplicittargetsandanactionplan.195Sincehospitalsareoften

themostexpensivecomponentofhealthsystems,performancemonitoringhaspotentialto

beusefulinallsettings.392

Wealso foundthatdelivery inaprivate-sector facilitywasassociatedwitha7-10%point

higherstandardofcarecomparedtodeliveringinapublic-sectorfacility.Thisisinlinewith

ourresultsfromtheQoCassessmentswhichfoundbetterQoCintheprivatesector.47These

resultsindicatethatmanagementpracticesdonotfullyexplainthedifferencesinqualityof

carebetweenpublicandprivatesectors.Itispossiblethattheprivatesectorattractsmore

competent,better-motivatedhealthworkerswithhigherremunerationwhointurnprovide

betterqualityofcare.Furtherresearchusingrobustmethodswouldbeusefultounderstand

whethermanagementinfluencesqualityofcareduringnormallabourandchildbirthacross

sectors.

8.5:LimitationsWenotethefollowinglimitationsofthestudy.First,oursampleofprivatesectorfacilitiesfor

theQoCassessmentswerelimitedbythefactthatwehadnoofficialsamplingframeforthe

privatesectorandundertakingacomprehensivecensusofprivatesectorfacilitieswasnot

feasible.Inaddition,13privatefacilitiesrefusedtoparticipateintheQoCstudy.Management

practicesandQoCatfacilitiesthatwerenotsampledorrefusedtoparticipatemayhavebeen

differentfromparticipatingprivatefacilities.Hence,ourfindingsontherelationshipbetween

Page151of248

managementpracticesandQoCarenotgeneralizabletoallfacilitiesprovidingmaternitycare

servicesinUttarPradesh.

Second,weinterviewedmanager’sface-tofaceunlikepreviousstudiesthathaveemployed

telephoneinterviews.Therefore,itispossiblethatassessorsmaybebiasedbytheappearance

ortheoperationsofaparticularfacility.Third,althoughweadaptedatoolthathaspreviously

beenusedinmultiplecountriesincludingIndia,thecontentandconstructvalidityofthetool

was not specifically tested which may have implications for findings. Since, correlation

betweenassessorswashighandourscoreswerecomparablewiththelargerIndianstudy,

reliabilityislessofaconcern.However,anadditionalvalidationstudyinasmallselectionof

participantswouldhavebeenusefultovalidatethestudyinstruments.Fourth,oursampleof

275observationsat26hospitalsisalsosmalltogeneratepreciseestimatesontherelationship

betweenmanagementandqualityofcare.Fifth,ourstudyinstrumentdidnotcaptureany

informationoncontextualdeterminants(political,social,economic,socio-cultural)thatmay

influencemanagersandfacilityperformanceinthissetting.Furtherresearchwouldbeuseful

toexaminetheseissuesindetail.

8.6:ConclusionsThis study is of interest to the maternal and newborn health academic and research

community,policymakers,programmemanagersandhospitaladministrators in resource-

constrained settings that are interested to improve quality of care during labour and

childbirth.Ourfindingssuggestthatmanagementbestpracticesarenotwidelyutilisedand

that considerablegaps in knowledgeand implementationexist atbothpublic andprivate

sectormaternityfacilities.Wefoundthattherelationshipbetweenmanagementpractices

andQoCfornormal labourandchildbirth iscomplexandmaynotbeapparent insettings

where both QoC and management are weak. However, we found a strong association

betweenperformancemanagementactivitiesandqualityofcare.Ourfindingsstrengthenthe

evidence-baseontheroleofactivitiessuchasauditsinlow-resourcesettingsthathavean

importantroleinimprovingqualityofcare.

ItislikelythatQoCduringlabourandchildbirthisdependentonindividualhealthworker’s

actions, competence and their motivations and health workers who are motivated will

providedhighqualitycaredespiteexistingconstraintsthattheyface.Furtherresearchinto

Page152of248

determinantsofmanagementpracticesathospitalsandvalidationofapproachestomeasure

managementpracticescomprehensivelyinresource-constrainedsettingswouldbeuseful.

Page153of248

Chapter9:Discussionoftheresultsoftheoveralldoctoralresearch

AddressingQoCatthetimeofbirthisanimportantpriorityforseveralglobalandnational

effortsthataimtoendavoidablematernaldeaths,neonataldeathsandstillbirths.4,103The

resultsofthreestudiespresentedinmyPhDprovideausefulcontributiontotheliterature

onQoCandmanagementpracticesatmaternity facilities inUttarPradesh, India.MyPhD

resultswillalsobeusefultoinformfuturematernalnewbornhealthprogrammesandsupport

thedesignofqualityimprovementeffortsinthestudydistricts.Atthegloballevel,myPhD

findingswillbeofinteresttotheglobalresearchcommunityworkingtodefinemetricsfor

qualityinmaternalnewbornhealth26andtodefineelementsofskilledattendanceatbirth

(SAB)393.

9.1:Summaryofkeyfindings

Theoverallpictureofmaternitycareprovisionthatemergesfromthestudydistrictsisofa

dysfunctionalcarepathwaywithlimitedadherencetoevidence-basedpracticesandahigh

prevalenceofcertainpracticesconsideredtobemistreatment.TheQoCatmaternityfacilities

in the three studied districts of Uttar Pradesh in 2015 was found to be generally poor.

Amongstalltheinvestigatedcharacteristicsofthewoman,thehealthfacilityandthetypeof

birthattendant,overallqualityofcarewasfoundtobebetterinprivatesectorfacilitiesand

forwomenthatwereadmittedduringtheworkweek(Monday-Friday).

Although I found that themajority of deliveries inmaternity facilitieswere conductedby

unqualified personnel in 2015, there were no statistical differences in care provided by

unqualifiedorqualifiedbirthattendants.Mistreatmentofwomen(definedaspresenceof

indicatorsofdisrespectandabuse,over-treatmentandunder-treatment)frequentlyoccurred

atmaternity facilities. Frommy investigation into the relationship betweenmanagement

practices and QoC, I found that there was no statistical association between total

managementscoresandQoC.Theonlymanagementdimensionthathadasignificantpositive

associationwithQoCwasperformancemanagement.

ThekeyfindingsoftheresultschaptersofmyPhDareelaboratedbelow.

Page154of248

9.1.1:Qualityofcarewasgenerallypooracrossthesampledpublicandprivatesector

maternityfacilitiesinUttarPradeshin2015

Theresultsofchapter6showedthatqualityofessentialcareduringlabourandchildbirthwas

poor inUttarPradesh, India.Onaverage,women received just36%of the recommended

seventeenpractices forcareat thetimeofbirthacrossthesampledmaternity facilities in

UttarPradesh.TheclinicalpracticesthatImeasuredwerethemostessentialandbasiccare

practicesrecommendedduringlabourandchildbirthsoinrelativetermsQoCwasfoundto

beverypoor.

TheoverallQoCwasfoundtobebetteramongstwomenattendingprivatesectorfacilities

where they received 45% of recommended practices compared to 33% amongst women

attendingthepublicsectorin2015.Notwithstandingthelimitationsofsamplingtheprivate

sector,Ifoundthatprivatesectorprovidedanoverallhigherstandardofcareduringlabour

andchildbirth(p=0.01)includingforbothobstetric(p=0.01)andneonatalcare(p=0.02).The

results fromthemultivariateanalysisconfirmedthatoverallQoCwas6percentagepoints

(95%CI:1-11%)higher(p=0.03)inprivatesectorfacilitiesthancorrespondingscoresinthe

publicsectoraftercontrollingforconfounders.Although,thisresultisstatisticallysignificant

thedifferenceinqualitybetweensectorsmaynotbeclinicallyrelevantastheeffectcanbe

assmallas1%.

Thereismixedresearchevidenceonwhetherprivatesectorprovidesbetterqualityhealth

services than the public sector. For example, two systematic reviews employing different

reviewmethodologieshavereporteddifferentresultsindicatingthattheunderlyingevidence

baseonthistopicisweak394.Intheirsystematicreview(2011)ofstudiesexaminingquality

of care in formal private versus public sector facilities in LMICs, Berendes et al. (2011),

concluded that “quality in both provider groups seems poor, with the private sector

performingbetterindrugavailabilityandaspectsofdeliveryofcare,includingresponsiveness

andeffort,andpossiblybeingmoreclientoriented”.365However,anothersystematicreview

publishedayearlaterbyBasuetal.(2012)concludedthat“studiesevaluatedinthisreview

donot support the claim that theprivate sector is usuallymore efficient, accountable or

medicallyeffectivethanthepublicsector”395.

Qualitativestudiesthathavesoughttoexplainthereasonsbehindpoorqualityinthepublic

sectorhavehighlightedreasonssuchasresourceconstraints, lowsalaries,highworkloads,

Page155of248

poorincentivesandconditionsofservices396,stafffavouringcertainpatients397,clientslacking

sufficientknowledgeabouttheappropriateuseofdrugsandtheirrightstochallengepoor

services.398-400

SpecifictoIndia,researchershavesuggestedthatpoorqualityofcareinIndiacanbepartly

explainedby thepoorqualityofmedical trainingsand theabsenceofnational continuing

medical educationand recertificationprogrammes.178 Other researchershave found that

providereffortisakeydeterminantforqualityandhealthworkersintheprivatesectorexert

moreeffort than thepublic sector.365 This isalso relevant in thecontextof LMICswhere

privatesectorpersonneloftenwanttodemonstratethattheyareprovidingbettervaluefor

moneyandexertgreatereffort.Researchevidencealsoindicatesthatprovidereffortcanbe

improvedbyprovidinghigherpayments,betterincentiveschemes,strengtheningmonitoring

andprovidingbettersupporttohealthworkersthroughpeer-networks.365

Quality of obstetric care, asmeasured by an index based on nine of themost important

practices,wasfoundtobelow(30.5%)acrosstheentiresample.Theobstetriccareindexwas

foundtobeloweramongstpublicsectorcases(28%)comparedtotheprivatesectorcases

(40%). Amongst obstetric care practices, regularmonitoring of labour using a partograph

(1.6%)wasrareandpartographswereusedinjust0.2%ofpublicsectorcasescomparedto

7.2%ofprivatesectorcases. Myfindingsaresimilartootherstudies in Indiawhichhave

foundpoorratesofpartographusewithinadequateattentiontoeitherfoetalormaternal

well-beingduringlabourandchildbirth.70,71Infactastudyexaminingtheimplementationof

partographsintheJSYprograminMadhyaPradeshfoundlowratesofpartographuse(6%)

andpoorcomptenceofhealthworkersinusingpartographscorrectly.67Inthisstudy,health

workersreceivedameanscoreof1.08(outof10)onclinicalvignettes,indicatingsubstantial

deficiencesinknowledgeofhealthworkers.67

Although,theuseofpartographisactivelypromotedbytheIndiangovernmentandnational

guidelinesalsorecommendthattrainingsonpartographsandessentialsuppliesshouldbe

providedatallbirthingfacilities,332,382myfindingsdemonstratethatpartographsarenotused

routinely. Other research evidence from LMICs has suggested that challenges for routine

partograph use include limited knowledge of health workers, limited availability of pre-

printed partographs, length of time needed to fully complete a partograph and high

workloadsofhealthworkers.16,67Asdiscussedabove,researchevidencealsoindicatesthat

Page156of248

deficiencies in intra-partum care such as inadequate monitoring of labour through

partographsoftenleadtopreventableintra-partumstillbirthsinIndia.401

Screeningmeasures for preeclampsia/ eclampsiawere found to be low (2.3%) across the

entiresamplewith2.2%ofpublicsectorcasesreceivingthesescreeningmeasurescompared

to2.5%ofprivatesectorcases.Theseresultssuggestthatsimplescreeningmeasuressuch

as detection of elevated blood pressure and presence of proteinurea are not routinely

assessedatbothpublicandprivatematernityfacilities.

Activemanagementofthirdstageoflabour(AMTSL)wasdoneinlessthanaquarterofall

cases, amongst a greater proportion (25.4%) of public sector cases compared to 21% of

private sector cases. These rates of AMTSL were higher than reported by another

observationalstudyinaneighbouringdistrictofUPwhichusedthesameWHOdefinitionI

used.77TheWHO(2014)definesAMTSLasthreecomponents:1.provisionofauterotonic

drug–Oxytocin(10IU,IV/IM)isrecommended;2.delayedcordclampingand3.controlled

cordtractioninsettingswhereSBAareavailable.402Uterinemassageisnotrecommendedin

WHOguidelines.402InfactarecentlargeclinicaltrialledbyWHO(2012)showedthatthe

most important component of AMTSL was the administration of the uterotonic drug.403

Encouragingly,Ifoundthatadministrationofuterotonicwashigh(above90%)andsimilarin

bothsectors.

Theneonatalcareindex,whichisasummaryindexforeightofthemostimportantneonatal

carepractices,wasfoundtobe41%acrosstheentiresample.Theneonatalcareindexwas

loweramongstcasesinthepublicsector(38.9%)thancomparableratesintheprivatesector

(51%). Assessmentof foetalviabilityafteradmissionbyassessing foetalpresentationand

fundalheightwasfoundtobedonein1.1%ofallobservedcases.Moreprivatesectorcases

(3.4%)receivedthisassessmentcomparedtopublicsectorcases(0.5%).Monitoringofthe

fetalheart rateat regular intervalswas foundtobedone in20%ofallcases, inagreater

proportion(73.3%)ofprivatesectorcasescomparedto6.6%ofpublicsectorcases.

ThemonitoringofApgarscoreat1and5minuteswasdoneinjust0.9%ofallobservedcases,

4.7% amongst private sector compared to none in the public sector. The Apgar score

assessment comprises of five components: colour, heart rate, reflexes, muscle tone and

respiration,eachofwhichisgivenascoreof0,1or2.Thescoreisreportedat1minuteand

Page157of248

5minutesafterbirthforallneonatesandat5-minuteintervalsthereafteruntil20minutesfor

infants with a score less than 7.404 Essentially, Apgar score is a convenient method for

reporting the status of the newborn infant immediately after birth and the response to

resuscitation if needed. Although, Apgar score measurement is recommended in WHO

guidelinesforcareatbirth114andtheIndianguidelines382asmyresultsdemonstratethese

arenotroutinelyassessed.However,someresearchershavealsoquestionedthevalidityof

theApgar score indicator sinceassessment comprisesofmany subjectiveelements. 405 In

addition, a range of factors including maternal sedation or anaesthesia, congenital

malformations,gestationalage,trauma,andinter-observervariabilitycanaffectthescore405

sotheseApgarscoresneedtobeinterpretedcautiously.

Myresultsonpoorqualityofcareforroutinenormallabourandchildbirthareinlinewith

other studies from India 70,71,78,406and fromotherLMICsettings inAfrica (Côted'Ivoire407,

BurkinaFaso,Ghana,Tanzania408)and fromArabcountries.118 In India,other researchers

have suggested that inadequate knowledge and skills, staffing shortages, poor quality in-

servicetrainings,lackofenablingenvironmentsandlimitedsupportivesupervisioncouldbe

underlyingcausesofpoorqualitycareatfacilities.66,71

Research evidence from countries such as Thailand, Malaysia and Sri-Lanka that have

achievedgoodprogress in improvingmaternalmortality indicates thatprogrammeefforts

needtogobeyond increasingcoverageof interventionsandaspecific focuson improving

qualityisrequiredwhichresearchershavereferredtoaseffectivecoverage.409,410

In the studydistricts, theQoCprovided– ineither thepublicorprivate sector–wasnot

significantlyrelatedtotheinvestigatedcharacteristicsofthebirthattendant,facilityorthe

woman’s age, caste, parity, referral status or socioeconomic status. The only covariate

associatedwithQoCwasadmissionataweekend,whichwasassociatedwith3-percentage

pointpoorerstandardofcare(p=0.03).Iwilldiscusssomeofthesefindingsingreaterdetail

below.

9.1.1.1:Poorerqualityofcareduringweekends

Manyresearchstudieshavereportedona“weekendeffect”inobstetricswithpoorQoCat

thetimeofbirthleadingtoadversematernalandperinataloutcomes.362,363,411Forexample:

a largeobservationalstudy fromtheUnitedKingomfoundhigher ratesofstillbirths,early

Page158of248

neonatal deaths, puerperal infections, injuries to the neonate, and increased three-day

neonatal admissions to the emergency room during weekends.362 Another study from

Scotlandreportedahigheradjustedoddsratioforweekendneonataldeathsof1.3(1.0to

1.6) compared with weekday within regular working hours.363 Specific to LMICs, a large

retrospectiverecordreviewstudy (2015) fromtheGambia foundthatnewbornsadmitted

during weekends were more likely to die than those admitted during the weekdays

(38%vs35%,P = 0.03).412Similarly,theriskofneonataldeathwasgreaterforthoseadmitted

out of hours than those admitted during during regular working hours

(38%vs33%,P = 0.004).412OtherresearchstudiesfromLMICshavealsoreportedfluctuations

inthenumbersofstaffsuchaslessnumbersofdoctorsornurseson-siteduringweekends

and at nights, that limits EmOC capability at hospitals. 413 In addition, laboratory, blood

transfusion, emergency referral and diagnostic services may also be limited during

weekends.412,414Reseachershavesuggestedthatdeficienciesinstructuralelementsofcare,

limitedresourcesandpoormanagementofmaternityservicesduringtheweekendsarethe

reasonsbehindpoorerobstetricandneonataloutcomesduringweekends.362,363

9.1.1.2:SimilarQoCcareprovidedbyunqualifiedandqualifiedmaternitycarepersonnel

My results indicate that the majority of deliveries (59%) were attended by unqualified

personnel in maternity facilities in UP. Research evidence from observational studies in

Rajasthan,whichisanotherstatewitharelativelysimilarhealthindicators,hasalsofound

thatunqualifiedpersonnelwereinvolvedinprovidingcareduringlabourandchildbirth,inup

tohalfofallobservedcases,andthatthereweresignificantdeficienciesinquality.70,71

However,inthemultivariateanalysis,IdidnotfindasignificantdifferenceinQoCprovided

by qualified and unqualified attendants. There could be many reasons that explain this

finding. First, my observations were limited to normal vaginal births which are a normal

physiological event and had I measured QoC for complications of pregnancy, perhaps,

maternity personnel’s qualificationsmayhaveemergedas a strongerpredictor for better

quality.

Second,thequalityoftrainingsreceivedbyqualifiedpersonnelmaybepoorandtheymay

notbeawareofup-to-datetechnicalguidelinesandthereforeunabletoprovidehighquality

care.ThequalityofmedicaleducationinbothpublicandprivatemedicalcollegesinIndiais

Page159of248

knowntobevariableandpreviousresearcheffortshavefoundthatthetechnicalcomptence

ofhealthworkerscanvarydependingonwheretheyreceivedtheirtrainings.178,415

Third, although, qualifiedpersonnelmayhave receivedhighquality trainings andpossess

goodknowledgeandskills,theymaynotbeabletoapplytheseintoregularclinicalpractice

as a resultofmanydifferent factors suchas resource-constraints, highwork-load, limited

incentivesandothersashighlightedpreviously.

Fourth,giventhattherearenoexistingmechanimsfortrainingunqualifiedmaternitycare

personnel(TBAs,ASHAs,BHWs,BSWs,cleaners)asapartofongoinggovernmentinitiatives,

perhaps, unqualified personnel learn informally on-the-job. My observations during field

workanddialoguewithIndiancolleaguesonthisissueconfirmsthisfinding.Since,maternity

carereliesheavilyonteamwork, theseunqualifiedpersonnelpickupessential skills from

qualifiedpersonnelasapartoftheirroutinework.Itmayalsobepossiblethatthroughthese

informalon-the-jobtrainingmechanims,unqualifiedpersonnelareabletogainequivalent

practicalskills,similartowhattheywouldobtainedthroughformaltraining.

Evidencefromameta-analysisofaudit-basedstudiesaimingtoidentifyavoidablefactorsfor

maternalandperinataldeathsinlow-resourcesettingshasidentifieddeficienciesincareby

healthworkersasthemostimportantfactorforavoidablematernalandperinataldeaths.416

SeveralstudiesfromLMICssuchasAfganistan417,Nigeria418,Pakistan75havereportedgapsin

knowledge and skills of SBAs, similar to those reported by Harvey et.al (2007) from

assessmentsinBenin,Ecuador,JamicaandRwanda364.Astudyusingstandadizedpatientsin

India also found limited differences in QoC provided by unqualified and qualified health

workers, although this study was not specifically focussed on maternity services. 178

Furthermore,itcanbeassumedthatqualificationsonpaperdonotguaranteethathealth

workers have adequate skills, up-to-date knowledge and clinical competence for proving

maternity services. Similarly, just because a doctor, nurse or a midwife meets theWHO

definedcriteriaforSBAsdoesnotmeanthattheyareadequatelyskilled.364Recevingaskilled

birthattendanttrainingcourse,feelingcompetentabouttheirexpertiseandapplyingthese

knowledgeandskillstodailyclinicalpracticeareseparateissues.

AsdemonstratedbytheLMICstudiesmentionedearlier,eventrainedSBAsoftenhavegaps

in their comptence and this may result in feeling under-qualified or uncomfortable in

Page160of248

managingconditionsduringlabourandchildbirth.Although,shortagesofadequatenumbers

ofSBAsandtheabsenceofformalmidwiferycadresareimportantissuesforIndia,ensuring

competenceofexistingSBAsalsoseemstobeanimportantbarrierforimprovingQoCatthe

timeofbirthinIndia.However,itisencouragingtonotethattheGovernmentofIndiaandits

partnersare implementinga rangeof schemes to improve thequalityof intrapartumand

immediatepostpartumcare.366

9.1.1.3:NoassociationbetweenQoCandcharacteristicsofthewomenandhospital

IdidnotfindasignificantrelationshipbetweenfacilitysizeandQoCatthetimeofbirthwhich

couldbeexplainedbythefactthatmyclinicalobservationswerelimitedtouncomplicated

vaginalbirthsandQoCinthissettingwasdeficientacrossallsampledfacilities.Previouscross-

sectionalstudiesfromAfrica(Tanzania)andSouthAsia(NepalandSrilanka)havefoundbetter

QoC at higher level facilities, potentially explaining why patients bypass lower level

facilities.348,419,420 In the study fromNepal (2013), themostpopular reasons identified for

bypassing smaller birthing centres to deliver at larger urban hospitals, despite incurring

additional costs, were found to be non-availability of operating theatres and inadequate

drugsandequipmentatsmallerbirthingcentres.420

Ialsodidnotfindasignificantrelationshipbetweenwomen’sage,caste,socio-economicor

referralstatusandQoCinthemultivariateanalysis.However,Ididfindagreatervariancein

QoCwithinindividualhealthworkersthanbetweenthemwhichsuggeststhathealthworkers

may not systematically follow standard protocols or provide preferential care to some

women.

9.1.2:MistreatmentofwomenfrequentlyoccurredatmaternityfacilitiesinUttarPradeshin

2015

Chapter 7 examined the nature and context ofmistreatment amongst women attending

public and private sectormaternity facilities in Uttar Pradesh. I found that all pregnant

womenencounteredatleastonepracticedefinedasmistreatment.Myestimatesaresimilar

toanothercross-sectionalstudyfromateachinghospitalinsouth-easternNigeriawhere98%

ofwomenreportedsomekindofmistreatmentduringchildbirth.374Similarly,anothercross-

sectional study in Ethiopia also found a high prevalence ofmistreatmentwhere 100% of

womenthatwenttoateachinghospitaland89.4%thatwenttoperipheralhealthcentres

encounteredsomeformofmistreatment.421

Page161of248

Asmentionedpreviously, there is now substantial researchevidencewhich indicates that

mistreatmentiswidespreadinbothhighandlowincomecountries.80,128-134Further,newer

research evidence is also emerging including from Uttar Pradesh which suggests that

mistreatmentmayalsobeassociatedwithmaternalhealthcomplicationsduringdeliveryand

thepost-partumperiod.422Hence,mistreatmentisnotjustarights-basedissuebutalsoa

medicalandpoorQoCissue.Moreover,weknowthatwomenwhoaremistreatedareless

likelytocometofacilitiesforfuturedeliveries80sothisisanimportantissuethatneedstobe

addressed.

9.1.2.1:Commonpracticesofmistreatmentinpublicandprivatesectorfacilities

Ifoundahigherprevalenceofverbalabuse(shout,threaten,talk-down)thanphysicalabuse

athealthfacilities.However,physicalabusewasparticularlyhigheramongwomenabove35

yearsofageandthoseattendingthepublicsectorfacilities.Myinformalobservationsduring

datacollectionwereconsistentwithother studies inMadhyaPradhesh66andRajasthan78,

that found labour roomenvironmentswerechaoticandhealthworkerscanbedominant,

abusiveandthreateningonoccasions.66Myimpressionsduringfieldworkalsosuggestthat

verbalabuseoccursmuchmorefrequentlythanphysicalabuse.

Ifoundthatthemostprevalentpracticesofmistreatmentwerenotofferingwomenachoice

ofbirthingposition(92%)andperformingroutinemanualexplorationoftheuterus(80.4%)

whichwere similar across facilities in both sectors. My estimates on healthworkers not

offeringwomenachoiceofbirthingpositionareinlinewithothercross-sectionalstudiesin

AfricaandAsia.421423Bohrenetal.’ssystematicreviewonbarrierstoinstitutionaldeliveries

identifiedthatwomenbeingaskedtoadoptunfamiliarpositionsandnothavingcontrolover

theirpositionduringchildbirthareimportantreasonsforwomenchoosingtodeliverathome.79InqualitativestudiesfromBangladeshandUganda,researchershavereportedthatsince

healthworkersarenottrainedtodeliverwomeninpositionsotherthanlyingontheirbacks,

theyarenotcomfortablewithofferingalternativebirthingpositions.424,425

Interpersonal communicationbetweenbirthattendantsand labouringwomenwas largely

non-existentasdemonstratedbythehighprevalenceofcaseswhereexplanationswerenot

providedtowomenpriortoinvasiveprocedures.Thesefindingsaresimilartothosereported

inotherIndianstatessuchasinRajasthanandMadhyaPradesh.66,70

Page162of248

9.1.2.2:Mixedpatternsofmistreatmentinpublicandprivatesectorfacilities

Privatesectorfacilitieswerefoundtoperformworsethanthepublicsectorfornotallowing

birthcompanionsandforperineal/pubicshaving.Thiscouldbebecauseofexistinglabour

roompoliciesinprivatehospitalswhichdonotallowbirthcompanionsastheymaynotbe

awareof the latest recommendationsonbirth companionshipor perhaps they think that

limiting the number of people in the labour room is better for infection prevention and

control.Itmayalsobepossiblethathealthworkersinprivatehospitalsfeelthatsincethey

alreadyprovidepersonalisedandcomprehensivematernitycare,birthcompanionsarenot

needed.IntheIndiansetting,birthcompanionsgenerallytendtobefamilymemberssuch

asmothers,mother-in-laws,sistersorthehusband.Asmentionedpreviously,evidencefrom

a systematic review indicates that continuous support fromachosen familymemberora

friendincreaseswomen’ssatisfactionwiththeirchildbearingexperience.380

Perineal/pubicshavinghasnoassociatedclinicalbenefits381andisnotrecommendedinthe

Indian skilledbirth attendance trainingmaterials382,which suggests that, perhaps, private

sectorhealthworkersmaynothavereceivedthesetrainingsorthatqualityofsuchtrainings

ispoor.

Ontheotherhand,thepublicsectorwasfoundtoperformworsethantheprivatesectorfor

notensuringadequateprivacy,notinformingwomanpriortoavaginalexamination,andfor

physical violence towards pregnant women. There could be many reasons such as

infrastructure-related deficiences (limited number of beds or screens), larger number of

clients,poorcommunication,normalisationofdisrespectandabuse79,80inthepublicsector

inUttarPradesh.

