REpORTON THE USE OF THE COMMUNITY DIAGNOSIS TO …

P Iu ~C~ ~t "20

! '"l ( "'I .. ~

FINAL

REpORT ON THE USE OF THE COMMUNITY DIAGNOSIS TO EXPLORE

SAFE MOTHERHOOD: A TWO-COUNTRY COMPARISON AND

METHODOLOGICAL CRITIQUE

TECHNICAL PAPER #6

Prepared by Nancy Nachbar for MotherCare IISeptember 15, 1997

libt,' itl

a::r::Ill1J'lilli'

U S Agency forInternational DevelopmentOffice of Health

, MotherCare-...-_'-li_:-:-J: John Snow, Inc

TABLE OF CONTENTS

ExecutIve Summary 1

IntroductIOn 5Background 5Report ObjectIves 5

SectIOn I-Report Methodology 7Report OvervIew 7CommunIty DIagnosIs Methodology 7Report Methodology 9

SectIOn II-Fmdmgs 11IntroductIOn 11Step 1 Problem RecogmtIOn Awareness and PerceptIOns of and Expenence WIth

Problems Durmg Pregnancy DelIvery, the Post-Partum Penod and WIth theNewborn 11Step 1 General Comments 11Step 1 PerceptIOns of Pregnancy and Pregnancy-related PractIces 12

Step 1 Awareness of and perceptIOns of Swellmg 14Step 1 Awareness of and PerceptIOns of AnemIa 14Step 1 Awareness of and PerceptIOns of Certam DelIvery Problems 15Step 1 Problem RecogmtIOn/PerceptIOn of Seventy (Post-partum) 17Step 1 Problem RecogmtIOn/PerceptIOn of Seventy (Neonatal) 19Step I PerceptIOns of Newborn deaths 21Step 1 Expenence of ComplIcatIOns 22

Step 2 DeclSlon-Makmg Around the Use of Health ServIces When ComplIcatIOns AnseDurmg Pregnancy, DelIvery, the Post-Partum Penod, or WIth the Neonate 23Step 2 DeclSlon-Makmg Who are the CntIcal Players? 23Step 2 DeclSlon-Makmg Whose Help IS Sought? 24Step 2 DeclSlon to Seek Care Who AccompanIes the Woman? 27

Step 3 Access to Care 28Step 3 Access to Care PerceptIOn of Care 28Step 3 Access to Care PotentIal BarrIers DIstance 30Step 31Step 3 Access to Care Help from the Extended FamIly 33Step 3 Access to Care Help from the Commumty 33

SectIOn IV-Strengths and Weaknesses of the SemI-structured IntervIews and Focus GroupsConducted m IndonesIa and BolIvIa 35

F \nnachbar\comdx\report\\finaI4 wpd

SEPTEMBER 15, 1997

SemI-Structured IntervIews m IndonesIa and m BolIvIaStrengthLImItatIOn Issues wIth Method ChOIceLImItatIOn Issues wIth SamplIngLImItatIOn Issues wIth Instrument DesIgnLImItatIOn Data Codmg and AnalysIs

Focus Groups m IndonesIa and BolIvIaGeneral CommentsLImItatIOn Response EffectsLImItatIOn Deference Effects

SectIOn -ConclusIOns and RecommendatIOnsFmdmgsStudy/Reportmg StrengthsStudy/ReportmgRecommendatIOns and ConclUSIOns Research Approach

MethodsInstrument DeSIgnContentData Management and AnalysIsTrammg, Momtonng, and SupervIsIOnReportmg

35353536373838383939

4040424243434344454545

AppendIX A FIshbem's Pathway to BehavIOr-A Framework for PossIble Use m FutureCommunIty DIagnosIs Research 47

AppendIx B IdentIficatIOn of Gaps/Areas of ConfuSIOn 48

AppendIx C RecogmtIOn of the Problem In-depth IntervIews WIth Mothers and SummaryofIndonesIalBohvIa Reports on Focus Groups 52

References

F \nnachbar\comdx\report\\fina14 wpd

SEPTHIBER 15 1997

89

EXECUTIVE SUMMARY

The development of effectIve mterventIOn strategIes desIgned to reduce unacceptably hIgh levelsof maternal and neonatal mortalIty and morbIdIty reqUIres an understandmg of the factorsmfluencmg the survIval of a woman or her newborn when a potentIally lIfe-threatemngcomplIcatIOn occurs To gather mdividual- and communIty-level mformatIOn on these factors,MotherCare conducted separate "communIty dIagnoses" m BolIVIa and IndoneSIa (SouthKalImantan)-two countnes where MotherCare has long-term projects The commumtydIagnOSIS research, whIch was qualItatIve m nature, used the Pathway to SurvIval as ItStheoretIcal base The Pathway to SurvIval IS a framework that delIneates the steps that maydetermme the survIval of a woman or her newborn when they expenence an obstetnc or neonatalcomplIcatIOn that can result m death (see figure 1) ThIs framework has four componentsproblem recogmtIOn, declSlon-makmg regardmg care, access to care, and qualIty of care Thefocus of the communIty dIagnOSIS research was on the first three elements

Figure 1Source Koblmsky 1995

Quality Care

!Survlva/

Life Threatening Illness

~Recognition of Problem

~Declslon-Makmg to Seek Care

~LOgiStiCS to Reach Quality Care

~

(Step,1

I Step)2

Three pnncipal mvestigatIve technIques were employed m the BolIVIan and IndoneSIancommumty dIagnosIs research (1) what were termed "semI-structured mtervlews (SSIs), (2)focus groups (FGs), and (3) dIrect observatIOns (DOs) In both countnes, the SSIs were held

WIth women, husbands, and healthPathway to Survival Maternal/Perinatal Mortality proVIders In BolIVIa, SSIs also

were conducted WIth responsablespopulares de salud (communItyhealth workers) and communItyleaders In BolIVIa and IndoneSIa,FGs were held WIth women andhusbands In IndoneSIa, male andfemale "key mformants"a alsopartICIpated m the focus groupsThe dIrect observatIOns wereconducted on health proVIders andhealth faCIlItIes DescnptIOns ofthe commumty mfrastructure alsowere gIVen

The report whIch follows contams a dISCUSSIOn of some key findmgs from the commumtydIagnOSIS research, comparmg IndoneSIa and BolIVIa The strengths and lImItatIOns of thecommumty dIagnOSIS approach as It was Implemented In the two countnes are outlmed andrecommendatIOns for the deSIgn of future commumty dIagnOSIS research are gIven

aThe charactenstlcs of the "key mformants" are not available


SEPTEMBER 15, 19971

For thIS report, some data (those Items whIch were closed-ended) from the SSIs m BolIvia andIndonesia were analyzed Focus group data and open-ended questIOns on the SSIs could not beexplored because raw data eIther were unavaIlable or were not translated mto EnglIshInformatIOn from open-ended questIOns m the SSIs and from the focus groups was obtamedthrough synthesIs of materIal contamed m the communIty diagnosIs report summarIes fromIndonesIa (Report Commumty Dzagnosls MotherCare Safe Motherhood Project SouthKahmantan, 1996 by Marsaban et al , 1996) and BolIVia [Dzagnostlco Barreras yVzablhzadores en fa AtenclOn de ComphcaclOnes Obstetrzcas y Neonatafes (EstudlO Cuahtatlvoen Comumdades y ServlclOs de Safud de Cmco Dlstntos de Safud en La Paz y Cochabamba bySeoane et al ,1996] Data from the dIrect observatIOns were not exammed

Results from the SSIs and focus groups reveal valuable mformatIOn m some mstances and areambIguous m others The ambIgUIty often IS due to Issues With samplmg, With questIOnnaIreconstructIOn, and to mcomplete mformatIOn gIven m the summary reports DespIte theambIgUIty of some of the data, the communIty diagnOSIS research yIelded useful mformatIOn forshapmg mterventIOn messages and provIder trammg and communIty mformatIOn, educatIOn, andcommunIcatIOn and counselmg (lEC/C) strategIes Results from the SSIs and FGs are bneflyoutlIned below

FIrst, m BolIVIa and IndoneSia, the degree to whIch women recogmze obstetnc and neonatalcomplIcatIOns vanes, dependmg on when the complIcatIOn occurs There IS lIttle recogmtIOn ofneonatal problems m both countrIes, but espeCially m BolIVIa Awareness of anemia or ItS SIgnsand symptoms IS high m IndoneSia and appears low m BolIVia IndoneSIan women seem toconSIder speCIfic problems durmg labor (e g ,prolonged labor, premature delIvery, andhemorrhage durmg delIvery) to be more senous than theu BolIVian counterparts, who generallyare more concerned With problems occurrmg after delIvery (e g ,sobreparto, translated as'relapse after bIrth") However, BolIVIan women often mentIOned mafparto ("bad bIrth") as aproblem dunng delIvery In general, both IndoneSian and BolIVIan women and theu famIlIesVIew hemorrhage to be serIOUS, though they have trouble dIstmgUIshmg between dangerous andless severe bleedmg

Second, m BolIVIa and IndoneSia, husbands play an Important role m the deCISIOn to seek careIn BohvIa, other famIly members such as mothers-m-Iaw are cntIcal players, partIcularly forproblems m pregnancy and post-partum Also m BohvIa, the women themselves often are theprmcipal declSlon-makers WhIle Bohvian study partICIpants were stratIfied accordmg towhether they came from hIgh or low servIce-use areas, women reported usmg both the tradItIonaland mstitutIOnal health systems m about equal numbers, regardless oftheir claSSIficatIOn byservIce use

Thud, though IndoneSIan women participatmg m the SSIs reported havmg confidence m theskIlls of the mIdWIves, lIttle other mformatIOn (e g , focus group data) IS aVailable regardmg


SEPTEMBER 15,19972

perceptIOns of care m IndonesIa Data from the SSIs m BolIvIa mdlcate that women and theIrfamIlIes generally are satIsfied WIth the health care they receIve, but data from the focus groupsgIve a dIfferent and probably more accurate pIcture of perceptIOns of qualIty of care andtreatment at health faCIlItIes m BolIvIa In the focus groups, treatment often was descnbed as"cold" and both women and theIr husbands expressed greater comfort m theIr mteractIOns WIthtradItIOnal bIrth attendants than WIth mstitutIOnal health provIders

IndonesIan SSI partICIpants generally reported knowmg where health faCIlItIes were and mostcould reach some faCIlIty withm about 15 mmutes InformatIOn on other pOSSIble barrIers toaccessmg care m IndonesIa IS scant SpeCIfically, data on cost of transportatIon and servIces andon aSSIstance from famIly members and the communIty are msufficlent In focus groups mBolIVia, cost emerged as a sIgmficant barrIer to "regular" care among partICIpants from hIghservIce-use areas and to both regular and emergency care among partICIpants from low servlceuse areas

The commumty dIagnosIs approach m BolIVIa and IndonesIa had several strengths Key amongthese was the fact that the research was gUIded by a conceptual framework (the Pathway toSurvIval) Importantly, thIS framework proVIdes a base for Identlfymg key mdividual- andcommumty-Ievel factors WIth the potentIal to mfluence the survIval of a women or her newbornIf potentially grave obstetrIC or neonatal complIcatIOns arIse The use of thIS framework ensuredthe collectIOn of cntIcal mformatIOn needed for the development of effectIve mterventIOnactIVItIes

The commumty dIagnOSIS research m both countnes also had some dIfficultIes SpeCIficallythere were problems WIth the deSIgn and wordmg ofthe data collectIOn mstruments, WIth datamanagement (mcludmg codmg), WIth gaps m the analySIS, and WIth reportmg At tImes, theproblems were severe enough to preclude meanmgful analysIs or to prohIbIt defimtiveconclusIOns from bemg drawn

SpeCIfic recommendatIOns for future commumty dIagnOSIS efforts are gIven m the body of thereport These mclude (1) use of multIple data collectIOn technIques, mcludmg but not lImIted tofocus groups and true m-depth mtervlews, (2) delIneatIOn of explICIt partICIpant selectIOn cntena,(3) development of clearly constructed questIOns that accurately reflect the concepts they aremtended to represent, (4) use ofmultIple questIOns to assess domams (e g, partICIpantsatIsfactIOn WIth care), (5) use of an open-ended questIOn format and extensIve probmgtechnlquesb

, (6) care m data codmg, partIcularly WIth regard to mergmg response categones(7) use of a framework to gUIde mstrument deSIgn (the current commumty dIagnOSIS researchused the Pathway to SurvIval future efforts may conSIder the use ofFIshbem's Pathway toBehaVIOr-see figure 2 In AppendiX C-though thiS framework has not been apphed or tested

bOpen-ended questIOns are a nch source of mfonnatlOn though they can present a challenge for analySIS


SEPTEMBER 15 19973

prevIously), (8) collectIOn ofmformatIOn on problem recogmtIOn, declSlon-makmg regardmgcare, perceptIOns of the qualIty of care, mtentIOn to use servIces, skIlls for performmg desIredbehavIOrs, and envIronmental constramts mhIbItmg the conduct of desIred behavIOrs, (9)ImplementatIOn of adequate trammg, momtonng, and supervlSlon for data collectIOn, and (10)detaIled descnptIOns of study methodology and results, mcludmg presentatIon of raw data whereappropnate


SEPTEMBER 15 19974

INTRODUCTION

BACKGROUND

PURPOSE OF THE MOTHERCARE COMMUNITY DIAGNOSIS

FormatIve research IS an Important component m the desIgn and ImplementatIOn of effectIvemterventIOns Recogmzmg thIS fact, m the wmter/sprmg of 1995-1996, MotherCare II undertook"communIty diagnoses" m Bohvla and m South KalImantan, Indonesia The purpose of thecommumty dIagnOSIS research was to (1) elucIdate IndonesIan and Bohvlan women'sreproductIve health behefs, values, and behavIOrs, (2) learn about the process by whIch famlhesdeCIde to seek care for pregnancy-related comphcatIOns, and (3) determme the bamers mhibItmgwomen's use of preventIve and emergency health servIces (Pareja and Galloway Tnp Report,May 27-June 19, 1996 and Pareja TrIp Report, Jan 24 - Feb 12, 1996)

A summary ()f the findmgs and a more detailed descnptIOn of the communIty diagnOSIS processm Bohvla appear m Dzagnostlco Barreras y Vzabllzzadores en la AtenclOn de ComplzcaclOnesObstetncas y Neonatales (EstudlO Cualztatlvo en ComunIdades y ServlclOs de Salud de CInCODlstntos de Salud en La Paz y Cochabamba) (Seoane et al ,1996) The document, ReportCommunIty Dzagnosls MotherCare Safe Motherhood Project South Kalzmantan 1996(Marsaban et al , 1996) provIdes a summary of the commumty dIagnOSIS research m SouthKahmantan

REPORT OBJECTIVES

The aims of thIS report are threefold The first ObjectIve IS to compare select results fromIndoneSIa and BolIVia m order to determme commonalItIes and dIfferences between the twocountnes m (1) the recogmtIOn of obstetnc and neonatal problems, (2) declSlon-makmgregardmg health-seekmg behaVIOr, (3) perceptIOns of the quahty of care, and (4) perceIvedbarners to accessmg care The second aim IS to reVIew the strengths and lImItatIOns of thecommunIty dIagnOSIS approach as It was employed m BohvIa and IndoneSia The final ObjectIveof thIS report IS to make recommendatIOns for the deSIgn of future MotherCare commumtydIagnOSIS tools and research methodologIes related to women's reproductIve health and to thehealth of the newborn

ThIS document IS dIVIded mto four sectIOns Section I contams a descnptIOn of the methodsused to gather and syntheSIze mformatIOn for thIS report Section II mcludes a dISCUSSIOn ofpertment findmgs from the two communIty diagnoses, comparmg BolIVia WIth IndoneSiaWhere relevant, ImphcatIOns for the deSIgn of future commumty diagnOSIS research (but not forprogram ImplementatIOn) also are dIscussed SectIOn III conSIsts of an overvIew of some of thestrengths and hmitatIOns of the semI-structured mterviews and focus groups, both as they wereconducted m the Bohvla and IndoneSIa more generally ThIS overvIew IS followed by SectIOnIV the conclUSIOns and recommendatIOns AppendIX A Illustrates a pOSSIble framework(Flshbem s) for gUIdmg future commumty diagnOSIS research AppendIX B outlInes some gapsand areas of confUSIOn regardmg the mformatIOn obtamed m the commumty diagnOSIS


SEPTEMBER 15 19975

Appendix C IS a matnx provIdmg a summary of the results from the SSIs and FGs m BolIViaand m IndonesIa It contams more m-depth mformatIOn on the strengths and lImItatIOns of thedata than IS provIded m SectIOn III