Research evidence from LMICs has identified factors such as unfavourable institutional

policies, resource and infrastructural constraints, socio-cultural factors, poor working

conditions, limited mentorship and supervision, limited knowledge and skills of health

workers of health workers as underlying causes for mistreatment of women which are

relevantinthissettingtoo.79,80,368

9.1.2.3:Somesocio-demographiccharacteristicsareriskfactorsformistreatment

Ifoundthattotalmistreatmentscoreswerehigheramongstwomenabovethan35yearsof

age (5.1),primiparous (5.2), those thatwere referred fromanother facility (5.0), amongst

Page163of248

womenbelongingtothe“scheduledcasteandtribe”(5.0),thoseinthefifth(richest)wealth

quintile(5.1),andamongstcasesadmittedduringregularwork-hours(5.0)onweekdays(5.0)

inthepublicsector(4.9).Across-sectionalstudyfromurbanslumsinUttarPradeshsimilarly

foundthatwealthierwomen,migrantwomenandwomenfromlowercastesself-reported

higher levels of disrespect and abuse, although this study was not based on actual

observations.375

The importanceofcaste iswelldocumented in Indiaandresearchershavesuggestedthat

since“scheduledcasteandtribe”womenarelessempowered,healthworkersaremorelikely

to think that they can get away withmistreatment of these women.375 However, in the

bivariate analysis, caste was only associated with episiotomy and women in the higher

“generalcaste”categorieswerefoundtohavegreaterproportionsofroutineepisiotomies

perhapsbecausetheyusedprivatesectorfacilitiesmoreoften.Womeninthefirstquintile

(poorest)were least likely tobe informedprior to a vaginal examination (p=0.002)which

suggests discriminatory care based onwealth status.378 However, women in the highest

wealth quintile (richest) were more frequently unaccompanied by a birth companions

(p=0.01), had higher rates of perineal shaving (p=0.001) and episiotomy (p=0.001) which

could perhaps reflect greater use of the private sector and consequent overtreatment of

womenthatattendprivatesectorfacilities.

9.1.2.4:Under-treatmentandover-treatmentofwomenatmaternityfacilitiesisalsomistreatment

InChapter7,Idemonstratedandarguedthatunder-treatmentsuchasthroughtheuseof

unqualifiedpersonnelwhomaynotbe capableofprovidingessential carepracticesorof

deliveries taking place in unhygienic conditions are against the rights of childbearing

women141andthereforeshouldbeconsideredmistreatment.Similarly,overtreatmentsuch

asthroughnon-adherencetoevidence-basedprotocolsorroutineuseofharmfulpractices

(uterinelavage,episiotomyorenemas)alsooccursfrequentlythatareagainsttherightsof

childbearingwomen141.Essentially,mistreatmentintersectsqualityofmaternalhealthcare

andrelatestocarethatisbothTooLittleTooLateandTooMuchTooSoon83.

TherecentWHOstatementondisrespectandabuse(2014)indicatesthatmistreatmentis

nowconsideredaserious issueattheglobal level.123TheUnitedNationshasalso issueda

resolution on preventablematernalmortality as a human rights violation426 and issued a

technicalguidanceontheapplicationofahumanrights-basedapproachtoreducematernal

Page164of248

deathsin2012.427Inaddition,bodiessuchasFIGO,ICM,WHOandothershavealsoinitiated

a“MotherandBabyFriendlyBirthingFacilities”initiativewhichamongstotherthingsstates

that “Every woman and every newly born baby should be protected from unnecessary

interventions,practices,andproceduresthatarenotevidence-based,andanypracticesthat

arenotrespectfuloftheirculture,bodilyintegrity,anddignity”.428

9.1.2.5:Mistreatmentandtypeofmaternitycarepersonnel

Although I found that the majority of pregnant women were cared for by unqualified

attendants(unweightedestimate:59%)andtheyweremoreprevalentinthepublicsector,

the aggregate scores formistreatment were higher for deliveries conducted by qualified

attendants(4.9)ascomparedtounqualifiedattendants(4.8)whichsupportsthenotionof

overtreatmentbyqualifiedpersonnel.

Uponcloserexamination,unqualifiedpersonnelweremorelikelynottoinformwomenprior

to a vaginal examination (p=0.01) and use unsterile gloves (p=0.04). This indicates poor

interpersonal communication and lack of knowledge amongst unqualified personnel.

However,qualifiedpersonnelweremorelikelyconductunnecessaryproceduressuchpubic/

perineal shaving (p=0.02) and episiotomy (p=0.003) which suggests either unfavourable

institutionalpoliciesoroutdatedknowledgeofhealthworkersresultinginovertreatment.

An important issue to note at this time, relates to the problems in conceptualising and

measuringmistreatment.Forexample,thesepracticesoutlinedabovesuchaspubic/perineal

shaving or routine episiotomy or fundal pressure are not evidence based and can be

harmful.83 However, health workers are often trained to do these things and they may

genuinelybelievethatthesepracticesareforthewoman’sbenefit.Therefore,itisimportant

think further about measurement of mistreatment, and whether the act or the harmful

practicewasintentionalornot.Iwillelaborateontheseissueslaterinthesectiononfuture

recommendations for research. Moreover, the research community will need to think

carefullyabouthowtoaddressmistreatmentcomprehensivelyandtakeabalancedapproach

withoutblaminghealthworkerswhoalsoworkindifficultsituations.

9.1.2.6:Informalpaymentsasalsoformofmistreatmentofwomeninmaternityfacilities.

I foundthat informalpaymentswere routinelydemandedbyhealthworkers in thepublic

sector. These informalpaymentsoftendetermined theQoC receivedbywomen inpublic

Page165of248

sectormaternityfacilitiesin2015.Informalpaymentscanrangefromgratuitypaymentsfrom

appreciative patients, payments to jump the queue, receive better or additional care, to

obtaindrugsandcommodities,orsimplytoreceiveanycareatall.385Informalpaymentsare

consideredtobeinequitableandconstituteinstitutionalisedbribery,whichmayhamperthe

entire health system.385,386 Further, they tend to be prevalent in settings where health

systems are under-funded, supervisory mechanisms are weak; where women are not

empoweredornotawareoftheirrights,andwhereprovidersareunlikelytofacedisciplinary

actionfortheirbehaviours.385

9.1.3:OverallmanagementscorewasnotassociatedwithQoCatmaternityfacilitiesinUttar

Pradeshin2015

Inchapter8,ImeasuredanddescribedmanagementpracticesatmaternityfacilitiesinUttar

Pradeshandexaminedwhethermanagementpracticeswereassociatedwithqualityoflabour

andchildbirthcare.TheresultsfromthisinvestigationfoundthattheQoCandmanagement

practiceswerebothpoor inmaternity facilities inUttarPradesh, India. In this setting,my

resultsindicatemanagementpracticesattheinstitutionalleveldonotinfluenceQoCduring

labourandchildbirth.Theonlymanagementdimensionthathadasignificantassociationwith

QoCwasperformancemanagementwhichwasfoundtobeassociatedwithuptoaseven

percentage point higher quality score. The key results from my investigation into

management practices and quality of care at the studied facilities in 2015 is summarised

below.

9.1.3.1:ManagementpracticeswerepooratthestudiedmaternityfacilitiesinUttarPradesh

IfoundthattheoverallmanagementscoresreceivedbymaternityfacilitiesinUttarPradesh

was low (1.6 on a scale of 1 to 5). Public sector facilities received a lower score of 1.5

comparedtotheprivatesectorfacilitieswhichreceivedascoreof2.0.Theprivatesectoralso

outperformed the public sector for allmanagement dimensions; operationsmanagement

(private:2andpublic:1.7),performancemanagement(private:1.9andpublic:1.5),targets

management(private:1.6andpublic:1.2)andpeoplemanagement(private:2.4andpublic:

1.2).

The lowperformanceof thepublic sector suggests that variousbottlenecksexists for the

implementationofthesemanagementbest-practicesinthepublicsector.Thesebottlenecks

couldincludeissuessuchaslimitedautonomyofmanagersinthepublicsectorwithbudgets,

Page166of248

human-resourcemanagementdecisions,andlimitedabilitytoincentivisebetterperformance

inthepublicsector.Thesefindingsareinlinewithpreviousstudiesthathavefoundhigher

managementscoresintheprivatesector.42Although,mydescriptiveresultsshowedhigher

scoresforallmanagementdimensionsandqualityofcareintheprivatesector,Ididnotfind

anoverallstatisticalassociationbetweenmanagementandqualityofcareinthemultivariate

analysis.

MyresultsarealsocompatiblewiththefindingsreportedbyalargerIndiansurveyin3,892

private sectorhospitals thatused the survey tool thatweadaptedbutwasdone through

telephoneinterviews.44ThisstudybyLemosetal.(2012)reportedatotalmanagementscore

of 1.9, which is comparable to our private sector score of 2.0. Similarly, scores for all

managementdimensionsobtainedbytheprivatesectorsamples inourstudywere in line

withthosereportedbytheLemosetalstudy(2012).Forexample,operationsscore(2.0to

2.1), performance score (1.9 to 2.0), target management score (1.6 to 1.6) and people

management(2.4to1.9).44

The scores obtained by the facilities in UP in 2015 were found to be poorer than

correspondingscoresobtainedbyhospitalsinUS(3.1),UK(2.9),Sweden(2.7),Germany(2.6),

Canada(2.5), Italy (2.5)andFrance(2.4)thatweredoneusingasimilartool. 44Specificto

UttarPradesh,myscoreswere0.2pointslowerthanwhatothershavefoundinUttarPradesh

usingthesametool44,however,Isurveyedmorepublicsectorfacilitieswhichmayexplain

thedifference.

9.1.3.2:Performancemanagementwastheonlymanagementdimensionassociatedwithbetter

qualityofcareatthestudiedmaternityfacilitiesin2015

Ifoundthatfacilitieswithbelowmedianmanagementscoresprovidedanaverageof39%of

all recommended seventeen interventions to women compared to 34% by facilities with

above median management scores but this difference was not statistically significant

(P=0.28).Similarly,forobstetriccare,bettermanagedfacilitieswerefoundtoprovide30%of

therecommendedinterventionscomparedto34%ofrecommendedinterventionsinpoorly

managed facilitiesbut thisdifferencewasnot statistically significant (p=0.4).Fornewborn

care,bettermanagedfacilitiesprovided39%oftherecommendedinterventionscompared

to poorly managed facilities that provided 44% of the recommended neonatal care

interventionsbutthisdifferencewasalsonotstatisticallysignificant(p=0.13).

Page167of248

Results fromthemixedeffectsmodels confirmed that therewasno statisticalassociation

betweentotalmanagementZscoreandQoCinboththeunadjusted(p=0.85)andadjusted

models (p = 0.55). This finding isnot consistentwithprevious research fromhigh-income

settings43,45,308; however, none of these studies were done in low-income settings or

specificallyfocussedonqualityofmaternitycare.

Amongstallmanagementdimensions,performancemonitoringwastheonlydimensionthat

hadasignificant relationshipwithQoC(adjustedpvalue=0.02)withone-unit increase in

performancemonitoring associated with a 7-percentage point higher quality score. Our

findings are in line with previous studies that have shown performance monitoring may

encourage the use of evidence-based-practices, improve supportive supervision of health

workers,encourageregularmonitoring,andreportingonperformanceindicators.293

TheeffectivenessofauditsandfeedbackwasevaluatedinaCochranereview,whichfound

thatauditsandfeedbackinterventionshavethepotentialforamodestimprovement(median

+4.3%)inhealthworkercompliancewithdesiredpractice.195 Inaddition,thereviewfound

thataudits and feedbackareparticularlyeffectivewhenbaselineperformance is low, the

sourceisasupervisororacolleague,itisdonemultipletimes,deliveredinbothverbaland

written formatsand includesexplicit targetsandanactionplan. 195Ashospitalsorhealth

facilitiesarethemostexpensiveandimportantcomponentsofanyhealthsystemwhetherin

LMICsorinHIC,performancemonitoringhaspotentialtobeusefulinallsettings.392

Lastly,afteraccountingforconfoundersinthemultivariateanalysis,Ialsofoundthatwomen

attendingprivatematernityfacilitiesreceiveda7-10%pointhigherstandardofcarewhichis

consistentwiththeresultsreportedinchapter6.

9.2:PlansfordisseminationUponcompletionofclinicalobservationsinindividualhealthfacilities,theresearchteamand

Iroutinelydebriefedwiththehealthfacilitymanagerand/orseniorclinicalstaff.Duringthese

debriefingmeetings,wediscussedouroverallimpressionsofQoCatthesefacilities.Wealso

discussedspecificcaseswheregrosslynegligentcare(forexamplemistreatmentofwomen)

orcaseswherelifethreateningmaternalandneonatalcomplications(suchasPPHorbirth

Page168of248

asphyxia)wereobservedbytheresearchteam.Hospitalauthoritieswerealsoinformedthat

if theywere interested, I would be happy to send them preliminary results from clinical

observational data collected at their facility. These results could potentially be useful for

initiatingqualityimprovementworkatindividualfacilities.

Ialso receivedanopportunity topresentpaperoneofmyPhDatapolicyseminarat the

World Health Organization, Switzerland on June 14, 2017.My paperwas included in the

BulletinoftheWorldHealthOrganization’sspecialthemeissueonqualityofcareintheera

of the Sustainable Development Goals (SDGs). Further details on the policy seminar are

available here. https://www.wider.unu.edu/event/policy-seminar-launch-who-bulletin-

theme-issue-measuring-quality-care

Myoverallresearchfindingswerealsosharedwiththefunderandimplementingpartnersin

aworkshopinSeptember2017andotherperiodicmeetings.Thereisanationaldissemination

eventplannedinFebruary2018wheretheresultsofthestudiesreportedinmyPhDwillbe

widelydisseminatedamongstnationalandstatelevelstakeholdersinIndia.Ihopethatthis

event isable tohighlight theurgentneedto improveexistingmaternitycarestandards in

UttarPradeshandotherstatesinIndia.Asmentionedpreviously,disseminationofevidence-

basedguidelinesandtheconceptsofrespectfulmaternitycareamongstallfront-lineworkers

wouldbeusefulinimprovingqualityofmaternitycare.

Lastly, thereareon-goingdiscussionsaboutadisseminationevent inLondon,planned for

April2018,wherethelearningfromallMETprojectswillbesynthesizedanddisseminatedto

theacademicandresearchcommunity.

9.3:Reflections,strengthsandlimitations

9.3.1:Forobjective1:QoCduringlabourandchildbirthatmaternityfacilitiesinUP

9.3.1.1:Reflections

Inchapter6,Iassessedanddescribedprocessesofcareduringlabourandchildbirththatwere

investigatedusingclinicalobservationsconductedat26maternityfacilitiesinUttarPradesh.

AnimportantconsiderationwhilemeasuringQoCishowfartogowhendefiningandselecting

appropriate QoC indicators. Although, conceptually, QoC has been thought by some to

encompassmultiple levels from patients to health systems and health policies86, for the

Page169of248

purposes of measuring actual services at health facilities, it is not ideal to measure

bottlenecksinhealthsystems16butratherpreferabletofocusonelementsofdirectservice

provisionandexperiencesofclients.

Hence,fortheQoCassessmentsinmyPhDstudy,IonlymeasuredprocessesofcarewhichI

definedastheapplicationofevidence-basedguidelinesandprovisionofrespectfulwoman-

centredmaternitycare.Althoughotheraspectsofqualitysuchasthoserelatedtostructure

(measured through an inventory assessment ofmedicines, infrastructure and supplies) or

outcomes(measuredthroughhospital-baseddataorspecialstudies)areimportant,theydo

not provide a comprehensive picture of quality of care during labour and childbirth that

womenreceive. Inaddition,thecontributionofmyPhDistoadvancethethinkingaround

measurement of processes of care for normal labour and childbirth which included the

developmentofthreeinnovativeindices.

TodeveloptheQoCindices,Ihadtothinkcarefullyabouthowtogroupvariousclinicalitems

intoclinicalpracticesandthendecideonhowtodevelopdifferentaggregateindices.While

developing these indices, I grouped individual items into clinical practices based on their

inherent clinical logic and their purpose rather than their relative importance in avoiding

adverseoutcomesastherewasnoscientificbasisfordoingso.Forexample,Ididnotapply

weightstointerventionssincemanyinterventionsusedforlabourandchildbirthdonothave

evidenceofefficacyastrialsonthesepracticeswouldbeunethicalforexample:monitoring

ofpost-partumbleedingorsterilecordcutting.Therefore,tokeeptheindicestransparent,I

gaveequalweighttoallindividualclinicalpractices.Iftherehadbeenevidenceforapplying

intervention-specificweights, the indiceswouldperhapshavebeenmore robust.Another

option would have been to generate a consensus on the importance of specific clinical

practicesbyundertakingamodifiedDelphi-methodapproach,asothershavedone,201but

suchmethodsarealsoknowntobesubjective429.Iwillelaborateontheseissuesfurtherin

recommendationsforfutureresearch.

9.3.1.2:Strengths

SomeofthestrengthsofthisQoCstudywerethatIdevelopedacomprehensiveassessment

toolwhichallowedforanintegratedassessmentofbothmaternalandnewborncarepractices

aroundthetimeofbirth,whichisoftenlackinginmanyoftheotheravailableassessment

tools. This tool is a direct output from this study and has also been used inUganda and

Page170of248

Rajasthan.ItishopedthatitwillalsobeusefulforfutureresearcheffortstomeasureQoCin

otherhigh-burdensettings.Sinceitfocusesonnormalbirths,itsapplicabilityalsoextendsto

smallerbirthingcentresinLMICsettings.Ialsoselectedallwomenwhocametodeliverinthe

studyfacilitiesandtherewerenorefusalsbywomentorecruitmentwhichisastrengthofthe

study.OtherstrengthsoftheQoCstudywerethatIdidnotrelyonself-reportedbehaviours

orfacilityrecordsandconductedclinicalobservationsround-theclockonallsevendaysof

theweek.

To overcomemeasurement errors, QoC assessment tool was developed as a structured,

standardizedtoolandwaspre-testedandpilotedpriortofieldapplication.Timeandeffort

was invested into training of observers so that they were competent in using the study

instruments.Thislimitedsubjectivitybyobservers.Further,allobserversusedatindividual

maternityfacilitieswereexternaltothatfacilitytominimiseanyinherentbiases.Toprevent

misclassificationofpregnanciesasnormal,allobservershadbeentrainedtofollowastrict

casedefinitionofnormalvaginalbirthsandcasesthatdidnotfulfilthiscasedefinitionwere

excludedfromthestudy.Dataqualitywasassuredthroughdailyqualitychecks.

Theessentialcareatbirthindexisaninnovativeanalyticalframeworkthatcapturestheuse

ofevidence-basedinterventions,useofrespectful,woman-centredcarepracticesaswellas

patternsofharmfulcare.Alltheindices-theessentialcareatbirthindex,neonatalcareindex

andobstetriccareindexcanalsobeusedindividuallyformonitoringmaternalandnewborn

healthprogrammeefforts.Thestrengthsofcreatingsuchaggregateindicesincludetheability

tocommunicatealargeamountofinformationandconveyacomprehensivepictureofQoC

inasuccinctmanner.SinceIwantedtheseindicestobeusefulforprogrammeimprovement

efforts,Idecidedtodevelopthreeseparateindices.Dependingonthespecificareaofconcern

forquality improvement, interestedresearchersandprogrammemanagersmayusethese

indices individually, as appropriate. However, interested researchers who use the

methodologydescribedinmyPhDtoconductclinicalobservationshavetobecarefulwhile

interpretingresultsfromsuchaggregateindices.Theywillhavetothinkcarefullyaboutwhat

arethespecificpracticesthatmakeuptheaggregateindexhaveaweak(er)orhigh(er)score.

Forexample:womancentredrespectfulcarepracticesmaybe theweakestelements that

bringdowntheentireobstetriccareindexorestablishingskintoskincontactmaybeoneof

thepracticesthatbringsdowntheentireneonatalcareindex.

Page171of248

Lastly,sincethesamplingforthepublic-sectorfacilitieswasdoneusingastratifiedrandom

samplingmethodology,andtheanalysisalsousedpopulationbasedweights,Iamconfident

thattheestimatesobtainedreflecttherealsituation inpublicsectorfacilities inthestudy

districts.

9.3.1.3:Limitations

Mymeasurementapproachisresourceandtime-consumingsoitmaynotbepossibletoscale

thisupbeyonddedicatedresearchprojects.However,efforts to improvemeasurementof

QoC at the time of birth are evolving rapidly despite the fact that WHO guidance on

measurementofQoCduringlabourandchildbirthdoesnotexisttilldate.

MystudywasonlyconductedinthreedistrictsofUttarPradeshwithintheframeworkofan

externalevaluationofthe“Matrika”projectsomyfindingsmaynotberepresentativeofthe

overallsituationinUttarPradesh.

Observerbias:Theremayhavebeenobserverbiasinthestudyduetothepopularperception

thattheprivatesectorissuperiorsinceithasbetterinfrastructure,bettertrainedpersonnel

whoarealsobetterpaid,leadingtoanover-estimationofqualityinprivatefacilities.

Selectionbias:Therewerechallengestosamplingtheprivatesector.Notonlydid13private

facilities refuse to participate, I had no official sampling frame fromwhich to select the

facilities. Therefore, it is possible that the QoC of the participating private facilities was

differentfromthosethatwereeithernotsampledorrefusedtoparticipate.Although,Ihad

the necessary permissions from the government and ensured confidentiality of any data

collected;asprivatelyownedfacilities,theywerenotobligedtoparticipateinmystudy.In

addition, obtaining detailed information from the private sector on their staff, their

qualifications,extrahoursofwork,numbersofcaesareansectionsandotherswasparticularly

problematicevenat facilitiesthatconsentedtotheclinicalobservations.Asreportedbya

qualitativestudythatinterviewedprivatesectorobstetriciansinthestatesofUttarPradesh

and Jharkhand in India, there is often a trust deficit between the private sector and the

government.430 My overall impressions during field work was that the private sector

authorities were very cautious with external entities given the volatile socio-political

environmentandmediastingoperationsthatfrequentlyoccurinUttarPradesh.

Page172of248

Ialsohadlimitednumberofcasesintheprivatesectorcomparedtothepublicsectorasmost

privatesectorbirthsresolvedascaesareansectionsandthisisanimportantlimitation.My

research focussedon examinationof normal labour and childbirth and I did notmeasure

qualityofcareforcomplicationsofpregnancysuchaspost-partumhaemorrhageorsevere

eclampsiaorforCEmOCprocedureslikecaesareanoperations.Itispossiblethatadifferent

pictureofqualitymayhaveemergedbetweenpublicandprivatesectorshadImeasuredQoC

for these issues. Perhaps, the public sectormayhave performedbetter for being able to

manage pregnancy complications or have fewer non-indicated caesarean operations

comparedtotheprivatesector.

However, there aremany challenges associatedwithmeasuringQoC for complications of

pregnancies.Someoftheseincludelimitedavailabilityofappropriate,validassessmenttools,

problemsinrecruitingspecialistssuchasobstetriciansandgynaecologiststoworkasclinical

observersthroughoutthedurationofthestudy,alongertimeframeisrequiredtogetthe

optimalnumberofobservationsandthereispotentialforobserverbias,aswithanyclinical

observation. A recent study from Afghanistan that usedmixed-methods including clinical

observationsofcaesareandeliveriesreportedthatdirectobservationswereafeasibleand

effective method for assessing QoC of caesarean deliveries in low resource settings.431

However,inthisstudy,researchersrecommendthatalongwithclinicalobservations,others

methodssuchasrecordreviewsandinterviewswithhealthworkersshouldbeundertakenso

thatacomprehensivepictureofqualitycanbeobtained.431

UsingaggregateindicesisusefultoreportcomprehensivelyonQoCbutitmasksdifferences

betweenindividualindicators.Also,anotherlimitationisthedifficultlyinidentifyingwhya

particularindexhasalowscoreoraparticularpracticeisweak,beyondcommonindividual

orfacilitybaseddeterminants.Inordertochoosethenecessaryactionsrequiredtoimprove

QoC,researcherswillneedto identifytheexactreasonsbehindpoorscores.Forexample,

poorratesofuterotonicdrugusewithin1minutemaybeduetomanyproblemssuchasnon-

availability, incorrect formulation, incorrect timing, lack of knowledge, poor injection

technique,poorlymotivatedstaff,poorworkingconditions,norefrigeratortostoreoxytocin

andothers.

TheHawthorneeffectreferstoaphenomenonwheresubjectsunderobservationmayalter

theirbehaviourpreciselybecausetheyarebeingobserved.181Theconcerninthisstudywas

Page173of248

thathealthworkersmayhaveperformedbetterwhenbeingobserved thanundernormal

conditions. To help address this problem, maternity care personnel were blinded to the

detailsoftheprimarymeasuresofthestudy.Informationsheetsprovidedtomaternitycare

personnelaspartoftheconsentproceduresemphasisedthatobservationswerenotmeant

toassess theirpersonalperformance, information collectedwill notbe liked to individual

providersandstudyfindingswillnotresultinpunitiveaction.Duringtheanalysis,Ididnot

examine individual performance of any specific maternity care personnel. However, all

observationsweretime-stampedsothatIcouldexplorethepresenceofHawthorneeffect

during analysis. It is encouraging to note that any Hawthorne effect is negligible in this

study.47

Lastly, Ididnot lookatmaternalandperinataloutcomes inmystudyas thatwouldhave

requiredlargersamplesizes,largertimeperiodfordatacollectionandadditionalfunding.As

highlightedpreviously, improvedprocessesof caredonotguaranteebettermaternal and

perinatal health outcomes. However, the global maternal health community is eagerly

awaitingtheresultsofacluster-randomizedcontrolledtrial inUttarPradeshknownasthe

betterbirthtrial.432 Inthistrial, researchersareevaluatingthe impactofa safechildbirth

checklist embedded within a quality improvement programme with a nurse “mentor”

providingsupportivesupervisionandreal-timefeedbackonQoCathealthfacilities.432Asper

their protocol, researchers are expected to report on a range of outcomes including

compositemeasures of maternal deaths, maternal severemorbidity, intrapartum-related

stillbirths,andnewborndeathoccurringwithin7daysafterbirth.432

In summary,myoverallexperiencewithclinicalobservationsof labourandchildbirthwas

promisingand,asmyresultshaveshown,withcarefuldesignandplanning,itispossibleto

conductarobustobservationalstudy.

9.3.2:Forobjective2:MistreatmentofwomenatmaternityfacilitiesinUP.

9.3.2.1:Reflections

Tomeasure and describemistreatment ofwomen atmaternity facilities, I used amixed-

methodsapproachusingquantitativedataobtainedfromclinicalobservationsandqualitative

data from unstructured observers comments. Other researchers have also measured

Page174of248

mistreatmentbyaskingwomenquestionsabouttheirmaternityexperiencesduringhospital-

exitinterviewsorhouseholdsurveys.376

Upon reflection, itmayhavebeenuseful to conduct some in-depth interviewsamongsta

sample of recently delivered woman and health workers as a part of my PhD study.

Understandingwomen’sinsightsandperceptionsofmistreatmentwouldhavebeenusefulto

understandtheculturalandcontextual issuesaroundmistreatment inUP. Interviewswith

health workers would have provided me with additional information on whether health

workersunderstandwhatmistreatmentis,whattheyperceiveasmistreatmentandwhether

they understand that poor experiences of womenmight affect future utilisation of their

services.However,theinnovativeaspectofmyPhDstudyisthatdataarebasedonactual

clinicalobservationsofmistreatmentincludingintheprivatesector.

Upon reflection, from a measurement point of view, it can be hard to distinguish

mistreatmentfromreceivingcarewhichisnotevidence-based,sincetheboundariesbetween

theseoftenoverlap.Forexample: isdeliverybyanunqualifiedpersonor inanunhygeinic

settingswithoutbasicamenitiesconsideredmistreatmentofwomensinceit isagainstthe

rightsofchildbearingwomen141orisitjustanindicatoroflackofresourcesorboth?.