Throughout thIs document, text boxes are used to Illummate the mam pomts m the adjacent textA broad overvIew of key consIderatIOns can be obtamed by perusmg the executIve summary andthe text boxes Readers wantmg greater detaIl about a particular tOPIC area (e g , women'srecogmtIOn of newborn complIcatIOns, strengths and lImItatIOns of the data on newborncomplIcatIOns), than can be found m the body of the report should see AppendIx C


SEPTEMBER 15, 19976

SECTION I-REPORTMETHODOLOGY

REpORT OVERVIEW

The mformatIOn contamed m thIS report was obtamed by analyzmg select data from the twocommunIty diagnoses and by syntheslzmg the findmgs presented m Dzagnostlco Barreras yVzabllzzadores en la AtenclOn de ComplzcaclOnes Obstetrlcas y Neonatales (EstudlO Cualztatlvoen Comumdades y ServlclOs de Salud de CinCO DlStrltoS de Salud en La Paz y Cochabamba)(Seoane et al , 1996) and m Report Commumty Dlagnosls MotherCare Safe MotherhoodProJect, South Kalzmantan, 1996 (Marsaban et al ,1996) To place thIS report m context, It ISnecessary first to gIve a bnef overvIew of the communIty dIagnosIs methodology as It wasapplIed m BolIvia and Indonesia C

COMMUNITY DIAGNOSIS METHODOLOGY

GENERAL

In both BolIvIa and IndonesIa, MotherCare staff and theIr sub-contractors held what were termed"semi-structured mtervlews" (SSIs) and focus groups (FGs) They also conducted directobservatIOns (DOs) of health care provIders and health care faCIlItIes InformatIOn from thesethree data collectIOn methods was supplemented by descnptIOns of the commumty mfrastructureand the enVIronment (e g , roads electncIty, health faCIlItIes, geography, clImate, and sanItatIOn)

The Pathway to Survival (see figure 1) was used as a framework to gUIde the development ofthe data collectIOn mstruments Other frameworks, such as the newly developed Pathway toBehaVIOr (see AppendIX A) also could be used The Pathway to SurvIval delIneates the steps thatmay determme the survIval of a woman or her newborn when they expenence a potentIally lIfethreatenmg obstetnc or neonatal comphcatIOn As descnbed by Koblmskyl, the Pathway toSurvIval has four components (1) problem recogmtIOn, (2) declSlon-makmg regardmg care, (3)access to care, (4) the qualIty of the care obtamed The SSIs and FGs addressed the first threesteps of the Pathway to SurvIval (1 e , those steps mfluencmg whether or not women or theIrnewborns Will present at a health faclhty when hfe-threatenmg complIcatIOns anse) The dIrectobservatIOns proVIded mformatIOn on the qualIty of care provIded at health faCIlItIes

BOLlUA

In BolIVia, data were collected m five dlstncts over approxImately four months, from January1996 through Apnl 1996 A total of 41 semI-structured mtervlews were conducted WIth women,30 WIth men, 10 wlthparteras (mIdWIves), 10 WIth responsables populares de salud (commumtyhealth workers) and 10 WIth commumty leaders

CSee DlQgnostlco Barreras y VlQblh::adores en fa AtenclOn de ComphcaclOnes Obstetncas y Neonatafes(EstudlO Cuahtatlvo en Comunidades y Servlclos de Salud de Cmco Distntos de Salud en La Pa:: y Cochabamba)(Seoane et al 1996) and Report Commumtv DIagnOSIs MotherCare Safe Motherhood Project South Kahmantan1996 (Marsaban et al 1996) for more detaIled descnptIOns of study methods


SEPTEMBER 15, 19977

To obtam data on problem recogmtion women and men participatmg m semI-structuredmtervlews were asked questIOns on theIr awareness of problems dunng pregnancy bIrth thepost-partum penod, and the newborn, on the degree to whIch they consIdered specIfic problemsto be senous, and on the kmds of problems they had expenenced

To get mformatIOn on the declSlon to seek care, mterviewees were asked about normativebehavIOr m the commumty SpecIfically, women were quened about theIr perceptIons of thecare they receIve at health faClhtIes, and about who plays the central role m the deCISIOn to seekcare Respondents who reported that they or theIr spouse had expenenced obstetnc or neonatalproblems were asked about theIr care-seekmg behaVIOr for those problems and about thenperceptIOn of the care they receIved when they went for treatment

To assess access to care, respondents were quened about the then perceptIOns of the cost,comfort, and acceSSIbIlIty of transportatIOn, about the cost of servIces and medicme, and aboutdIstance to a health facIhty PartICIpants m the SSIs were requested to descnbe any aSSIstancethey receIved from members of the commumty when they expenenced obstetnc or neonatalcomphcatIOns The SSIs also contamed some questIOns regardmg knowledge of varIOUS famIlyplannmg methods, method preference, method use, awareness of sexually transmItted dIseases(STDs), and preferred STD treatment modahtIes

The same tOpIC areas mentIOned above were dIscussed m the 20 focus groups (10 WIth womenand 10 WIth men) held m the five dIstncts

INDONESIA

In IndoneSIa, the semI-structured mterview and focus group components ofthecommumtydiagnOSIS were dIVIded mto two parts One focused on "Safe Motherhood" Issues (1 e , all tOPICScovered m the BolIVIa research, WIth the exceptIOn of anemia, famtly plannmg, and STDs) Thesecond component focused largely on anemia, WIth some questIOns on famtly plannmg andsexually transmItted dIseases/reproductIve tract mfectIOns (RTls)

Data collectIOn occurred m four VIllages m three dlstncts durmg March and Apnl of 1996 Forthe Safe Motherhood component, 90 women who had chIldren under five years of age or whowere currently pregnant partICIpated m the SSIs A total of twelve focus groups (SIX WIth womenand SIX WIth men) addressmg Safe Motherhood Issues were conducted m the four commumtiesAddItIonally, SIX focus groups were held (three WIth men and three WIth women) WIthpartICIpants who were conSIdered "key mformants" For the component dealmg WIth anemIa,famIly plannmg, and STDs/RTls, SSIs were conducted WIth 90 women who had chIldren underfive years of age SIX focus groups (1-2 m each of the four VIllages) also were held Focus grouppartICIpants were women (some of whom were pregnant) WIth chIldren under five years of age


SEPTEMBER 15, 19978

AddItIOnally, SSIs were conducted WIth a total of29 bldans (mIdWIves) or bldan dl desas(mIdwIves-most of whom are young-who have graduated from an accelerated mIdWIferyprogram) The mterview gUIde contamed questIOns on perceptIOns of care m servIce delIverysItes, belIefs about commuruty perceptIOns of servIce qualIty, anemIa, famIly plannmg, andpartner notIficatIOn for STDs/RTIs

REPORT METHODOLOGY

G£1VERAL

In thIS document, mformatIOn on problem recogrutIOn, declSlon-makmg regardmg care, andperceptIOns of health servIce qualIty were obtamed through data analysIs and throughsynthesIzmg the matenal contamed m DIagnostlco Barreras y VIabllzzadores en la AtenclOn deComplzcaclOnes Obstetncas y Neonatales (EstudlO Cualztatlvo en Comumdades y ServlclOs deSalud de Cmco Dlstntos de Salud en La Paz y Cochabamba) (Seoane et al , 1996) and m ReportCommumty DIagnosIs MotherCare Safe Motherhood Project South Kalzmantan, 1996(Marsaban et al , 1996)

ThIS report does not attempt to be comprehensIve Only data from the Safe Motherhoodcomponent of the SSIs WIth women and husbands are dIscussed and only mformatIOn stemmmgfrom the focus groups dealmg WIth pregnancy-related complIcatIOns IS presented Data andfindmgs from the dIrect observatIOns and from the IndoneSIan SSIs focusmg on anemIa, famIlyplannmg, and RTIs are not explored

D4T4. AlV4.L}SIS 4./YD S}NTHESIS

The field offices m BolIVia and IndoneSIa proVIded the data from the SSIs These data had beenentered and coded m-country and were submItted on dIskette The SSIs m BolIVIa and IndoneSiacontamed both open- and closed-ended questIOns The open-ended responses eIther were notmcluded on the dIskettes, were mcomplete or were m Sparush or Bahasa IndonesIanTherefore, only the closed-ended questIons were analyzed

Before begmnmg data analysIs, the SSI mstruments used m IndoneSIa and BolIVIa werecompared WhIle the Items contamed m the SSIs m both countnes were framed around thePathway to SurvIval, the actual questIOns often dIffered m sigruficant ways For example, mboth countnes sectIOns of the SSI were devoted to assessmg women's awareness of problemsoccurnng dunng pregnancy, delIvery the post-partum penod, or WIth the newborn In BolIVIa,an unprompted format was used Women were asked to lIst the pregnancy-related problems WIthwhIch they were famIlIar In IndoneSia, answers were prompted Women were read a lIst ofproblems and for each problem on the lIst, were asked to state whether or not It was senousBecause these and other Items on the two questIOnnaireS were not IdentIcal, data could not bepooled, nor could direct comparIsons be made Instead, those Items which were SimIlar III


SEPTEMBER 15 19979

both countries, or WhICh addressed the same tOPIC area were selected for separate analysesand InterpretatIOn

Only those Items correspondIng wIth the first three steps of the Pathway (I e , problemrecogmtIOn, decision-makmg, and access to care) were selected for analyses Under problemrecognztwn, questIOns chosen dealt With awareness of complIcatIOns, perceIved seventy ofpregnancy-related problems, and prevIOUS expenence With complIcatIOns To explore declswnmakmg, selected Items addressed who the key players were m health-care seekmg decisIOnmakmg, where help first IS sought, and bIrth-place preferences Under access to care, questIOnsselected for analyses concerned perceIved qualIty of care, dIstance/tIme to the health faCIlIty costof transportatIOn and servIces, and commumty aSSIstance

Once Items were selected, data were recoded as necessary and frequency dIstnbutIOns wereobtamed Where appropnate, data were stratIfied by key variables (e g , whether or notrespondents had expenenced pregnancy complIcatIOns, place of reSIdence) BIVanate analyseswere performed ChI-square statIstICS also were obtamed But, the very small cell-SIzesprecluded meanmgful mterpretatlOn and the results ofthe chI-square tests are not reported heremBecause of small sample SIzes and also because many questIOns had low response rates(partIcularly for BolIVIa), multivanate analyses were not conducted

Raw data from the focus groups were not translated from SpanIsh or Bahasa IndoneSIan mtoEnglIsh nor were transcnpts sent to MotherCare, DC SInce raw data from the focus groupswere unavaIlable, analyses could not be performed Instead, focus group InfOrmatIOn asdocumented by the authors of the commumty dIagnOSIS reports from BolIVIa (Seoane et al ,1996) and IndoneSIa (Marsaban et al , 1996) were syntheSIZed, and the results of thatsyntheSIS are dIscussed In theIr report on the commumty diagnOSIS m BolIVIa Seoane andcolleagues proVIde theIr mterpretatlOns of the focus group dISCUSSIOns The authors support theIrstatements WIth many dIrect and very reveahng quotes from partICIpants Marsaban andcollegues also dISCUSS focus group findmgs m theu report on the commumty diagnOSIS m SouthKalImantan However, the report contams very few quotes from focus group partICIpantsConsequently whIle results ofthe SSIs m BolIVIa could be supplemented WIth select quotesfrom the focus groups, thIS was not pOSSIble WIth the IndoneSIan data In Section II, focus groupfindmgs from IndoneSia are dIscussed, but WIth the caveat that almost no data were aVailable forsyntheSIS


SEPT£I1..fBER 15 199710

SECTIONII-FINDINGS

INTRODUCTION

ThIS sectIOn contams a summary of the findmgs from the SSIs and FGs In keepmg wIth theformat of the Pathway to SurvIval, the dIScussIon begms WIth results from the sectIon onproblem recognztwn, contmues WIth declswn-malang, and concludes WIth access to care At thebegmnmg of each subsectIOn, a general overvIew of findmgs and methodologIcal or analytIcalIssues IS gIven Fmdmgs are then dIscussed III more detaIl Where approprIate, ImplIcatIOns forfuture commumty dIagnosIs research are hIghlIghted Appendix B contams more detml on thelImItatIOns of each Item appearmg m the SSIs and FGs A general dISCUSSIOn of methodologIcalIssues IS contamed m SectIOn IV However, a few key caveats must be mentIOned here FIrstand foremost, because purposive samplIng was used for the SSIs and the FGs, the resultscannot be considered representative of the broader populatIOn, and so, cannot begeneralIzed Also, whIle the terms "IndoneSIan" women and familIes are used throughoutthe report, study partIcipants were from South KalImantan only

In IndoneSIa, problems With S8IquestIOnnaire construction mean thatresults regardmg awareness, perceptIons,and prevIous experIence are notconclUSIve In BolIVia, respondents wereaware of many problems, but m somecases, had trouble dlstmgUlshmg severityIn both countries, women and husbandswere less lIkely to report awareness ofnewborn problems than they were toreport awareness of maternal problemsThey also were less lIkely to considernewborn complIcatIons to be severe (ascompared to maternal complIcatIOns)

STEP 1 PROBLEM RECOGNITION AWARENESS AND PERCEPTIONS OF AND EXPERIENCE WITH

PROBLEMS DURING PREGNANCY,

DELIVERY, THE POST-PARTUM PERIOD

AND WITH THE NEWBORN

STEP 1 GEVERA.L COU!lvfElVTS

Data from the SSIs do not give a clearmdlcatlOn of the degree to whichIndoneSIan women are aware of the signsof obstetric and neonatal complIcatIOnsThe ambIgUIty IS due to Issues WIthsamplmg and WIth constructIOn of thoseSSI Items deSIgned to garner mformatIOnon problem recogmtIOn In partIcular,some women were selected forpartICIpatIOn m the SSIs and FGsspeCIfically because they had expenencedpregnancy-related complIcatIons Thesewomen would be expected to have a hIgherlevel of awareness and knowledge of certam comphcatIOns than women WIthout comparableexpenences AddItIOnally women were asked to hst the problems they had expenenced but notto hst the problems of whIch they were aware Only If the respondents reported they had not hadany comphcatIOns were they asked to name problems women mIght expenence 2 Thus, theItems representmg the domam of problem recogmtlOn lack content valIdity (I e , thequestIOns were not representatIve of the concepts they were mtended to reflect)

F \nnachbar\comdx\report\\finaI4 wpdSEPTEHBER 15 1997

11

Some data mdlcate that BoliVian andIndoneSian women do not vIew pregnancy as atime reqUlrmg speCial care or attentionHowever, InformatIOn IS neIther defimtlve norcomplete

In BolIvia, selectIOn cntena for the women participatmg m the SSIs are not gIven So, It IS notpossIble to determme the degree to whIch selectIOn Issues may have affected the resultsHowever, the BoliVian SSI Items pertaInIng to problem recogmtlOn had greater contentvalidity than those In the IndoneSian Instrument Usmg an unprompted questIOn format,women were asked to lIst the pregnancy-related problems With WhICh they were famIliar Thus,In BoliVia, awareness could be assessed more easily and accurately than In IndoneSia

Data from the semI-structured questIOnnaIreS and focus groups mdicate that women recogmzesome but not all SIgns of complIcatIOns and that they often have trouble dIstmgmshmg problemseventy (e g , hemorrhage versus hght bleedmg) 3 AddItIonally, focus group reports suggest that

whIle Some problems are IdentIfied easIly, others are not

WhIle It IS pOSSIble, even hkely, that IndoneSian women (and other respondents) have troubledetermmmg when bleedmg IS lIfe-threatemng, the report by Marsaban and colleagues on thecommunIty diagnOSIS m IndoneSIa does not contam any focus group mformatIOn on seventydIstmctIOn for hemorrhage or other complIcatIOns

In both IndoneSia and BolIVia, results from the SSIs with women show variability In theextent to which problems are recognIZed and the degree of seventy attnbuted to theseproblems, dependIng on whether the women are speakIng about pregnancy, delivery, thepost-partum period, or the health of their newborns SpeCIfic findmgs demonstratmg thISvanabilIty are dIscussed below

ImplzcatlOns for Future Commumty DzagnosesTo ensure content validity, semi-structured Interview Items deSigned to get Information onproblem recogmtlon and perceptIOn of problem seventy need to be carefully constructed

STEP 1 PERCEPTIONS OF PREGNANC}

4.'vD PREGA -4 'vC}-REL-4TED PRACTICES

IndoneszaIn theIr summary report, Marsaban andcolleagues suggest that women are notpartIcularly concerned about theIrpregnanCIes and that they mamtam anormal workload (e g , grocery shoppmg, cookmg, washmg, cleanmg, and chIld-rearmg) whIlepregnant 4 At the same tIme, they wnte that women are dIscouraged-though they do not say bywhom-from IIftmg heavy loads later m pregnancy 5 In theIr dISCUSSIOn of the anemIa focusgroups, Marsaban and colleagues also state that some partICIpants Said women should eat "morenutntIOUS food" when pregnant 6 The SSIs dId not contam any Items on prescnbed or proscnbed