Lynn Freedman has suggested that a definition of mistreatment should try to capture

individualdisrespectandabuse–thatisspecificbehavioursofthehealthworkersthatare

intendedtobedisrespectfulorhumiliatingsuchasslappingorscoldingthewoman.Butalso

theroleofsystemicdeficienciesthatmaycreateadisrespectfulandabusiveenvironment,for

example, anovercrowdedandunderstaffedmaternitywardwherewomendeliveron the

floor,alone,inunhygienicconditions.81

While defining and measuring mistreatment, we also need to think about whether

mistreatmentwasintentionalornot.Forexample,somepractices,suchasfundalpressureor

routineepisiotomyarenotevidencebasedandcanbeharmful,83,114butoftenhealthworkers

havebeentrainedtodothesethingsandthinktheyareforthewoman’sbenefit.Arethese

indicatorsofmistreatmentorofpoorqualityofcare?Although,healthworkersmayhave

beentaughttousetheseinterventionsinthepast,theseharmfulinterventionsarenolonger

recommended.Hence,thisconceptofintentionalitycomplicatesmeasurementefforts.83,114

Page175of248

Therefore, further conceptual clarity on the boundaries betweenmistreatment and poor

qualityofcaremaybeneeded.

Ialsohadlimitedqualitativedatafromtheobserver’scommentsanduponreflection,itmay

havebeenbettertohaveadedicatedobserverassignedspecificallyfortakingdetailedfield

notesasemployedbyotherqualitativestudies.433

Lastly,myQoCassessment tool shouldhave specifically captureddetailed informationon

informalpaymentsinthepublicsectorwhichseemtobewidespreaddespitethepresenceof

programmessuchasJSYwhichshouldintheoryensurethattherearenofinancialimplications

towomenthatchoosetodeliveratinstitutions.

9.3.2.2:Strengths

Iconductedacomprehensiveassessmentofmistreatmentofwomenatmaternityfacilities

using actual clinical observations. My PhD findings advances the understanding and

measurement of mistreatment at maternity facilities. I operationalised indicators of

mistreatmentasthoserelatedtointentionaldisrespectandabuse,overtreatmentandunder-

treatment by using a rights-based approach to conceptualise mistreatment. This

comprehensiveapproachtomeasurementisastrengthofthestudy

RatherthantakeastrictquantitativeapproachasIdidinChapter6,Ifeltthatinchapter7it

would be more insightful to explore the nature and context of care provision using the

availableinformationfromopen-endedcomments.Thisprovidesausefulcontributiontothe

literatureonmistreatmentparticularlybecausedataarebasedonactualobservationsand

weremoreobjectivecomparedtoself-reportedperceptionsofwomenasemployedbythe

vastmajorityofstudies.375,376

I also looked at public and private sector differences in the nature and patterns of

mistreatmentwhichisakeystrengthandinnovationofthisPhD.

Themixedmethodsapproachtakentotriangulatequalitativeandquantitativefindings,data

collection round-the-clock on all seven days of theweek, and the use of clinical practice

observationsweresomeofthekeystrengthsofthestudy.

Page176of248

9.3.2.3:Limitations

Thisstudyuseddatafromanobservationalstudydesignedtocapturedescriptiveinformation

onelementsofQoCfornormallabourandchildbirth.Thestudywasnotspecificallypowered

tomeasureandexplainmistreatmentasaseparatecategoryofpoorqualityofcarewhichis

akeylimitationofthisstudy.

Itisalsohardtointerpretdatafromalimitedsetofunstructuredobservers’comments.While

usingunstructuredcomments,itisimportanttonotethatfindingsmaybehardtoreplicate,

sinceobserversmayonlyrecordeventsthatwereinterestingorparticularlystrikingtothem,

whichisakeylimitationofmymethodology.Theselimitationscouldhavebeenovercomeif

Ihadusedmultipleobservers433oradditionaldatacollectionmethodsasdiscussedabove.

Inafewinstances,Ialsofoundthatitwashardforobserverstofindaprivatespacetorecord

downtheirobservations.Sometimeshealthworkersatthestudiedfacilitieswouldtrytolook

atwhattheobservershadwrittendown.Insuchcases,observershadtowaituntiltheywere

aloneorwaittilltheendofthedayandrelyontheirmemorytonotedowntheirobservations

Observerbiascouldalsooccurifobserversbecometooinvolvedoraffectedbythehospital

enviorenmentordetailsofaparticularcase. Ianticipatedmanyofthese issuesand inthe

trainingsfocussedontheimportanceofbeingsilentobserversandnotinterferingorbeing

involvedwithanyaspectsofcareprovision. Inaddition,comments recordedbyobservers

perhaps reflects theirownprofessional experiences, trainingandknowledgeof respectful

carepracticeswhichisalimitation.Duringfieldwork,Ialsonotedthatyoungerobserverswere

morelikelytotakedowndetailednotescomparedtotheolderobservers,whoweremore

experienced,andperhaps,moreinclinedtoacceptmistreatmentasanormaloccurrence.

9.3.3:Forobjective3:ManagementanditsrelationshipwithQoC

9.3.3.1:Reflections

Inchapter8,IassessedanddescribedmanagementpracticesatmaternityfacilitiesinUttar

Pradeshandexaminedwhethermanagementpracticesareassociatedwithqualityofcare.

Measuringmanagementpracticesathospitalsisanemergingfieldandtherearewidespread

opinionsaboutthedefinitions,scopeandmeasurementofdifferentmanagementpractices.

Asdiscussedearlier,mystartingpointwasapre-existing tool for themanagementsurvey

whichIusedbecauseofthewideapplicationofthetoolwhichsupportscomparisonsandan

Page177of248

opportunitytoundertakethisworksoonaftertheQoCassessmentsinthestudiedfacilities.

Ialsopilotedandfurtheradaptedthistoolpriortousingitfordatacollectionwhichprovided

manyusefulinsights.

Thekeyreflectionfromthemanagementstudy is that if Ihadthetimeandtheresources

available,Iwouldhaveundertakenamuchmorethoroughexplorationoftheconceptsand

determinantsofmanagementpracticesparticularlyinthepublicsector.Detailedformative

researchwouldhaveprovidedmewithbetterinsightsintothenatureanddeterminantsof

managementpracticesinUttarPradesh.Asaresultofthis,myassessmenttoolmaynothave

captured information on the contextual determinants of management for example local

politics,socio-economicfactorsandothers,whichmaydrivepublic-sectorperformancein

LMIC settings. This could mean that I may not have measured all the determinants of

managementathospitalsinUttarPradeshwhichhasimplicationsfortheresultsobtainedby

mystudy.

Ifoundthattheresearch-assistants(field-supervisorqualifications)oftenstruggledwiththe

fundamentals of hospital management, perhaps because none of us had any academic

traininginhospitalmanagementorbusinessadminstration.ThisiswhyIdecidedtoconduct

all of the interviews myself, although, it is encouraging to note that there was a high

correlationbetweenscoresgivenbymeandscoresgivenbythesecondsilentrater.While

orientingtheresearch-assistantstothemanagementtool,Ialsofoundthattheystruggledto

conceptualise hospital management as a separate entity from wider health systems

management,perhapsbecauseinmanywayshospitalsarealsohealthsystemsthemselves.

Thescoringmethodologyforindividualquestionscouldalsohavealsobeensimplified.For

example,insteadofaskingopen-endedquestionsandprovidingaratingbetween1to5,I

couldhavesimplifiedtheresponsetoayesornoresponsewhichmayhavebeeneasierto

implement.

Ifoundthatthestudyinstrumentwascomprehensiveandtriedtomeasureallthepractices

thatrepresentgoodmanagementinbothpublicandprivatesectors.However,itisimportant

tonotethatthetooloriginatesfrommanufacturingsectoranditsmainpurposehasbeento

make cross-country and cross-sector comparisions. Therefore, there is some benefit in

designinga tailored tool that ismuchmore relevent formaternity careprovision in LMIC

Page178of248

settings. For example, the new assessment tool could specifically examine management

practicesassociatedwith implementingaseamlessmaternitycarepathway includingdrug

andsupplychainmanagement,organisationofclinicalteamsandothers.

Manyof thequestions in the assessment tool for example those related to the layoutof

patient flow, performance management and target management, human resource

managementand incentivesmanagementwerecomparativelypoorer inthepublicsector.

Thisraisessomedeeperquestionsonwhatitmeansforapublicsectorinstitutiontobewell

managed,particularlyinLMICssettingswherethereisnoautonomyorauthoritygiventothe

managertoimplementsomeofthesemanagementbestpractices.

I found that most hospitals in Uttar Pradesh do not have standardized maternity care

pathways.Although,protocolsfor labourandchildbirthareavailableinmostpublicsector

hospitals,theyarenotspecificallykeptinthelabourrooms,andthesetendtobeabsentin

privatehospitals. Monitoringtheuseofstandardisedprotocolswasnotroutinelydone in

bothsectorsandmanagerswereoftennotsurewhetherstaffwerefollowingtherelevant

protocols. I also found that while specialists such as doctors tend not moved across

specialities,nursesarefrequentlymovedacrossspecialitieswhichhasimplicationsforquality

ofservices.Thepublicsectorwasfoundtobeinflexibleintermsofdeploymentofstaffand

oftenstruggledwithrecruitment,selectionandretentionofstaff.

Although, the private sector was found to be relatively better for human-resource

management,managersfrequentlycomplainedthatfindingqualifiedstafftocomeandwork

intheseruralareaswaschallenging.Ididnotfindsystemsfortrackingperformanceindicators

routinely in the public sector, whereas the larger private sector hospitals often tracked

indicatorsrelatedtofinancialearningssuchasoutpatientquantitity,surgeryquantitity,bed

turnover rates and length of stay. Conversations about hospital performance were not

regularlydoneandprocessesforexposingoperationalproblemswererareinbothsectors.

Duringmyinterviewswiththemanagers,Ialsofoundthattherearenoimmidiateordirect

consequences of poor performance for staff in the public sector. For targets, the private

sectoronlyhadtargetsforrevenueandnotforqualitywhereasinthepublicsector,targets

werelimitedtotheonesprescribedbythecentralgovernment.

Page179of248

9.3.3.2:Strengths

Keystrengthsofthemanagementstudyincludedobtainingdataonmanagementpractices

from facilities where actual clinical observations had taken place. To the best of my

knowledge,nostudytodatehasinvestigatedtherelationshipbetweenmanagementandQoC

insuchaLMICsetting,primarilybecauseofthedifficultiesinobtainingsuchprimarydata.

Wealsodoublescoredallresponsestotheinterviewquestionsandfoundgoodcorrelation

betweenscoresgivenbytwodifferentassessorswhichstrengthenstheinternalvalidityofmy

study.

I could also obtain better response rates for themanagement survey and interviewed all

administratorsandclinicalleadersat33maternityfacilities(10privateand23publicsector)

unlikeintheQoCstudywhereIcouldjustgetobservationsat26facilities.Managerswere

alsoappreciativeof the fact that Iwentback toengagewith themon this issueafter the

completionofQoCassessmentswhichisastrength.

I adapted a previously used survey instrument which supported comparability of results

acrossdifferentsettings.Inaddition,myresultsparticularlyfromprivatesectorsampleswere

similartothosereportedbythelargerstudyinIndiawhichisencouragingintermsofexternal

validity.ThepreviousstudybyLemosetal(2013)44inIndiawasonlyconductedintheprivate

sector.Therefore,mystudymakesausefulcontributionto the literatureonmanagement

practicesandqualityofcareinbothprivateandpublicsectors.Thisisakeycontributionof

myPhD.

9.3.3.3:Limitations

Forthemanagementsurvey,limitationsincludedpurposivesamplingofmaternityfacilitiesin

threedistrictsinUPbecauseofwhichmyfindingsarenotgeneralizabletoalldistrictsofUttar

Pradesh.

I also interviewed manager’s face-to face unlike previous studies that have employed

telephone-interviews. Therefore, it is possible that other assessors and I may have been

biasedby theappearanceor theoperationsofaparticular facility. However, face-to-face

interviews rather than telephone-based interviewsare a good researchpractice. Personal

face-tofaceinterviewswerealsoessentialtoensurethatIobtainedagoodresponserateand

managersalsoappreciatedthefollowupvisit.

Page180of248

AlthoughIadaptedatoolthathaspreviouslybeenusedinmultiplecountriesincludingIndia,

the content and construct validityof the toolwasnot specifically testedwhichmayhave

implicationsforfindings.Since,correlationbetweenassessorswashighandourscoreswere

comparable with the larger Indian study, reliability is less of a concern. However, an

additional validation study in a small selection of participantswould have been useful to

validatethestudyinstruments.

Lastly, my sample of 275 observations at 26 hospitals is also small to generate precise

estimates on the relationship between management and quality of care and therefore,

follow-upstudywithalargersamplesizewouldbeuseful.

9.4:Implicationsofthedoctoralstudy

MyPhDstudyadvancestheevidence-baseonqualityofcareduringlabourandchildbirthin

in at least four ways with important implications, as I will discuss. First, I conducted a

comprehensiveassessmentofQoCfornormallabourandchildbirthincludinginprivatesector

facilitiesandthisisoneofthefirsteffortstodosoinUttarPradesh,whichisahigh-priority

statefortheIndiangovernment.MyfindingsofpoorQoCatbothpublicandprivatesector

maternity facilities shines an important light on this issue and demands a strong policy

responsetoimproveQoCduringlabourandchildbirthinUP.

Second, I demonstrated that given the high prevalence of mistreatment of women in

maternity facilities it is important to consider disrespect and abuse, over-treatment and

under-treatment innationalandglobaldebatesonpoorqualityofcare.Regardlessof the

terminologyused,mistreatmentofwomenfallsunderpoorqualityofcareandit isbotha

rights-basedandamedicalissue.

Third,thisPhDadvancesmeasurementeffortsbydevelopingthreetransparentindiceswhich

canbeusedtoevaluateandmonitoroverallQoCduringnormallabourandchildbirth,and

QoC for obstetric and neonatal care. These indices could be utilised by other quality

improvementprojectsandresearchers.

Fourth,Idemonstratethatmanagementbestpracticesarenotutilisedatmaternityfacilities

inUttarPradeshandinsuchsettings,performancemanagementactivitiessuchasauditshave

Page181of248

apositiveimpactonqualityofcare.Tothebestofmyknowledge,thisisalsooneofthefirst

effortstoinvestigatetheroleofmanagementpracticesonQoCduringlabourandchildbirth

atmaternityfacilitiesinalow-resourcesetting.

Thespecificrecommendationsofmydoctoralstudyonresearch,programmesandpoliciesis

discussedinthesectionsbelow.

9.4.1:Recommendationsforthefutureresearchagenda

TheresultsfrommyPhDshowthatQoCwaspoorinbothpublicandprivatesectormaternity

facilities inUttar Pradesh. Therefore, a systematic effort tomeasure and identify existing

qualitygapsduringlabourandchildbirthisrequired,especiallyinIndia’shigh-burdenstates

andinsimilarsettingselsewhere.Theseresearcheffortsshouldalsoincludeprivate-sector

facilities,whichprovideasubstantialandincreasingproportionofthematernitycareinIndia

andinotherLMICs.

Since,Iwasnotabletoobtainasamplingframefortheprivatesectorandconductingalarge

censusofprivatesectorfacilitieswasnotfeasiblewithinthetimeframeofmyPhD,myprivate

sectorestimatemaynotbeprecise.Therefore,thereisaneedtoconductacensusofprivate

sector facilities inUP so that future research and government efforts to engagewith the

privatesectorcanbemoreeffective.ResearchersworkinginIndiawidelyacknowledgethat

obtaining participation from the private sector in research is a fundamental challenge.

Therefore, the research communitywill need to think carefully about innovative research

strategies to improve participation of the private sector in future research efforts. It is

importanttohighlightthatinvolvingtheprivatesectorinfuturelarge-scaleresearchactivities

mayonlybepossiblethroughrobustgovernmentalregulation,orasapartofgovernment

purchasing of private sector services or private-led initiatives amongst insurers or large

hospitalgroups.

TherearemanybenefitsofconductingalargerstudyusingthemethodsdescribedinmyQoC

study.A largerstudywillprovideestimatesthataremorerepresentativeforthewholeof

UttarPradeshand inparticular for theprivate sector. Future researchefforts should also

anticipate thedifficulties of sampling associatedwith theprivate sector; invest additional

Page182of248

resourcesandhavealongerperiodfordatacollectionsothatagreaternumberofnormal

labourandchildbirthcasescanberecruited,especiallyintheprivatesector.

My PhD findings showed that the quality of care was generally poor during labour and

childbirthandthatbeingqualifieddidnotguaranteeprovisionofahigherstandardofcare.

Theseresultssuggestthatimprovingtheknowledge,skillsandcompetenceofqualifiedhealth

workers is important.Mixed-methods implementation research studies can be utilised to

investigate the extent to which the quality of existing trainings influence health workers

knowledge, skills and actual performance on-the job. Kirkpatrick’s model of training

programmeevaluationrecommendsacomprehensiveassessmentofreaction(ofthetrainees

tothetrainingprogram),learning(theknowledge/skillgain),behaviour(theactualchangein

practice)andresults(thefinaloutcomeduetothetraining)434.

Other innovative research questions include examining the effectiveness of innovative

trainingapproacheslikesimulation-basedtrainings/obstetricskillsanddrillsmethods435in

improving health workers knowledge, skills and confidence. Similarly, implementation

research to investigate whether training and retraining of health workers linked to re-

accreditationorcertificationwithprofessionalcouncils(medicalornursingcouncils)havethe

potentialtoimproveQoCathealthfacilitiescouldalsobeinvestigatedfurther.

Criterion-basedclinicalauditsareconsideredasausefulmethodtoauditqualityinmaternal

and newborn health. Process indicators used during these audits can help to assess the

adherence to evidence-based practices.436 Audit approaches can often use a structured

problem-solving methodology (for example: Plan, Do, Study, Act cycles) where teams of

providersareorganizedandsupportedtoidentifyandtestchangesintheprocessestheywish

to improve and tomeasure the impact of the changes against quantitative indicators of

quality.388Thesemethodshavenotbeenutilisedextensively inhospitals inUttarPradesh.

Therefore,thereisscopeforfutureresearchonexaminingtheeffectivenessofintroducing

suchapproachesinUttarPradesh.Moreover,thetrialbyDumontetal.(2013)inSenegaland

Mali offers an interesting example that researchers could replicate in India226. Additional

implementation research questions could also include issues such as feasibility, cost-

effectivenessandsustainabilityoftheseauditmechanisms.

Page183of248

Since,Ifoundthattherewasahighprevalenceofunqualifiedpersonnelinmaternityfacilities

inUPin2015,itisimportanttounderstandthereasonsbehindthisphenomenon.Whatare

the contextual factors and determinants that lead to such a high reliance on unqualified

personnel in maternity facilities? Is this caused by staffing shortages, monetary issues,

sociocultural factors, deficiencies in knowledge, skills or competence of qualified health

workersorarequalifiedhealthworkersoverburdened?Understandingtheseissuesindetail,

willbeusefultoformulateappropriatepolicyresponses,thereforefurtherresearchonthis

issuewouldbeuseful.Thereisalsoanimportantglobalevidencegaparoundoptimalstaffing

levelsandSBAstaffingmodelsinhospitals.437Furtherresearchonidentifyingoptimalstaffing

levelswouldalsobeveryusefultosupportsafemotherhoodinIndia.

AsdemonstratedinmyPhDstudy,clinicalobservationsareafeasiblewaytoassessprocess

measures of quality of care for labour and childbirth. Future research efforts could also

employ clinical observations to investigate theQoC for complications of pregnancies and

caesarean sections431 in both public and private sector facilities in LMIC settings. The

differencesinQoCformanagingcomplicationsamongstqualifiedandunqualifiedpersonnel

wouldalsobeaninterestingquestionforresearch.

Thereshouldalsobefutureresearchoncharacterisingandimprovingorganizationalculture

to enhance patient safety at maternity facilities.438 These research efforts could employ

multidisciplinary approaches to investigate the effectiveness of interventions to improve

patient and provider safety such as interventions to improve hand-hygiene, improve

adherence to evidence-based practices or adherence to infection prevention and control

proceduresandothers.439,432

Furtherconceptualworkisrequiredaroundmeasurementofmistreatmentofwomenduring

labourandchildbirth in India,particularly,as I foundahighprevalenceofcertainharmful

practices which were also associated with socio-demographic characteristics of women.

Defining theboundaries formeasurementbetweenpoorqualityandmistreatment is also

requiredsincesomeindicatorsofover-treatmentandunder-treatmentareagainsttherights

ofchildbearingwomen141butalsoconsideredindicatorsofpoorQoC.Identifyingimproved

waystoincorporatetheconceptofintentionalityinmeasurementeffortsofmistreatmentas

describedearlierisalsorequired.

Page184of248

Forensuringrespectfulmaternitycare,furtherresearchisrequiredtoidentify,testandscale

upeffectiveaccountabilitymechanisms.Manypromisingpilotsofaccountabilitymechanisms

suchassocialaudits,publichearings,citizen’sreportcards,verbalandsocialautopsiesand

partnership-defined-quality have been utilised but these require further implementation

researchinordertosupportscale-upinhighburdencountries.440-442Inaddition,westillneed

furtherresearchtoidentifyeffectiveinterventionsforvariouscontextsparticularlytoidentify

interventions needed to improve interpersonal care and social support for women at

maternity facilities, without blaming healthworkerswho are alsoworking under difficult

situations.

Futureresearchcouldalsoexaminetheprevalenceanddeterminantsofinformalpayments

atmaternityfacilitiesinIndia.Informalpaymentsseemwidespreaddespitenationalschemes

suchastheJSYschemeinIndia.Thisalsorelatestoalargerresearchagendaoncorruptionin

thehealthsectorinIndia.Detailedinformationonlevelandnatureofinformalpaymentscan

be collected throughobservations, household surveys, focus groupswithwomenor from

reports of other health providers385. Additional research questions could also explore

effective approaches to empower women and families so that they can refuse informal

paymentsinfacilities.

Future research efforts could examine ways to assign intervention-specific weights to

different elements of care provided during normal labour and childbirth. These research

efforts could employ methods for establishing and developing consensus such as Delphi

techniques,consensuspanels,ornominalgroupprocesses.429Theseapproachesareoften

usedincombinationandusebothquantitativeandqualitativemethods.Essentially,agroup

ofexpertsinaparticularfieldareconvenedandaskedtodecideontheimportanceofspecific

issuesofinterest.429However,somecriticshavearguedthatthereisselectionbiaswiththese

consensusmethods.429Forexample,expertsinvitedtoparticipatemaynotberepresentative

of the wider research community ormay not be front-line health workers. There is also

debateaboutthevalidityandreliabilityoftheseconsensusmethodsandnoagreementabout

whichmethodisthemostappropriate.429Thecurrentthinkingappearstobethattheresults

ofconsensusmethodsshouldbeinterpretedcautiouslyandtestedfortheirvalidityagainst

observationswhichareconsideredthegoldstandardformeasuringprocessesofcare.429To

illustrate thispoint inanexample, inorder toundertake suchanexercise for careduring

Page185of248

routinelabourandchildbirth,expertswouldbeconvenedandaskedtorankitemsforvarious

dimensionsofcare.Whatdotheythinkarethemostimportantinterventionsforevidence-

based obstetric or neonatal care, for interpersonal care or for harmful care procedures?

Basedontheirratingsoftheimportanceofspecificitems,indicatorswouldthengroupedand

candidateindicesdevelopedbasedonexpert’srankings.201

The global community has now recognized the importance of QoC in achieving further

reductionsinmaternalandneonatalmortalityandstillbirths.Inordertoachievethis,weneed

validwaystoassessQoCatthetimeofbirth.ValidationoftheindicesdevelopedinthisPhD

study couldbeundertaken - face validity assessed throughexpert feedback. Content and

criterion validity assessed by using data collected from clinical observations.201 Additional

researchquestionscouldalsofocusonfeasibility,reliability,andperceptionoftheseindices

byend-userssuchasprogrammemanagersandnationalmonitoringandevaluationexperts

inhigh-burdensettings.Otherresearchersthathaveemployedasimilarmethodologyhave

foundgoodspecificityandsensitivityofindicesdevelopedonqualityofcareduringlabour

andchildbirth.201

Therearealsostatisticalmethods likeprincipal componentanalysis thatcouldbeused to

develop such indices. Principal component analysis (PCA) have generally been used to

constructmeasuresof socio-economic status fromhouseholdownershipof assets.443PCA

essentiallyemploysmathematicalalgorithmstoretainvariationsinthedataset444andare

usedtoreducealargevolumeofindicatorstoasingleindexappropriatelyconstructedfrom

thecommonvarianceofaspecificsetofindicators.443 Oneexampleofthisapproachhas

beenusedbyresearchersinvestigatinghealthservicereadinessbyusingdatafromaservice

provision assessment survey in Tanzania. 443 However, there are also limitations of this

approachsincePCAmaygivelowimportancetoindicatorsthatarecommonestinthedataset

regardless of their clinical importance. For example, uterotonic within 1 minute is an

importantclinicalindicatorbutsinceitwasrelativelycommoninmydataset,PCAwillnotgive

adequateimportancetothatindicator.

Asshownbymyresults,performancemanagementactivitieshaveapositiveeffectonquality

of care at the time of birth. Therefore, further implementation research on ways to

institutionalise and implementmechanism such as criterion-based audits,maternal death

Page186of248

reviews,confidentialenquiries,near-missauditsandmaternalorperinataldeathsurveillance

and responsemechanisms, is important for all high burden settings. These activities can

provide powerful information that can guide actions to end preventable maternal and

neonataldeathsinhigh-burdensettings.

9.4.2:Recommendationforprogrammes

Currently, knowledge about best practices during normal labour and childbirth in LMIC

settingsislimited.AssessmentoftheQoCduringlabourandchildbirthneedstobeinstituted

systematicallyinallhigh-burdenstatesofIndia.Theavailabilityofassessmenttools,suchas

thosedevelopedinthisstudymaybeuseful.Programmesshouldtargettheirresourcesto

improve measurement and improve existing QoC at facilities in both public and private

sectors.Ashighlightedinthisstudy,quality improvementeffortsneedstobecentral inall

effortstoendpreventablematernal,neonataldeathsandstillbirthsinmaternityfacilities.

Disseminationofevidence-basedguidelinesandconceptsofrespectfulmaternitycareneeds

tobedoneextensivelyamongstallfront-lineworkersinIndiaandothersimilarsettings.This

isparticularlyimportantsinceIfoundthatQoCandlevelsofmistreatmentweresimilarfor

bothqualifiedandunqualifiedpersonnel.Thiscouldbeaninnovativeareaofworktodevelop

suitable training programmes for both qualified and unqualified personnel, design

appropriate skill-development activities and improve linkageswith specialists and higher-

centres.

Since, I found that QoC during weekends and outside normal working hours was poor

comparedtoweekdaysandwithinregularworkinghours.Managersatmaternity facilities

shouldensurethatoptimalstaffinglevelsandancillaryresourcesareavailableduringthese

times. Overall, given the poor quality of care inmaternity facilities inUP, improving the

knowledge and competence of all maternity care personnel is urgently needed. Use of

appropriatetrainingmethodswithadequateopportunitiesforsupervisedpracticaltraining

sessions and further on-the-job supportive supervision and refresher trainings would be

useful,asdescribedearlier.

Innovativeprojectstoimprovedeficienciesinknowledgeandskillsamongsthealthworkers,

improvementofthequalityoftrainings,mechanismtoaddresshealthworkershortagesand

Page187of248

developmentofinnovativementorshipandsupervisionmechanismswouldhelpinadvance

theevidence-baseon these issuesand support furtherdeclines inmaternal andnewborn

mortality.

9.4.3:Recommendationsforpolicy

My results indicate that unqualified personnel such as TBAs, sweepers, dais, ASHAswere

frequently involved in providing care in maternity facilities in UP in 2015. This is a

troublesomefindingwithmanyimportantglobalandnationalimplications.First,itmaybe

possible that thematernal health community in India and globally, is over-estimating the

proportionofwomenthatdeliverwithSBAsparticularlyinhigh-focusstatesofIndiasuchas

Uttar Pradesh. Globalmonitoring efforts often employ advancedmathematicalmodelling

methodsthatrelyon indicatorssuchaspopulationcoverageofSBAbirths, therefore, it is

possiblethatglobalestimatesforMMRdeclinesinIndiacouldbeunder-estimatingtheactual

burdenofmaternalmortality.