SEPTEMBER15199712

food and drInk m pregnancy and the focus group field gUIde contamed only one questIOnregardmg perceptIOns of a normal pregnancy In sum, the commuDIty diagnosIs m Indonesiadoes not appear to have gathered defiDItive mformatIon on pregnancy-related norms,beliefs, or mdividual practices, nor does It appear that women were asked how they feelabout their pregnancies

BohvwBoliVian women appear to treat pregnancy III much the same way as their IndoneSiancounterparts WhIle women are encouraged not to "lIft heavy thmgs," they contmue to work mmuch the same manner as before theIr pregnancy 7 Seoane et al use quotes from female focusgroup partIcIpants to support thIS statement As was the case m IndoneSia, It does not appearthat questIOns speCific to diet or speCIal practices were asked

o oss-countryIn both countrIes, women participatmg m focus groups reported that theIr husbands helped wIthtasks requmng heavy lIftmg (e g ,fetchmg water, or washmg bedspreads) However, m theIrsummary document, Marsaban and colleagues state that women reported feelIng embarrassed IftheIr husbands were seen as helpmg too much 8 No comparable sentiment was mentIOned m theBolIVIan report by Seoane and colleagues However, m a separate qualItatIve research study onanemIa m pregnant women m BolIVIa, the majorIty of the women questIOned stated that theIrpregnancIes adversely affected theIr qualIty of lIfe 9 Some reported feelmg useless because theywere unable to work as much as they had before theIr pregnancy AddItionally, some womenreported that theIr husbands were neIther supportIve nor understandmg of theIr lImItatIOns 10

ImphcatlOns for Future Commumty DwgnosesFuture commuDIty diagnosIs research should collect more mformatlOn on how women andcommuDItIes view normal pregnancies, mcludmg normative beliefs, and practices (e g ,regardmg diet, workload, health-care seekmg behaVior, and women's roles whIlepregnant) Efforts also should be made to determme the role and expectatIOns of husbandswhen their wives are pregnant, delivermg, or post-partum InformatIOn on spousal roles mthe care of the newborn also should be gathered d

dLength of mtervlews and/or focus groups must be conSidered m the constructIon of any field gUIdeWhIle mformatlOn on newborn care IS Important, It may be approprIate to collect thiS (and other data on newborns)separately from data gathered on pregnancy and obstetrIC complIcatIOns


SEPTEMBER 15 199713

Women In IndoneSIa generally are notaware of the term, anemIa But, comparedwIth BolIVIan women, they appear moreaware of anemIa's sIgns and symptoms.

STEP 1 A WARENESS OF 4ND PERCEPTIONS OF SWELLINGIndoneslG r---------------------,

Some IndonesIan women do dIstInguIShIn the semI-structured questIOnnaIreS, one-thIrd of the IndonesIan women stated that among the vanous types of swellIng (I e, of

the feet,. face, or hands) Some BolIVIanhavmg swollen feet IS a senous problemAbout half the women mtervlewed saId that women are aware of swellIng, but few

consIder It to be senous (even when Itswelhng of the face or hands IS a senous

Involves the face).problem The rest of the women SaId thatswellmg of the feet, face, or hands IS notsenous, or that they were unsure of about the level of seventy The results mdlcate that a fewwomen do appear to dlstmgUIsh among types of swelhng SpecIfically, these women perceIvepuffiness of the face and hands (a sIgn assocIated wIth pre-eclampsIa) as more severe thanpuffiness of the feet (a common problem In pregnancy) However, overall, IndoneSianwomen do not appear to view swellIng as a danger sIgn requIrIng ImmedIate attentIOn

BohVlGIn BolIVia, the findIngs are SImIlar to those In IndoneSIa WhIle some women recognIZeswellIng of the feet or face, few conSIder any kmd of swellIng to be serIOUS Hardly anyhusbands mentIOned swellmg as a problem Accordmg to the report by Seoane et ai, neIthermen nor women aSSOCIate the SIgns of eclampSIa WIth swellmg 11 Further, as Illustrated by thefollowmg quote from a focus group partICIpant, swellIng appears to be conSidered a normalpart of pregnancy

In each pregnancy, puffiness IS normal, If IS not a problem(Artesan, 26 years, 1 son, Barno N HonzonteslLow UselLPZye

STEP 1 A WARENESS OF AND PERCEPTIONS OF

ANEMJ.1.

IndoneslGIn the IndoneSian focus groups, women dIdnot mentIOn anemIa or ItS SIgns andsymptoms WIthout promptmg 12 WhIleunaware of the term, "'anemia," women dId recogmze the term, kwang darah, meanmg "notenough blood" Women m the focus groups conSIdered kurang darah to be the converse of hIghblood pressure, and the term often was used synonymously With tekanan darah rendah, meamnglow blood pressure The confUSIOn between the terms IS noteworthy and, accordmg to the reportby Marsaban et ai, extends to Puskesmas (commumty health center) staff, bidans, and bidan dldesas b

e Seoane et ai, 1996 Page 90


SEPTEMBER 15, 199714

Most women m IndoneSia conSider longlabor, premature delivery, andhemorrhage durIng delivery to be serIOUS,but In BoliVia, these problems appear to beof less concern than many of thoseoccurrmg after delivery

WhIle Marsaban and colleagues state that few women reported knowmg the cause ofkurang darah, all the focus group respondents were aware that Iron tablets, or obat tambaltdaralt (medicIne to add blood) could be taken to alleviate Its signs and symptoms 14 The factthat women dId not spontaneously mentIOn anemIa or Its sIgns and symptoms, but whenprompted, dIscussed kurang darah and Its treatment mdIcates the Importance of gathenngmformatlon on local termmology

In the Safe Motherhood SSIs, women were asked whether kurang daralt IS a serIOUSproblem Three-quarters ofthe respondents stated that It IS a serIOUS problem However,gIVen that women often consider kurang daralt and tekanan daralt rendalt to besynonymous, It IS not possible to conclude that women consider anemia per se to be aserIOUS problem

BolzVlaIf the number of women mentIOnIng a problem (unprompted) can be taken as an Indicatorof the degree to which that problem IS recognIZed, or thought Important, then women InBoliVia do not appear to consider anemia to be particularly SignIficant When asked to nameproblems they consIdered senous m pregnancy only a handful of BolIVIan women mentIOnedanemIa or ItS assocIated SIgns and symptoms, although those who dId so consIdered the SIgns andsymptoms to be senous There IS lIttle mformatIOn pertammg to anemIa III the report by Seoaneet aI, because a detaIled qualItatIve analyses on the tOpIC was conducted by the OMNI project 15

ImplzcatlOnsFuture commumty diagnOSIs mstruments should contam questIOns on perceived causalityof signs and symptoms of anemia Data collectIOn mstruments also should contam ItemsdeSigned to obtam local termmology for anemia and ItS associated signs or symptoms

STEP 1 AH 4.REIIIESS OF 4111DPERCEPTIONSOFCERTAIN DELiVER} PROBLEMS

lndoneslaIn the IndoneSian SSIs, when women were askedwhether they conSidered prolonged labor, prematurerupture of the membranes, and cesarean sections tobe serIOUS, about three-quarters said that they didMore than half the women said that fever durmgdelivery IS a serIOUS problem and more than threequarters of the respondents also conSideredhemorrhage durmg delivery to be serIOUS 16 WhIletests of statIstIcal sIgmficance cannot be performed, the results mdIcate that women do notconSIder all problems to be equal


SEPTEMBER 15,199715

Wlule retained placenta IS recognIZed bywomen In both IndonesIa and ID BolIvia,women In both countnes have UnIqueperceptions regardIng Its causes andconsequences

In Indonesia, the SSIs did not contam anyquestIOns specific to retamed placenta Intheir diSCUSSion regardmg the focusgroups, however, Marsaban and colleaguesstate that It IS one of the most commonlyexperienced and well-knowncomplicatIons 17 While the author does not supply any focus group data, she states thatwomen belIeve oral contraceptives are responsIble for the perceIVed rIse m the number ofcases of retamed placenta f,18

BohvzaIn Bolivia, all SSI respondents were asked to name the labor/delivery problems with whichthey were familiar In general, women were less likely to mention problems durmg thistIme period than they were to cite problems occurrmg m the post-partum penod Thosewomen who dId mentIOn delIvery-related problems generally spoke of premature delIvery,hemorrhage, or rna/parto (bad bIrth), whIch can encompass, among others, any of theaforementIOned problems as well as prolonged labor 19 No addItional mformatIon on rna/partoIS available (e g , perceived causes and consequences, or what signs or symptoms rna/partoencompasses)

Less than one-quarter of women who partIcipated m the SSIs mentIOned prematuredelivery as a problem, though almost all who dId so thmk It IS serious Some BolIVIanwomen mentIOned premature labor, and those who dId generally conSIder It to be senous WhIlefocus group respondents dIscussed prolonged labor, no women participatmg m the SSIsmentIOned It as a problem, and only two spoke about hemorrhage dunng delIvery ThIS couldmdicate that the dIstmction between those problems occurrmg m labor/delIvery and thoseoccurnng m the post-partum penod may be somewhat artIfiCIal The fact that only two womenmentIOned hemorrhage m delIvery also may mdicate that the respondents do not dIfferentIatebetween between hemorrhage and bleedmg (smce some bleedmg durmg delIvery IS to beexpected) AlternatIvely, It also IS pOSSIble that hemorrhage durmg delIvery was CIted far lessoften than post-partum hemorrhage because women accurately conSIder the latter to be moresenous

[Accordmg to Mary Kroeger, a midWife m Java 60 percent of maternal deaths m IndoneSia are attnbutedto post-partum hemorrhage She believes the high hemorrhage rate IS due, m part to the fact that TBAs are tramednot to touch the placenta and that m many births the placenta needs to be gUided out Therefore, the Issue may notbe retamed placentas per se but ungUided placentas


SEPTE\/BER 15, 199716

In the 88Is, about one-quarter of the women stated that retaIned placenta IS a problemHowever, of these, only half consider It serIOUS Data from the focus groups reveal thatwomen, particularly those from rural areas, bebeve retaIned placenta to be bfe-threatenIngHowever, these women do not associate retaIned placenta with hemorrhage, but withInfectIon and heart problems In additIOn, many women bebeve retaIned placenta IS causedby premature cuttIng of the umblbcal cord

ImplzcatlOns for Future Community DzagnosesFuture research efforts should allow for probIng on culturally-defined problems durIngdelIvery (e g , the perceived lInk between oral contraceptIOn and retaIned placenta InIndoneSia and ma/parto In Bobvla) Data collectIon Instruments should be modified toInclude Items on speCific aspects (I e , perceptIOns and behaViors) of normal delIveries

STEP 1 PROBLEM RECOGNITION/PERCEPTION OF SEVERITY (POST-PARTUM)Indonesza ,-------------------,

In both IndoneSia and In BolIVia, womenFor many condItIOns dIscussed m the SSIs,

view hemorrhage as a serIOUS, potentiallyabout two-thIrds to three-quarters of thewomen m IndoneSIa conSIder varIOUS IIfe-threatemng problem However, how

women and their famIlies distIngUIshproblems to be senous For SIgns ofmfection the numbers were lower Just over between mIld and severe bleedIng IS not

clearhalf the women mtervIewed reported thatabdommal pam IS senous and about one-quarter SaId that maloderous vagmal dIscharge IS senous When asked about post-partumhemorrhage, Virtually all the women IntervIewed saId It IS serIOUS More women SaIdhemorrhage IS senous than SaId the same of bleedmg These responses mdicate that at least someof the respondents dIstmgUIshed between the two problems In theIr IndoneSIa trIp report of May27-June 19, 1996, PareJa and Galloway state that for hemorrhage dUrIng dehvery, womenconSIder bleedmg to be exceSSIve, or mdIcatIve of hemorrhage If three sarongs are filled WIthblood 20 It IS not clear whether thIS same cntena IS used for hemorrhage occurrmg at other tImesFurther, It IS not clear how women dIstmgUIsh hemorrhage dUrIng dehvery from hemorrhage mthe post-partum perIod, except that It IS pOSSIble that women conSIder post-pactum hemorrhage tobe a more serIOUS problem because It occurs more frequently

BohvzaWith the exceptIOn of sobreparto (relapse after bIrth), post-partum hemorrhage wasmentIOned WIthout promptmg by more women m BolIVIa than any other problem Most ofthese women conSIdered It to be a serIOUS problem, though a few bebeved It IS not serIOUSIn theIr report, Seoane and colleagues state that several women m the focus groups attnbutedhemorrhage dUrIng pregnancy to hftmg heavy thmgs 21 Data on the perceIved causes of postpartum hemorrhage are not aVaIlable, nor IS mformatIOn on how women dIstmgUIsh betweenmIld or severe bleedmg



In BolIVia, more women mentIOnedsobreparto (relapse after bIrth) than anyother problem. ThIS phenomenon, whichincludes a Wide range of complIcatIons andsymptoms is attributed to the mother'scareless or mapproprlate behaVIor. Whilewomen often seek assIStance forsobreparto;they do not turn to themstItutional health system. Nocomparable problem emerged from thedata gathered m IndoneSIa.

Sobreparto, whIch can mclude symptomsremInISCent of sepSIS or hemorrhage, or WhIChmay mclude swollen feet, or pams resemblmgthose occurnng durmg labor,22 generally ISconsIdered a senous problem by most womenwho mentIOned thIS phenomenon However,a substantIal number of women stated that ItIS not senous The range of responses WIthregard to seventy IS most lIkely accounted forby the range of symptoms sobrepartoencompasses As the focus group data reveal,and as Seoane et al report, sobreparto ISperceIved to occur as a result of a woman'sbehaVior, particularly exposure to cold air, cold foods, or cold waterdoes not appear to be a comparable phenomenon

In IndoneSia, there

IrnplzcatlOns for Future Cornrnumty DzagnosesSInce InformatIOn on behaVIOrs durIng the post-partum period and on normal recoveryfrom delIvery IS lackIng, field gUides should Include Items deSigned to capture thesedomaInS



Both BolIVIan and IndoneSian women areless familIar With problems occurrIng Intheir newborns than they are wIthproblems lD pregnancy, delIVery, or In thepost-partum perIod

STEP 1 PROBLEM RECOGNITION/PERCEPTIOlv OF SEVERIn (NEONAT4L)

IndonesiaAll women participatmg m the SSIs wereasked to mdicate whether they consIdered apartIcular problem that "can happen to ababy"g to be senous or not Only about onequarter belIeve havmg twms or tnplets IS asenous problem About half the womenasked saId that the follOWIng are notserIOUS PushIng before the expected due date, havIng the umbIlIcal cord wrapped aroundthe baby's neck, If the baby doesn't cry, or If the baby IS cold More than half the womensaId havIng a small baby IS not a problem More than half the women saId that It IS aserIous problem If the baby looks blue and If the baby's eyes ooze, though a substantIalnumber of women saId they dIdn't know whether or not the latter was serIOUS

BolzVlaResults from focus groups, SSIs (despIte problems wIth the data) IndIcate that studypartICIpants generally are unaware of potentIal complIcatIOns wIth the newborn In theSSIs, BolIVIan women were asked to name the problems WIth "the bIrth ofthe newborn" ofwhIch they were aware The wordmg of the questIon IS ambIguous It IS unclear whether the"bIrth of the newborn" pertams to newborn health or to the delIvery process h The questIOn lackscontent valIdIty and the resultmg data must be mterpreted WIth cautIOn WIth those caveats, fewwomen mentIOned any neonatal problem The three problems cIted were premature bIrth,stIllbIrths or neonatal deaths, and problems WIth the umbIlIcal cord Ofthese, prematurebIrth was mentIOned by a handful of women Only three of 41 women mentIOnedstIllbIrths/neonatal deaths and only one woman CIted problems WIth the umbIlIcal cord Thosewho mentIOned prematunty belIeve It IS a senous problem ThIS conflIcts WIth results from thefocus groups, where prematunty and low birthweight are seen as tranSIent condItIOns of lIttleImport (though the low number of responses m the SSIs preclude meanmgful mterpretatIOn ofthe data)

AccordIng to MotherCare staff, temperature changes, breathIng problems, lIstlessness, andpoor suckmg are neIther looked for nor seen as dangerous 23 Further, In theIr report,Seoane and colleagues conclude that neonatal problems (e g, fetal dIstress) whenmentIOned, are seen as Important for theIr potentIal to threaten the mother's lIfe, ratherthan the lIfe of the newborn 24 The fact that so few women mentIOned neonatal problemsappears to mdIcate a low level of awareness of the pOSSIbIlIty of complIcatIOns and of theIr

blndonesmn Safe Motherhood SSI Instrument

hSohvlan Safe Motherhood SSI Instrument

F \nnachbar\comdx\report\\finaI4 wpdSEPTEMBER 15 1997

19

potential seventy ThIS lack ofunderstandmg of newborn complIcatIOns does not necessarIlyImply a lack of concern for the well-bemg ofthe mfant, however

ImpllcatlOns for Future Commumty DwgnosesData collectIOn mstruments should be reVIsed so that ambIguously worded questIOns regardmgthe newborn are changed AddItIOnally, mformatIOn should be gathered on perceptIons ofnormal newborn behavIOr and on treatment of the newborn


SEPTEMBER 15,199720

How women perceive a newborn death ISunclear EVidence mdlcates that mBolIVIa, sUrvIval of the newborn ISconsIdered subordmate to survIval of themother. In IndonesIa, neonatal deathsoften are ascrIbed to fate.