TheIndianGovernmentrecommendsprovisionofmaternityservicesbyappropriatelytrained

and qualified skilled birth attendants at health facilities. However, given various context

specificchallengesdescribedearlier, theprospectofallbirthsbeingcaredforbyqualified

personnelathealthfacilitiesisanimportantchallenging,particularlyinruralareas.Therefore,

it is importantforpolicymakerstoissueclearandcomprehensiveguidanceontheroleof

unqualifiedprovidersatmaternityfacilities.

If national authorities decide against using unqualified personnel to provide institutional

services, theymust design and implement robustmonitoringmechanisms to ensure that

unqualifiedpersonnelarenot involved inservicedelivery.Up-todate jobdescriptionsare

required,sothatthere isnoconfusionaboutrolesandresponsibilities.Womenthatgoto

institutions have a right to expect care from qualified personnel irrespective of public or

privatesectoranditisthedutyofthegovernmenttoensureandprotectthatrightforwomen.

Policymakersmustinvestindesigningappropriatecareerdevelopmentpathwaysforyoung

doctors,nursesandmidwivessothattheyjointhepublicsector.Thiswillalsorequirebetter

remuneration packages to attract and retain qualified health workers particularly those

servinginruralareas.Ultimately,thereisaneedtodevelopcomprehensivenationalhuman

Page188of248

resourceplanswithstrategiestoensureadequaterecruitment,rationaldeployment,ongoing

retention and capacity building of qualified health workers providing maternity care

services.16

Policymakers should consider the importance of long-term investments in strengthening

health systems and improving work conditions for front-line health workers. Given the

immenseshortagesofskilledhumanresourcesformaternitycareinIndia,focusedeffortsto

establishaprofessionalcadreofmidwivescouldbebeneficialand long-terms investments

arerequiredtoincreaseproductionofqualifiedmaternitycarepersonnel.

Policymakersalsoneedtotacklethewidespreadexistenceofinformalpaymentsbydesigning

better polices for supervision, disseminate patient charters, and institute disciplinary

mechanismsforhealthworkersinvolvedinsuchcorruptpractices.

Policymakersneedtoinvestinparticipatoryaccountabilitymechanismsashighlightedinthe

recommendationsforresearchsectionearliersothatevidence-basedandrespectfulcareis

providedtoallwomanandtheirbabiesatmaternityfacilities.

Finally,thereisnowincreasingrealisationthatgovernmentsalonemaynotbeabletodeliver

allservicestomeettheirpopulation’sneeds.Further,somewomenoftenprefertoseekcare

intheprivatesectorandthereforeitisquiteimportanttoregulateandimprovequalityinthe

private sector aswell. Comprehensive regulationon theprivate sectors’ role in providing

healthservices includingdetailedqualitystandardsexpected inprivatesector facilities for

maternity care need to be developed.Once regulations and quality standards have been

finalised,permissivepolicieswillberequiredtoimplementandtestinnovativepublic-private

partnershipmodelstoimproveefficiency,effectivenessandQoCofmaternityservices.

Page189of248

Chapter10:Conclusions

The results of my PhD study indicate that in maternity facilities in Uttar Pradesh, the

personnelprovidingnormallabourandchildbirthcarewereoftenunqualified;adherenceto

evidence-basedobstetricandneonatalcareprotocolswasgenerallypoor;and,allwomen

encounteredatleastonepracticeconsideredtobemistreatment.Mistreatmentofwomen

atmaternityfacilitiesfallsunderpoorqualityofcareandneedsgreaterattentioninnational

andglobaldebates.

MyPhDresultsindicatethatasystematicandurgenteffortisneededtomeasureandimprove

QoCatthetimeofbirthinpublicandprivatesectorfacilitiesinhigh-burdenstatesinIndia.

Appropriatecontext-specificstrategiesandinterventionsneedtobedevelopedforimproving

careduringlabourandchildbirth.

Lastly,Ididnotfindanassociationbetweenoverallmanagementscoresandqualityofcare

during labour and childbirth. The only management dimension that positively influenced

qualitywasperformancemanagementandhence,performancemanagementactivitiessuch

asdifferenttypesofauditsshouldbeimplementedinallmaternityfacilitiesinhigh-burden

settings.

Page190of248

11.Listofreferences1. AkachiY,KrukME.Qualityofcare:measuringaneglecteddriverofimprovedhealth.BulletinoftheWorldHealthOrganization2017;95(6):465-72.2. GrahamW,WooddS,ByassP,etal.Diversityanddivergence:thedynamicburdenofpoormaternalhealth.Lancet2016;388(10056):2164-75.3. HortonR.Offline:Ensuringhealthylivesafter2015.Lancet2013;381(9882):1972-.4. WorldHealthOrganization.Strategiestowardsendingpreventablematernalmortality(EPMM).2015.5. AlkemaL,ChouD,HoganD,etal.Global,regional,andnationallevelsandtrendsinmaternalmortalitybetween1990and2015,withscenario-basedprojectionsto2030:asystematicanalysisbytheUNMaternalMortalityEstimationInter-AgencyGroup.Lancet2016;387(10017):462-74.6. EveryWomanEveryChild.TheGlobalStrategyforWomen’s,Children’sAndAdolescents’health(2016-2030).SustainableDevelopmentGoals2015.7. KrukME,LarsonE,Twum-DansoNA.Timeforaqualityrevolutioninglobalhealth.LancetGlobHealth2016;4(9):e594-6.8. WorldHealthOrganization.WorldHealthStatistics2015.:WHOPress;2015.9. UnitedNations.TheMillenniumDevelopmentGoalsReport2014.NewYork:UnitedNations2014.10. Powell-JacksonT,MazumdarS,MillsA.Financialincentivesinhealth:NewevidencefromIndia'sJananiSurakshaYojana.JHealthEcon2015;43:154-69.11. LimSS,DandonaL,HoisingtonJA,JamesSL,HoganMC,GakidouE.India'sJananiSurakshaYojana,aconditionalcashtransferprogrammetoincreasebirthsinhealthfacilities:animpactevaluation.Lancet2010;375(9730):2009-23.12. OkekeE,ChariAV.CanInstitutionalDeliveriesReduceNewbornMortality?EvidencefromRwanda.SantaMonica,CA:RANDCorporation,2014.13. LeslieHH,FinkG,NsonaH,KrukME.ObstetricFacilityQualityandNewbornMortalityinMalawi:ACross-SectionalStudy.PLoSMed2016;13(10):e1002151.14. HodginsS.Achievingbettermaternalandnewbornoutcomes:coherentstrategyandpragmatic,tailoredimplementation.GlobHealthSciPract2013;1(2):146-53.15. CampbellOM,CalvertC,TestaA,etal.Thescale,scope,coverage,andcapabilityofchildbirthcare.Lancet2016;388(10056):2193-208.16. SharmaG,MathaiM,DicksonKE,etal.Qualitycareduringlabourandbirth:amulti-countryanalysisofhealthsystembottlenecksandpotentialsolutions.BMCPregnancyChildbirth2015;15Suppl2(Suppl2):S2.17. DicksonKE,Simen-KapeuA,KinneyMV,etal.EveryNewborn:health-systemsbottlenecksandstrategiestoacceleratescale-upincountries.Lancet2014;384(9941):438-54.18. LawnJE,BlencoweH,OzaS,etal.EveryNewborn:progress,priorities,andpotentialbeyondsurvival.Lancet2014;384(9938):189-205.19. BhuttaZA,DasJK,BahlR,etal.Canavailableinterventionsendpreventabledeathsinmothers,newbornbabies,andstillbirths,andatwhatcost?TheLancet2014;384(9940):347-70.20. WHOEssentialInterventions.CommoditiesandGuidelinesforReproductive,Maternal,NewbornandChildHealth:Aglobalreviewofthekeyinterventionsrelatedtoreproductive,maternal,newbornandchildHealth.Geneva:WHO2011.21. WorldHealthOrganization.MakeEveryMotherandChildCount:TheWorldHealthReport2005.22. UNFPA;WHOandUNICEF.AMDD:Monitoringemergencyobstetriccare:ahandbook..2009.23. GabryschS,SimushiV,CampbellOM.Availabilityanddistributionof,andgeographicaccesstoemergencyobstetriccareinZambia.IntJGynaecolObstet2011;114(2):174-9.

Page191of248

24. LumbiganonP,LaopaiboonM,GulmezogluAM,etal.MethodofdeliveryandpregnancyoutcomesinAsia:theWHOglobalsurveyonmaternalandperinatalhealth2007-08.Lancet2010;375(9713):490-9.25. ShahA,FawoleB,M'ImunyaJM,etal.CesareandeliveryoutcomesfromtheWHOglobalsurveyonmaternalandperinatalhealthinAfrica.IntJGynaecolObstet2009;107(3):191-7.26. MadajB,SmithH,MathaiM,RoosN,vandenBroekN.Developingglobalindicatorsforqualityofmaternalandnewborncare:afeasibilityassessment.BulletinoftheWorldHealthOrganization2017;95(6):445-52I.27. MillsA,BrughaR,HansonK,McPakeB.Whatcanbedoneabouttheprivatehealthsectorinlow-incomecountries?BulletinoftheWorldHealthOrganization2002;80.28. HansonK,GilsonL,GoodmanC,etal.Isprivatehealthcaretheanswertothehealthproblemsoftheworld'spoor?PLoSMedicine2008;5(11):e233.29. RosenthalG,NewbranderW.Publicpolicyandprivatesectorprovisionofhealthservices.IntJHealthPlannManage1996;11(3):203-16.30. JakovljevicMB.ResourceallocationstrategiesinSoutheasternEuropeanhealthpolicy.Springer;2013.31. MorganR,EnsorT.TheregulationofprivatehospitalsinAsia.IntJHealthPlannManage2016;31(1):49-64.32. MadhavanS,BishaiD,StantonC,HardingA.Engagingtheprivatesectorinmaternalandneonatalhealthinlowandmiddleincomecountries:Futurehealthsystems(FHS);2010.33. CampbellOM,BenovaL,MacLeodD,etal.Familyplanning,antenatalanddeliverycare:cross-sectionalsurveyevidenceonlevelsofcoverageandinequalitiesbypublicandprivatesectorin57low-andmiddle-incomecountries.TropMedIntHealth2016;21(4):486-503.34. AfifiNH,BusseR,HardingA.Regulationofhealthservices.Privateparticipationinhealthservices2003:219-334.35. AlmeidaSd,BettiolH,BarbieriMA,SilvaAAMd,RibeiroVS.SignificantdifferencesincesareansectionratesbetweenaprivateandapublichospitalinBrazil.CadernosdeSaúdePública2008;24(12):2909-18.36. SharmaG.ThechangingparadigmoflabourandchildbirthinIndiancities:anenquiryintoincreasingratesofcaesareandeliveries.IntJEpidemiol2016;45(5):1390-3.37. SarkerBKHJ,MridhaMK,etal..CaesareandeliveryinurbanslumsofDhakaCity:indicationsandconsequences..2012.;ManoshiWorkingPaper18.38. ParkhurstJO,RahmanSA.Lifesavingormoneywasting?PerceptionsofcaesareansectionsamongusersofservicesinruralBangladesh.HealthPolicy2007;80(3):392-401.39. NeumanM,AlcockG,AzadK,etal.PrevalenceanddeterminantsofcaesareansectioninprivateandpublichealthfacilitiesinunderservedSouthAsiancommunities:cross-sectionalanalysisofdatafromBangladesh,IndiaandNepal.BMJOpen2014;4(12):e005982.40. PomeroyA,KoblinskyM,AlvaS.WhogivesbirthinprivatefacilitiesinAsia?Alookatsixcountries.Healthpolicyandplanning2014;29(suppl1):i38-i47.41. DaireJ,GilsonL,ClearyS.Developingleadershipandmanagementcompetenciesinlowandmiddle-incomecountryhealthsystems:areviewoftheliterature.CapeTown:ResilientandResponsiveHealthSystems(RESYST)2014.42. BloomN,GenakosC,SadunR,VanReenenJ.ManagementPracticesAcrossFirmsandCountries.AcademyofManagementPerspectives2012;26(1):12-33.43. DorganS,LaytonD,BloomN,HomkesR,SadunR,VanReenenJ.Managementinhealthcare:whygoodpracticereallymatters.London:McKinseyandCompany/LondonSchoolofEconomics2010.44. LemosR,ScurD.Couldpoormanagementbeholdingbackdevelopment?DescribingpracticesinthepublicandprivatesectorsinIndia:IGCWorkingPaper,2013.45. AgarwalR,GreenR,AgarwalN,RandhawaK.BenchmarkingmanagementpracticesinAustralianpublichealthcare.JHealthOrganManag2016;30(1):31-56.

Page192of248

46. TsaiTC,JhaAK,GawandeAA,HuckmanRS,BloomN,SadunR.Hospitalboardandmanagementpracticesarestronglyrelatedtohospitalperformanceonclinicalqualitymetrics.HealthAff(Millwood)2015;34(8):1304-11.47. SharmaG,Powell-JacksonT,HaldarK,BradleyJ,FilippiV.Qualityofroutineessentialcareduringchildbirth:clinicalobservationsofuncomplicatedbirthsinUttarPradesh,India.BulletinoftheWorldHealthOrganization2017;95(6):419-29.48. UN-IGME.Levelsandtrendsinchildmortality-report2013.2013.49. LawnJ,BlencoweH,PattinsonR,etal.Stillbirths:Where?When?Why?Howtomakethedatacount?Lancet2011;377(9775):448-1463.50. LawnJE,BlencoweH,KinneyMV,BianchiF,GrahamWJ.Evidencetoinformthefutureformaternalandnewbornhealth.BestPractResClinObstetGynaecol2016;36:169-83.51. YouD,HugL,EjdemyrS,etal.Global,regional,andnationallevelsandtrendsinunder-5mortalitybetween1990and2015,withscenario-basedprojectionsto2030:asystematicanalysisbytheUNInter-agencyGroupforChildMortalityEstimation.TheLancet2015;386(10010):2275-86.52. LawnJE,BlencoweH,WaiswaP,etal.Stillbirths:rates,riskfactors,andaccelerationtowards2030.Lancet2016;387(10018):587-603.53. SouzaJP,GulmezogluAM,VogelJ,etal.Movingbeyondessentialinterventionsforreductionofmaternalmortality(theWHOMulticountrySurveyonMaternalandNewbornHealth):across-sectionalstudy.Lancet2013;381(9879):1747-55.54. SayL,ChouD,GemmillA,etal.Globalcausesofmaternaldeath:aWHOsystematicanalysis.LancetGlobalHealth2014;2(6):E323-E33.55. WHOGlobalHealthObservatory.[Internet]2014.56. UNInter-agencyGroupforChildMortalityEstimation(IGME).Report:Levelsandtrendsinchildmortality2015.57. GrahamWJ,VargheseB.Quality,quality,quality:gapsinthecontinuumofcare.Lancet2012;379(9811):e5-6.58. StorengKT,BaggaleyRF,GanabaR,OuattaraF,AkoumMS,FilippiV.Payingtheprice:thecostandconsequencesofemergencyobstetriccareinBurkinaFaso.Socialscience&medicine(1982)2008;66(3):545-57.59. vandenBroekNR,GrahamWJ.Qualityofcareformaternalandnewbornhealth:theneglectedagenda.BJOG2009;116Suppl1(Suppl1):18-21.60. NesbittRC,LohelaTJ,ManuA,etal.Qualityalongthecontinuum:ahealthfacilityassessmentofintrapartumandpostnatalcareinGhana.PLoSOne2013;8(11):e81089.61. ErimDO,KolapoUM,ReschSC.ArapidassessmentoftheavailabilityanduseofobstetriccareinNigerianhealthcarefacilities.PLoSOne2012;7(6):e39555.62. SabdeY,DiwanV,RandiveB,etal.TheavailabilityofemergencyobstetriccareinthecontextoftheJSYcashtransferprogrammeinMadhyaPradesh,India.BMCPregnancyChildbirth2016;16(1):116.63. RandiveB,ChaturvediS,MistryN.Contractinginspecialistsforemergencyobstetriccare-doesitworkinruralIndia?BMCHealthServRes2012;12:485.64. RandiveB,DiwanV,DeCostaA.India’sConditionalCashTransferProgramme(theJSY)topromoteinstitutionalbirth:Isthereanassociationbetweeninstitutionalbirthproportionandmaternalmortality?PLoSOne2013;8(6):e67452.65. ChaturvediS,UpadhyayS,DeCostaA.CompetenceofbirthattendantsatprovidingemergencyobstetriccareunderIndia'sJSYconditionalcashtransferprogramforinstitutionaldelivery:anassessmentusingcasevignettesinMadhyaPradeshprovince.BMCPregnancyChildbirth2014;14(1):174.66. ChaturvediS,DeCostaA,RavenJ.DoestheJananiSurakshaYojanacashtransferprogrammetopromotefacilitybirthsinIndiaensureskilledbirthattendance?AqualitativestudyofintrapartumcareinMadhyaPradesh.Globalhealthaction2015;8.

Page193of248

67. ChaturvediS,UpadhyayS,DeCostaA,RavenJ.ImplementationofthepartographinIndia'sJSYcashtransferprogrammeforfacilitybirths:amixedmethodsstudyinMadhyaPradeshprovince.BMJOpen2015;5(4):e006211.68. Adair-RohaniH,ZukorK,BonjourS,etal.Limitedelectricityaccessinhealthfacilitiesofsub-SaharanAfrica:asystematicreviewofdataonelectricityaccess,sources,andreliability.GlobHealthSciPract2013;1(2):249-61.69. MakingPregnancySafer.Makingpregnancysafer:thecriticalroleoftheskilledattendant-JointStatementbyWHO,ICMandFIGO.WorldHealthOrganization:Geneva2004.70. IyengarSD,IyengarK,SuhalkaV,AgarwalK.Comparisonofdomiciliaryandinstitutionaldelivery-carepracticesinruralRajasthan,India.JHealthPopulNutr2009;27(2):303-12.71. IyengarK,JainM,ThomasS,etal.AdherencetoevidencebasedcarepracticesforchildbirthbeforeandafteraqualityimprovementinterventioninhealthfacilitiesofRajasthan,India.BMCPregnancyChildbirth2014;14:270.72. HusseinJ,BellJ,NazzarA,etal.TheSkilledAttendanceIndex:Proposalforanewmeasureofskilledattendanceatdelivery.ReproductiveHealthMatters2004;12(24):160-70.73. tenHoope-BenderP,CampbellJ,FauveauV,MatthewsZ.Thestateoftheworld'smidwifery2011:deliveringhealth,savinglives.IntJGynaecolObstet2011;114(3):211-2.74. AdegokeA,UtzB,MsuyaSE,VanDenBroekN.SkilledBirthAttendants:whoiswho?AdescriptivestudyofdefinitionsandrolesfromnineSubSaharanAfricancountries.PloSone2012;7(7):e40220.75. AriffS,SoofiSB,SadiqK,etal.EvaluationofhealthworkforcecompetenceinmaternalandneonatalissuesinpublichealthsectorofPakistan:anAssessmentoftheirtrainingneeds.BMCHealthServRes2010;10(1):319.76. FootmanK,BenovaL,GoodmanC,etal.Usingmulti-countryhouseholdsurveystounderstandwhoprovidesreproductiveandmaternalhealthservicesinlow-andmiddle-incomecountries:acriticalappraisaloftheDemographicandHealthSurveys.TropMedIntHealth2015;20(5):589-606.77. StantonCK,DeepakNN,MallapurAA,etal.Directobservationofuterotonicdruguseatpublichealthfacility-baseddeliveriesinfourdistrictsinIndia.IntJGynaecolObstet2014;127(1):25-30.78. IyengarSD,IyengarK,MartinesJC,DashoraK,DeoraKK.ChildbirthpracticesinruralRajasthan,India:implicationsforneonatalhealthandsurvival.JPerinatol2008;28Suppl2:S23-30.79. BohrenMA,HunterEC,Munthe-KaasHM,SouzaJP,VogelJP,GulmezogluAM.Facilitatorsandbarrierstofacility-baseddeliveryinlow-andmiddle-incomecountries:aqualitativeevidencesynthesis.ReprodHealth2014;11(1):71.80. BohrenMA,VogelJP,HunterEC,etal.TheMistreatmentofWomenduringChildbirthinHealthFacilitiesGlobally:AMixed-MethodsSystematicReview.PLoSMed2015;12(6):e1001847;discussione.81. FreedmanLP,RamseyK,AbuyaT,etal.Definingdisrespectandabuseofwomeninchildbirth:aresearch,policyandrightsagenda.BulletinoftheWorldHealthOrganization2014;92(12):915-7.82. CampbellOM,CegolonL,MacleodD,BenovaL.LengthofStayAfterChildbirthin92CountriesandAssociatedFactorsin30Low-andMiddle-IncomeCountries:CompilationofReportedDataandaCross-sectionalAnalysisfromNationallyRepresentativeSurveys.PLoSMed2016;13(3):e1001972.83. MillerS,AbalosE,ChamillardM,etal.Beyondtoolittle,toolateandtoomuch,toosoon:apathwaytowardsevidence-based,respectfulmaternitycareworldwide.Lancet2016;388(10056):2176-92.84. MurrayCJ,FrenkJ.Aframeworkforassessingtheperformanceofhealthsystems.BulletinoftheWorldHealthOrganization2000;78(6):717-31.

Page194of248

85. BerwickDM.Continuousimprovementasanidealinhealthcare.NEnglJMed1989;320(1):53-6.86. PeabodyJW,TaguiwaloMM,RobalinoDA,FrenkJ.Improvingthequalityofcareindevelopingcountries.Diseasecontrolprioritiesindevelopingcountries2006;2:1293-308.87. SilimperiDR,FrancoLM,VeldhuyzenvanZantenT,MacAulayC.Aframeworkforinstitutionalizingqualityassurance.IntJQualHealthCare2002;14Suppl1(suppl1):67-73.88. BarkerPM,ReidA,SchallMW.Aframeworkforscalinguphealthinterventions:lessonsfromlarge-scaleimprovementinitiativesinAfrica.ImplementSci2016;11(1):12.89. Tuncalp,WereWM,MacLennanC,etal.Qualityofcareforpregnantwomenandnewborns-theWHOvision.BJOG2015;122(8):1045-9.90. BorchertM,GoufodjiS,AlihonouE,etal.Canhospitalauditteamsidentifycasemanagementproblems,analysetheircauses,identifyandimplementimprovements?Across-sectionalprocessevaluationofobstetricnear-misscasereviewsinBenin.BMCpregnancyandchildbirth2012;12(1):109.91. ThaddeusS,MaineD.Toofartowalk:maternalmortalityincontext.SocSciMed1994;38(8):1091-110.92. AdegokeAA,HofmanJJ,KongnyuyEJ,vandenBroekN.Monitoringandevaluationofskilledbirthattendance:aproposednewframework.Midwifery2011;27(3):350-9.93. AdegokeAA,vandenBroekN.Skilledbirthattendance-lessonslearnt.BJOG2009;116Suppl1(s1):33-40.94. YanqiuG,RonsmansC,LinA.TimetrendsandregionaldifferencesinmaternalmortalityinChinafrom2000to2005.BulletinoftheWorldHealthOrganization2009;87(12):913-20.95. CampbellC,GipsonR,IssaAH,etal.NationalmaternalmortalityratioinEgypthalvedbetween1992-93and2000.BulletinoftheWorldHealthOrganization2005;83(6):462-71.96. LiJ,LuoC,DengR,JacobyP,deKlerkN.MaternalmortalityinYunnan,China:recenttrendsandassociatedfactors.BJOG2007;114(7):865-74.97. McClureEM,GoldenbergRL,BannCM.Maternalmortality,stillbirthandmeasuresofobstetriccareindevelopinganddevelopedcountries.IntJGynaecolObstet2007;96(2):139-46.98. WorldHealthOrganization.MakingPregnancySafer:theCriticalRoleoftheSkilledAttendant.JointStatementbyWHO,ICMandFIGO.Geneva:WHO;2004.99. FreedmanLP,KrukME.Disrespectandabuseofwomeninchildbirth:challengingtheglobalqualityandaccountabilityagendas.Lancet2014;384(9948):e42-4.100. HarveyS.Areskilledbirthattendantsreallyskilled?Ameasurementmethod,somedisturbingresultsandapotentialwayforward.BulletinoftheWorldHealthOrganization2007;85(10):783-90.101. HarveySA,AyabacaP,BucaguM,etal.Skilledbirthattendantcompetence:aninitialassessmentinfourcountries,andimplicationsfortheSafeMotherhoodmovement.IntJGynaecolObstet2004;87(2):203-10.102. UNICEF.Countdownto2015:adecadeoftrackingprogressformaternal,newbornandchildsurvival—the2015report.WHOPress,WorldHealthOrganization,Geneva,Switzerland;2015.103. WHO,UNICEF.EveryNewborn:Anactionplantoendpreventablenewborndeaths.104. UNICEFM.Multipleindicatorclustersurvey(MICS).2000.105. MEASUREDHS.Demographicandhealthsurveys:Calverton:MEASUREDHS;2013.106. WorldHealthOrganization.The2014update.GlobalHealthWorkforceStatistics2014.107. KoblinskyM,MatthewsZ,HusseinJ,etal.Goingtoscalewithprofessionalskilledcare.Lancet2006;368(9544):1377-86.108. GereinN,GreenA,PearsonS.Theimplicationsofshortagesofhealthprofessionalsformaternalhealthinsub-saharanAfrica.ReprodHealthMatters2006;14(27):40-50.109. LehmannU,DielemanM,MartineauT.Staffingremoteruralareasinmiddle-andlow-incomecountries:aliteraturereviewofattractionandretention.BMCHealthServRes2008;8(1):19.

Page195of248

110. BustreoF,SayL,KoblinskyM,PullumTW,TemmermanM,Pablos-MendezA.Endingpreventablematernaldeaths:thetimeisnow.LancetGlobalHealth2013;1(4):E176-E7.111. AlthabeF,BergelE,CafferataML,etal.Strategiesforimprovingthequalityofhealthcareinmaternalandchildhealthinlow-andmiddle-incomecountries:anoverviewofsystematicreviews.PaediatrPerinatEpidemiol2008;22Suppl1(s1):42-60.112. PronovostPJ.Enhancingphysicians'useofclinicalguidelines.JAMA2013;310(23):2501-2.113. NeilsonJP.Evidence-basedintrapartumcare:evidencefromtheCochranelibrary.IntJGynaecolObstet1998;63Suppl1(S1):S97-102.114. WorldHealthOrganization.CareinNormalBirth:APracticalGuide.Geneva:WorldHealthOrganization;1996.115. EasonE,LabrecqueM,WellsG,FeldmanP.Preventingperinealtraumaduringchildbirth:asystematicreview.ObstetGynecol2000;95(3):464-71.116. LudkaLM,RobertsCC.Eatinganddrinkinginlabor.Aliteraturereview.JNurseMidwifery1993;38(4):199-207.117. TuranJM,BulutA,NalbantH,OrtaylıN,ErbaydarT.Challengesfortheadoptionofevidence-basedmaternitycareinTurkey.SocialScience&Medicine2006;62(9):2196-204.118. KhalilK,ElnouryA,CherineM,etal.HospitalPracticeVersusEvidence-BasedObstetrics:CategorizingPracticesforNormalBirthinanEgyptianTeachingHospital.Birth2005;32(4):283-90.119. Sandin-BojoA-K,KvistL.CareinLabor:ASwedishSurveyUsingtheBolognaScore.Birth2008;35(4):321-8.120. GroupS-OS,LaopaiboonM,LumbiganonP,etal.Useofevidence-basedpracticesinpregnancyandchildbirth:SouthEastAsiaOptimisingReproductiveandChildHealthinDevelopingCountriesproject.PLoSOne2008;3(7):e2646.121. NeilsonJP.Evidence-basedintrapartumcare:evidencefromtheCochranelibrary.IntJGynaecolObstet1998;63Suppl1:S97-102.122. InternationalFederationofG,Obstetrics,InternationalConfederationofM,etal.Mother-babyfriendlybirthingfacilities.IntJGynaecolObstet2015;128(2):95-9.123. WorldHealthOrganization.Thepreventionandeliminationofdisrespectandabuseduringfacility-basedchildbirth:WHOstatement.2014.124. RenfrewMJ,McFaddenA,BastosMH,etal.Midwiferyandqualitycare:findingsfromanewevidence-informedframeworkformaternalandnewborncare.Lancet2014;384(9948):1129-45.125. BowserD,HillK.Exploringevidencefordisrespectandabuseinfacility-basedchildbirth:reportofalandscapeanalysis.USAID-TRActionProject,Washington,DC;2010.126. D'GregorioRP.Obstetricviolence:anewlegaltermintroducedinVenezuela.Elsevier;2010.127. MisagoC,KendallC,FreitasP,etal.From‘cultureofdehumanizationofchildbirth’to‘childbirthasatransformativeexperience’:changesinfivemunicipalitiesinnorth-eastBrazil.InternationalJournalofGynecology&Obstetrics2001;75:S67-S72.128. VedamS,StollK,RubashkinN,etal.TheMothersonRespect(MOR)index:measuringquality,safety,andhumanrightsinchildbirth.SSM-PopulationHealth2017;3:201-10.129. DeclercqE,SakalaC,CorryM,ApplebaumS,HerrlichIIIA.ListeningtoMothersIII:ReportoftheThirdNationalUSSurveyofWomen’sChildbearingExperiences.NewYork:ChildbirthConnection;2013.2013.130. BakerSR,ChoiPYL,HenshawCA,TreeJ.'IfeltasthoughI'dbeeninjail':Women'sexperiencesofmaternitycareduringlabour,deliveryandtheimmediatepostpartum.Feminism&Psychology2005;15(3):315-42.131. IsholaF,OwolabiO,FilippiV.DisrespectandabuseofwomenduringchildbirthinNigeria:Asystematicreview.PLoSOne2017;12(3):e0174084.132. WarrenC,NjukiR,AbuyaT,etal.Studyprotocolforpromotingrespectfulmaternitycareinitiativetoassess,measureanddesigninterventionstoreducedisrespectandabuseduringchildbirthinKenya.BMCPregnancyChildbirth2013;13:21.

Page196of248

133. SheferawED,MengeshaTZ,WaseSB.Developmentofatooltomeasurewomen'sperceptionofrespectfulmaternitycareinpublichealthfacilities.BMCPregnancyChildbirth2016;16(1):67.134. ChadwickRJ,CooperD,HarriesJ.NarrativesofdistressaboutbirthinSouthAfricanpublicmaternitysettings:aqualitativestudy.Midwifery2014;30(7):862-8.135. HultonL,ZoëMatthews,andR.WilliamStones..Aframeworkfortheevaluationofqualityofcareinmaternityservices.2000.136. CalnanM.Layevaluationofmedicineandmedicalpractice:reportofapilotstudy.IntJHealthServ1988;18(2):311-22.137. OkaforCB,RizzutoRR.Women'sandhealth-careproviders'viewsofmaternalpracticesandservicesinruralNigeria.Studiesinfamilyplanning1994;25(6Pt1):353-61.138. BruceJ.Fundamentalelementsofthequalityofcare:asimpleframework.StudFamPlann1990;21(2):61-91.139. LohrKN,DonaldsonMS,WalkerAJ.Medicare:astrategyforqualityassurance,III:Beneficiaryandphysicianfocusgroups.QRBQualityreviewbulletin1991;17(8):242-53.140. VeraH.Theclient'sviewofhigh-qualitycareinSantiago,Chile.Studiesinfamilyplanning1993;24(1):40-9.141. WhiteRibbonAlliance.Respectfulmaternitycare:theuniversalrightsofchildbearingwomen.Availableat:http://whiteribbonallianceorg/wp-content/uploads/2013/10/Final_RMC_Charterpdf2012;Accessedon4.12.2016.142. MannavaP,DurrantK,FisherJ,ChersichM,LuchtersS.Attitudesandbehavioursofmaternalhealthcareprovidersininteractionswithclients:asystematicreview.GlobalHealth2015;11(1):36.143. JOV,BateP,ClearyP,etal.Qualitycollaboratives:lessonsfromresearch.Quality&safetyinhealthcare2002;11(4):345-51.144. InstituteofMedicine;CommitteeonQualityofHealthCareinAmerica.Crossingthequalitychasm:Anewhealthsystemforthe21stcentury,:NationalAcademiesPress;2001.145. DonabedianA.Thequalityofcare:Howcanitbeassessed?Jama1988;260(12):1743-8.146. DonabedianA.ExplorationsinQualityAssessmentandMonitoring:TheDefinitionofQualityandApproachestoitsAssessment..1980.147. RoemerMI,CarlosMontoya-Aguilar,andWorldHealthOrganization.Qualityassessmentandassuranceinprimaryhealthcare.1988.148. LohrKN,InstituteofMedicineMedicare:astrategyforqualityassurance.Vol.1..NationalAcademiesPress1990.149. WilsonLL.Qualityanditsmeasurement.JQualClinPract1998;18(3):163-70.150. GodleeF.Effective,safe,andagoodpatientexperience.Bmj2009;339:b4346.151. DonabedianA.Thequalityofcare:Howcanitbeassessed?Jama1988;260(12):1743-8.152. MedicineIo,AmericaCoQoHCi.Crossingthequalitychasm:Anewhealthsystemforthe21stcentury.2001.153. HultonL,MatthewsZ,StonesR.Aframeworkfortheevaluationofqualityofcareinmaternityservices.2000.154. RavenJH,TolhurstRJ,TangS,VanDenBroekN.Whatisqualityinmaternalandneonatalhealthcare?Midwifery2012;28(5):e676-e83.155. GermainA;OrdwayJ.PopulationControlandWomen'sHealth:BalancingtheScales.1989.156. EngenderHealth.COPEhandbook:aprocessforimprovingqualityinhealthservices..2003.157. MaxwellRJ.Dimensionsofqualityrevisited:fromthoughttoaction.Qualityinhealthcare:QHC1992;1(3):171-7.158. SrivastavaA,AvanBI,RajbangshiP,BhattacharyyaS.Determinantsofwomen'ssatisfactionwithmaternalhealthcare:areviewofliteraturefromdevelopingcountries.BMCPregnancyChildbirth2015;15:97.

Page197of248

159. MendozaAldanaJ,PiechulekH,al-SabirA.ClientsatisfactionandqualityofhealthcareinruralBangladesh.BulletinoftheWorldHealthOrganization2001;79(6):512-7.160. GoodmanP,MackeyMC,TavakoliAS.Factorsrelatedtochildbirthsatisfaction.Journalofadvancednursing2004;46(2):212-9.161. HarveyS,RachD,StaintonMC,JarrellJ,BrantR.Evaluationofsatisfactionwithmidwiferycare.Midwifery2002;18(4):260-7.162. WaldenstromU,HildingssonI,RubertssonC,RadestadI.Anegativebirthexperience:prevalenceandriskfactorsinanationalsample.Birth2004;31(1):17-27.163. WorldHealthOrganization.Standardsformaternalandneonatalcare.2007.164. WorldHealthOrganization.Everybody'sbusiness--strengtheninghealthsystemstoimprovehealthoutcomes:WHO'sframeworkforaction.2007.165. WorldHealthOrganization.KeepingPromises,MeasuringResults:CommissiononInformationandAccountabilityforWomen'sandChildren'sHealth..2012.166. MunosMK,StantonCK,BryceJ,CoreGroupforImprovingCoverageMeasurementforM.Improvingcoveragemeasurementforreproductive,maternal,neonatalandchildhealth:gapsandopportunities.JGlobHealth2017;7(1):010801.167. DonabedianA.Thequalityofcare:Howcanitbeassessed?.Jama1988;260(no.12):1743-8.168. WorldHealthOrganization.Serviceavailabilityandreadinessassessment(SARA):anannualmonitoringsystemforservicedelivery:referencemanual.2013.169. MEASUREDHS.DemographicandHealthSurveys.TheServiceProvisionAssessment(SPA)..2011.170. MEASUREEvaluation;USAID.HealthFacilityAssessmentTechnicalWorkingGroup(2005)Profilesofhealthfacilityassessmentmethods..2005.171. WangW,WinnerM,Burgert-BruckerCR.LimitedServiceAvailability,Readiness,andUseofFacility-BasedDeliveryCareinHaiti:AStudyLinkingHealthFacilityDataandPopulationData.GlobalHealth:ScienceandPractice2017:GHSP-D-16-00311.172. USAIDandMCHIP.MaternalandNewbornQualityofCareSurveys..173. PeabodyJW,LuckJ,GlassmanP,etal.Measuringthequalityofphysicianpracticebyusingclinicalvignettes:Aprospectivevalidationstudy.AnnalsofInternalMedicine2004;141(10):771-80.174. AustinA,LangerA,SalamRA,LassiZS,DasJK,BhuttaZA.Approachestoimprovethequalityofmaternalandnewbornhealthcare:anoverviewoftheevidence.ReprodHealth2014;11Suppl2(Suppl2):S1.175. AungT,MontaguD,SchleinK,KhineTM,McFarlandW.Validationofanewmethodfortestingproviderclinicalqualityinruralsettingsinlow-andmiddle-incomecountries:theobservedsimulatedpatient.PLoSOne2012;7(1):e30196.176. MohananM,Vera-HernandezM,DasV,etal.Theknow-dogapinqualityofhealthcareforchildhooddiarrheaandpneumoniainruralIndia.JAMAPediatr2015;169(4):349-57.177. KwenaZA,SharmaA,MugaC,WamaeN,BukusiEA.ManagementofsimulatedpatientswithsexuallytransmittedinfectionsbystaffofretailpharmaciesinKiberaslumsofNairobi.EastAfrMedJ2008;85(9):419-24.178. DasJ,HollaA,DasV,MohananM,TabakD,ChanB.InurbanandruralIndia,astandardizedpatientstudyshowedlowlevelsofprovidertrainingandhugequalitygaps.HealthAff(Millwood)2012;31(12):2774-84.179. PeabodyJW,LuckJ,GlassmanP,DresselhausTR,LeeM.Comparisonofvignettes,standardizedpatients,andchartabstraction:aprospectivevalidationstudyof3methodsformeasuringquality.JAMA2000;283(13):1715-22.180. DanielsB,DolingerA,BedoyaG,etal.UseofstandardisedpatientstoassessqualityofhealthcareinNairobi,Kenya:apilot,cross-sectionalstudywithinternationalcomparisons.BMJGlobalHealth2017;2(2):e000333.

Page198of248

181. WickströmG,BendixT.The"Hawthorneeffect"—whatdidtheoriginalHawthornestudiesactuallyshow?Scandinavianjournalofwork,environment&health2000:363-7.182. PeabodyJW,TozijaF,MunozJA,NordykeRJ,LuckJ.Usingvignettestocomparethequalityofclinicalcarevariationineconomicallydivergentcountries.HealthServRes2004;39(6Pt2):1951-70.183. StylesM,CheyneH,O'CarrollR,GreigF,Dagge-BellF,NivenC.TheScottishTrialofReferorKeep(theSTORKstudy):midwives'intrapartumdecisionmaking.Midwifery2011;27(1):104-11.184. LuckJ,PeabodyJW,LewisBL.Anautomatedscoringalgorithmforcomputerizedclinicalvignettes:evaluatingphysicianperformanceagainstexplicitqualitycriteria.IntJMedInform2006;75(10-11):701-7.185. LeonardKL,MasatuMC.Theuseofdirectclinicianobservationandvignettesforhealthservicesqualityevaluationindevelopingcountries.SocSciMed2005;61(9):1944-51.186. RethansJJ,SturmansF,DropR,vanderVleutenC,HobusP.Doescompetenceofgeneralpractitionerspredicttheirperformance?Comparisonbetweenexaminationsettingandactualpractice.Bmj1991;303(6814):1377-80.187. LandryE,PettC,FiorentinoR,RuminjoJ,MattisonC.Assessingthequalityofrecordkeepingforcesareandeliveries:resultsfromamulticenterretrospectiverecordreviewinfivelow-incomecountries.BMCPregnancyChildbirth2014;14(1):139.188. WorldHealthOrganization.Beyondthenumbers:reviewingmaternaldeathsandcomplicationstomakepregnancysafer:WorldHealthOrganization;2004.189. PattinsonR,KerberK,WaiswaP,etal.Perinatalmortalityaudit:counting,accountability,andovercomingchallengesinscalingupinlow-andmiddle-incomecountries.IntJGynaecolObstet2009;107Suppl1(Supplement):S113-21,S21-2.190. FilippiV,RonsmansC,GohouV,etal.Maternitywardsoremergencyobstetricrooms?Incidenceofnear-misseventsinAfricanhospitals.ActaObstetGynecolScand2005;84(1):11-6.191. HutchinsonC,LangeI,KanhonouL,FilippiV,BorchertM.Exploringthesustainabilityofobstetricnear-misscasereviews:aqualitativestudyintheSouthofBenin.Midwifery2010;26(5):537-43.192. FilippiV,RichardF,LangeI,OuattaraF.Identifyingbarriersfromhometotheappropriatehospitalthroughnear-missauditsindevelopingcountries.BestPractice&ResearchinClinicalObstetrics&Gynaecology2009;23(3):389-400.193. FilippiV,BrughaR,BrowneE,etal.Howtodo(ornottodo)...Obstetricauditinresource-poorsettings:lessonsfromamulti-countryprojectauditing'nearmiss'obstetricalemergencies.HealthPolicyandPlanning2004;19(1):57-66.194. CrombieI,DaviesH,AbrahamS,FloreyCdV.Theaudithandbook.Improvinghealthcarethroughaudit.NewYork:JohnWiley&Sons;1997.195. IversN,JamtvedtG,FlottorpS,etal.Auditandfeedback:effectsonprofessionalpracticeandhealthcareoutcomes.CochraneDatabaseSystRev2012;(6):CD000259.196. BirchL,JonesN,DoylePM,etal.Obstetricskillsdrills:evaluationofteachingmethods.NurseEducToday2007;27(8):915-22.197. CroS,KingB,PaineP.Practicemakesperfect:maternalemergencytraining.BritishJournalofMidwifery2001;9(8):492-6.198. HardenRM,GleesonFA.AssessmentofClinicalCompetenceUsinganObjectiveStructuredClinicalExamination(Osce).MedicalEducation1979;13(1):41-54.199. deGraft-JohnsonJ,VeselL,RosenHE,etal.Cross-sectionalobservationalassessmentofqualityofnewborncareimmediatelyafterbirthinhealthfacilitiesacrosssixsub-SaharanAfricancountries.BMJOpen2017;7(3):e014680.200. RosenHE,LynamPF,CarrC,etal.Directobservationofrespectfulmaternitycareinfivecountries:across-sectionalstudyofhealthfacilitiesinEastandSouthernAfrica.BMCPregnancyChildbirth2015;15(1):306.

Page199of248

201. TripathiV,StantonC,StrobinoD,BartlettL.DevelopmentandValidationofanIndextoMeasuretheQualityofFacility-BasedLaborandDeliveryCareProcessesinSub-SaharanAfrica.PLoSOne2015;10(6):e0129491.202. SinghalN,LockyerJ,FidlerH,etal.HelpingBabiesBreathe:globalneonatalresuscitationprogramdevelopmentandformativeeducationalevaluation.Resuscitation2012;83(1):90-6.203. SholkamyHKK,CherineM,ElnouryA,BreebaartM,HassaneinN..AnObservationalChecklistforFacility-BasedNormalLabourandDeliveryPractices:TheGalaaStudy.MonographsinReproductiveHealthNo5204. OrganizationWH.Integratedmanagementofpregnancyandchildbirth.2003.205. LindbackC,KcA,WrammertJ,VitrakotiR,EwaldU,MalqvistM.Pooradherencetoneonatalresuscitationguidelinesexposed;anobservationalstudyusingcamerasurveillanceatatertiaryhospitalinNepal.BMCPediatr2014;14(1):233.206. Linder-PelzS,StrueningEL.Themultidimensionalityofpatientsatisfactionwithaclinicvisit.JCommunityHealth1985;10(1):42-54.207. SitziaJ,WoodN.Patientsatisfaction:areviewofissuesandconcepts.Socialscience&medicine1997;45(12):1829-43.208. SawyerA,AyersS,AbbottJ,GyteG,RabeH,DuleyL.Measuresofsatisfactionwithcareduringlabourandbirth:acomparativereview.BMCPregnancyChildbirth2013;13(1):108.209. AghaS,KarimAM,BalalA,SoslerS.TheimpactofareproductivehealthfranchiseonclientsatisfactioninruralNepal.HealthPolicyPlan2007;22(5):320-8.210. BramadatIJ,DriedgerM.Satisfactionwithchildbirth:theoriesandmethodsofmeasurement.Birth1993;20(1):22-9.211. AvisM,BondM,ArthurA.Satisfyingsolutions?Areviewofsomeunresolvedissuesinthemeasurementofpatientsatisfaction.Journalofadvancednursing1995;22(2):316-22.212. StaniszewskaS,AhmedL.Theconceptsofexpectationandsatisfaction:dotheycapturethewaypatientsevaluatetheircare?Journalofadvancednursing1999;29(2):364-72.213. WaldenstromU,RudmanA,HildingssonI.IntrapartumandpostpartumcareinSweden:women'sopinionsandriskfactorsfornotbeingsatisfied.ActaObstetGynecolScand2006;85(5):551-60.214. HodnettED.Painandwomen'ssatisfactionwiththeexperienceofchildbirth:asystematicreview.Americanjournalofobstetricsandgynecology2002;186(5):S160-S72.215. BrownS,LumleyJ.Satisfactionwithcareinlaborandbirth:asurveyof790Australianwomen.Birth1994;21(1):4-13.216. LomasJ,DoreS,EnkinM,MitchellA.TheLaborandDeliverySatisfactionIndex:thedevelopmentandevaluationofasoftoutcomemeasure.Birth1987;14(3):125-9.217. KnappL.Childbirthsatisfaction:theeffectsofinternalityandperceivedcontrol.TheJournalofPerinatalEducation1996;5(4):7-16.218. Wilde-LarssonB,LarssonG,KvistLJ,Sandin-BojoAK.Womens'opinionsonintrapartalcare:developmentofatheory-basedquestionnaire.JClinNurs2010;19(11-12):1748-60.219. BrittonJR.Theassessmentofsatisfactionwithcareintheperinatalperiod.JPsychosomObstetGynaecol2012;33(2):37-44.220. RudmanA,El-KhouriB,WaldenstromU.Women'ssatisfactionwithintrapartumcare-apatternapproach.Journalofadvancednursing2007;59(5):474-87.221. GreenJM,CouplandVA,KitzingerJ.Greatexpectations:aprospectivestudyofwomen’sexpectationsandexperiencesofchildbirth.Cambridge:UniversityofCambridge;1988.222. RedshawM,MartinCR.Validationofaperceptionsofcareadjectivechecklist.JEvalClinPract2009;15(2):281-8.223. GrahamWJ,AhmedS,StantonC,Abou-ZahrC,CampbellOM.Measuringmaternalmortality:anoverviewofopportunitiesandoptionsfordevelopingcountries.BMCMed2008;6:12.224. AbouZahrC,WardlawT.Maternalmortalityattheendofadecade:signsofprogress?BulletinoftheWorldHealthOrganization2001;79(6):561-73.

Page200of248

225. ShakurH,RobertsI,FawoleB,etal.Effectofearlytranexamicacidadministrationonmortality,hysterectomy,andothermorbiditiesinwomenwithpost-partumhaemorrhage(WOMAN):aninternational,randomised,double-blind,placebo-controlledtrial.Lancet2017;389(10084):2105-16.226. DumontA,FournierP,AbrahamowiczM,etal.Qualityofcare,riskmanagement,andtechnologyinobstetricstoreducehospital-basedmaternalmortalityinSenegalandMali(QUARITE):acluster-randomisedtrial.Lancet2013;382(9887):146-57.227. RaoKD,ShahrawatR,BhatnagarA.CompositionanddistributionofthehealthworkforceinIndia:estimatesbasedondatafromtheNationalSampleSurvey.WHOSouthEastAsiaJPublicHealth2016;5(2):133-40.228. RaoM,RaoKD,KumarAK,ChatterjeeM,SundararamanT.HumanresourcesforhealthinIndia.Lancet2011;377(9765):587-98.229. SathpathySVS.HumanresourcesforhealthinIndia’snationalruralhealthmission:dimensionsandchallenges..RegionalHealthForum2006;(10:29–37.).230. RaoKD,BhatnagarA,BermanP.Somany,yetfew:HumanresourcesforhealthinIndia.HumResourHealth2012;10(1):19.231. RaoKDBA,BermanP.IndiaHealthBeat,Volume1.India’shealthworkforce:size,compositionanddistribution.;2009.232. BanerjeeA,DeatonA,DufloE.HEALTH,HEALTHCARE,ANDECONOMICDEVELOPMENT:Wealth,Health,andHealthServicesinRuralRajasthan.AmEconRev2004;94(2):326-30.233. NaryanaK.TheUnqualifiedMedicalPractitionersMethodsofPracticeandNexuswiththeQualifiedDoctors,2008.234. MavalankarD,VoraK,PrakasammaM.Achievingmillenniumdevelopmentgoal5:isIndiaserious?BulletinoftheWorldHealthOrganization2008;86(4):243-A.235. Jhpiego.AddressingIndia’sHumanResourceforHealth(HRH)Challenge:GlobalEvidence,LocalEfforts.236. GillK.AprimaryevaluationofservicedeliveryundertheNationalRuralHealthMission(NRHM):findingsfromastudyinAndhraPradesh,UttarPradesh,BiharandRajasthan.PlanningCommissionofIndia,GovernmentofIndia2009.237. DirectorateofFamilyWelfareandSIFPSA.GapAnalysisofFirstReferralUnits(FRUs)inUPbySIFPSA..2013.238. GhodkiPS,SardesaiSP.Obstetrichemorrhage:Anestheticimplicationsandmanagement.AnaesthPainIntensiveCare2014;18:405-14.239. HultonLA,MatthewsZ,StonesRW.ApplyingaframeworkforassessingthequalityofmaternalhealthservicesinurbanIndia.SocSciMed2007;64(10):2083-95.240. SabdeY,DeCostaA,DiwanV.Aspatialanalysistostudyaccesstoemergencyobstetrictransportservicesunderthepublicprivate"JananiExpressYojana"programintwodistrictsofMadhyaPradesh,India.ReprodHealth2014;11(1):57.241. StantonC,ArmbrusterD,KnightR,etal.Useofactivemanagementofthethirdstageoflabourinsevendevelopingcountries.BullWorldHealthOrgan2009;87(3):207-15.242. StantonCK,DeepakNN,MallapurAA,etal.Directobservationofuterotonicdruguseatpublichealthfacility-baseddeliveriesinfourdistrictsinIndia.Internationaljournalofgynaecologyandobstetrics:theofficialorganoftheInternationalFederationofGynaecologyandObstetrics2014;(0).243. IyengarK,JainM,ThomasS,etal.AdherencetoevidencebasedcarepracticesforchildbirthbeforeandafteraqualityimprovementinterventioninhealthfacilitiesofRajasthan,India.BmcPregnancyandChildbirth2014;14(1):270.244. IyengarSD,IyengarK,MartinesJC,DashoraK,DeoraKK.ChildbirthpracticesinruralRajasthan,India:implicationsforneonatalhealthandsurvival.Journalofperinatology:officialjournaloftheCaliforniaPerinatalAssociation2008;28Suppl2((Suppl2)):S23-30.

Page201of248

245. AlehagenSA,FinnstromO,HermanssonGV,etal.Nurse-basedantenatalandchildhealthcareinruralIndia,implementationandeffects-anIndian-Swedishcollaboration.RuralRemoteHealth2012;12(3):2140.246. SodaniP,SharmaK.Trainingneedsassessmentofpublichealthprofessionalsinmaternalandneonatalcare.IndianJournalofMaternalandChildHealth2011;13(4):8.247. AfsanaKBM,CrossS,MavalankarD,RahmanA,RoyT,etal..Cleandelivery:asituationanalysisofhygieneonmaternitywardsinIndiaandBangladesh..2014.248. ThindA,MohaniA,BanerjeeK,HagigiF.Wheretodeliver?AnalysisofchoiceofdeliverylocationfromanationalsurveyinIndia.BMCPublicHealth2008;8(1):29.249. FerrinhoP,BugalhoAM,VanLerbergheW.Isthereacaseforprivatisingreproductivehealth?Patchyevidenceandmuchwishfulthinking.SafeMotherhoodStrategies:aReviewoftheEvidence2000.250. SharanM,StrobinoD,AhmedS.IntrapartumoxytocinuseforlaboraccelerationinruralIndia.IntJGynaecolObstet2005;90(3):251-7.251. JefferyP,DasA,DasguptaJ,JefferyR.UnmonitoredIntrapartumOxytocinUseinHomeDeliveries:EvidencefromUttarPradesh,India.ReproductiveHealthMatters2007;15(30):172-8.252. BaquiAH,WilliamsEK,DarmstadtGL,etal.NewborncareinruralUttarPradesh.IndianJPediatr2007;74(3):241-7.253. OfficeoftheRegistrarGeneral&CensusCommissioner.AnnualHealthSurvey2012-2013FactSheet,UttarPradesh..2012-2013.254. UNICEFIndia.Coverageevaluationsurvey2009.NewDelhi:UnitedNationsChildrenFund2010.255. RegistrarGeneralofIndia.SpecialBulletinonMaternalMortalityinIndia2010-12.NewDelhi,India:GovernmentofIndia;2013.256. SocialStatisticsDivisionMoSaPI,GovernmentofIndia.MillenniumDevelopmentGoals,IndiaCountryReport2014..2014.257. RegistrarGeneralofIndia.SampleRegistrationSystem:MaternalMortalityinIndia1997-2003;Trends,CausesandRiskFactors,.2006.258. LiuL,OzaS,HoganD,etal.Global,regional,andnationalcausesofchildmortalityin2000-13,withprojectionstoinformpost-2015priorities:anupdatedsystematicanalysis.Lancet2015;385(9966):430-40.259. India.RGo.Sampleregistrationsystem(SRS)statisticalreport2013,.2013.260. SankarMJ,NeogiSB,SharmaJ,etal.StateofnewbornhealthinIndia.JPerinatol2016;36(s3):S3-S8.261. WHO-MCEEestimatesforchildcausesofdeath.2000-2015.262. Korde-NayakVNGP.CausesofStillbirth..JObstetGynecolIndia2008.263. McClureEM,WrightLL,GoldenbergRL,etal.Theglobalnetwork:aprospectivestudyofstillbirthsindevelopingcountries.AmJObstetGynecol2007;197.264. WorldHealthOrganization.Buildingleadershipandmanagementcapacityinhealth:Aidememoire.WHO/EIP/healthsystems/2005.1.2005.265. SmithR.Thescientificbasisofhealthservices.BMJ1995;311(7011):961-2.266. BloomN,SadunR,VanReenenJ.Doesmanagementreallywork?Harvardbusinessreview2012;90(11):76-82.267. BertaWB,BakerR.Factorsthatimpactthetransferandretentionofbestpracticesforreducingerrorinhospitals.HealthCareManagementReview2004;29(2):90-7.268. KollbergB,DahlgaardJJ,BrehmerP-O.Measuringleaninitiativesinhealthcareservices:issuesandfindings.InternationalJournalofProductivityandPerformanceManagement2006;56(1):7-24.269. KujalaJ,LillrankP,KronströmV,PeltokorpiA.Time-basedmanagementofpatientprocesses.JournalofHealthOrganizationandManagement2006;20(6):512-24.