STEP 1 PERCEPTIOTVS OF NEWBORN DEATHS

GeneralData from BolIvIa and Indonesia show thatwomen and theIr farrlllIes are notknowledgeable about neonatal complIcatIOns,and that they appear to belIeve lIttle can bedone to prevent neonatal deaths A betterunderstandmg of women's perceptIOns ofstIllbIrths and neonatal deaths IS necessary todetermme what messages are appropnate for motivatmg women and theIr famIlIes to look forproblems and seek help when neonatal complIcatIOns arIse

IndoneszaAccordmg to PATH-Indonesia, familIes "seem to accept mfant and chIld death as 'normal'and as dictated by fate, "'by the will ofAllah ",2~ In theIr final report, Marsaban andcolleagues also state that whIle some respondents report an mtense feelIng of gnef m the penodImmediately followmg a stIllbIrth or neonatal death,26 thIS feelIng IS short-lIved, and that thewomen belIeve when a newborn dIes the chIld goes straight to heaven and can help pull themother there when her tIme comes On the surface, It appears that mfant and chIld mortalItyIS considered fated-a common response given 10 qualItative research It should be noted,however, that through probmg techmques, It often IS possible to elICit addItIOnal responses10 which deaths are attributable to multiple causes

BolzvzaIn their report, Seoane and colleagues state that among many BolIVIan familIes, the birthprocess IS seen as such a threat to the woman's lIfe, that her survival IS viewed as atrIUmph, even If the mfant dies In thIS context, the authors suggest, a neonatal death IS notconsIdered a tragedy 27 At the same tIme m the SSIs, most BolIVian women Said thatmalpOSItIOn of the baby IS a senous problem In theIr analysIs of thIS findmg, Seoane et alsuggest that malpOSItIOn IS vIewed as senous not Just because of ItS assocIatIOn WIth a dIfficultand potentially dangerous bIrth for the woman, but because women belIeve malpositIoned mfantsgenerallv dIe durmg the bIrth process Further whIle less than one-quarter of the BolIVIanwomen mterviewed mentIOn abortIOn as a complIcatIOn of pregnancy, of these, almost allconSIder It to be a senous problem These results mdicate that BolIVIan women are, m fact,concerned about bnngmg theIr chIld to term and m good health

ImplzcatlOns f01 Future Communzty DzagnosesFuture research should probe more deeply mto belIefs and perceptions for neonatal deathsCare also should be taken m the wordmg of questIOns related to the neonate (1 e , Items shouldclearly dIfferentIate between the bIrth process and the neonate) to ensure content valIdIty ofmdividual questIOn Items


SEPTEMBER 15 199721

Among women partlclpatmg m the SSIs mIndonesia and m BolIvia, more reportexpenencmg comphcatIons durmgpregnancy, birth, or m the post-partumperiod than With the newborn In BolIvia,problems With questIOn wordmg may havemfluenced results

STEP 1 EXPERIENCE OF COMPLIC4TIO"lS

GeneralIn both BolIvIa and Indonesia, fewerwomen reported experIencmg acomplIcatIOn with their newborn thandurmg pregnancy, delIvery, or the postpartum period About one-thud of theIndonesIan women mterviewed stated thatthey had a problem m pregnancy, dehvery, ordunng post-partum, whereas only one-quartersaId that theIr baby expenenced a comphcatIOn after bIrth In BohvIa, even fewer women saIdthat theIr newborn had a problem, yet almost two-thuds of the women stated that they had acomplIcatIOn dunng pregnancy, about one-thIrd reported problems durmg delIvery, and less thana quarter reported post-partum comphcatIOns

The low levels of reported newborn complIcations may reflect actual experience (I e , thatfew newborns experience problems) However, gIVen that newborn complIcations are notan mfrequent occurence, the fact that few women reported experIencmg problems may,mstead, mdlcate low levels of awareness


SEPTEMBER 15 199722

STEP 2 DECISION-MAKING AROUND THE USE OF HEALTH SERVICES WHEN COMPLICATIONS

ARISE DURING PREGNANCY, DELIVERY, THE POST-PARTUM PERIOD, OR WITH THE NEONATE

STEP 2 DECISION-MAKING WHO ARE THE CRITICAL PLA}ERS?General r---------------------,It IS eVident from the SSIs and focus Husbands playa critical role m the

decIsIOn to seek care m both Bolivia andgroups In BolivIa and IndonesIa that Indonesia Data from the SSIs in bothhusbands playa critical role m health-

countries are mcomplete, and/or sufferseekmg declslon-makmg However,

from problems with content valIdity andproblems with content validity of SSI Items

codmg. In Bolivia, focus group data revealand with data codmg preclude meanmgful that the woman herself and the mother-m-m-depth analysIs of the declslOn-makmg

law also can play an Important declslon-process makmg role.

IndoneszaIn Indonesia, about half the women mtervIewed stated that they would make the decIsIon to seekcare themselves or that they decIded where they would seek care (for problems occurrmg m anystage) When data were analyzed for only those women who reported expenencmgcomplIcatIOns m delIvery, the post-partum penod, or wIth theIr newborns, results were notdIscernIbly dIfferent About three-quarters of the women who reported expenencmgcomplIcatIOns dunng pregnancy SaId that they deCIded to seek care or that they deCIded whosecare to seek (the questIOn was worded ambIguously) The ambIgUIty of the questIOn, m tum,makes mterpretatIOn dIfficult Accordmg to Marsaban and colleagues, husbands deCIded whetherto take theIr WIves to the health faCIlIty for problems occurrmg durmg delIvery I However, m theSSIs, about half the women stated that they had made or would make the deCISIOn themselves

In theIr report on focus groups, PATH states that women and theIr husbands rely on theJudgement of the tradItIOnal bIrth attendant as to whether or not a mIdWIfe should be called or thewoman brought to a health faCIlIty From the report, It IS not pOSSIble to determme whether thISmformatIOn was denved from focus group dISCUSSIOns or IS PATH's mterpretatIOn

BobvzaIn BoliVIa, there were problems With codmg or mlssmg data Fmdmgs must be mterpretedWith cautIOn Among women who reported expenencmg pregnancy complIcatIOns, most SaIdthat theIr husbands made the deCISIOn regardmg care-seekmg However, nearly half the womenwho SaId that they had a problem dunng pregnancy were coded as "don't know/no response"Among women expenencmg complIcatIOns m delIvery, most women Said that eIther they or theIrhusbands made the deCISIOn Almost no women reportmg post-partum complIcatIOns responded

'The report does not contam any mfonnatIOn regardmg declsIOn-makmg at other tImes


SEPTEMBER 15 199723

In IndoneSia, most women report that theysought the help (or would seek the help ofa midWife or Village midWife) whencomplicatIons arose (or If complicatIonsarose) dUring any stage In BolIVia, datafrom the SSIs yielded questIonable results,some of which conflict With focus groupreports

to the questIOn on the decIsIon to seek care Very few women reported havmg problems WIththeIr newborn, but those women who dId reported eIther that theIr husband made the deCISIonalone or consulted WIth them Because so much data are mlssmg or coded ambiguously, andbecause the Items themselves were ambiguous, the SSIs were not particularly useful fordetermmmg who plays the most slgmficant role m the declslon-makmg process

Reports from the focus groups, however, demonstrate the husband's Important role m thedecIsion to seek care Both male and female focus group partIcIpants CIted the husband'soblIgatIOn to care for hIS famIly, hIS perceIved knowledge of the communIty, hIS authOrIty todeCIde major household expendItures, hIS concern for hIS WIfe, and the knowledge that when hISWIfe IS sufferIng, she may not be able to make the deCISIon

Data from the focus groups also mdlcate that the woman, as an mtegral family memberalso takes part m the decIsIOn In theIr report, Seoane and colleagues state that a woman'sdecisIOn-makmg authOrIty IS greatest dunng pregnancy DUrIng delIvery the husband andsometImes other famIly members, such as the woman or mother-m-Iaw take on a more dommant,decision-makmg role In the post-partum perIod, the woman agam becomes central Seoane etal state that the husband's role m the deCISIon to seek care for neonatal complIcatIOns IS almostnon-eXIstent ThIS IS m contrast to the findmgs from the SSIs

ImpllcatlOns for Future Community DzagnosesIt IS cntical to obtam valId data on the declSlon-makmg process surroundmg the use of healthservIces QuestIOns on SSIs need to reflect the concepts they are mtended to representAddItIOnally, "no response" and "don't know" answers should be recorded and coded as separatecategones

STEP 2 DECISION-M-4KING WHOSE HELP IS SOUGHT?

GeneralIn both IndoneSia and BoliVIa, resultsof the focus groups and SSIs shed somelight on the ratIOnale for selectmg oneform of health services over anotherwhen complicatIOns arise, but they donot give the full picture In theIndoneSIan SSIs, women report usmg orhavmg the deSIre to use the servIces ofmIdWIves, but the role of the TBA m thebIrth process and m deCISIons to seek outthe care of a mIdWIfe IS not elucIdated clearly In BolIVia, data from focus groups and SSIs aremconsistent Also, because of low response rates m the SSIs, results should be mterpreted WIth


SEPTEMBER 15,199724

cautIOn Furthermore, the communIty diagnoses gathered lIttle mformatIOn on how decIsionsregardmg health care under normal CIrcumstances (1 e , unrelated to the expenence ofcomplIcatIOns) are made

IndoneslaIn the SSIs, women report seekIng the help of midWIVes when complIcatIOns arIse Abouthalf the women partIcipatmg m the SSIs stated that they sought care from mIdWIves (bzdans) andabout one-quarter Said they went to VIllage mIdWIves (bzdan dl desas) A few mentIOned gomgto the doctor or tradItIOnal birth attendant (TBA), and only a handful Said they went to the healthpost/sub-health center (pustu) or health center (puskesmas) Results were vIrtually Identicalwhen women were asked where they went or would go for complIcatIOns dUrIng delIvery, thepost-partum penod, or With the newborn Though most births occur at home and though TBAsattend a hIgh percentage of these blrths,28 few women partlclpatmg m the SSIs reported gomg toTBAs for help (pOSSIbly because the TBAs already may have been present already) Raw datafrom the focus groups may help elUCIdate thIS SItuatIOn With regard to thIS tOpIC, however,Marsaban and colleagues-m theIr report-only state that women and theIr famIlIes rely on theTBA to tell them when a tramed midWIfe should be summoned 29 When women were askedwhose care IS sought when complIcatIOns anse, the questIOn appears to have been open-endedWomen may have assumed the questIOn pertamed to the formal health system only, which wouldaccount for the hIgh percentage of women reportmg that they went or would go to a bldan orbldan dl desa Alternatively, It IS pOSSIble that women are m fact seekmg out the care of thebldans and bldan dl desas However, more 1OformatlOn IS needed to determ10e theCIrcumstances under which IndoneSIan women and their famIlies seek care, whose care ISsought, and for which problems and why

BohVlaSeoane and colleagues state that most women focus group participants from areas wherethere IS low use of 1Ostltutlonal health services go first to parteras, 10 whom they haveconfidence 30 Whereas, women FG participants from high use areas go first to the1OstitutlOnal health system, because they belIeve their problems are more lIkely to beresolved through that channel than With the partera 31 That women chose the traditIonalsystem m low use areas and the mstitutIOnal system m high use areas IS to be expected, smcefocus group participants were selected on the baSIS of thiS charactenstic However, analyses ofthe SSIs reveal potentially conflIctmg results Women from high and low use areas who reportedhavmg complIcatIOns dUrIng delIvery Said that they used the mstitutIOnal health system, or wentto a partera m about equal numbers, regardless of their preassigned service use category Theseresults should be mterpreted With cautIOn however, because of low response rates Data are notavaIlable for the other stages either because of questIOnnaire deSign or because low responserates make mterpretatIOn difficult In their dISCUSSIOn of the focus groups, Seoane and colleaguessuggest that husbands and Wives, regardless of whether they come from high or low service useareas have confidence that doctors have the knowledge and trammg to resolve complIcatIOns 32


SEPTEMBER 15,199725

However, the authors state that such confidence IS not apphed unIversally across all healthsystem levels In partIcular, health center phySICIans are perceIved as mexpenenced and lesscompetent compared to those at the dlstnCt hospItal Seoane et al assert that thIS perceptIOn ISpartIcularly sahent m communItIes where servIce use IS low

ImphcatlOns for Future Communzty DzagnosesEfforts should be made to gather more data on declSlon-makmg behaVIOr VIs-a-VIS healthservIces use m pregnancy, dehvery, the post-partum penod and WIth the newborn DatacollectIOn mstruments should have an expanded sectIOn on the decision-makmg process-bothunder normal CIrcumstances and when comphcatIOns arIse



In IndoneSia, most women say theirneighbors accompany them when they goto the post, center, or hospital (m general,not specIfically for problems). In BolIVia,when women had complIcatIOns, they weremost often brought to the health center,post, or hospital by their husbands

STEP 2 DEClSIOIIf TO SEEK CARE WHO ACCOMPANIES THE WOM4N?

GeneralData from the SSIs show that whenIndonesIan women go to health facIlities mgeneral (I e , not specifically forcomplicatIOns), they are accompamed byneighbors In Bolivia, husbands generallywent with their wives to the health facIlitywhen the women experiencedcomplicatIOns Women participatmg m theSSIs m IndonesIa were not asked whoaccompamed them when they had comphcatIOns In contrast, Bohvian women were asked onlyabout theIr compamons when they went to a health facIhty for comphcatIOns dunng pregnancy,dehvery, the post-partum penod or wIth theIr newborn and not about who accompanIes themunder normal condItIOns Smce the questIOns were dIfferent, the two countnes cannot becompared However the central role of neighbors and husbands In IndoneSia and BoliVIarespectively should be noted

ImphcatlOnsSectIOns of the field gUIdes desIgned to assess the same mformatIOn across countnes should, tothe extent possIble contam Items that are standardIzed InformatIOn should be obtamed on whoaccompames the woman to the health facIhty and whIch famIly members mteract wIth proVIdersunder normal CIrcumstances and when pregnancy or obstetnc problems occur


SEPTEMBER 15 199727

STEP 3 ACCESS TO CARE

STEP] ACCESS TO CARE PERCEPTION OF CAREGeneral r---------------------,

Neither the SSI lD IndonesIa, nor theMarsaban and colleagues do not report summary report by Marsaban and othersany FG data on perceptions of care lD provide much lDformatIon on perceIvedIndoneSia Results from one closed-ended quahty ofcare The httle data avaIlableItem on the SSIs show women report lDdicate women have confidence lD thehavmg confidence m the skIll of midwIVes b,dans and bldan d, desas, but theseGIVen that an entIre concept cannot be

findmgs should be Interpreted withassessed usmg only one Item, these findmgs caution. The SSIs In BolIVIa contaInedshould be Interpreted with cautIOn No

more Items on perceived qualIty, but theseother data on perceptIOns of care are aVailable