Page202of248

270. LiLX,BentonW,LeongGK.Theimpactofstrategicoperationsmanagementdecisionsoncommunityhospitalperformance.JournalofOperationsManagement2002;20(4):389-408.271. TuckerAL,EdmondsonAC.Whyhospitalsdon'tlearnfromfailures:organizationalandpsychologicaldynamicsthatinhibitsystemchange.Californiamanagementreview2003;45(2):55-72.272. GoldsteinSM,WardPT.Performanceeffectsofphysicians’involvementinhospitalstrategicdecisions.JournalofServiceResearch2004;6(4):361-72.273. RambaniR,OkaforB.Evaluationoffactorsdelayingdischargeinacuteorthopedicwards:aprospectivestudy.EuropeanJournalofTraumaandEmergencySurgery2008;34(1):24-8.274. ScottI,PooleP,JayathissaS.Improvingqualityandsafetyofhospitalcare:areappraisalandanagendaforclinicallyrelevantreform.Internalmedicinejournal2008;38(1):44-55.275. ProudloveN,BoadenR.Usingoperationalinformationandinformationsystemstoimprovein-patientflowinhospitals.Journalofhealthorganizationandmanagement2005;19(6):466-77.276. RotterT,KinsmanL,JamesE,etal.Clinicalpathways:effectsonprofessionalpractice,patientoutcomes,lengthofstayandhospitalcosts.CochraneDatabaseSystRev2010;3(3).277. BloomN,PropperC,SeilerS,VanReenenJ.Managementpracticesinhospitals.Manuscript,LondonSchoolEcon2009.278. MinistryofHealthandFamilyWelfareGoI,Maternalhealthdivision..MaternalandNewbornHealthToolkit.Jan2013279. SatpathySK.Indianpublichealthstandards(IPHS)forcommunityhealthcentres.IndianJPublicHealth2005;49(3):123-6.280. BoyneGA,ChenAA.Performancetargetsandpublicserviceimprovement.JournalofPublicAdministrationResearchandTheory2007;17(3):455-77.281. MicheliP,NeelyA.PerformanceMeasurementinthePublicSectorinEngland:SearchingfortheGoldenThread.PublicAdministrationReview2010;70(4):591-600.282. AgarwalR,GreenR,AgarwalN,RandhawaK.BenchmarkingmanagementpracticesinAustralianpublichealthcare.Journalofhealthorganizationandmanagement2016;30(1):31-56.283. ChunYH,RaineyHG.GoalambiguityandorganizationalperformanceinUSFederalagencies.JournalofPublicAdministrationResearchandTheory2005;15(4):529-57.284. MannionR,DaviesHT,BuetowS.Pay-for-performanceinNewZealandprimaryhealthcare.JournalofHealthOrganizationandManagement2008;22(1):36-47.285. KabeneSM,OrchardC,HowardJM,SorianoMA,LeducR.Theimportanceofhumanresourcesmanagementinhealthcare:aglobalcontext.HumResourHealth2006;4(1):20.286. WestMA,BorrillC,DawsonJ,etal.Thelinkbetweenthemanagementofemployeesandpatientmortalityinacutehospitals.InternationalJournalofHumanResourceManagement2002;13(8):1299-310.287. PattersonM,RickJ,WoodSJ,CarrollC,BalainS,BoothA.Systematicreviewofthelinksbetweenhumanresourcemanagementpracticesandperformance.2010.288. AdzeiFA,AtingaRA.MotivationandretentionofhealthworkersinGhana'sdistricthospitals:addressingthecriticalissues.JournalofHealthOrganizationandManagement2012;26(4):467-85.289. BorrillC,WestM,ShapiroD,ReesA.Teamworkingandeffectivenessinhealthcare.BritishJournalofHealthcareManagement2000;6(8):364-71.290. HunterE,NicolM.Systematicreview:evidenceofthevalueofcontinuingprofessionaldevelopmenttoenhancerecruitmentandretentionofoccupationaltherapistsinmentalhealth.TheBritishJournalofOccupationalTherapy2002;65(5):207-15.291. BrooksRG,WalshM,MardonRE,LewisM,ClawsonA.Therolesofnatureandnurtureintherecruitmentandretentionofprimarycarephysiciansinruralareas:areviewoftheliterature.AcademicMedicine2002;77(8):790-8.292. WangY,CollinsC,TangSL,MartineauT.Healthsystemsdecentralizationandhumanresourcesmanagementinlowandmiddleincomecountries.PublicAdministrationandDevelopment2002;22(5):439-53.

Page203of248

293. SmithPC.Performancemeasurementforhealthsystemimprovement:experiences,challengesandprospects:CambridgeUniversityPress;2009.294. HafnerJM,WilliamsSC,KossRG,TschurtzBA,SchmaltzSP,LoebJM.Theperceivedimpactofpublicreportinghospitalperformancedata:interviewswithhospitalstaff.IntJQualHealthCare2011;23(6):697-704.295. MannionR,DaviesHT,GrossR,ElhaynayA,FriedmanN,BuetowS.Pay-for-performanceprogramsinIsraelisickfunds.Journalofhealthorganizationandmanagement2008;22(1):23-35.296. FergusonB,LimJN.Incentivesandclinicalgovernance:moneyfollowingquality?JManagMed2001;15(6):463-87.297. PetersonED,RoeMT,MulgundJ,etal.Associationbetweenhospitalprocessperformanceandoutcomesamongpatientswithacutecoronarysyndromes.Jama2006;295(16):1912-20.298. ChangL-c,LinSW,NorthcottDN.TheNHSperformanceassessmentframework:a“balancedscorecard”approach?Journalofmanagementinmedicine2002;16(5):345-58.299. GiuffridaA,GravelleH,RolandM.Measuringqualityofcarewithroutinedata:avoidingconfusionbetweenperformanceindicatorsandhealthoutcomes.BritishMedicalJournal1999;319(7202):94-8.300. ParandA,DopsonS,RenzA,VincentC.Theroleofhospitalmanagersinqualityandpatientsafety:asystematicreview.BMJOpen2014;4(9):e005055.301. ConwayJ.Gettingboardsonboard:engaginggoverningboardsinqualityandsafety.JtCommJQualPatientSaf2008;34(4):214-20.302. GautamKS.Acallforboardleadershiponqualityinhospitals.QualManagHealthCare2005;14(1):18-30.303. GoeschelCA,WachterRM,PronovostPJ.Responsibilityforqualityimprovementandpatientsafety:hospitalboardandmedicalstaffleadershipchallenges.Chest2010;138(1):171-8.304. ØvretveitJ.Leadingimprovementeffectively:reviewofresearch.2009.London:TheHealthFoundationLondonGoogleScholar.305. FlinR,YuleS.Leadershipforsafety:industrialexperience.QualSafHealthCare2004;13Suppl2(suppl2):ii45-51.306. WestE.Managementmatters:thelinkbetweenhospitalorganisationandqualityofpatientcare.QualityinHealthCare2001;10(1):40-8.307. BloomN,VanReenenJ.WhyDoManagementPracticesDifferacrossFirmsandCountries?JournalofEconomicPerspectives2010;24(1):203-24.308. BloomN,SadunR,VanReenenJ.Doesmanagementmatterinhealthcare.LondonSchoolofEconomicsWorkingpaper2013.309. CastroPJ,DorganSJ,RichardsonB.AhealthierhealthcaresystemfortheUnitedKingdom.McKinseyQuarterly2008.310. BloomN,Propper,C.,Seiler,S.&Reenen,J.V..TheImpactofCompetitiononManagementQuality:EvidencefromPublicHospitalsNationalBureauofEconomicResearchInc2010;NBERWorkingPapers:16032.311. BloomN,SadunR,VanReenenJ.Doesmanagementreallywork?Howthreeessentialpracticescanaddresseventhemostcomplexglobalproblems.Harvardbusinessreview2012;90(11):76-82.312. McConnellKJ,LindroothRC,WholeyDR,MaddoxTM,BloomN.Managementpracticesandthequalityofcareincardiacunits.JAMAInternMed2013;173(8):684-92.313. McConnellKJ,HoffmanKA,QuanbeckA,McCartyD.Managementpracticesinsubstanceabusetreatmentprograms.JournalofSubstanceAbuseTreatment2009;37(1):79-89.314. RegistrarGeneralofIndia.PopulationprojectionsforIndiaandStates:2001-2026.2010.315. RegistrarGeneralofIndia.CensusofIndia(2011).2011.316. RegistrarGeneral&CensusCommissioner.AnnualHealthSurveyBulletin2011-2012UTTARPRADESH.2011..317. RegistrarGeneralofIndia.SampleRegistrationSystem,(2007-09).2007

Page204of248

318. IndiaRGo.SampleRegistrationSystem.2009319. RegistrarGeneralofIndia.SampleRegistrationSystem,.2008.320. IIPSandMacroInternational.NationalFamilyHealthSurvey(NFHS-3),2005-06,India:KeyFindings..2007.321. SinghS,DoyleP,CampbellOM,MathewM,MurthyGV.ReferralsbetweenPublicSectorHealthInstitutionsforWomenwithObstetricHighRisk,Complications,orEmergenciesinIndia-ASystematicReview.PLoSOne2016;11(8):e0159793.322. MOHFWIndia.RuralHealthStatisticsinIndia2013–14..2014.323. MOHFWIndia.NUHM:Implementationframework.2013.324. MackintoshM,ChannonA,KaranA,SelvarajS,CavagneroE,ZhaoH.Whatistheprivatesector?Understandingprivateprovisioninthehealthsystemsoflow-incomeandmiddle-incomecountries.TheLancet2016;388(10044):596-605.325. Bhate-DeosthaliPKR,WagleS..Poorstandardsofcareinsmall,privatehospitalsinMaharashtra,India:implicationsforpublic-privatepartnershipsformaternitycare..ReproductiveHealthMatters2011;19(37).326. GovernmentofIndia;MinistryofHealthandFamilyWelfare.OperationalguidelinesonMaternal&NewbornHealth,.2010,.327. SrivastavaA,BhattacharyyaS,ClarC,AvanBI.EvolutionofqualityinmaternalhealthinIndia:Lessonsandpriorities.InternationalJournalofMedicineandPublicHealth2014;4(1):33.328. MinistryofHealthandFamilyWelfare.NationalPopulationPolicy..2000.329. MinistryofWomenandChildDevelopment.NationalPolicyforEmpowermentofWomen..2001.330. MinistryofHealthandFamilyWelfare.NationalHealthPolicy..2002331. MinistryofHealthandFamilyWelfare.ExecutiveSummaryofNRHMprogramme.2015.332. MinistryofHealthandFamilyWelfare.JananiSurakshaYojana:RevisedGuidelinesforimplementation2006.NewDelhi:GovernmentofIndia;2006.333. NagarajanS,PaulVK,YadavN,GuptaS.TheNationalRuralHealthMissioninIndia:itsimpactonmaternal,neonatal,andinfantmortality.SeminFetalNeonatalMed2015;20(5):315-20.334. WhiteRibbonAlliance.Respectfulmaternitycare:theuniversalrightsofchildbearingwomen.Available:Accessed2012;3.335. DonabedianA.Thequalityofmedicalcare.Science1978;200(4344):856-64.336. WorldHealthOrganization.Managingcomplicationsinpregnancyandchildbirth:aguideformidwivesanddoctors.2000.337. Choices,ChallengesinChangingChildbirthResearchN.Routinesinfacility-basedmaternitycare:evidencefromtheArabWorld.BJOG2005;112(9):1270-6.338. InternationalInstituteforPopulationSciences(IIPS)andMacroInternational.NationalFamilyHealthSurvey(NFHS-4),2014-2015.2014.339. NationalInstituteforClinicalExcellence.Intrapartumcare.Careofhealthywomenandtheirbabiesduringchildbirth.ClinicalGuideline2007;6.340. HayesR,MoultonL.Clusterrandomisedtrials.Taylor&Francis;2009.341. VozdolskaR,SanoM,AisenP,EdlandSD.Theneteffectofalternativeallocationratiosonrecruitmenttimeandtrialcost.ClinTrials2009;6(2):126-32.342. MinistryofHealthandFamilyWelfare.NHMHealthmanagementinformationsystem(HMIS)portal(https://nrhm-mis.nic.in/).2015.343. VyasS,KumaranayakeL.Constructingsocio-economicstatusindices:howtouseprincipalcomponentsanalysis.HealthPolicyPlan2006;21(6):459-68.344. SnijdersTA.Multilevelanalysis:Springer;2011.345. UNICEF.UNICEFData:MonitoringtheSituationofChildrenandWomen2016.346. HinderakerSG,OlsenBE,BergsjoPB,etal.AvoidablestillbirthsandneonataldeathsinruralTanzania.BJOG2003;110(6):616-23.

Page205of248

347. AbouchadiS,AlaouiAB,MeskiFZ,BezadR,DeBrouwereV.PreventablematernalmortalityinMorocco:theroleofhospitals.TropMedIntHealth2013;18(4):444-50.348. KrukME,MbarukuG,McCordCW,MoranM,RockersPC,GaleaS.Bypassingprimarycarefacilitiesforchildbirth:apopulation-basedstudyinruralTanzania.HealthPolicyPlan2009;24(4):279-88.349. ParkhurstJO,SsengoobaF.Assessingaccessbarrierstomaternalhealthcare:measuringbypassingtoidentifyhealthcentreneedsinruralUganda.HealthPolicyPlan2009;24(5):377-84.350. RoweAK,deSavignyD,LanataCF,VictoraCG.Howcanweachieveandmaintainhigh-qualityperformanceofhealthworkersinlow-resourcesettings?TheLancet2005;366(9490):1026-35.351. WorldHealthOrganization.Strategiesforassistinghealthworkerstomodifyandimproveskills:developingqualityhealthcare:aprocessofchange.2000.352. WorldHealthOrganization.Trendsinmaternalmortality:1990-2015:estimatesfromWHO,UNICEF,UNFPA,WorldBankGroupandtheUnitedNationsPopulationDivision.2015.353. MinistryofHealthandFamilyWelfare.MaternalandNewbornHealthToolkit.NewDelhi,India:MaternalHealthDivision,GovernmentofIndia;2013.354. SinghA,MavalankarDV,BhatR,etal.ProvidingskilledbirthattendantsandemergencyobstetriccaretothepoorthroughpartnershipwithprivatesectorobstetriciansinGujarat,India.BulletinoftheWorldHealthOrganization2009;87(12):960-4.355. EvansCL,MaineD,McCloskeyL,FeeleyFG,SanghviH.Wherethereisnoobstetrician--increasingcapacityforemergencyobstetriccareinruralIndia:anevaluationofapilotprogramtotraingeneraldoctors.IntJGynaecolObstet2009;107(3):277-82.356. PotterJE,BerquoE,PerpetuoIHO,etal.UnwantedcaesareansectionsamongpublicandprivatepatientsinBrazil:prospectivestudy.BritishMedicalJournal2001;323(7322):1155-8.357. MurraySF,HunterBM,BishtR,EnsorT,BickD.Effectsofdemand-sidefinancingonutilisation,experiencesandoutcomesofmaternitycareinlow-andmiddle-incomecountries:asystematicreview.BMCPregnancyChildbirth2014;14:30.358. PhadungkiatwattanaP,TongsakulN.AnalyzingtheimpactofprivateserviceonthecesareansectionrateinpublichospitalThailand.ArchGynecolObstet2011;284(6):1375-9.359. PereiraSK,KumarP,DuttV,etal.ProtocolfortheevaluationofasocialfranchisingmodeltoimprovematernalhealthinUttarPradesh,India.ImplementSci2015;10(1):77.360. GetachewA,RiccaJ,CantorD,etal.Qualityofcareforpreventionandmanagementofcommonmaternalandnewborncomplications:astudyofEthiopia’shospitals.Baltimore:Jhpiego2011.361. WorldHealthOrganization.Integratedmanagementofpregnancyandchildbirth:WorldHealthOrganization.;2003.362. PalmerWL,BottleA,AylinP.Associationbetweendayofdeliveryandobstetricoutcomes:observationalstudy.BMJ2015;351:h5774.363. PasupathyD,WoodAM,PellJP,FlemingM,SmithGC.Timeofbirthandriskofneonataldeathatterm:retrospectivecohortstudy.Bmj2010;341:c3498.364. HarveySA,BlandónYCW,McCaw-BinnsA,etal.Areskilledbirthattendantsreallyskilled?Ameasurementmethod,somedisturbingresultsandapotentialwayforward.BulletinoftheWorldHealthOrganization2007;85(10):783-90.365. BerendesS,HeywoodP,OliverS,GarnerP.Qualityofprivateandpublicambulatoryhealthcareinlowandmiddleincomecountries:systematicreviewofcomparativestudies.PLoSMed2011;8(4):e1000433.366. PressInformationBureauMinistryofHealthandFamilyWelfare.Achievementsundermillenniumdevelopmentgoals.2015.367. NgM,MisraA,DiwanV,AgnaniM,Levin-RectorA,DeCostaA.AnassessmentoftheimpactoftheJSYcashtransferprogramonmaternalmortalityreductioninMadhyaPradesh,India.GlobHealthAction2014;7:24939.

Page206of248

368. VogelJP,BohrenMA,TuncalpO,etal.Howwomenaretreatedduringfacility-basedchildbirth:developmentandvalidationofmeasurementtoolsinfourcountries-phase1formativeresearchstudyprotocol.ReprodHealth2015;12(1):60.369. ChassinMR,GalvinRW.Theurgentneedtoimprovehealthcarequality.InstituteofMedicineNationalRoundtableonHealthCareQuality.JAMA1998;280(11):1000-5.370. SainiV,BrownleeS,ElshaugAG,GlasziouP,HeathI.Addressingoveruseandunderusearoundtheworld.Lancet2017.371. OscarssonME,Amer-WahlinI,RydhstroemH,KallenK.Outcomeinobstetriccarerelatedtooxytocinuse.Apopulation-basedstudy.ActaObstetGynecolScand2006;85(9):1094-8.372. CoffeyD.CostsandconsequencesofacashtransferforhospitalbirthsinaruraldistrictofUttarPradesh,India.SocSciMed2014;114:89-96.373. AmbedkarBR.CastesinIndia:Theirmechanism,genesisanddevelopment.ReadingsinIndianGovernmentAndPoliticsClass,Caste,Gender2004:131-53.374. Okafor,II,UgwuEO,ObiSN.Disrespectandabuseduringfacility-basedchildbirthinalow-incomecountry.IntJGynaecolObstet2015;128(2):110-3.375. SudhinarasetM,TreleavenE,MeloJ,SinghK,Diamond-SmithN.Women'sstatusandexperiencesofmistreatmentduringchildbirthinUttarPradesh:amixedmethodsstudyusingculturalhealthcapitaltheory.BMCPregnancyChildbirth2016;16(1):332.376. KrukME,KujawskiS,MbarukuG,RamseyK,MoyoW,FreedmanLP.DisrespectfulandabusivetreatmentduringfacilitydeliveryinTanzania:afacilityandcommunitysurvey.HealthPolicyandPlanning2014:czu079.377. MitraA.ThestatusofwomenamongthescheduledtribesinIndia.TheJournalofSocio-Economics2008;37(3):1202-17.378. SubramanianSV,DaveySmithG,SubramanyamM.IndigenoushealthandsocioeconomicstatusinIndia.PLoSMed2006;3(10):e421.379. GuptaJ,HofmeyrG,SmythR.Positioninthesecondstageoflabourforwomenwithoutepiduralanaesthesia(Review).2007.380. HodnettED,GatesS,HofmeyrGJ,SakalaC.Continuoussupportforwomenduringchildbirth.CochraneDatabaseSystRev2013;7:CD003766.381. BaseviV,LavenderT.Routineperinealshavingonadmissioninlabour.CochraneDatabaseSystRev2014;(11):CD001236.382. MinistryofHealthandFamilyWelfare.SkilledBirthAttendance-AHandbookforAuxiliaryNurseMidwives/LadyHealthVisitorsandStaffNurses.2010.383. AmoranO,OmokhodionF,DairoM,AdebayoA.JobsatisfactionamongprimaryhealthcareworkersinthreeselectedlocalgovernmentareasinsouthwestNigeria.Nigerianjournalofmedicine:journaloftheNationalAssociationofResidentDoctorsofNigeria2004;14(2):195-9.384. Bosch-CapblanchX,GarnerP.Primaryhealthcaresupervisionindevelopingcountries.TropMedIntHealth2008;13(3):369-83.385. LewisM.Informalpaymentsandthefinancingofhealthcareindevelopingandtransitioncountries.HealthAff(Millwood)2007;26(4):984-97.386. GaalP,BelliPC,McKeeM,SzocskaM.Informalpaymentsforhealthcare:definitions,distinctions,anddilemmas.JHealthPolitPolicyLaw2006;31(2):251-93.387. OverM,WatanabeN.Evaluatingtheimpactoforganizationalreformsinhospitals.InnovationsinHealthServiceDelivery:TheCorporatizationofPublicHospitalsHumanDevelopmentNetwork2003.388. HansonC,WaiswaP,MarchantT,etal.ExpandedQualityManagementUsingInformationPower(EQUIP):protocolforaquasi-experimentalstudytoimprovematernalandnewbornhealthinTanzaniaandUganda.ImplementSci2014;9(1):41.389. LaingAW,ShiroyamaC.Managingcapacityanddemandinaresourceconstrainedenvironment:lessonsfortheNHS?JManagMed1995;9(5):51-67.

Page207of248

390. ButlerTW,LeongGK,EverettLN.Theoperationsmanagementroleinhospitalstrategicplanning.JournalofOperationsManagement1996;14(2):137-56.391. TrisoliniMG.Applyingbusinessmanagementmodelsinhealthcare.TheInternationaljournalofhealthplanningandmanagement2002;17(4):295-314.392. l'EuropeOmdlsBrd,ShawCR.Howcanhospitalperformancebemeasuredandmonitored?:WHORegionalOfficeforEurope;2003.393. DuffE.Internationalnews–June2017.Midwifery2017;49:A1-A5.394. CoarasaJ,DasJ,GummersonE,BittonA.Asystematictaleoftwodifferingreviews:evaluatingtheevidenceonpublicandprivatesectorqualityofprimarycareinlowandmiddleincomecountries.GlobalHealth2017;13(1):24.395. BasuS,AndrewsJ,KishoreS,PanjabiR,StucklerD.Comparativeperformanceofprivateandpublichealthcaresystemsinlow-andmiddle-incomecountries:asystematicreview.PLoSMed2012;9(6):e1001244.396. LindelowM,SerneelsP.TheperformanceofhealthworkersinEthiopia:resultsfromqualitativeresearch.SocSciMed2006;62(9):2225-35.397. TuranJM,BulutA,NalbantH,OrtayliN,AkalinAA.Thequalityofhospital-basedantenatalcareinIstanbul.StudFamPlann2006;37(1):49-60.398. HoaNQ,OhmanA,LundborgCS,ChucNTK.Druguseandhealth-seekingbehaviorforchildhoodillnessinVietnam-Aqualitativestudy.HealthPolicy2007;82(3):320-9.399. PaphassarangC,PhilavongK,BouphaB,BlasE.Equity,privatizationandcostrecoveryinurbanhealthcare:thecaseofLaoPDR.HealthPolicyPlan2002;17Suppl(suppl_1):72-84.400. SchneiderH,PalmerN.Gettingtothetruth?Researchinguserviewsofprimaryhealthcare.HealthPolicyPlan2002;17(1):32-41.401. McClureEM,PashaO,GoudarSS,etal.Epidemiologyofstillbirthinlow-middleincomecountries:aGlobalNetworkStudy.ActaObstetGynecolScand2011;90(12):1379-85.402. WorldHealthOrganization.Activemanagementofthethirdstageoflabour:newWHOrecommendationshelptofocusimplementation.2014.403. GulmezogluAM,LumbiganonP,LandoulsiS,etal.Activemanagementofthethirdstageoflabourwithandwithoutcontrolledcordtraction:arandomised,controlled,non-inferioritytrial.Lancet2012;379(9827):1721-7.404. AmericanAcademyofPediatricsandAmericanHeartAssociation.TextbookofNeonatalResuscitation..6theditioned;2011.405. PediatricsAAo.Neonatalencephalopathyandneurologicoutcome.Pediatrics2014;133(5):e1482-e8.406. NagpalJ,SachdevaA,SenguptaDharR,BhargavaVL,BhartiaA.Widespreadnon-adherencetoevidence-basedmaternitycareguidelines:apopulation-basedclusterrandomisedhouseholdsurvey.BJOG2015;122.407. DelvauxT,Ake-TanoO,Gohou-KouassiV,BossoP,CollinS,RonsmansC.QualityofnormaldeliverycareinCoted'Ivoire.AfrJReprodHealth2007;11(1):22-32.408. DuysburghE,ZhangWH,YeM,etal.QualityofantenatalandchildbirthcareinselectedruralhealthfacilitiesinBurkinaFaso,GhanaandTanzania:similarfinding.TropMedIntHealth2013;18(5):534-47.409. PathmanathanI,LiljestrandJ.Investinginmaternalhealth:learningfromMalaysiaandSriLanka:WorldBankPublications;2003.410. BryceJ,ArnoldF,BlancA,etal.MeasuringcoverageinMNCH:newfindings,newstrategies,andrecommendationsforaction.PLoSMed2013;10(5):e1001423.411. MathersCD.BirthsandperinataldeathsinAustralia:variationsbydayofweek.JEpidemiolCommunityHealth1983;37(1):57-62.412. OkomoUA,DibbaseyT,KassamaK,etal.Neonataladmissions,qualityofcareandoutcome:4yearsofinpatientauditdatafromTheGambia'steachinghospital.PaediatrIntChildHealth2015;35(3):252-64.

Page208of248

413. HossainJ,RossSR.TheeffectofaddressingdemandforaswellassupplyofemergencyobstetriccareinDinajpur,Bangladesh.InternationalJournalofGynecology&Obstetrics2006;92(3):320-8.414. KimYM,TappisH,ZainullahP,etal.Qualityofcaesareandeliveryservicesanddocumentationinfirst-linereferralfacilitiesinAfghanistan:achartreview.BMCPregnancyChildbirth2012;12(1):14.415. BhatR.CharacteristicsofprivatemedicalpracticeinIndia:aproviderperspective.HealthPolicyPlan1999;14(1):26-37.416. MeraliHS,LipsitzS,HeveloneN,etal.Audit-identifiedavoidablefactorsinmaternalandperinataldeathsinlowresourcesettings:asystematicreview.BMCPregnancyChildbirth2014;14(1):280.417. Partamin,KimYM,MungiaJ,FaqirM,AnsariN,EvansC.Patternsintraining,knowledge,andperformanceofskilledbirthattendantsprovidingemergencyobstetricandnewborncareinAfghanistan.InternationalJournalofGynecology&Obstetrics2012;119(2):125-9.418. IjadunolaKT,IjadunolaMY,EsimaiOA,AbionaTC.Newparadigmoldthinking:ThecaseforemergencyobstetriccareinthepreventionofmaternalmortalityinNigeria.BMCWomensHealth2010;10:6.419. LeonardKL,MligaGR,MariamDH.BypassinghealthcentresinTanzania:Revealedpreferencesforquality.JournalofAfricanEconomies2002;11(4):441-71.420. KarkeeR,LeeAH,BinnsCW.Bypassingbirthcentresforchildbirth:ananalysisofdatafromacommunity-basedprospectivecohortstudyinNepal.Healthpolicyandplanning2013;30(1):1-7.421. AsefaA,BekeleD.Statusofrespectfulandnon-abusivecareduringfacility-basedchildbirthinahospitalandhealthcentersinAddisAbaba,Ethiopia.ReprodHealth2015;12(1):33.422. RajA,DeyA,BoyceS,etal.AssociationsBetweenMistreatmentbyaProviderduringChildbirthandMaternalHealthComplicationsinUttarPradesh,India.MaternChildHealthJ2017.423. SheferawED,BazantE,GibsonH,etal.RespectfulmaternitycareinEthiopianpublichealthfacilities.ReprodHealth2017;14(1):60.424. AfsanaK,RashidSF.ThechallengesofmeetingruralBangladeshiwomen'sneedsindeliverycare.ReprodHealthMatters2001;9(18):79-89.425. KyomuhendoGB.LowuseofruralmaternityservicesinUganda:Impactofwomen'sstatus,traditionalbeliefsandlimitedresources.ReproductiveHealthMatters2003;11(21):16-26.426. UnitedNationsHighCommissioner.ReportoftheOfficeoftheUnitedNationsHighCommissionerforHumanRightsonpreventablematernalmortalityandmorbidityandhumanrights.2010.427. UnitedNationsHighCommissioner.Technicalguidanceontheapplicationofahumanrights-basedapproachtotheimplementationofpoliciesandprogrammestoreducepreventablematernalmorbidityandmortality.2012.428. MillerS,LalondeA.Theglobalepidemicofabuseanddisrespectduringchildbirth:History,evidence,interventions,andFIGO'smother-babyfriendlybirthingfacilitiesinitiative.IntJGynaecolObstet2015;131Suppl1:S49-52.429. BowlingA.Researchmethodsinhealth:investigatinghealthandhealthservices:McGraw-HillEducation(UK);2014.430. YadavV,KumarS,BalasubramaniamS,etal.Facilitatorsandbarrierstoparticipationofprivatesectorhealthfacilitiesingovernment-ledschemesformaternityservicesinIndia:aqualitativestudy.BMJOpen2017;7(6):e017092.431. EvansCL,KimYM,YariK,AnsariN,TappisH.UsingdirectclinicalobservationtoassessthequalityofcesareandeliveryinAfghanistan:anexploratorystudy.BMCPregnancyChildbirth2014;14(1):176.432. SemrauKE,HirschhornLR,KodkanyB,etal.EffectivenessoftheWHOSafeChildbirthChecklistprograminreducingseverematernal,fetal,andnewbornharminUttarPradesh,India:studyprotocolforamatched-pair,cluster-randomizedcontrolledtrial.Trials2016;17(1):576.