Items had problems wIth content vabdityfor IndoneSia Focus groups In BoliVIa

Focus groups were a much richer source ofprOVided valuable mformatlOn mdlcatmg data Results indIcate that famlhes are notthat treatment (mterpersonal) at health

comfortable wIth male prOVIders, and seefacilities IS perceived to be poor In sum, th t t t " ld" Th d t Ielr rea men as co e a a a sothe SSIs were not a good source of show that women who had attendedmformatlOn about perceived quality ofcare at health facilities prenatal care had more confidence m the

health system than those who had not

IndoneslQThe SSI contamed one closed-ended, numeriC questIOn on confidence m the VillagemidWives' skIll More than three-quarters of the respondents reported that they were'convInced" of the VIllage mIdWIves' SkIll The SSI also contaIned some open-ended, shortanswer questIons related to perceIved servIce qualIty Because these data were non-numenc (andIn Bahasa IndoneSIan), for the purposes of thIS report, they could not be analyzed Hence theonly mformatIOn aVailable on women s perceptIOn of the qualIty of care receIved comes from thereport on the commumty dIagnOSIS by Marsaban et al However, the authors only gIve the resultsof the numenc questIOn regardIng confidence In the SkIll of the bldans/bldan dl desas 33 WhIlethe focus group Instrument contaIned questIOns on perceIved problemsJ WIth care at healthfaCIlItIes, these are not reported In the summary by Marsaban et al Smce perceived qualIty ofcare cannot be assessed usmg only one Item, much regardmg thiS domam remamsunknown

IN B Even If more InformatIon on perceIved qualIty of care were avaIlable from the focus groups, the FGItems related to perceIved qualIty of care were bIased and leadIng For example, women were not asked to talkabout the thIngs they lIked or dIslIked about theIr treatment at health faCIlItIes, or about what they thought ofthebldans Instead the women were asked If they noticed any problems or defiCIenCIes In the health servIces, If therewas somethIng they really dIslIked about the health faCIlItIes and If they conSIdered the btdans and btdan dl desasto be too young


SEPTEt/BER 15 199728

BolzvzaFocus groups were a much ncher source of mformatIOn regardmg perceptIOns of care than werethe SSIs In the FGs, many women expressed dIscomfort wIth male health care provIdersand stated that theIr husbands were not happy that they were exposmg theIr '"mtImate parts"34 tomale health care workers Further, accordmg to Seoane et al both husbands and WIves feltmore comfortable mteractmg WIth tradItIOnal bIrth attendants (parteras) because they have"better mterpersonal trust"35 WIth parteras and because women can remam clothed

In the focus groups, both women and men, partICularly those from low servIce use areas,expressed the expectatIOn of '"bad," "mhuman," or' cold" treatment m the health faCIlItIes Onewoman stated that the doctors '"don't respect our customs," and that they wash women WIth coldwater and gIve cold food 36 GIven the WIdespread perceptIOn that exposure to cold can lead tosobreparto, such treatment may be perceIved as threatemng

Accordmg to the data from the SSIs, women appear satIsfied wIth the qualIty of care theyreceIved at health facIlItIes However, these findmgs stIll should be mterpreted WIthcautIOn Items deSIgned to assess qualIty were vague and may not adequately have reflected theconcept they were mtended to measure WhIle Items from the SSIs WIth husbands m BolIVIa alsowere explored, large numbers of mIssmg data make mterpretatIOn dIfficult In some cases somuch data were mIssmg, that varIables could not be analyzed Husbands dId report bemggenerally pleased WIth the treatment they receIved at the health center, post, or hospItal whentheIr WIves went there for problems m pregnancy, though not all felt they were seen m a tImelymanner or that theIr problems were properly resolved Though lImIted data are aVaIlable forother stages of pregnancy, those husbands who dId respond to questIOns about qualIty of care forproblems m delIvery post-partum or WIth the newborn expressed less satlsfactIOn However,these results must be mterpreted WIth cautIOn for the reasons dIscussed above The Itemsassessmg qualIty are of questIOnable valIdIty and deference effects-m whIch respondents gIveanswers they belIeve the mterviewer wants to hear-also are pOSSIble

1mplzcatlOns for Future Commumty DzagnosesFocus groups should contmue to be used to obtam mformatIOn on perceIved qualIty of care Indepth mtervIews can be a valuable source of mformatIOn, but data collectIOn mstruments mustassess perceptIOns of care usmg multIple questIOns AddItIOnally, clear dIstmctIOns should bemade between perceptIOns of regular care and perceptIOns of emergency care durmg pregnancy,delIvery, the post-partum penod, and WIth the newborn



In IndoneSia, study commuDitles werestratified accordmg to distance from thehealth center Most women reported thatthey could reach the sub-health center(health post) wIthm 15 mmutes. Mostwomen also perceived the hospital orhealth center as far away, (though abouthalf said they could reach the healthcenter m less than 30 minutes). In BolIVia,most women, regardless ofwhether theycame from a commuDity wIth high or lowservice use, said that the health center orpost was close-by These findmgs suggeststhat distance to the health faCIlIty may notbe the most critical factor m deClSlonmakmg

STEP 3 ACCESS TO CARE POTENTIAL B4RRJERS DIST4NCE.---------------------,IndoneSIaWithm each dIstnct, the commumtIesparticipatmg m the commumty dIagnosIswere stratified accordmg to whether thehealth center was nearby or far away Thecntenon for clasSIficatIOn was not consIstentCommumtIes were termed "near" If thedIstance to the health center was withm "easydIstance" by foot or by pubhc transport "Far"commumties were those that were one or twohours from the health center by pubhctransport ~7 However, Marsaban andcolleagues also stated that "a commumty mavbe only two kIlometers away from the healthcenter and IS labeled 'far' because It IS thefurthest commumty m that area "38

Commumties also were stratIfied accordmg tomtensity of servIce use In theIr report,Marsaban and colleagues state that level of attendance was determmed by staff at theMotherCare South Kahmantan office, but she gIves no further mformatIOn on stratIficatIOncntena Thus, some areas were far from health facIhtIes, but had hIgh servIce use and otherswere close to health facIhtIes, but had low servIce use

J\ I~r, \.l.Regardless of the ktratdicatlOn, most women from two of the districts said the sub-healthcenter (puskesma~u)was close-by, and most women m all three districts saId theycould reach the sub-health center wIthm 15 mmutes In all dIstncts, a surpnsmg number ofwomen (more than a quarter ofthe total sample) Said they dIdn't know whether the sub-healthcenter was close-by or not, but very few reported not knowmg whether the health center(puskesmas) was near or far These findmgs may mdicate a problem WIth the questIOn, or It maybe reflective of low serVIce use It also IS pOSSIble that women used servIces at the sub-healthcenter and SImply dId not perceIve dIstance as Important DespIte stratIficatIOn by dIstance, mostwomen m two dIstncts reported that the health center was far ThIS suggests that the dIstancestratIficatIOn cntena may have led to maccurate claSSIficatIOns Most women, regardless ofdIstrict, stated that the hospItal was far from them Very few reported not knowmg whetherthe hospItal was near or far In IndoneSia, women reported gomg to the sub-health center on footand about one-quarter reported walkmg to the health center as well About half saId they couldreach the health center In less than 30 mInutes and about one-quarter saId they could reachthe hospItal In 30 mInutes All the women who reported that It took more than an hour to reachthe hospItal were from one dIstnct



BolzVIaIn BolIVIa, SSI partIcIpants were asked how long It took for them to get to the health facIlItyusmg whatever source of transportatIon they had aVaIlable Women from low service use areasreported that It took longer for them to reach their destmatIon than It did for women fromhigh service use areas However, because many responses were mlssmg, findmgs should bemterpreted with cautIOn When women were asked theIr perceptIOn of the dIstance betweenthe commumty and the health center or post, most women (regardless of whether they came froma hIgh use or low use commumty) SaId that the health center or post was "close by" ThIS IS oneof the Items for WhICh very lIttle data were mlssmg, lendmg credence to the hypothesIs thatperceIved dIstance to the health faCIlIty IS less Important m the process of decisIOn-makmg thanare other factors (e g , cost)

ImplzcatlOns for Future Communzty DIagnosesWhen gathenng mformatIOn on dIstances to health facIlItIes, questIOns should be desIgned tocollect data on and dlstmgUIsh betwen real and perceIved dIstances

In IndoneSia, the SSIs did not prOVidegood mformatlOn on cost of emergencyservices and Marsaban and colleagues donot cite any cost-related focus group datam their summary report While theaverage cost of delivery services mIndoneSia IS known, no conclUSIOns can bedrawn regardmg the degree to which costIS a barrier to health services utilizationIn BoliVia, the SSIs prOVided littlemformatIon regardmg cost oftransportatIOn and services. However,data from the focus groups demonstratethat cost of services and related expensesare a critical barrier to "regular" andemergency care m areas with low healthservices utIlization and to emergency carem areas With high health servicesutilIZation

STEP 3 ACCESS TO CARE COST 4ND REGUL 4RlTl OF TRANSPORTATIOlV, COST OF SER~ICES

IndoneSIaData on transportatIOn costs to healthfaCIlities suffer from large numbers ofmlssmg responses Of those women whodid respond, only a few said that It costmore than Rp 1,000 (less than 50 cents) toget to a health faCIlity

Women were asked about the cost ofantenatal care services and not about costof services when compllcatlOns arise Morethan three-quarters of the womenrespondmg said these services cost betweenRp 0-1000 (less than 50 cents) GIven thatthe per capIta mcome m IndonesIa m 1994was $880 US, both servIce andtransportatIOn costs appear affordable (thougheven mlmmal costs may serve as barrIers tothe very poor) These results should bemterpreted WIth cautIOn due to problems WIthmlssmg data, and gIven that the questIOns dIdnot address emergency servIces In theIr report on the commumty dIagnosIs, the authors reportfindmgs from the SSI, but no focus group mformatIOn IS proVIded Marsaban and colleaguesstate that the maIO barrier to health seekIng behaVIOr IS "the lack of recogmtlOn of hfe-


31

threatemng complicatIOns and the lack of money for transportation and treatment at thehospItal "k However, based on the SSIs and summary report, It does not appear that thecommumty diagnosIs yIelded enough mformatIon to determme how cost IS a factor mdeclslon-makmg

BolIvzaThe SSI questIOns on transportatIOn cost and regularIty, and perceIVed comfort contamedtoo much mlssmg data to conduct meanmgful analyses The large number of mIssmg data ISdue m part to the fact that the pertment Items were asked only to women who had used pnvate orpubhc transportatIOn to go to a health facIhty for comphcatIOns dunng pregnancy, labor/delIverythe post-partum penod, or wIth theIr newborn Of the four women from low servIce-use areaswho responded to the questIOns on cost, two reported costs of less than U S $1 00 and tworeported spendmg US $400 or more Of the women from hIgh servIce-use areas who answeredquestIOns on cost, none reported spendmg more than US $200 Because so few womenresponded to questIOns on transportatIOn, extreme cautIOn should be used m mterpretatIOn

In the SSIs m BohvIa, only those women who went to a health post, center, or hospItal wereasked about the cost of servIces and medicmes In both hIgh and low servIce use areas, resultswere very sImIlar In each area, about half the women who had used the servIces stated thatthey could afford servIces or drugs, and about half saId they could not No data on thesevarIables were avaIlable for husbands As was the case wIth many questIOns, a substantIalnumber of responses were mIssmg

WhIle the SSIs yIelded only hmited mformatIOn, the focus groups were a good source of data oncost and perceptIons of cost Accordmg to Seoane et al women from areas where servIceutIhzatIOn was hIgh conSIdered fees for regular care (not for complIcatIOns) to be affordableWomen from areas where servIce use was low eIther beheved the cost was too hIgh, or Said Itwas unaffordable at any pnce As one woman stated

The money zs the most difficult to obtain, that's why we don't go tothe health center, beszdes, there are other babzes at home to be fedThe MInzstry ofHealth should gzve free delzvery care but zt zs notSOl

When discussmg costs of care for obstetrIc complIcatIOns, women from both hIgh and low usecommunItIes perceIve the costs to be hIgh Accordmg to one husband

For lack ofmoney we sometzmes don't take the woman to theHospztal, zt s that there, they charge us for everything, even the azr

"Marsaban, 1996 (p 2 ExecutIve Summary)

ISeoane et al , 1996 (p 122)



we breathe That's why sometImes we have to resIgn ourselves tolosmg our companzon because we are poor m

In sum, cost-related data from the Bolivian SSIs are mconcluslve, but the focus groups aremore revealing Cost of services and related expenses (e g, drugs), emerges as an Importantbarrier to "regular" care and to emergency care m areas where service use IS low Costalso IS an Important barrier to emergency care where service use IS high

In IndoneSIa, family members other thanthe husband sometimes help care forchIldren when women generally go tohealth facilItIes. In many cases, thechIldren accompany the mother InBoliVIa, women report leavmg theIrchIldren with husbands, parents, andother famIly members. Data from thefocus groups show that famIly membersoften provide finanCial aSSIstance

STEP 3 ACCESS TO C4RE HELP FROU THE EITENDED

F4MIL}

IndoneSIaData from SSIs WIth women m IndoneSIa show thatwhen women go to health faCIlItIes, family members(generally the woman's mother, or an older child)sometimes playa role m carmg for young childrenOnlv a few women Said that theIr husbands or any otherfamIly member look after the chIldren when they go tothe health center However, almost half the womenrespondmg Said that they take theIr chIldren WIth themWhIle the focus group gmdes contam a sectIOn on commumty aSSIstance and chIld carearrangements, the subject IS not dIscussed m the summary report by Marsaban et al

BohvIaIn BoliVia, women reported leavmg their children With their husbands, parents, or otherfamIly members (all wIth about equal frequency) Focus group data reveal the Importance ofthe extended famIly m provIdmg help not Just m carmg for young chIldren, but m canng for themother dunng the post-partum penod AddItIOnally, famIly members sometImes prOVIde moneyto help defray some of the costs of health servIces, partIcularly when the costs are exorbItant,whIch may be the case WIth certam complIcatIOns 39

STEP3 ACCESSTOC4RE HELPFROHTHE

COMML \IT>

IndoneSIaIn IndoneSIa, more than one-quarter of thewomen partIclpatmg In the SSIs reported

Accordmg to the SSIs m IndoneSIa andBolIVIa, most women report they dId notreceIVe help from the commuDIty InBolIVIa, those who saId they dId receIveaSSIstance generally got help In findmgtransportatIOn.

mSeoane et al 1996 (p 123)


SEPTEHBER 15 199733

that they receIved help from a neIghbor or someone In theIr commUnIty A lIttle more than aquarter of the women reported that they dId not receIve any such help WhIle women were askedto descnbe the kmd of help they receIved, thIS data remams In text form and could not beanalyzed herem Marsaban and colleagues do not dISCUSS commumty aSSIstance In theIr report

BohvzaIn BohvIa, more than half the women who were mtervIewed and who reported problemsdurmg pregnancy, dehvery, the post-partum perIod, or WIth the newborn saId that they dIdnot receIve aSSIstance from anyone m theIr communIty Those who dId report gettmg helpsaid that they were assIsted by a member of the health professIOn, or by a famIly member otherthan theIr husband Only two women said they got help from a frIend or neIghbor In theIrreport, Seoane and colleagues state that there IS no formalIzed system of communIty aSSIstancefor obstetnc emergencIes 40 Data from the focus groups support thIS claim Help fromcommunIty members conSIsts prImarIly of aSSIstance m obtammg transportatIOn or In

findmg a health provIder Further, Seoane et al state that whIle commumty leaders consIder IttheIr role to organIze commumty aSSIstance, they fail to do so


SEPTE\1BER 15,199734

Strength Some valuable mformatIOn onproblem awareness was obtamed throughthe semI-structured mtervlews

L,m,tatwn Issues With Method ChOiceThe "semI-structured mtervlews" reallywere structured mtervlews If there ISscant mformatIOn about a tOPIC area andIf data collectIOn IS Just begmnmg,structured mtervlews are not Ideal

SECTION IV-STRENGTHS AND WEAKNESSES OF THE SEMI-STRUCTURED

INTERVIEWS AND Focus GROUPS CONDUCTED IN INDONESIA AND BOLIVIA

In thIS sectIOn, the strengths and weaknesses of the communIty dIagnosIs methodology,mstruments, and data management procedures are outlmed DetaIled mformatIOn regardmg theconstructIOn or analyses of specIfic Items m the mstruments can be found m AppendIX A ThesectIOn below also contams a general overvIew of some methodologIcal conSIderatIOns mherentm usmg the qualItatIve methods (1 e , SSIs and FGs employed m the communIty dIagnOSIS) ThedIScussIon of the strengths and lImItatIons IS not comprehensIve Issues of cost, burden of datacollectIOn and analYSIS, and detaIls on mtervlewer trammg are not dIscussed Furthermore, notall methodologIcal conSIderatIOns are raIsed

SEUI-STRLCTURED Ilv TER VIEWS IN IIVDONESIA AND IN BOLl~14

Strength The SSIs m BolIVIa and IndoneSIayIelded some Important mformatIOn regardmgperceIved seventy of certam neonatal andpregnancy-related complIcatIOns The semIstructured mtervlews also were useful forIdentIfymg some of the problems womenknew about, but WhICh fall outSIde the purvIew of western medlcme (e g ,sobreparto)

LlmlfatlOn Issues wlth Method CholceSemI-structured mtervlews are partIcularlyuseful when the study partIcIpant cannot beremtervlewed,41 as was the case m bothBolIVIa and m IndoneSIa In semI-structuredmtervlews a pre-prepared gUIde IS used toshape the mtervlew format and to obtammformatIOn on a select number of tOpICSTOPICS generally are covered m a partIcularorder The gUIde IS not ngld IntervIewees are not asked to respond to a partIcular set of stImulIor to closed-ended questIOns In addItIon, the mtervlew structure IS suffiCIently fleXIble to allowdISCUSSIOn to be gUIded by the study partICIpant 42

Structured mterviews follow a more ngid pattern As stated by Bernard, the goal of structuredmtervIewmg IS to "control the mput that tnggers each mformant's responses so that output can be


SEPTEMBER 15 199735

LimitatIOn Issues With SamplmgPurposive samplmg, while appropriatefor true m-depth mtervlews and semistructured mtervlews, yIelds results thatcannot be generalIZed to the populatIOn atlarge The use of thIS samplIng techmquem conjunction With admlDlstratIon of aquestionnaire contammg mostly closedended questIOns was not Ideal

relIably compared"o Structured mtervIewmg can take on many forms (e g , free-lIstmg or pIlesortmg) The SSI mstruments used m the communIty dIagnoses were Just one example of astructured mterview Study partIcIpants responded to closed-ended questIOns and to open-ended,short-answer questIOns Structured mterviews are best when a sufficIent amount of trustworthymformatIOn about a tOPIC area already eXIsts 43 They may be used to gather more detaIledmformatIOn about a partIcular tOpIC area and to refine understandmg of a partIcular Issue (e g , acertam belIef, behaVIOr, or process)

Based on a vanety of conSIderatIOns and constramts, the decIsIon was made to conduct SSIsusmg a structured questIOnnaIre However, as dIscussed above thIS form of data collectIOn IS notIdeal for gammg a basIc understandmg of Important values, norms, belIefs, and behaVIOrs WhIChcannot readIly be observed Respondents must mold theIr responses to fit the researcher spreconceIved categorIes As Patton states, the respondents true meanmg can be dIstorted becausetheIr response choIces are constramed 44

Lzmltatzon Issues wIth SamplmgThe aim of qualItatIve research IS toIllummate the questIOns of mterest, (e g , tounderstand how women VIew the post-partumpenod, what behaVIOrs are prohIbIted orencouraged, what behaVIOrs they engage mand why) Samplmg m qualItatIve research ISconducted to maXImIze mformatIOnrIchness 45 Study partICIpants are selected notbecause they are representatIve of the broaderpopulatIOn, but because they are lIkely toprOVIde Important and valId mformation 0 InqualItatIve research, "the valIdIty, meanmgfulness, and mSIghts generated from qualItatIvemqmry have more to do wIth the mformatIOn-rIchness of the cases selected and theobservatIOnal/analytIcal capabIlItIes of the researcher than wIth sample SIze "P

O(Bernard p 237)

o In fact, m many qualItatIve research technIques (e g key-mformant mtervlews) mformants often areselected because they are not representatIve of the broader populatIOn (e g they may be communIty leaders or theymay be UnIquely mSlghtful, analytIcal, and artIculate) These mformants generally have the abIlIty to work wlthmand understand multIple cultures As Crabtree states "These mdlVlduals by theIr very abIlIty and wIllIngness tostraddle two cultures often do not represent the natIve culture" (Crabtree, p 75)

P(Patton p 185)


SEPTHfBER 15, 199736

The samplIng strategIes employed m qualItatIve and quantItatIve data collectIOn methods aredIfferent because the research goals dIffer In quantItatIve research, the aim IS to generalIze studyfindmgs to a broader populatIOn Hence, the study sample must be representatIve of that broaderpopulatIOn Such a sample IS obtamed through some method of random samplIng

The samplIng strategy used m the commumty diagnosIs was qUIte appropnate for true semIstructured, or unstructured mtervIewmg, but was less SUIted for the structured technIqueemployed m the SSI The SSIs yIelded substantial quantItatIve mformatIOn, but the purpOSIvesamplmg (and small sample SIze) lImIted ItS value As mentIOned earlIer, results cannot beconsIdered representatIve of the broader populatIOn of BolIvian or IndonesIan women ofreproductIve age Exact percentages Call1lot be reported smce they are not mdicatIve of the truefrequency dIstnbutIOns m the broader populaTIon Hence, the value of the quantItatIvemformatIOn was dimmished

Some concepts may not have beencaptured adequately In the SSIsbecause questIOns were wordedambiguously or because therewere too few Items used torepresent a domaIn

•

LimitatIOn Issues With Instrument DeSign• There were differences In the

questIOn wordIng and InstrumentdeSign of the SSIs In BoliVia andIndoneSia ThiS variationprecluded direct comparisonbetween the two countries

LlmltatlOn Issues WIth Instrument DeSign(A) comparabllzty Because MotherCarewanted to generate mformatIOn that could becompared across study SItes, staff tned todevelop two SSI mstruments for bothcountnes, each contammg the same Items andm the same order However, m the end, themstruments vaned sIgmficantly, thus lImItmgcomparabIlIty of the quantItatIve mformatIOn

(B) content vahdity Some of the SSI Itemshad problems WIth content valIdIty AsdIscussed prevIOusly, content valIdIty refersto the extent to WhICh Items m thequestIOnnaire are representatIve of theconcepts they are meant to reflect In the SSI,the content valIdIty of certam Items was lowfor two reasons FIrst, questIOns sometImeswere worded ambIguously Second, because researchers were concerned about thequestIOnnaire'S length, efforts were made to lImIt the number of Items asked As a consequence,m some cases, the number of questIOn Items was msufficient to cover a concept adequately

(C) ambIgUity In both the IndoneSIan and BolIVian verSIOns of the SSIs, the meanmg of somequestIOns was unclear, or subject to multIple mterpretatIOns For example m BolIVIa, wherepost-partum famIly pla11llmg questIOns were mcluded m the Safe Motherhood questIOnnaIre,women were asked, "If a woman wants to use a method ImmedIately followmg delIvery, whIchmethod do you thmk IS the most appropnate?" Because the penod "ImmedIately followmg

F \nnachbar\comdx\report\\final4 wpd


delIvery" IS open to mterpretatIOn, the Item IS neIther relIable nor valId, and the responses cannotbe analyzed m a meanmgful way

LimItatIOn Data Codmg and AnalysIsIn Indonesia, most Items contamed only a fewmIssmg responses In BolIvia, a sIgmficantlylarge proportIOn of the responses weremIssmg, thus precludmg meanmgful analysesFor example, when women were asked wherethey first sought help for problems mpregnancy, responses were obtamed for onlyone-quarter of the women who reportedexpenencmg a pregnancy complIcatIOn

LimitatIOn Data Coding and AnalySIS• For some questIOns, partIcularly

m BolIvIa, large numbers ofmIssmg data rendered certamItems mvalId

• Some responses were codedambIguously, thus prohlbltmgmeamngful analyses andmhlbltmg mterpretatIOn

In BolIVIa, the focus groups were anextremely rIch source of mformatIOn

WIthout seemg the data, It IS not pOSSIbleto determme whether the focus groupswere usefulm IndoneSIa

Sometimes, whIle the questIOn wasstraightforward, the response categones wereambIguous, or were too broad For example, m BolIVIa, whIle the SSI mstrument separated thecategones "don't know" and "no response," these categones were combmed when the data werecoded One goal of the SSI was to determme people's knowledge By combmmg nonresponders WIth those who reported "don't know," mformatIOn on knowledge IS lost Mergmgthe two response categones creates other problems as well When the "non-response' category IStoo large, the valIdity ofthe Item declmes When "don't know" responses are mcluded wIth' noresponse" answers, It IS not pOSSIble to tell whether non-response IS a problem

Focus GROUPS Ilv INDONESIA AND BOLlVI 4

General CommentsMotherCare DC does not have the transcnptsof the focus groups m BolIVIa, but the reportby Seoane et a1 contams a great deal of focusgroup data On the baSIS of thIS report It ISclear that the focus groups m BolIVIagenerally were a good source of mformatIOnregardmg perceptIOns of pregnancycomplIcatIOns, the role of varIOUS actors mthe declsIOn-makmg process, perceptIOns of the effectIveness of the tradItIOnal and mstItutIOna1health systems m handlIng complIcatIOns, and for Illustratmg how commumty members perceIvethe servIces they receIve at health facIlIties



LimitatIOn Focus groups are subject toresponse effects and deference effects,

MotherCare DC does not have the transcnpts of the focus groups from IndoneSIa The report byMarsaban and others contams vIrtually no focus group data Thus, It IS not pOSSIble to determmewhether or not thIS data collectlOn technIque provIded valuable and new mformatlOn

LlmltatlOn Response EffectsIn focus groups, charactenstics (and hence thevahdIty) of the data obtamed may vary,dependmg on the charactenstics of thepartIcIpants, the mterviewers, and theenVIronments m WhICh the focus groups are conducted LIttle mformatlOn IS gIVen m reports byMarsaban et al , and by Seoane et al , regardmg the actual charactenstics of each group and of themterviewers It IS reasonable, however, to expect that response effects were present, and thatthey may, to varymg degrees, have mfluenced the vahdity of the data obtamed

LzmltatlOn Deference EffectsDeference effects occur when partIcIpants gIve answers they thmk the mterviewer or other groupmembers want to hear Response effects can mteract wIth deference effects For example, focusgroup partICIpants may gIve the "polIte" response to a questlOns If the mterviewer comes fromtheIr commumty, whereas they mIght gIve a dIfferent statement to an outSIder How questlOnsare worded can have a sIgmficant Impact on the deference effect Peterson found that fornonthreatenmg questlOns, slIght changes m wordmg yIelded very lIttle change m the response 46

But as Bernard states, when askmg about more sensItIve tOpICS, small changes m wordmg canhave sIgmficant effects on partIcIpant responses 47

GIven the personal nature of some of the questlOns (e g , those dealmg WIth famIly plannmg,sexually transmItted dIseases and the perceIved SkIll of the health care workers), It IS pOSSIblethat deference effects mfluenced the qualIty of some of the data


SEPT£'I,fBER 15 199739

SECTION-CONCLUSIONS AND RECOMMENDA TIONS

The followmg sectlOn contams (1) a bnefsummary of the findmgs from the commumtydIagnosIs research m BolIVIa and IndonesIa, (2) hIghlIghts of study strengths and weaknesses,and (3) recommendatIOns and conc1uslOns regardmg methods, mstrument desIgn, datamanagement, trammg, momtonng, and supervISIOn, reportmg, and content of field gUIdes forfuture commumty dIagnOSIS research The suggestIOns are not exhaustIve In some cases theywere developed m response to partIcular problems expenenced m the commumty dIagnoses mIndonesIa and m BolIVIa In other mstances, they follow gUIdelInes for the conduct of qualItatIveresearch more generally

FINDINGS

(1) Knowledge/Awareness ofProblems The degree to whIch women and theIr fmmhes In

BolIvIa and IndonesIa recogmze the SIgns of complIcatIOns varIes, dependmg on whetherthe problem of mterest occurs m pregnancy, dehvery, the post-partum penod, or WIth thenewborn In both countnes, there appears to be less awareness of newborn comphcatlOnsthan of complIcatlOns at other stages

In IndonesIa, swellmg IS recogmzed but generally not VIewed as senous The SIgns ofanemIa are wIdely known, though the term "anemIa" IS not Women are aware that Irontablets can alleVIate anemIa-related problems Women m IndoneSIa also know of andconSIder the followmg dehvery problems to be senous long labor, premature bIrth, andhemorrhage Hemorrhage durmg the post-partum penod also IS conSIdered potentIallylIfe-threatenmg though women have trouble dIstmgUIshmg dangerous bleedmg from lesssenous bleedmg

In BolIVIa swellmg IS conSIdered normal Few women mentIOned anemIa or ItSassocIated SIgns and symptoms BolIVIan women were less lIkely to mentlOn problems mlabor and dehvery than m the post-partum penod Premature dehvery, hemorrhage, andrnalparto (bad bIrth) were among the most frequently CIted problems durmg thIS tImepenod

Sobrepm to (relapse after bIrth) IS the complIcatIOn Bohvian women mentlOned more thanany other Sobreparto IS VIewed as havmg varymg degrees of seventy, most hkelyreflectmg the fact that the problem encompasses a WIde range of symptoms

(2) Health-Seekmg DeclslOn-Makmg In both IndoneSIa and BolIVIa, husbands are verymfluential m the deCISIon to seek care In IndoneSIa, most women report seekmg orwantmg to seek the help of mIdWIves or VIllage mIdWIves for comphcatIOns However,data on thIS tOpIC are mcomplete and httle mformatIOn was gathered on the role of


SEPTEMBER 15 199740

tradItIOnal bIrth attendants When IndonesIan women go to health facIlItIes (notnecessanly for treatment of complIcatIOns) neIghbors generally accompany them

In BolIVIa, husbands appear to playa particularly Important role m the declSlon to seekcare as do the women themselves (particularly durmg pregnancy) Mothers-m-Iawalsocan be mfluentIal BolIVIan data on use of tradItIOnal and mstitutIOnal health servIces aresomewhat ambIguous Reports on focus group results from BolIVia mdicate that faith m agIven system of care eIther tradItIOnal or mstItutIOnallargely IS what determmes thedegree to whIch they are used Data from the BolIVIan SSIs, however, mdicate thatregardless of whether women came from hIgh or low mstitutIOnal serVIce-use areas, theytend to use these facIlItIes and tradItional servIces m about equal numbers Fmally, mBolIVIa, husbands most often accompany theIr WIves to the health facIlIty whenpregnancy or obstetnc complIcatIOns anse

(3) Access to Care There IS lIttle mformatIOn aVailable on perceIved qualIty of care mIndoneSIa WhIle some data mdICate that IndonesIan women have confidence m thebldans and bldan dl desas, mformatIOn on thIS tOPIC IS sparse (partIcularly smce no focusgroup data on thIS subject were reported) and may not be relIable Data from the SSIsshow that IndonesIan women generally know the locatIOn of varIOUS health facIlItIes andcan reach some health facIlIty withm 15 mmutes, though most women reported that thehospItal IS far away In IndonesIa, SSI data on cost of servIces and aSSOCiated expensessuffer from problems wIth questIOn wordmg and large numbers of mIssmg responsesFocus group data on the subject were not reported and therefore, conclusIOns regardmgthe Importance of cost as a potential bamer to care cannot be drawn WhIle IndonesIanwomen sometimes report receIvmg help from neIghbors when complIcatIOns arose, thenature of thIS assIstance IS not clear

Data from the BolIVIan SSIs mdicate that women and theIr husbands are reasonablysatisfied wIth care receIved at health facIlIties However, data from the focus groups tella dIfferent and probably more accurate story In the focus groups, treatment wasdescnbed as "mhuman" and "cold" Women and theIr famIlIes also mdicated dIscomfort

'"wIth male proVIders and greater comfort m theIr mteractIOns wIth tradItIOnal bIrthattendants than With mstitutIOnal health personnel WIth regard to dIstance from healthfaCIlIties, BolIVIan SSI partICIpants from low servIce-use areas reported longertransportatIOn tImes to health facIlItIes than dId theIr hIgh servIce-use counterparts,though large numbers of mIssmg responses reqmre that the data be mterpreted cautIOuslyThe semI-structured mterviews m BolIVia revealed lIttle about transportatIOn or cost ofservIces and related expenses However, focus group data mdlcate that cost IS a cntIcalbamer to "regular" care and emergency care m areas wIth low health serVIces use and toemergency care m areas where health servIces utIlIzatIOn IS hIgh In BolIVIa, data fromthe focus groups show the Importance of assIstance from the extended famIly m


SEPTEMBER 15 199741

provldmg chIld care, tendmg to the mother dunng the post-partum penod, and mprovIdmg money to defray the cost of health servIces Women m BolIvIa generally dIdnot report receIVmg assIstance from theIr communIty, though some reported gettmg helpm findmg transportatIOn

STUDy!REPORTING STRENGTHS

• Overall, the communIty diagnosIs exerCIse IS a useful process for IdentIfymg trammg,servIce-delIvery, and mdlvldual/communlty-Ievel needs With regard to maternal andneonatal health m developmg country contexts, and for developmg key IEC/C messagesThe semI-structured mtervlews and focus groups provIded pertment mformatIOn onknowledge/recogmtIOn of problems, declSlon-makmg regardmg health-care seekmgbehaVIOr, access to care, and perceptIOns of the qualIty of care ThIS mformatIOn prOVIdesa base from whIch mterventIOns are bemg developed