Page209of248

433. LangeIL,KanhonouL,GoufodjiS,RonsmansC,FilippiV.Thecostsof'free':Experiencesoffacility-basedchildbirthafterBenin'scaesareansectionexemptionpolicy.SocialScience&Medicine2016;168:53-62.434. KirkpatrickDonald.GreatIdeasRevisited.TechniquesforEvaluatingTrainingPrograms.RevisitingKirkpatrick'sFour-LevelModel.TrainingandDevelopment1996;50(1):54-9.435. UtzB,KanaT,vandenBroekN.Practicalaspectsofsettingupobstetricskillslaboratories--aliteraturereviewandproposedmodel.Midwifery2015;31(4):400-8.436. GrahamWJ.Criterion-basedclinicalauditinobstetrics:bridgingthequalitygap?BestPractResClinObstetGynaecol2009;23(3):375-88.437. DasJK,KumarR,SalamRA,LassiZS,BhuttaZA.Evidencefromfacilitylevelinputstoimprovequalityofcareformaternalandnewbornhealth:interventionsandfindings.ReprodHealth2014;11Suppl2(Suppl2):S4.438. SingerSJ,FalwellA,GabaDM,etal.Identifyingorganizationalculturesthatpromotepatientsafety.HealthCareManagementReview2009;34(4):300-11.439. GonG,AliSM,TowrissC,etal.Unpackingtheenablingfactorsforhand,cordandbirth-surfacehygieneinZanzibarmaternityunits.HealthPolicyandPlanning2017.440. PuriM,LahariyaC.SocialauditinhealthsectorplanningandprogramimplementationinIndia.IndianJCommunityMed2011;36(3):174-7.441. JoshiA.Dotheywork?Assessingtheimpactoftransparencyandaccountabilityinitiativesinservicedelivery.DevelopmentPolicyReview2013;31(s1).442. WaiswaP,KalterHD,JakobR,BlackRE,SocialAutopsyWorkingG.Increaseduseofsocialautopsyisneededtoimprovematernal,neonatalandchildhealthprogrammesinlow-incomecountries.BulletinoftheWorldHealthOrganization2012;90(6):403-A.443. JacksonEF,SiddiquiA,GutierrezH,KanteAM,AustinJ,PhillipsJF.EstimationofindicesofhealthservicereadinesswithaprincipalcomponentanalysisoftheTanzaniaServiceProvisionAssessmentSurvey.BMCHealthServRes2015;15(1):536.444. JolliffeI.Principalcomponentanalysis:WileyOnlineLibrary;2002.

Page210of248

12.ListofAppendicesAppendix1:QoCassessmenttoolfornormallabourandchildbirthinUttarPradeshin2015

Part 1: IDENTIFICATION

I01.NAMEANDCODEOFDISTRICT_______________________I02.NAMEANDCODEOFBLOCK_______________________I03.FACILITYTYPE 1. COMMUNITYHEALTHCENTRES(CHCs)2. DISTRICTHOSPITAL(DH)3. MEDICALCOLLEGEANDTEACHINGHOSPITAL4.PRIVATEMATERNITYCENTREI04.OWNER/MANAGERNAME_____________________________________

I05.HEALTHWORKERNAME

(Attendingnurse/doctorwhoisconductingthedelivery)

_____________________________________

[___][___]

[___][___]

[___][___]

RESEARCHER VISITS I06.DATE

I07.TIMEOBSERVATIONSTARTED

I08.TIMEOBSERVATIONFINISHED

I09.INTERVIEWER’SNAME

I10.SUPERVISOR’SNAME

Part2:SummarySheet

CIRCLEALLTHATAPPLIES

I11.UniqueIdentificationcode

FacilityNumber(FF):

ObservationNumber:

I12.Participation Agreedtoparticipate A. Refusedtoparticipate B. Didnotfulfileligibilitycriteria C. Developedacomplicationafterenrolmentandobservationended(Pleasespecifyreason)

D.

I13.Accompanyingperson Yes 1

Page211of248

No 2I14.Consentreceived?

Wittenconsentbywoman A. Writtenconsentfamily B. Oralconsentbyresearcher C. ThumbPrint D.

SectionA:Screeningquestionnaire

Instructionstotheresearcher:Approachthewardnursestoidentifyallpregnantwomeninthelabourwards,admissionsdepartment,orotherplaceswherepregnantwomanmaybeadmitted.Fromtheirmedicalrecords,completeSectionAtoassesstheireligibilityforinclusioninthestudy.Evenifthewomanisdeemedeligibleshemaydevelopacomplicationduringlabourandchildbirth.Insuchacaseobservecareprovideduptothatpointintime.

Casedefinition:

Anormalvaginaldeliveryisonethatis:• Spontaneousinonset• Low-risk at the start of labour (nohistory ofmedical conditions in thepast or problems in the current

pregnancy,nohistoryofpreviousobstetricandneonatalcomplications,nofoetalcomplicationsincurrentpregnancyandnopreviousgynaecologicalhistory)

• Asingleinfantisbornspontaneouslyinavertexposition• Gestationalagebetween37and42(+0)completedweeksofpregnancy.• Thewomanshouldbebetween18-49yearsofage.UniqueID:

A1. Ageofthewoman Completeinyears___________

A2. Gestationalageofthewomen,indicateinweeksand

days

Verifywithmedicalrecords

______/Weeks_______Days

A3. Gravidity

1. Numberofbabiesbornalive

2. Numberofbabiesborndead?

3. Numberofabortions/Miscarriage

1.

2.

3.

A4. Parity(1+2)

A5. Isthelabourspontaneous Yes No DK

A6. Wasinductionoflabourconducted? Yes No DK

A7. IfyesinA6,pleaseprovidedetailsofthemethodused?(Specify

drugorproceduresused)

A8. Isthereanyhistoryofmedical/obstetriccomplicationsin

previouspregnancies?

Yes

No

DK

A9. Ifyes,Pleasespecify:

Page212of248

A10. Arethereanyfoetalcomplicationsinthecurrentpregnancy? Yes No DK

A11. Isthisasingletonpregnancy? Yes No DK

A12. Isthebabyinvertexposition? Yes No DK

A13. Cervicaldilation (________cms)

A14. Wasthewoman’sBloodpressuremeasured?

Ifyes,pleasespecifythereading._______________(mm/hg)

Yes

No

DK

A15. Wasthewoman’stemperaturemeasured?

Ifyes,pleasespecifythereading./_____________(Degree

Celsius)

Yes

No

DK

A16. Wasurinetestedforpresenceofprotein? Yes No DK

A17. Didthehealthworkerperformthefollowingstepsforabdominal

examination

a. Checksfundalheightwithameasuringtape Yes No DK

b. Checksfetalpresentationbypalpationofabdomen Yes No DK

c. ChecksFoetalHeartRate Yes No DK

SectionB:Demographic,Socio-economicandEducationalStatus

Instructionstotheobserver:Ifthewomanisinactivelabour,approachcompanionsorfamilymembersofthepregnant woman to complete this section after obtaining consent. If the woman doesn’t have anyaccompanyingperson,collectinformationdirectlyfromheratasuitabletimeafterdelivery.

No. QUESTION CATEGORIES

CODE(Circle)

B1. Clienthospitalmedicalnumber B2. Addressoftheclient B3. Areyouabookedcase? Yes 1

No 2B4. Whereistheclientcomingfrom?

Directlytofacility

1

Referredfromanotherfacility

2

B5. Howmuchtimedidtheclient/familytaketotravelfromhome/elsewheretothisfacility

a. Fromhometofirstfacilityorthisfacilityifcomingdirectly.

(___/___)Timeinhh/mm

b. Fromreferralfacilitytothisfacility (___/___)Timeinhh/mm

c. Totaltimetoreachfacility (___/___)Timeinhh/mm

B6. Whatisyourreligion? Hindu 1.

Page213of248

No. QUESTION CATEGORIES

CODE(Circle)

Muslim 2. Christian 3. Noreligion 4. Other 5. Don’tKnow 6.

B7. Whatisyourcaste/category?

Brahmin 1. Rajput/Thakur 2.

Kayasthi/Srivastava/Lala 3. Chamar 4. Dusadh 5.

Musahar 6. Pasi 7. Dhobi 8. Chaupal 9. Yadav 10. Vaishya/Bania 11. Kurmi/Katiyar 12.

Shah 13. Nocaste/Tribe 14.

Other; 15. Don’tKnow 16.

B8. Note:ifthecasteisascheduledcaste,scheduledtribe,otherbackwardcaste

Scheduledcaste 1. Scheduledtribe 2. Otherbackwardcaste 3. GeneralCaste 4.

Other 5.

B9. DoesthewomanorherfamilyhaveaBelowPovertyLineCard(verifyBPLcard)?

Yes 1. No 2.

Don’tKnow 3.

B10. Whatistheoccupationofthepregnantwoman?

Self-employed,Agriculture 1. Wageemployed 2. Agriculturallabourer 3. Salariedworker 4. Self/Employed,Business 5. Skilledworker 6. Retired 7. Lookingforwork 8. Notworkingandnotlookingforwork 9. Other; 10.

Don’tKnow 11.

B11. Kachha 1.

Page214of248

No. QUESTION CATEGORIES

CODE(Circle)

Howhasyour(thewoman’shouse)beenconstructed?

Semi-Pucca 2.

Pucca 3. Don’tKnow 4.

B12. Whatisthemainsourceofdrinkingwaterforthemembersofwoman’shousehold?

Pipedwaterintodwelling 1. Pipedintoyard/plot 2.

Publictaps/Standpipe 3. PublicHand-pump 4. Privatehand-pump 5. Tubewellorborehole 6. Dugwell 7. Rainwater 8. Tanker/truck 9. Surfacewater(Rover,Lake,Pond,Stream,Canal,Irrigationchannel)

10.

Don’tKnow 11. B13. Whatkindoftoiletfacilitiesdoesthe

householdhave?Probeindetail

Flushorpourflushtoilet 1.

Flushtopipedsewersystem 2. Flushtoseptictank 3. Flushtopitlatrine 4. Flushtosomewhereelse 5. PitLatrine 6. Ventilatedimprovepitbiogaslatrine 7. Pitlatrinewithslab/openpit 8. Twinput/Compostingtoilet 9. DryToilet 10. Nofacilities/usedopenspaceorfield 11. Don’tKnow 12. Other(Pleasespecify)……….

B14. Whatkindoffueldoesthehouseholduseforcookingmostofthetime?

Selectonlyoneoption

Electricity 1. LPG/Naturalgas 2.

Biogas 3.

Kerosene 4.

Wood 5. Agriculturecropwaste 6. Dungcakes 7. Others(Pleasespecify) 8.

Don’tknow 9.

B15. Doesthepregnantwoman’shouseholdhave?(circleallasappropriate)

Mattress 1. Pressurecooker 2.

Chair 3.

Cot/Bed 4.

Page215of248

No. QUESTION CATEGORIES

CODE(Circle)

Table 5.

Almirah/Dressingtable 6. Electricfan 7.

Radio/transistor 8.

ColourTV 9.

VCR/VCD/DVD/CDplayer 10.

Sewingmachine 11.

Mobiletelephone 12. Anyothertelephone 13.

Computer/Laptop 14.

Refrigerator 15.

Watch/clock 16.

Bicycle 17.

Motorcycle/scooter 18. Animal-drawncart 19.

Car 20. Waterpump 21.

Tractor 22. B16. Doesyourhouseholdhaveelectricity

Yes 1.

No 2.

Don’tKnow 3.

B17. Whatisthehighestlevelofeducationthewomanhasattained?

Literatewithoutformaleducation 1. LiteratewithFormalEducationBelowPrimary

2.

Primary(Upto5thstandard) 3. Middle(6thto8thStandard) 4. Secondary/MetricsClassX 5.

Hr.Secondary/Sr.Secondary/Pre-University(ClassXII)

6.

Graduate/BBA/B.TECH/Equivalent 7.

PostGraduate/MBA/EquivalentorHigher 8.

TechnicaDiploma 9. Non-technical/CertificateCourse 10. Other 96Illiterate 00

Page216of248

SectionC:DirectObservationofnormalvaginaldeliveries

Instructionstotheresearcher:

Findthehealthworkerinvolvedintheprovisionofcaretothewoman.Ifthisisnotanewrespondent,proceeddirectlytopart2.Pleaseobtaininformedconsentfromboththeclientandthehealthworkerbeforebeginningtheclinicalpracticeobservations.Ensurethattheproviderknowsthatyouarenottheretoevaluatehimorherandthatyouarenotanexperttobeconsultedduringthesession.

ProvideInformationandConsentsheetstothehealthworker.

Part1:

C2.Whoisconductingthedelivery?(Circleasappropriate;severalresponsespossible)

Healthworkerqualification Categorycode

Healthworkerqualification

Categorycode

Doctor(MBBS) 1 Nursingprofessionals(post-bachelor) 7Doctor(BAMS) 2 Midwiferyprofessionals(post-bachelor) 8Obstetricianandgynaecologists 3 AuxiliaryNurseMidwife 9Paediatriciansandneonatologists 4 GeneralNurseMidwife 10Anaesthetists 5 Neonatalnurse 11Nursingprofessional(Bachelor) 6 Others(specify):…..………..………………….. 12

Instructionstotheresearcher:Providetheinformationsheetandconsentformtotheclient,nextofkinorfamilymemberpriortobeginningPart2.Didtheclient/family/accompanyingpersonprovideaninformedconsentandagreetolettheresearcherbepresentduringlabouranddelivery?

Part2:ClinicalpracticeObservationFirststageofLabour

Yes

No

DK

C4.Didahealthworkerexplaintheprocessoflabourtothewomanorcompanionatleastoncebeforethestartofactivelabour?

1 2 8

C5.Observer:Wasacompanionallowedtobewiththewomanduringlabour?

1 2 8

C6.VaginalExaminationperformedusingsterilegloves? 1 2 8C7.Womaninformedbeforevaginalexaminationperformed? 1 2 8C8.Waspartographusedtomonitorlabour?

IfNo,thenskiptoC11.1 2 8

C9.Ifactionlinereachedonpartograph,wasanydefinitiveactiontaken?

1 2 8

C10.Whatdefinitiveactionwastaken:(circleasappropriate) a. Consultwithspecialist 1 2 8b. Refertootherfacilityforspecialist 1 2 8

C1.ConsentGivenbyhealthworker:�Yes:AftercompletingthePartOneproceedtoPart2.�No:Finishtheinterview.

C3.ConsentGivenbywoman:�Yes:ContinuetoPart2.�No:Finishinterview.

Page217of248

c. Prepareforassisteddelivery 1 2 8d. PrepareforC-section 1 2 8e. Other(pleasespecify__________________________)

C11.WasFoetalheartbeatmonitoredatregularintervals? 1 2 8C12.Iffoetalheartbeatindicated,writeinthebox? ……../bpmC13.Oralfluidofferedtothewomanonrequest? 1 2 8C14.VisualPrivacyofthepregnantwomanensured? 1 2 8C15.Womenencouragedtomovearoundbytheprovider 1 2 8C16.Waslabouraugmentationdone?

IfNo,thenskiptoC20.1 2 8

C17.Whywaslabouraugmentationperformed? a) Inefficientuterineactivity Ab) Cervicaldilatationoflessthan2cmin4hours Bc) Notknown Cd) Other(Pleasespecify)……..

C18.Didahealthworkerexplaintothemotherwhylabouraugmentationwasbeingdone?

1 2 8

C19.Howwaslabouraugmentationdone? a) Artificialruptureofthemembranes 1 2 8b) Useofsyntocinon/Oxytocin 1 2 8c) Others (Please Specify the name of the injection apart from

syntocinon/Oxytocin)

Questionsonexaminationandprocedures C20.Handwashingdonepriortoanyexaminationofthewoman 1 2 8C21.Healthworkerwearssterilesurgicalgloves 1 2 8C22.Cleansthevulvaandperineumwithantisepticsolution 1 2 8C23.Drapeswoman(onedrapeunderbuttocks,oneoverabdomen) 1 2 8Preparationfordelivery Checktoseeifthefollowingequipmentandsuppliesarelaidoutinpreparationfordelivery.Ifsomesuppliesareinadeliverykit,look/asktodeterminewhatitemsareincluded.

C24.Preparesuterotonic(Oxytocin)forActivemanagementofthirdstageoflabour

1 2 8

C25.Timer(clockorwatchwithsecondshand) 1 2 8C26.Self-inflatingventilationbag(250or500mL) 1 2 8C27.Newbornfacemask(size0,1) 1 2 8C28.MucusExtractor,suctiontube/Suctionbulb 1 2 8C29.Catheter 1 2 8C30RadiantWarmer 1 2 8C31.WeighingScale C32.Atleasttwocloths/blanketsfornewborn(onetodry;onetocover)

1 2 8

C33.Umblicalcordtiesorclamps 1 2 8C34.Sterilescissorsorblade 1 2 8C35.Hasthewomancompletedthefirststageoflabour? 1 2 8EnsurethatthefirststageoflabouriscompletebeforemovingdowntoSectionDbelow. SectionD:ObservationofSecond&ThirdStageofLabour PREPARATIONFORDELIVERY D1.Motherinformedofthestageoflabour 1 2 8

Page218of248

D2.Wasepisiotomyperformed 1 2 8D3.Wasalocalanestheticinjectionadministeredbeforetheepisiotomy

1 2 8

D4.Positionofthewomanduringchildbirth a) Lithotomy(onherback) Ab) Squatting Bc) Leftlateral Cd) Other(Pleasespecify)....................…………. D

D5.Motheraskedaboutchoiceofpositionfordelivery 1 2 8D6.Whoperformedthedelivery a) Doctor Ab) Nurse Bc) Midwife Cd) Interndoctor De) Studentnurse Ef) Studentmidwife Fg) Other(pleasespecify)………………………………………DELIVERY&UTEROTONIC(OXYTOCIN) D7.Asbaby'sheadisdelivered,supportsperineum 1 2 8D8.Recordtimeofthedeliveryofthebaby 1 2 8D9.Checksforanotherbabypriortogivingtheuterotonic(Oxytocin) 1 2 8D10.Administersuterotonic(oxytocin)?IfNo,thenskiptoD13

1 2 8

D11.Whichuterotonicwasgiven?(circleasappropriate)VERIFY

a) Oxytocin Ab) Ergometrine Bc) Syntometrine Cd) Misoprostol D

D12.Timingofadministrationofuterotonic/oxytocin(circleasappropriate)a) Atdeliveryofanteriorshoulder Ab) Within1minofdeliveryofbaby Bc) Within3minofdeliveryofbaby Cd) Morethan3minafterdeliveryofbaby D

D13.Tiesorclampscordwhenpulsationsstop,orby2-3minutesafterbirth(notimmediatelyafterbirth)

1 2 8

D14.Cutscordwithsterilebladeorsterilescissors

1 2 8

D15.Appliestractiontothecordwhileapplyingsuprapubiccountertraction

1 2 8

D16.Performsuterinemassageimmediatelyfollowingthedeliveryoftheplacenta

1 2 8

D17.Assessescompletenessoftheplacentaandmembranes 1 2 8D18.Assessesforperinealandvaginallacerations 1 2 8

D19.Wasvaginalbleedingmonitoredafterthedelivery? 1 2 8D20.WasanydruggiventopreventPPH?IfYes,PleaseSpecify(_______________________________________)

1 2 8

D21.Womaninformedaboutthesexofthenewborn 1 2 8

Page219of248

D22.Outcomeofthebaby

Alive

Complication

s

Dead

D23.Outcomeofthewoman

Alive

Complication

s

Dead

POTENIALLYHARMFULPRACTICES D24.Didyouobserveanyofthefollowingharmfulpracticesdonebyanyhealthworkerinvolvedintheprovisionofcare?(Circleallthatapply)

a. Useofenema Ab. Pubicshaving Bc. Applyfundalpressuretohastenthedeliveryofthebabyorthe

placentaC

d. Uterinelavageafterdelivery De. Slapthenewborn Ef. Holdthenewbornupsidedown Fg. Shoutinsultorthreatenthewomanatanytimeduringlabourand

childbirthG

h. Slap,hitorpinchthewomanatanytimeduringlabourandchildbirth HD25.Didyouseeanyofthefollowingpracticesthatweredonewithoutappropriateindication(Circleallthatapply)

a) Manualexplorationoftheuterusafterdelivery Ab) Useofepisiotomy Bc) Aspirationofthemouthandnoseassoonasthenewbornis

bornC

d) Restrictfoodandfluidduringlabour De) Noneoftheabove. E

ESSENTIALNEWBORNCARED26.Immediatelydriesbabywithtowel 1 2 8D27.Discardsthewettowel 1 2 8D28.Isthebabybreathingorcrying? 1 2 8D29.Placesbabyonmother’sabdomen“skintoskin” 1 2 8D30.Coversbabywithdrytowel 1 2 8D31.Apgarscorecheckedafteroneminute,Ifyes,pleaseindicateintheboxASKANDVERIFYFROMRECORDS

1 2 8

D32.Apgarscorecheckedafterfiveminutes?Ifyes,indicateintheboxASKANDVERIFYFROMRECORDS

1 2 8

D33.Motherandnewbornkeptinsameroomafterdelivery(rooming-in)

1 2 8

D34.Wasbreastfeedinginitiatedwithinthefirsthourafterbirth 1 2 8D35.Wastheweightofthebabymeasured?Ifyes,thenSpecify(_____________________)

1 2 8

Fieldnotes:

Page220of248

Appendix2:ToolforassessmentofmanagementpracticesinmaternityfacilitiesinUttarPradeshin2015InterviewDetails HospitalandManager’sInformationa) HospitalName: b) Name:c) HospitalID d) Position:e) InterviewerName: f) Specialty:g) Date(DD/MM/YY): h) Tenureinpost(numberofyears):i) Time(24-hourclock): j) Tenureinhospital(numberofyears):k) Runninginterview Listeningtointerview

l) Howoldisyourhospital(numberofyears)?ManagementQuestions

1. LayoutofPatientFlow

Testshowwellthematernity

carepathwayisconfiguredat

thefacilityandwhetherstaff

pro-activelyimprovetheirown

work-placeorganization

Score:

1 2 3 4 5 99

a) Canyoubrieflydescribethepregnantwomen’sjourneyatthefacility?

b) Howcloselylocatedarewards,theatres,diagnosticscentresandconsumables?

c) Howoftendoyourunintoproblemswiththecurrentlayoutandpathwaymanagement?

Score1:Lay-outofhospitaland

organizationofworkplaceisnot

conducivetopatientflow(e.g.wardison

differentlevelfromtheatreor

consumablesareoftennotavailablein

therightplaceattherighttime)

Score3:Lay-outofhospitalhasbeenthought-

throughandoptimizedasfaraspossible;work

placeorganisationisnotregularly

challenged/changed(orviceversa)

Score5:Hospitallayouthasbeen

configuredtooptimizepatientflow;

workplaceorganizationischallenged

regularlyandchangedwheneverneeded

2) Isthereastandardisedmaternitycarepathwayatthefacility?Ifyes,whatwastherationaleforIntroducingStandardisation/PathwayManagement?

Teststhemotivationand

impetusbehindchangesto

operationsandwhatchange

storywascommunicated

Score:

a) Howdidyoumakeimprovementstothematernitycarepathway?Canyoudescribearecentexampletome?

b) Howoftendoyouchangethematernitycarepathway?

c) Whatfactorsledtotheadoptionofthesepractices?

d) Whotypicallydrivesthesechanges?

Score1:Changeswereimposedtop-down

orbecauseotherdepartmentswere

Score 3: Changes were made because of

financialpressureandtheneedtosavemoney

Score5:Changesweremadetoimprove

overallperformance,bothclinicaland

Page221of248

1 2 3 4 5 99

making(similar)changes;rationalewas

notcommunicatedorunderstoodbyall.

or as a (short-term) measure to achieve

governmentand/orexternaltargets

financial,withbuy-infromallaffected

staffgroups;thechangeswere

communicatedinacoherentchangestory.

3) Standardisationand

ProtocolsTestsifthereare

standardisedprocedures,

guidelinesandprotocolsfor

managementoflabourand

childbirththatareapplied

andmonitored

systematically

Score:

1 2 3 4 5 99

a) Howstandardisedarethemainclinicalprocesses?

b) Howclearareclinicalstaffsonhowspecificproceduresshouldbecarriedout?

c) Whattoolsandresourcesdoestheclinicalstaffemploy(e.g.checklists)toqualitycareduringlabourandchildbirth?

d) Howaremanagersabletomonitorwhetherclinicalstaffarefollowingestablishedprotocols?

Score 1: Little standardisation and few

protocolsexists

Score3:Protocolshavebeencreated,butare

notcommonlyusedbecausetheyaretoo

complicated,haven’tbeendisseminatedand

notmonitoredadequately

Score5:Protocolsareknownandused

byallclinicalstaffandregularlyfollowed

uponthroughsomeformofmonitoring

oroversight

4) GooduseofHuman

Resources

Testswhetherstaffaredeployed

todowhattheyarebest

qualifiedfor,butnevertheless

helpoutelsewherewhenneeded

Score:

1 2 3 4 5 -99

a) Withrespecttoyourstaff,whathappenswhenthereisahighvolumeofwomencomingtodeliveratyourhospital?

b) Howdoyouknowwhichtasksarebestsuitedtodifferentstaff?Fore.g.:whoconductsnormaldeliveriesorcaesareansor

providesanaesthesia?