• The communIty diagnosIs uses the Pathway to SurvIval as the baSIS for mstrumentdeSIgn The Importance of usmg a framework to gUIde the data collectIOn process cannotbe overemphaSIzed

• In theIr summary report of the commumty diagnOSIS m BolIVia, Seoane and colleaguesmake good use of focus group data to support theIr conclUSIOns

STUDY!REPORTING WEAKNESSES

• Wordmg and formattmg of questIOns and questIOn flow were not consIstent betweenBolIVia and IndoneSIa, so dIrect comparIsons between the two countnes often could notbe made

• There were several problems WIth Items on the SSI and FG field gUIdes SpeCIfically,questIOns were not always worded clearly, smgle Items sometImes were used to representan entIre domam, and questIOns often were closed-ended when the mtentIOn was to elICItdialogue between the mtervlewer and respondent

o

• In the codmg process, mergmg certam response categones led to dIfficultIes mmterpretmg results

• The reportmg from BolIVia and IndoneSia was variable In theIr summary document, theBoliVIan authors present much detaIl regardmg study results, though more mformatIOn onmethodology would have strengthened the document The IndoneSian summary reportcontams mcomplete mformatIOn regardmg findmgs and study methodology Examples ofproblems that can be found m one or both of the reports mclude (1) lack of detailregardmg partICIpant selectIOn cntena (e g , for focus groups), mtervlewer trammg andqualIty assurance efforts, and codmg schemes, (2) mcomplete reportmg (or data analySIS)



of certam responses and response categorIes, (3) msufficient use of raw data (e g , quotesfrom focus groups or open-ended questIOns on the SSIs), (4) reportmg of percentages fordata obtamed from small, purposIve samples and (5) mIssmg mstruments (data collectIOnmstruments were not mcluded as appendIces m the report documents)

RECOMMENDATIONS AND CONCLUSIONS RESEARCH ApPROACH

In some cases, the comments below reflect general prmcipies m the conduct of qualItatIveresearch In others, they reflect Issues specIfic to the commumty diagnosIs research m BolIvIaand IndoneSIa

Methods• A combmatIOn of data collectIOn methods should be employed, e g , both m-depth

mterviews and focus groups Other technIques such as ratmgs, rankmgs, trIads, pIlesortmg and free lIstmg also should be consIdered

• Where quantitatlve mformatIOn IS sought (e g , where populatIOn-based percentages arereqmred), true quantitatlve data collectIOn methods (e g , surveys) should be used Inthese mstances samplmg should be random

• To lImIt deference effects, very sensItlve tOpICS (e g , questIOns on STDs/RTIs or famIlyplanmng) often are best explored wIth m-depth mterviews as opposed to focus groups 4849

Furthermore, mformatIOn on these tOpICS often IS more successfully obtamed m latermterviews wIth the same respondent SpecIfically, Bentley recommends usmg the firstmterview to estabhsh a good rapport, gam trust, and gather mformatIOn on less sensItIveIssues SensItIve tOpICS can then be addressed m a second or thIrd mterview 50

• PartIcIpant selectIOn CrIterIa should be clearly speCIfied and should remam consIstentFor focus groups, efforts should be made to ensure partIcIpants are homogenous acrossthe characterIstIcs of mterest (e g , women who have experIenced a pregnancy or obstetnccomphcatIOn m theIr last pregnancy) Focus groups also should contam at least SIX andnot more than 10 partIcIpants

Instrument DeSign• IndIVIdual researchers and program planners workmg m a gIven settmg WIll need to

modIfy and adapt data collectIOn mstruments to meet theIr mterests and reqmrementsHowever, m order to make cross-country comparIsons, where pOSSIble, MotherCareshould work to ensure Items are Identlcal across mstruments In cases where thIS IS notpOSSIble, MotherCare should stnve for sImIlar wordmg and formattmg of questIOns andquestIOn flow


43

• Efforts should be made to maXImIze questIOn clanty and content vahdity Items shouldbe carefully screened for ambIgUity, should employ local termmology, should representthe concepts ofmterest and be clearly understood by the respondent AddItIOnally,domams should be measured by multIple questIOns rather than smgle Items

• QuestIOns should be open-ended and should not lead the respondent DIchotomousquestIOns, partIcularly m the begmnmg of an mterview or focus group WIll mhibit talkmgand sharmg of mformatIOn and should be aVOided

• If closed-ended questIOns are used wIth prescnbed response categorIes, "don't know'responses should be separated from "no response" answers Double-barreled Items, mwhIch two or more questlOns are embedded m one, should not be used

• Probes should be used hberally They should be clear and precIse "Why" probes shouldbe aVOided, as they Imply a questIOnmg of the vahdity of the mterviewees response

• The format of the mstrument should be such that questlOns flow from the general to thespecIfic ImtIally, tOpICS should be non-controversIal and non-sensItIve, to encouragerespondents to talk and to estabhsh trust and a good rapport It often IS useful to begmmterviews or focus groups With deSCrIptive questlOns, followed by querIes on opmlOnsfeelmgs, and mterpretatlOns QuestlOns on knowledge or SkIlls should be saved untIl arapport between the mterviewer and respondent has been estabhshed 51 DemographIcquestIOns should be reserved untIl the end of the mterview However, If certamdemographIc mformatIOn IS reqUired to ensure that study partIcIpants meet selectIOnCrIterIa an mitIal screenmg questIOn on the demographIc Item ofmterest can be usedbefore the start of the actual mterview

Content• WhIle mstruments need to be developed and modIfied to be context specIfic, theIr deSIgn

should be gUided by a framework or model (e g , the Pathway to SurvIval or the Pathwayto BehavlOr) Efforts m IndoneSia and BohvIa used the Pathway to SUrvIVal as aframework, thIS approach gave valuable mformatIOn on problem recogmtIOn, perceptlOnsof care, declSlon-makmg regardmg care and some envIronmental or structural constramtssuch as cost and transportatIOn Other frameworks also can be used For example,Fishbem's Pathway to BehavIOr (see AppendIx A) emphasIzes these and other factors,such as mtentlOns to perform a gIven behavIOr and the SkIlls needed to do so WhIle theframework has yet to be tested, It mIght prove useful

• General categorIes to consIder for mclusIOn m field gUides are (l) recogmtlOn of SIgns ofcomphcatIOns, (2) key players m the process by WhICh care IS sought, (3) perceptIOns ofthe quahty of care m health facIhties and from tradItIOnal health care provIders, (4) mtent



to use servIces (IncludIng InfOrmatIOn on attItudes, knowledge, behefs, skIlls, and norms),(5) skills (e g , abIhty to negotIate With provIders), and (6) envIronmental constraInts(e g cost, transportatIOn, aVaIlabIhty of servIces, hours of operatIOn) to behavIOrperformance

Data Management and AnalysIs• Careful attentIOn must be gIven to the codIng process If quantItatIve InfOrmatIOn IS

obtamed, "don't know" and "no response" categones should not be merged and, wherepossIble, an "other" category should be aVOIded or mInImIZed In quahtative data, codersalso should be wary of mergIng responses For example, hemorrhagIng and bleedIngshould be coded separately For quahtative InfOrmatIOn, a codIng scheme should bedocumented WIth a descnptIOn of the code and ItS InclUSIOn cntena

• If data are collected, they should be coded and analyzed ThIS IS an extremely tImeconsummg process TIme and resource constraInts must be consIdered In deCISIonsregardIng research methodology and sample SIze

Trammg, Momtormg, and SupervlSlon• The use of effectIve probmg technIques IS cntIcal to successful quahtatlve IntervIeWIng

IntervIewers and focus group moderators should receIve IntensIve traInIng on probIng andshould receIve contInuous performance feedback throughout the data collectIon process

• Note-takIng durIng IntervIews and focus groups IS cntIcal, even when seSSIOns arerecorded on cassettes Notes should Include-among other thIngS-IndIcatIOns ofgestures and tones, mood, who IS speakIng, who IS not speakIng, a descnptIOn of thesettmg and the partICIpant, and dIrect quotes from the respondent Notes should bereVIewed on a regular baSIS and appropnate feedback gIven to the IntervIewers or focusgroup recorders

• Where respondents are remtervlewed one or two tImes to obtaIn more senSItIveInfOrmatIOn, the InItIal IDI cassette and notes should be reVIewed so that the IntervIewercan follow-up on Issues that reqUIre more probIng or clanficatIOn

• The data management process must be supervIsed to ensure accurate codIng of dataWhere possIble, as a method of quahty control, two people should code the same segmentof matenal, thIS segment should be compared, and-If appropnate-modlficatIOns IncodIng procedures should be made

Reportmg


SEPTEMBER 15 199745

• Adequate descnptIOns of study methodology (e g , partIcIpant selectIon cntena focusgroup composItIOn, IntervIewer traInIng and supervIsIon, codIng technIques) must becontaIned In the summary report of commumty dIagnosIs efforts

• Care should be taken In the reportIng of percentages PurposIve samplIng and the smallsample SIzes often found In qualItatIve research generally preclude meanIngful reportIngof exact percentages


SEPTEMBER 15 199746

ApPENDIXA FISHBEIN'S PATHWAY TO BEHAVIOR-A FRAMEWORK FOR POSSIBLE

USE IN FUTURE COMMUNITYDIAGNOSIS RESEARCH

Pathway to Behavior

emotion~ -

~B

EXTERNAL behavioral '- attitudesVARIAB_LE~ II beliefs skills E

• demographics H• mdlvldual

differences '- normsJ '- -I- Anorms Intention• personality beliefs

• perceived risk

\ !JV

• attitude andvalues efficacy self environmental

I• culture -beliefs efficacy constraints 0

R~self -

Fillur.2 standards

SouIce Fishbein ne7

F \nnachbar\comdx\report\final\appenda1SEPTEMBER 15, 1997 47

ApPENDIXB. IDENTIFIED GAPs/AREAS OF CONFUSION

After havmg conducted the commumty dIagnosIs and after gammg addItIonal expenence mBolIvIa and IndonesIa, It IS possIble to reVIew MotherCare's commumty dIagnosIs research andshed new lIght on questIOns or tOpIC areas related to the Pathway to SurvIval that eIther were notaddressed m the qualItatIve research or were not answered defimtIvely

Below IS a lIst of those gaps or areas of confuSIOn AddItIOnal data collectIOn IS not necessary toobtam answers to every questIOn In fact, no one research endeavor needs to or should addressall the tOpICS dIscussed below In some mstances, answenng the questIOns raIsed WIll reqmrenew research In others, prevIOUS research (unrelated to the commumty dIagnosIs), or extantlIterature may proVIde adequate mformatIOn The lIst IS neIther narrow nor exhaustIve, somedomams or Issues are more Important for program desIgn and ImplementatIOn than others Also,as addItIOnal research IS conducted and programs are deSIgned, some questIOns WIll be answeredand many more w111 lIkely anse

N B , thIS lIst of gaps pertams to mformatIOn related to the Pathway to SurvIval Were thePathway to BehaVIOr also used as a base for mqmry, other gaps could be IdentIfied, partIcularlythose related to the mtentIOn to perform a behaVIOr and the skIlls reqmred to do so

PROBLEM RECOGNITION

Domam IdentIfied GaplArea of Confusion

Pregnancy PerceptlOns/Practices · Workload nonns and proscnptlOns

· Health promotmg/mhlbltmg behavIOrs, (e g , benefiCial orhannful traditIOnal practices, prenatal care)

· InfonnatlOn sources (I e, how do women know what they shouldand shouldn't do durmg pregnancy?)

· Diet What foods are proscnbed/prescnbed and why?

· Signs and symptoms that are conSidered nonnal" (e gswellIng)

Swelling PerceptIOns/Practices · What women do about swellIng

AnemIa Perceptlons/Practlces · Indonesian women's belIefs about the causes of kurang darah

· IndoneSian women's belIefs about the causes of tekanan darahrendah

· How (and It) IndoneSian women dlstmgUlsh between kurangdarah and tekanan darah rendah

· Reasons why women won't take Iron pills throughout pregnancy

· Whether women Will eat more dunng pregnancy

· What IndoneSian women do about kurang darah, tekanan darahrendah

· What BoliVian women do about anemia or ItS symptoms

48

Domam

Certam Delivery ProblemsPerceptlons/Practlces

Certam Post-Partum ProblemsPerceptions/Practices

Certam Neonatal ProblemsPerceptions/Practices

Neonatal DeathsPerceptions/Practices

IdentIfied Gap/Area of Confusion

Symptoms and perceived causes of malparto m BoliviaHow women/others dlstmgUlsh between delivery and post-partum(as opposed to clImcal defmltlOns)How women/others dlstmgUlsh between hemorrhage andbleedmg (or between "normal" bleedmg and "excessive"bleedmg)Perceived link between retamed placenta and oral contraceptives(m Indonesia)Importance of retamed placentaHealth promotmg/Inhlbltmg behavIOrsWhat women do about hemorrhage, bleedmg, or other deliverycomplicatIOns

What women do to treat sobreparto (m Bolivia)List of symptoms associated With sobreparto and their perceivedseverityInformatIOn on other problems (from the woman's or family'sperspective) Examples mclude pasmo, arrebato, andsuspensIOn de la sangre!InformatIOn on dietary prescriptIOns/proscriptIOnsHealth promotmg/mhlblting behavIOrsWhat women do about other post-partum problems

PerceptIOns of "normal" newborn characteristics (e g , color,breathmg)InformatIOn on newborn careInformatIOn on mfant feedmg practlces/behefs (e g colostrumtlmmg, techmques)InformatIOn on other health promotmg/mhlbltmg behaVIOrsWhat families do about neonatal problems

How (and It) women and their families believe neonatal deathscan be preventedCurrent practices to prevent neonatal deathsWJilmgness to do certam thmgs (e g warmmg, exclUSive breastfeedmg clean cord care, etc) to prevent death

I These Illnesses and their cures are deSCribed by CIAES m their report to MotherCare (The Center forHealth Research ConsultatIOn and EducatIOn (CIAES) 1991 Qualztatlve Research on Knowledge Attztudes andPractices Related to Women s Reproductive Health MotherCare Project Cochabamba, BoliVia)

49

DECISION-MAKING

Domam IdentIfied Gap!Area of Confusion

DeClSIon-Makmg Who are the · The husband's role m declSlon-makmg when neonatalCntical Players? complIcatIOns anse

DecISlon-Makmg Whose help IS · InformatIOn on whose help IS sought (and m what circumstances)sought? when the TBA IS already present

· InformatIOn on whose help IS sought (and m what circumstances)when the TBA IS not present

· SpecIfic mformation on when a Bldan or Bldan dl Desa ISconsidered the best optIOn (IndoneSia)

Declslon-Makmg Who · InformatIOn on who goes WIth the woman or newborn when IIfe-Accompames the Woman to threatenmg complIcatIOns anse (IndoneSIa)Health FaCIlIties? · InformatIOn on who mteracts With health proViders when a

woman goes for regular or emergency care

DecISlon-Makmg Other · The most cntlcal factors mhIbltmg the use of health servicesunder normal and emergency CIrcumstances (IndoneSia)

F\nnachbar\comdx~eport\appendbl

SEPTEMBER 15, 1997 50

ACCESS

Domam IdentIfied Gap/Area of ConfusIOn

Access to Care PerceptIOn of Care • InfonnatIOn on hnles between practices at health facl1lt1es andsobreparto or other lIlnesses (BollVla)

· Specific InfOnnatIOn on the perceived benefits and problems ofhome births versus births In a health facllity for nonnal andcomplicated cases

• How care provIded by Bldans and Bldan dl Desas IS perceived(Indonesia)

· How the care provIded by other health care workers IS perceived(IndonesIa)

· Willingness to move (Influence of perceived quahty of care andeffectiveness of servIces m the deCISion to seek care, Ie, howperceived quahty of servIces mfluences the decIsIOn to seek care)

Access to Care Potential Barners · How dIstance consIderatIOns affect deCISion-making for life-(Distance) threatemng conditions

Importance of dIstance conSiderations verus avallabl1lty, cost, orother factors

Access to Care Potential Barners · How cost conSIderatIOns affect deCISion-making for life-(Cost and Regularity of threatenmg conditIOnsTransportatIOn Cost of Services) · Importance of transportatIOn costs versus avmlabl1lty or other

factors

Access to Care Help from the · Role women want other faml1y members to play, and whenExtended Family

Access to Care Help from the · Current role of the commumty m promotmg maternal health andCommumty In preventmg maternal deaths when hfe-threatenIng

comphcatlOns anse

· Role the commumty IS expected to play

· Role famlhes want the commumty to play

F \nnachbar\comdx\report\appendb 1SEPTEMBER 15 1997 51

ApPENDIX C RECOGNITION OF THE PROBLEM IN-DEPTH INTERVIEWS WITH MOTHERS AND SUMMARY OF INDONESIA/BoLIVIA

REPORTS ON Focus GROUPS

Domam/Item

tA O-RECOGNITION OFTHE PROBLEMKnowledge and Perceptton ofSeventy

Indonesia (90 women mtervlewed) Bolivia (41 women mtervlewed)

SSI Responses were unprompted Virtuallyall the problems mentioned were cited by lessthan one-quarter of the women mtervlewedThe most frequently cited problems werehemorrhage and 50bleparto About half thewomen mentIOned these problems

Comments/Limitations

General (SSls)(I) Sample selectIOn (purposive sampling) andthe small sample size mean that data are notnecessarily representative of the broaderpopulatIOn and that results cannot begeneralized

(2) There were Important differences betweenBoliVia and IndoneSia m how questIOns onproblem recognitIOn and perceived severitywere asked The numbers reported forIndoneSia are from Items where responseswere prompted (e g , "IS anemia a seriousproblem or not?") The numbels reported forBoliVia stem from unprompted responses(e g, what problems occurrmg durmg deliveryare you familiar With?)