Score1:Staffoftenendupundertaking

tasksforwhichtheyarenotqualifiedor

over-qualifiedwhentheycouldbeused

elsewhere;staffarenotutilised

effectively,andtendtobegenerally

underutilised

Score3:Seniorstafftrytousetherightstaff

fortherightjob,butdonotgotogreat

lengthstoensurethis;staffmaymovebut

ofteninanuncoordinatedmanner

Score5:Staffrecogniseeffectivehuman

resourcedeploymentasakeyissueand

willgotosomelengthstomakeit

happen;shiftingstafffromlessbusyto

busyareasisdoneroutinelyandina

coordinatedmanner,basedonthe

documentedskills

Page222of248

5) ContinuousImprovement

Testsprocessesforandattitudes

towardscontinuous

improvement,andwhether

learningsarecapturedand

documented

Score:

1 2 3 4 5 -99

a) Howdoproblemstypicallygetexposedandfixedatthismaternityfacility?

b) Canyoutalkmethroughtheprocessforarecentproblemthatyoufaced?

c) Whenprocessesdochange,whatisthemaindriverofchange?

d) Whowithinthehospitaltypicallygetsinvolvedinchangingorimproving?Howdo/candifferentstaffgroupsgetinvolvedin

thisprocess?Canyouthinkofanyexamples?

Score1:Processimprovementsare

madeonlywhenproblemsoccur,oronly

involveonestaffgroup

Score3:Improvementsaremadeinirregular

meetingsinvolvingallstaffgroups,to

improveperformanceintheirareaofwork

(e.g.wardortheatre)

Score5:Exposingproblemsina

structuredwayisintegraltoan

individualsresponsibilitiesand

resolutioninvolvesallstaffgroups,along

theentirepatientpathway;exposingand

resolvingproblemsisapartofaregular

businessprocessratherthanbeingthe

resultofextraordinaryefforts

6) Whodecideshowworkisallocatedacrossclinicalstaff?Allmanagers Mostlymanagers Aboutthesame Mostlyclinicalleaders Allclinical

leaders -9

7) PerformanceTracking:

Testswhetherperformanceis

trackedusingmeaningful

metricsandwithappropriate

regularity

Score:

1 2 3 4 5 -99

a) Whatkindofperformanceorqualityindicatorswouldyouuseforperformancetracking?

b) Howfrequentlyarethesemeasured?

c) Whogetstoseethesedata?

d) If I were to walk through your hospital wards and operating rooms, could I tell how you were doing against your

performancegoals?

Score1:Measurestrackeddonot

indicatedirectlyifoverallobjectivesare

beingmet(onlygovernmenttargetsare

tracked);trackingisanad-hocprocess

(certainprocessesaren’ttrackedatall)

Score3:Mostimportantperformanceor

qualityindicatorsareformallytrackedand

overseenbyseniorstaff

Score5:Performanceorquality

indicatorsarecontinuouslytrackedand

communicatedagainstmostcritical

measures,bothformallyandinformally,

Page223of248

toallstaffusingarangeofvisual

managementtools

8) PerformanceReview:

Tests whether performance is

reviewed withappropriate

frequency and communicated

to staff

Score:

1 2 3 4 5 -99

a) Howdoyoureviewyourmainperformanceindicators?

b) Canyoutellmeaboutarecentreviewmeeting?

c) Whoisinvolvedinthesemeetings?Whogetstoseetheresultsofthisreview?

d) Whatisatypicalfollow-upplanthatresultsfromthesemeetings?

Score1:Performanceisreviewed

infrequentlyorinanun-meaningfulway

(e.g.onlysuccessorfailureisnoted)

Score3:Performanceisreviewed

periodicallywithbothsuccessesandfailures

identified;resultsarecommunicatedto

seniorstaff;noclearfollowupplanis

adopted

Score5:Performanceiscontinually

reviewed,basedontheindicators

tracked;allaspectsarefollowedupon,to

ensurecontinuousimprovement;results

arecommunicatedtoallstaff

9) PerformanceDialogue:

Tests the quality of review

meetings.

Score:

1 2 3 4 5 -99

a) Howarethesereviewmeetingsstructured?Howistheagendadetermined?Couldyougivemearecentexample?

b) Duringthesemeetings,doyoufindthatyougenerallyhaveenoughinformationforreview?

c) Howusefuldoyoufindthesemeetings?Whattypeoffeedbackoccursinthesemeetings?

d) Foragivenproblem,howdoyougenerallyidentifytherootcause?

Score1:Therightinformationfora

constructivediscussionisoftennot

presentorthequalityistoolow;

conversationsfocusoverlyondatathat

isnotmeaningful;aclearagendaisnot

knownandpurposeisnotexplicitly

stated;nextstepsarenotclearlydefined

Score3:Reviewconversationsareheldwith

theappropriatedatapresent;objectivesof

meetingsarecleartoallparticipatinganda

clearagendaispresent;conversationsdonot,

drivetotherootcausesoftheproblems;next

stepsarenotwelldefined

Score5:Regularreview/performance

conversationsfocusonproblemsolving

andaddressingrootcauses;purpose,

agendaandfollow-upstepsareclearto

all;meetingsareanopportunityfor

constructivefeedbackandcoaching

Page224of248

10) Consequence

Management:

Testswhetherdifferinglevelsof

performance(NOTpersonalbut

plan/processbased)leadto

differentconsequence.

Score:

1 2 3 4 5 -99

a) Let’ssayyou’veagreedtoafollow-upplanatoneofyourmeetings,whatwouldhappeniftheplanweren’tenacted?

b) Howlongisitbetweenwhenaproblemisidentifiedtowhenitissolved?Canyougivemearecentexample?

c) Howdoyoudealwithrepeatedfailuresinobstetriccare?

Score1:Failuretoachieveagreed

objectivesdoesnotcarryany

consequences

Score3:Failuretoachieveagreedresultsis

toleratedforaperiodbeforeactionistaken

Score5:Afailuretoachieveagreed

targetsdrivesretraininginidentified

areasofweaknessormovingindividuals

towheretheirskillsareappropriate

11) TargetBalance:

Testswhethertargetscovera

sufficientlybroadsetofmetrics

Score:

1 2 3 4 5 -99

a) Whattypesoftargetsaresetforthematernityunit?

b) Arethereanygoalsthatarenotsetexternally(e.g.bythegovernment,regulators)?

Score1:Goalsfocusedonlyon

governmenttargetsandachievingthe

budget

Score3:Goalsarebalancedsetoftargets

(includingquality,waitingtime,operational

efficiency,andfinancialbalance);goalsform

partoftheappraisalforseniorstaffonlyor

donotextendtoallstaffgroups;realinter

dependencyisnotwellunderstood

Score5:Goalsareabalancedsetof

targetscoveringallfourdimensions(see

Score3);interplayofallfourdimensions

isunderstoodbyseniorandjuniorstaff

(cliniciansaswellasnursesand

managers)

12) TargetInter-Connection

Testswhethermaternityunits

targetsaretiedtooverall

hospitalobjectivesandcascade

downtodifferentstaffgroupsor

members.

Score:

1 2 3 4 5 -99

a) Whatisthemotivationbehindthesegoals?

b) Howarethesegoalscascadeddowntothedifferentstaffgroupsortoindividualstaffmembers?

c) Howarematernityunittargetslinkedtooverallhospitalperformanceanditsgoals?

Score1:Goalsdonotcascadedownthe

organisation

Score 3: Goals do cascade, but only to some

staffgroups(e.g.nursesonly)

Score 5: Goals increase in specificity as

they cascade, ultimately defining

individualexpectationsforallstaffgroups

Page225of248

13) TimeHorizonofTargets

Testswhetherhospitalhasa‘3

horizons’approachtoplanning

andtargets

Score:

1 2 3 4 5 -99

a) Whatkindoftimescaleareyoulookingatwithyourtargets?

b) Whichgoalsreceivethemostemphasis?

c) Arethelong-termandshort-termgoalssetindependently?

d) Couldyoumeetallyourshort-rungoalsbutmissyourlong-rungoals?

Score1:Thestaff’smainfocusison

achievingshort-termtargets

Score3:Thereareshortandlong-termgoals

foralllevelsoftheorganisation;goalsareset

independentlyandthereforearenot

necessarilylinkedtooneanother

Score5:Long-termgoalsaretranslated

intospecificshort-termtargetssothat

short-termtargetsbecomea‘staircase’to

reachlong-termgoals

14) TargetStretch:

Testswhethertargetsare

appropriatelydifficultto

achieve

Score:

1 2 3 4 5 -99

a) Howtoughareyourtargetsformaternitycare?Howpushedareyoubythetargets?

b) Onaverage,howoftenwouldyousaythatyoumeetyourtargets?Howareyourtargetsbenchmarked?

Score1:Goalsareeithertooeasyor

impossibletoachieve,atleastinpart

becausetheyaresetwithlittleclinician

involvement(e.g.simplyoffhistorical

performance)

Score3:Seniorstaffpushforaggressivegoals

basedonexternalbenchmarksbutwithlittle

buy-infromclinicalstaff.

Score5:Goalsaregenuinelydemanding

forallpartsoftheorganisationand

developedinconsultationwithsenior

staff(e.g.toadjustexternalbenchmarks

appropriately)

15) ClarityandComparability

ofTargets:

Testshoweasilyunderstandable

performancemeasuresareand

whetherperformanceisopenly

communicated

Score:

1 2 3 4 5 -99

a) IfIaskedsomeoneonyourstaffdirectlyaboutindividualtargets,whatwouldheorshetellme?

b) Doesanyonecomplainthatthetargetsaretoocomplex?

c) Howdopeopleknowhowtheirownperformancecomparestootherpeople’sperformance?Isthispublishedorpostedinany

way?

Score1:Performancemeasuresare

complexandnotclearlyunderstood,or

onlyrelatetogovernment/regulator

targets;individualperformanceisnot

madepublic

Score3:Performancemeasuresarewell

definedandcommunicated;performanceis

publicatalllevelsbutcomparisonsare

discouraged

Score5:Performancemeasures arewell

defined, strongly communicated and

reinforced at all reviews; performance

and rankings aremade public to induce

competition

Page226of248

16) RewardingHigh

Performers

Testswhethergoodperformance

isrewardedproportionately

Score:

1 2 3 4 5 -99

a) Howdoesyourstaffappraisalsystemwork?Canyoutellmeaboutyourmostrecentone?

b) Howdoesyourstaff’spayrelatetotheresultsofthisreview?Howdoesthebonussystemwork?

c) Aretherenon-financialrewardsforthebestperformersacrossallstaffgroups?

d) Howdoesyourrewardsystemcomparetothatatothercomparablehospitals?

Score1:Staffmembersarerewardedin

thesamewayirrespectiveoftheirlevel

ofperformance

Score3:Thereisanevaluationsystemforthe

awardingofperformancerelatedrewards

thatarenon-financialattheindividuallevel;

rewardsarealwaysorneverachieved

Score5:Thereisanevaluationsystem

whichrewardsindividualsbasedon

performance;thesystemincludesboth

personalfinancialandnon-financial

awards;rewardsareawardedasa

consequenceofwell-definedand

monitoredindividualachievements17) RemovingPoor

Performers

Testswhetherhospitalisableto

dealwithunderperformers

Score:

1 2 3 4 5 -99

a) Ifyouhadaclinicianoranursewhocouldnotdohis/herjob,whatwouldyoudo?Couldyougivemearecentexample?

b) Howlongisunder-performancetolerated?Howdifficultisittoterminateanurse/clinician?

Score1:Poorperformersarerarely

removedfromtheirpositionsScore3:Suspectedpoorperformersstayina

positionformorethanayearbeforeactionis

taken

Score5:Wemovepoorperformersoutof

theunitortolesscriticalrolesassoonas

aweaknessisidentified

18) PromotingHigh

Performers

Testswhetherpromotionis

performancebased

Score:

1 2 3 4 5 -99

a) Canyoutellmeaboutyourcareerprogression/promotionsystemwithinthehospital?

b) Howdoyouidentifyanddevelopyourstarperformers?Whattypesofprofessionaldevelopment/trainingopportunities

areprovided?

c) Howdoyoumakedecisionsregardingprogression/promotionswithintheunit/hospital?

d) Arebetterperformerslikelytobepromotedfasterorarepromotionsgivenonthebasisoftenure/seniority?Score1:Peoplearepromotedprimarily

onthebasisoftenure(yearsofservice)Score3:Peoplearepromoteduponthebasis

ofperformance

Score5:Weactivelyidentify,developand

promoteourtopperformers

Page227of248

19) ManagingHR/Talent

Testswhatemphasisisputon

talent/Humanresource

management

Score:

1 2 3 4 5 -99

a) Doyouhaveauthoritytohireordismissadditionalhealthworkers?

b) Howdoyouensureyouhaveenoughstaff/nursesoftherighttypeinthehospital?

c) Howdoseniormanagersshowthatattractingtalentedindividualsanddevelopingtheirskillsisatoppriority?

d) Doseniorstaffmembersgetanyrewardsforbringinginandkeepingtalentedpeopleinthehospital?Score1:Seniorstaffdonot

communicatethatattracting,retaining

anddevelopingtalentthroughoutthe

organisationisatoppriority

Score3:Seniorstaffbelieveand

communicatethathavingtoptalent

throughouttheorganisationiskeytogood

performance

Score5:Seniorstaffareevaluatedand

heldaccountableonthestrengthofthe

talentpooltheyactivelybuild

20) RetainingTalent:

Testswhetherhospitalwillgo

outofitswaytokeepitstop

talent

Score:

1 2 3 4 5 -99

a) Ifyouhadatopperformingmanager,nurseorclinicianthatwantedtoleave,whatwouldthehospitaldo?

b) Couldyougivemeanexampleofastarperformerbeingpersuadedtostayafterwantingtoleave?

c) Couldyougivemeanexampleofastarperformerwholeftthehospitalwithoutanyonetryingtokeepthem?Score1:Wedolittletotryandkeepourtoptalent

Score3:Weusuallyworkhardtokeepour

toptalent

Score5:Wedowhateverittakesto

retainourtoptalentacrossallstaff

groups

21) AttractingTalent

Teststhestrengthofthe

employeevalueproposition

Score:

1 2 3 4 5 -99

a) Whatmakesitattractivetoworkatthishospital,asopposedtoothersimilarhospitals?

b) IfIwasatopnurse/clinicianandyouwantedtopersuademetoworkatyourhospital,howwouldyoudothis?

c) Whatdoyouthinkpeoplemaynotlikeaboutworkingatyourhospital?Score1:Competinghospitalsoffer

strongerreasonsfortalentedpeopleto

jointheirorganizations

Score3:Ourvaluepropositioniscomparable

tothoseofferedbyotherhospitals

Score5:Weprovideauniquevalue

propositiontoencouragetalented

individualstojoinourhospitalcompared

toourcompetition

a) Canyoutellmeabouttherolethatclinicians(e.g.doctors/consultants)haveinimprovingperformanceandachieving

targets?

b) Howareindividualcliniciansresponsiblefordeliveryoftargets?Doesthisapplytocosttargetsaswellasquality

targets?.c) Howdoclinicianstakeonrolestodelivercostimprovements?Aretheyselectedforthisroleordotheyvolunteer?Can

youthinkofexamples?

Page228of248

22) ClearlyDefined

Accountabilityfor

Clinicians

Testswhetherthereisformal

leadershiprolesand

accountabilityamongclinicians

fordeliveryofhospitaltargets

andobjectives

Score:

1 2 3 4 5 -99

Score1:Formalaccountabilityfor

clinicalperformance(quality)

only

Score3:Thereissomeaccountabilityfordelivery

beyondclinicalqualitybutthismightbediffusedwithin

ateamornotcarrysignificantconsequences;clinical

performancestillconsideredtobethemainpartofthe

job

Score5:Formalaccountabilityacross

qualityserviceandcostdimensions

witheffectiveperformance

managementandconsequencesfor

good/poorperformance

Post-Interview

23) Interviewduration(minutes)_________________

24) Intervieweeknowledgeof

managementpractices

Score:1 2 3 4 5

Score1:Someknowledge

aboutmanagementof

maternityfacilities.

Score3:Expertknowledgemanagementofmaternity

unit

Score5:Expertknowledgeabouthis

specialtyandalsotherestofthe

hospital.

25) Intervieweewillingnessto

revealinformation

Score:1 2 3 4 5

Score1:Veryreluctantto

providemorethanbasic

information

Score3:Providesallbasicinformationandsomemore

confidentialinformation

Score5:Totallywillingtoprovide

anyinformationaboutthehospital!

26) Intervieweepatience

Score:1 2 3 4 5

Score1:Littlepatience-wants

toruntheinterviewasquickly

aspossible.Ifeltheavytime

pressure

Score3:Somepatience-willingtoproviderichnessto

answersbutalsotimeconstrained.Ifeltmoderatetime

pressure

Score5:Lotofpatience-willingto

talkforaslongasrequired.Ifeltno

timepressure27) Didthemanagermention

thatthehospitalwasa

teachinghospital?

Yes No

28) Ageofinterviewee

(don'task)-guessifnot

told

29) Numberoftimesrescheduled(0=never

rescheduled)

30) Genderofinterviewee:

Male Female

31) Seniorityofinterviewee:

1. CEO

2. Multi-specialtymanager

3. SpecialtyManager

32) Didtheintervieweehaveadegree-guessifnot

told

33) Interviewlanguage

Hindi

English

Page229of248

4. Withinspecialtymanagement

5. Technicianwithoutmanagementrole(e.g.nurseorjunior

doctor)

Page230of248

Appendix3:Informationsheetsandconsentforms

3.1:InformationSheetforhealthworkersforclinicalpracticeobservation

DearMadam/Sir,

Weare conducting a study tounderstand thequality ofmaternal andnewbornhealth servicesprovided atseveralhealthfacilitiesinUP,Indiaincludingthishealthfacility.Thissheetprovidesyouwithinformationaboutthisresearch.ThisstudyhasbeenapprovedbythePublicHealthcareSociety(PHS)EthicsReviewBoardinIndiaandtheLondonSchoolofHygiene&TropicalMedicineintheUK(LSHTMEthicsRef:8858).ThestudyprotocolalsoreceivedclearancefromtheNationalHealthMissioninUttarPradesh.Wehavealsoobtainedpermissionfromthefacilityincharge/hospitaldirectortoobservethecareprovidedatthisfacilitytoday.

Whyisthisimportant?

Asyouknow,manywomenandbabiesdieduetocomplicationsduringpregnancyandchildbirthinUP.Hence,weareconductingthisresearchtounderstandmoreaboutthequalityofservicesofferedathealthfacilitiessothatwecanimprovethequalityofobstetricandneonatalcareservices.

Whoiscarryingoutthestudy?

ThisstudyisfundedbyMerckforMothers.ItisbeingrunbyasmallteamofresearchersfromSambodhiandtheLondonSchoolofHygieneandTropicalMedicine(LSHTM).

Whatisinvolved?

Aclinicallyqualifiedresearcherwillobservethequalityofservicesofferedtowomenandneonatesduringlabourandchildbirthandtheimmediatepostnatalperiod.Theobserverisnottheretosupportyouorinterferewithanyaspectsofclinicalcareprovision.

Isthisresearchconfidential?

Yes.Anyinformationobtainedfromthisresearchisconfidentialandwillonlybeseenbythemembersoftheresearch team. All information will be stored securely. This means any findings obtained from the clinicalobservationswillnotbelinkedtoanyindividualhealthworkerorfacility.

Whatarethebenefitsoftakingpartinthisresearch?

There arenodirect benefits to you for participating in this research.However,wewill use the informationobtainedfromthissurveytoimprovethehealthcareservicesatselectedhealthfacilitiesinUP,India.

Whataretherisksintakingpart?

Therearenorisksbecauseoftakingpartinthisresearch.Yourpersonalidentitywillbeprotectedatalltimesandthiswillhavenoimpactonyourworkatthishealthfacility.

Page231of248

DoIhavetotakepartinthisresearch?

No.Ifyoudecidenottoparticipateinthestudy,itwillnothaveanyeffectonanyoftheservicesthatyoureceive.

Howwilltheresearchfindingsbeused?

ThefindingsoftheresearchwillbeusedtodevelopareportwhichwillhighlighttheexistingqualityofmaternalandneonatalhealthservicesprovidedatselectedhealthfacilitiesinthreedistrictsinUttarPradesh,India.

Thankyoufortakingthetimetoreadthisinformation.Wereallyappreciateyourparticipationinthisresearch.

Theresearchwillonlyproceedonceyouhaveaskedanyotherquestionsthatyoumayhaveandhavesignedtherelevantconsentforms.Youcankeepthisinformationsheetwithyou.

Ifyouhaveanyquestionsoropinionsaboutthisstudy,pleasecontact:

LSHTM:GAURAVSHARMA,+ +918601882687;Gaurav.Sharma@lshtm.ac.uk

Page232of248

3.2:Consentformforhealthworkeronclinicalpracticeobservation

Instructions:Whenhealthworkerarrivestoconductadelivery,pleaseprovidehimorherwiththeinformationsheet.Itisessentialthatyouobtainaninformedconsentfromthehealthworkerbeforebeginningtheobservation.Pleaseaskthehealthworkertosignanddatetheconsentformoncetheyhavereadtheinformationbelowandagreetoparticipateinthestudy.

I_________________________________________agreetotakepart inthisstudyonthequalityofmaternalandnewborncareservicesprovidedatthishealthfacility.

Iunderstandthat:

• Iamagreeingtoallowaclinicallyqualifiedresearchertoobserveaspectsofclinicalcareprovision.• All the findings from this research are confidential andwill not be linked tomy name or any personal

information.• Myparticipationiscompletelyvoluntaryandrefusaltoparticipatewillnothaveanyimplicationsonmeor

myworkatthishealthfacility.• Ihavebeenprovidedwiththenecessaryinformationaboutthisresearchandhavealsohadanopportunity

toclarifyallmyquestions.

Myquestionshavebeenansweredby___________________________________

Signatureofthehealthworker_______________________

Date:_______________________________

Page233of248

3.3:WomanconsentformforclinicalpracticeobservationfpfdRlh; O;ogkj vkCtosZ'ku ds fy, lsokxzkgh ¼Dyk,aV½ dh lgefr

INSTRUCTIONSTOOBSERVER

vkCtoZj ds fy, funsZ'k

Whenapregnantwomanarrivesattheemergencyroomorwaitingroomofthelabouranddeliveryward,pleaseprovidehertheinformationsheetbeforeenrollingherinthestudy.Itisessentialthatyouobtainaninformedconsentfromtheclientbeforebeginningtheobservation.Iftheclientcannotreadathumbprintshouldbeobtained.Consentforclientcannotbegivenbyhealthworkerorfacilityin-charge.

tc dksbZ xHkZorh efgyk vkikRdkyhu d{k esa ;k izlo vkSj izlwfr okMZ ds izrh{kk d{k esa vk;s rks ;bl v/;;u esa ukekafdr djus ls igys mls ;g tkudkjh 'khV nsaA ;g vko';d gS fd vkCtosZ'ku ¼i;Zos{k.k½ djus ls igys vki lsokxzkgh ¼Dyk,aV½ dh tkudkjh;qDr lgefr izkIr dj ysaA vxj lsokxzkgh i<+ fy[k ugha ldrh rks mlds vaxwBs dk fu'kku fy;k tkuk pkfg,A Dyk,aV ds fy, LokLF; dk;ZdrkZ ;k LokLF; lqfo/kk dk izHkkjh lgefr ugha ns ldrkA

Iunderstandthat:

eSa le>rh gwa fd%

IamagreeingtoallowaclinicallyqualifiedresearchertoobservethequalityofservicesthatIreceiveatthishealthfacilitytoday.

eSa ,d fpfdRlh; :i ls ;ksX; 'kks/kdrkZ dks vkt bl lqfo/kk esa eq>s izkIr gksus okyh lsokvksa dh xq.koÙkk dks vkCtoZ djus ;k ns[kus dh vuqefr ns jgh gwaA

Allthefindingsfromthisresearchareconfidentialandwillnotbelinkedtomynameoranypersonalinformation.

bl 'kks/k ds lHkh fu"d"kZ xksiuh; gSa vkSj mUgsa esjs uke ;k fdlh O;fDrxr tkudkjh ls ugha tksM+k tk;sxkA

MyparticipationiscompletelyvoluntaryandwillnothaveanyimplicationsontheservicesthatIreceivetoday.

esjh Hkkxhnkjh iwjh rjg ls LoSfPNd gS vkSj eSa tks lsok,a izkIr dj jgh gwa mu ij bldk dksbZ izHkko ugha iM+sxkA

Ihavebeenprovidedwiththenecessaryinformationaboutthisresearchandhavealsohadanopportunitytoclarifyallmyquestions.

eq>s bl 'kks/k ds ckjs esa vko';d tkudkjh ns nh xbZ gS vkSj eq>s iz'u iwNus dk volj Hkh feyk gSA

Page234of248

Myquestionshavebeenansweredby___________________________________esjs iz'uksa ds mÙkj

IattestthatIreadtheconsentformtotheparticipantandshehasagreedtoparticipate.eSa ;g izekf.kr djrk gwa fd eSaus lgHkkxh dks lgefr i= i<+dj lquk;k gS vkSj og Hkkx ysus ds fy, lger gSA

Thumbprint____________________________vaxwBs dk fu'kku

Researcher’ssignature:_______________________'kks/kdrkZ ds gLrk{kj%

Signed_______________________gLrk{kj

Date:_______________________________frfFk%

Date:_______________________________frfFk

Page235of248

Appendix4:Ethicalapprovallettersandpermissions

 

                                             

Observational / Interventions Research Ethics Committee

    LSHTM

18 May 2015 

Dear 

Study Title: Quality of Care for normal labour and childbirth at maternity facilities in Uttar Pradesh, India: A Cross‑Sectional Study  

LSHTM Ethics Ref: 8858 

Thank you for responding to the Observational Committee’s request for further information on the above research and submitting revised documentation.

The further information has been considered on behalf of the Committee by the Chair. 

Confirmation of ethical opinion

On behalf of the Committee, I am pleased to confirm a favourable ethical opinion for the above research on the basis described in the application form, protocol and supporting documentationas revised, subject to the conditions specified below.

Conditions of the favourable opinion

Approval is dependent on local ethical approval having been received, where relevant. 

Approved documents

The final list of documents reviewed and approved by the Committee is as follows:

Document Type File Name Date Version

Covering Letter Covering letter after resubmission 14.5.15 14/05/2015 2

Protocol / Proposal Consent and info sheet combined 5.5.15 14/05/2015 2

Information Sheet Consent and info sheet combined 5.5.15 14/05/2015 2   

After ethical review

The Chief Investigator (CI) or delegate is responsible for informing the ethics committee of any subsequent changes to the application.  These must be submitted to the Committee for reviewusing an Amendment form.  Amendments must not be initiated before receipt of written favourable opinion from the committee.  

The CI or delegate is also required to notify the ethics committee of any protocol violations and/or Suspected Unexpected Serious Adverse Reactions (SUSARs) which occur during the projectby submitting a Serious Adverse Event form. 

At the end of the study, the CI or delegate must notify the committee using an End of Study form. 

All aforementioned forms are available on the ethics online applications website and can only be submitted to the committee via the website at: http://leo.lshtm.ac.uk

Additional information is available at: www.lshtm.ac.uk/ethics

Yours sincerely,

Professor John DH PorterChair

ethics@lshtm.ac.ukhttp://www.lshtm.ac.uk/ethics/ 

Page 1 of 2

Page236of248

Page237of248

Scanned by CamScanner

Page238of248

Page239of248

Page240of248

Appendix5:PublishedmanuscriptforChapter6

Page241of248

Page242of248

Page243of248

Page244of248

Page245of248

Page246of248

Page247of248

Page248of248

Appendix6:Tableshowingfrequencyofmistreatmentbysector

Itemsofmistreatment Publicsector Privatesector N(%) N(%)1. Anyitemofmistreatment 211(100.0) 64(100.0)2. Twoitemsofmistreatment 10(4.7) 3(4.7)3. Threeitemsofmistreatment 41(19.4) 10(15.6)4. Fouritemsofmistreatment 44(20.9) 17(26.6)5. Fiveitemsofmistreatment 43(20.4) 19(29.7)6. Sixitemsofmistreatment 32(15.2) 9(14.1)7. Sevenitemsofmistreatment 21(10.0) 4(6.3)8. Eightitemsofmistreatment 14(6.6) 0(0.0)9. Nineitemsofmistreatment 3(1.4) 2(3.1)10. Tenitemsofmistreatment 3(1.4) 0(0.0)