IndoneSia (S51) Women were ask.ed to list theproblems they had experienced durmgpregnancy, delivery post-partum, or With thenewborn If the women reported havmgexpellenced no complicatIOns m given stage,they were asked to name the problems Withwhich they were familiar Because these datawere open-ended (and In text form), they couldnot be analyzed for thiS report Only c1osedended SSI questions are discussed

F \nnachbar\comdx\report\final\appendc I


II Domam/Item

RECOGNITION OF THEPROBLEMKnowledge and PerceptIOn ofSeventy (contmued)

IA I-Hemorrhage (all stages)

IA 2-0ther (all stages)

Indonesia

SSI Not calculable for Indonesia

Report (FGs) Discussed for dehvery(see IA 19)

SSI Not available for IndoneSia (notasked m closed-ended questIOns)

BohvIa

SSI Responses were unpromptedHemorrhage (10 any of the stages) was themost frequent problem mentIOned (cited bynearly half the women mtervlewed) TheperceptIOn of the seventy vaned While mostwomen who mentIOned hemorrhage behevedIt was senous, some did not

Report See IA II, IA 19, IA 24

SSI A few women mentIOned problems thatwere coded as "other'

Comments/LimItatlOns

IndoneSia (Report-FGs) It IS unclear fromthe report whether women associatehemorrhage with retamed placenta or whetherIt IS PATH who makes thiS association

Bohvla (SSI) Hemorrhage and bleedmg werecoded together Ifwomen were talkmg aboutbleedmg (as opposed to hemorrhage), thosewho mdlcated "hemorrhage" was not senousmay accurately have assessed the sevellty ofthe problem

Bohvl3 (SSI) It IS not pOSSible to determmewhat conditions were coded as "other '

Report (FGs) Temporary leg paralySIS I Report Not discussedwas mentIOned by some women whoassociated thiS conditIOn with difficultor prolonged labor

IA 3-Pam (all stages) SSI Not available for IndoneSia (notasked 10 closed-ended questIOns)

Report (FGs) Not discussed

SSI Pam was mentIOned by about one-fifthof the women Half of these women thoughtIt was a senous problem while the other halfdid not

Report Pam IS mentIOned 10 the context ofmalparto and sobleparto (see IA 14 andIA 22)

BoliVia (SSI) From the data, It IS not pOSSibleto determme the locatIOn or Significance of"pam



Domam/ltem


IA 4-Spontaneous abortion(mlscarnage)

IA 5-Baby m bad pOSitIOn(BolIvia and IndoneSia)

Baby upSide-down (IndoneSIa)

IndoneSia

SSI Not available for IndoneSia (notasked m closed-ended questions)

Report (SSls) Accordmg to theauthors, most respondents mentionedmlscarnage (unprompted) andconSidered It dangerous


SSI Most women (about three-quarters)Said baby m a bad pOSitIOn was a serIOUSproblem

SSI About three-quarters of the womensaid an upSide-down baby was a serIOUSproblem


BolIVia

SSI Responses were unprompted A handfulof women mentioned spontaneous abortIOnOf these, all but one conSidered It serIOUS

Report Not discussed

SSI Responses were unprompted BadpOSitIOn of the baby was mentioned by lessthan one-quarter of the women A few ofthem Said It wasn't serIOUS

Report Accordmg to the authors,malpOSition of the baby IS conSidered serIOUSand IS seen as a problem m birth that stemsfrom complIcatIOns m pregnancy Theprimary reason women give for antenatalVISitS to a par/era or doctor IS to be SUI e thebaby IS m a good pOSitIOn


IndoneSia (SSI) Women were not askedabout spontaneous abortIOn m IndoneSia Theresults reported by the authors may mdlcatethat mlscarnage IS seen as a threat to thewoman s health It also may be Viewed asserIOUS because It IS a loss Unless thiS tOpiCwas discussed m the focus groups, furtherresearch IS needed to understand theSignIficance women aSCrIbe to mlscarnage

IndoneSia (SSI) The consistency m responsesbetween the two Similar questIOns m IndoneSiasuggests the Items have good content valIdity

BohVIa The findmgs from the FGs arenoteworthy and would have been missed byrelymg on the SSls alone


SEPTEMBER 15 1997 54

Domam/Item

RECOGNITION OF THEPROBLEMKnowledge and Perception ofSeventy (coutmned)

IA 6-Swellmg offace/eyes/feet (Bohvla)

Swelhng of face/hands(Indonesia)

Swellmg of feet (Indonesia)

IA 7-Common problems(pregnancy)

Indonesia

SSI About half the women mtervlewedfelt that swollen hands andface was aserious problem Most others said Itwas not seriOUS, though some statedthey didn't know If It was serious or not

SSI More than a quarter of the womenfelt that swollen feet was a seriousproblem Most others said It was notserious A handful stated that theydldn t know If It was serious or not


SSI Not available for Indonesia (notasked m closed-ended questIOns)

Report (FGs) Anemia was discussed mfocus groups Problems mentionedwere nausea, vomltmg, headaches,dlzzmess, fatigue, weakness, lack ofenergy, and stomach aches/pams

Boltvla

SSI Swelhng was mentIOned by more thanone-quarter of the women, most of whombeheved thiS was not a serious conditIOnVery few husbands mentIOned swellmg

Report Data from the focus groups mdlcatewomen do not see swellmg as dangerousSome view It as mdlcatlve of a favorablepregnancy

SSI 'Common problems" was mentIOned byonly one woman who said It was not serious


Comments/Llmltattons

IndonesIa (SSI) Some women do seem todlstmgUlsh between swollen face/hands andswollen feet

Bohvaa (SSI and Report) Swollenfeet/face/hands were coded together, makmg Ithard to determme whether women dlstmgUlshbetween swellmg that IS mdlcatlve of preeclampSIa/eclampSia and swellmg that may bemdlcatlve of pressure edema (a commonproblem In pregnancy) At the same time, FGdata support the findmg that women do not seeswelhng (mcludmg that of the face and hands)as serious

Bohvla (SSI) Although "common problemswas menttoned by only one woman, It may beuseful to determme what women conSider, common problems' to be



Domain/item

RECOGNITION OF THEPROBLEMKnowledge and PerceptIon ofSeventy (contmued)

IA 8-Anemla (BoliVIa)

AnemIa/Low Blood Pressure(IndoneSIa)

IA 9-Weak/tlred/unpowerful/yelloweyes

1A 1O-Nausea and vomItIng

Indonesia

SSI Most (three-quarters) of the womenIdentified this (anemia/low bloodpressure) as a senous problem Almostone-quarter saId It was not senous

Report (From the Anemia SSI) Morethan one-quarter of the women dId notaSSOCIate symptoms of anemIa (e g, lowenergy, weakness) WIth a speCIfic term(e g , kurang dmah) Over one-quarterof the women associated anemIasymptoms With kurang darah Somewomen gave othel terms for the samesymptoms

SSI The maJonty of respondents SaidthIs was a senous problem About onequarter reported that It was not senousSome did not know


SSI Most women leported that this wasnot senous

Report (FGs) This was mentIOned as acommon problem In plegnancy

Bolivia

SSI Very few women (only 4 of 41)mentIOned anemia All those who mentIOnedIt consldeled It to be a sellOUS problem

Report Not dIscussed (anemia research wasconducted as a separate InvestigatIOn)

SSI Not avaIlable fOl BolIVIa (not CIted Inunprompted questIOns or not coded)

Report Not dIscussed

SSI Not aVailable for BoliVia (not Cited In

unprompted questIOns or not coded)


Comments/LimitatIons

Indonesia (SSI) From the Safe MotherhoodSSls, most women seem to feel anemIa/lowblood pressure (LBP) IS senous, but mostwomen particIpating In the anemIa SSls werenot aware of any speCIfic effects ofanemra onpregnancy/delivery (N B , It IS not clear flomthe report whethel women were asked abouteffects of anemia or If local termInology wasused) InformatIOn on women's perceptIOns ofthe differences between anemIa and LBP IS notavailable

BoliVia (SSI) Fainting, weakness, and anemIawere coded together Even so only a fewwomen mentIOned anemIa or ItS associatedsIgns or symptoms Fewer women mentIOnedthis problem than many others

IndoneSia (SSI) The purpose of this questIonmay have been to get InformatIOn on hepatItiS

IndoneSia (SSI) Responses Indicate that thewomen do dlstmgUlsh problem seventy TheIrresponses also mdlcate some degree of contentvalidIty In the questIOnnaire Items dealing Withproblem seventy

I

F \nnachbar\comdx\report\final\appendc 1


Domam/Item

RECOGNITION OF THEPROBLEMKnowledge and Perception ofSeventy (contmued)

IA II-Hemorrhage mpregnancy

IA 12-Attacks/faIntmg(BolivIa)

ConvulSIOns m Pregnancy(IndonesIa)

DELIVERY

IA 13-Convulslons m dehvery(Indonesia)

Indonesia

SSI Most women (about three-quarters)said hemorrhage m pregnancy was asenous problem The rest reported thatthey either did not know whether It wassenous or not or did not believe It wassenous

Report (FGs) Discussed forhemorrhage durmg delivery (seeIA 19)

SSI Almost three quarters of thewomen said that convulSions mpregnancy were seilOUS Almost onetenth stated they didn't know whether Itwas senous or not


SSt SlIghtly more than three-quartelsof the women mtervlewed said thatconvulSions m delivery were senousThiS figure IS slightly more than thenumber of women who Said thatconvulSIOns In pregnancy were senous


Bolivia

SSI About one-fifth of the womenmentIOned hemorrhage m pregnancy as aproblem Of these, most conSidered It to besenous

Report Data from the focus groups revealthat hemorrhage dunng pregnancy often ISperceived to be caused by liftIng heavythIngs Focus group partiCipants had troubledlstmgUlshmg between lIght and moreabundant bleedmg

SSI Only one woman mentIOnedattacks/famtmg as a problem She conSideredIt to be very senous

Report The authors state that people don'trelate attacks/famtmg with swellingAttacks/famtmg were mentIOned by only oneperson (a partew)

SSI Not available for BoliVia (not Cited munprompted questIOns or not coded)


CommentslLlmltatlons

IndoneSia In the SSIs, It IS unclear If womenwere dlstIngUlshmg between hemorrhagIngand bleedmg In their reports on the SSIs andfocus groups, the authors state that "retaInedplacenta and the resultIng hemorrhagmg wereperceived to be the most dangerous [ofproblems]" It IS not clear whether studypartiCipants made the connection betweenretamed placenta and hemorrhage or whetherthiS aSSOCiatIOn was made by the authors

Bohvla (SSI) Attacks and faIntmg werecoded together, makmg It difficult to tellwhether the woman mentlOnmg thiS problemwas thmkmg more of anemia or of eclampSia



Domain/item IndonesIa BolivIa Comments/LImItatIOns


IA I4-Bad bIrth (Ma/parto In SSI Not avaIlable for IndonesIa (not SSI Almost one-fifth of the women Boltvia (SSI) The lack of agreementBoltvla) asked In closed-ended questIons) mentIOned ma/parto as a problem but there regardmg problem seventy may reflect the fact

was no agreement about Its seventy that ma/parto encompasses a wIde range ofReport (FGs) Bad bIrth IS not sIgns and symptomsdIscussed, though prolonged labor IS Report Responses of vanous focus group(see IA 18) partIcIpants indIcate that ma/parto consIsts of

a vanety of sIgns and symptoms (e ghemorrhage, premature deliverydIfficult/painful delivery, and prolongedlabor)

IA I5-Fever (of woman) SSI About two-thIrds of the women SSI Fever was mentIOned by only onedunng bIrth saId that fever dunng bIrth was a senous woman who thought It was senous

problem Most others saId It was notsenous Report Not dIscussed

Report (FGs) Not dIscussed

IA 16-Early water break SSI Most women (about till ee-quartel s) SSI Not avaIlable for Boltvla (not CIted In

saId thIS was a senous problem The unprompted questIons or not coded)vast maJonty of the remaining womensaId It was not a senous problem Report Not dIscussed




II Domam/Item


IA 17-Cesarean sectIOn

IA IS-Labor> 12 hours

IA 19-Hemorrhage m deltvery

IndonesIa

SSI Most women (three-quartels) felthavmg a cesarean sectIOn was a serIOUSproblem The rest eIther dIdn't know orsaId It was not serIOUS


SSI About three-quarters of the womensaId that labor> 12 hours was a serIOUSproblem Almost all the rest saId It wasnot

Report (FGs) Some women reportedlyassOCIated prolonged labor WIthstIllbirths and neonatal deaths

SSI The vast majorIty of women (wellover three-quarters) saId thiS was aserIOUS problem Only one person SaIdshe didn't know If was serIOUS or not

Report (FGs) Accordmg to theauthors, m the focus groupshemorrhage emerged as one of the mostcommonly perceived problems and themost dangerous comphcatlon Theauthors state that hemorrhage IS hnkedto retamed placenta and that men andwomen blame the perceived mcrease mmCldence of retamed placenta on the useof oral contraceptives

Bohvla

SSI Not avatlable for BohvIa (not CIted munprompted questIons or not coded)


SSI Not avaIlable for Bohvla

Report Some focus group respondentsconsIdered prolonged labor (durationundefined) to be a part of malparto

SSI Only two women mentioned hemorrhagedurmg bIrth as a problem One belteved Itwas serIOUS, while the other dId not

Report Some focus group respondentsconsIdered hemorrhage dUrIng deltvery to bea part of malparto

Comments/LImItatIons

IndoneSIa (Report-FGs) Use of raw datafrom the focus groups would have been usefulto support the authors' contentIOn that famlhesassocIate the perceIved rIse m rates of retaInedplacenta WIth the use of oral contraceptIves

Bohvla (SSI) WhIle only two womenmentIoned hemorrhage durIng deltvery, asubstantIal number mentIOned post partumhemorrhage (see IA 24)



Domam/ltem IndonesIa BolIvIa Comments/LImItatIOns


POSTPARTUM

IA 20-lnfectlOn postpartum SSI Not avaIlable for IndonesIa (not SSI About one-sIxth of the women Bolivia See sobleparto (lA 22)asked 10 closed-ended questIOns though mentIoned postpartum mfectlOn InfectIOnquestIOns were asked on associated was generally perceived as senoussigns and symptoms For example,slIghtly more than half the women Report Data from the focus groups showmtervlewed considered abdommal pam that sobreparto mcludes many of the sIgnsto be senous whIle most of the rest did and symptoms of post-parturn mfectlonnot About one-quarter of the women Sobreparto was one of the most Widelymtervlewed thought that vagmal known complicatIOnsdIscharge wIth a bad odor was senous,whIle most of the rest did not

IA 21-Retamed placenta SSI Not aVailable for IndoneSia (not 881 Retamed placenta was mentIOned byasked 10 closed-ended questIOns) about one-sIxth of the women Half the

women Said thIS was not a senous problemReport (FGs) Accordmg to theauthors, retamed placenta IS reported to Report Data from the focus gloups showbe one of the most frequently that the placenta IS very SignIficant 10 theexpenenced and well-known birth process and that retamed placenta IScomplicatIOns Oral contraceptIves ale seen as the result of premature cuttmg of thecommonly blamed for what people umbilIcal COldconsider to be a nse 10 mCldence ofretamed placenta



II Domam/Item


IA 22-Sobleparto

IA 23-Bleedmg postpartum

IndoneSIa

SSI Not avaIlable for IndonesIa(BohvIan concept)

Report (FGs) Not dIscussed (BolIvIanconcept)

SSI More than three-quarters of thewomen saId post-partum bleedmg wasnot a senous problem As results fromthe questIon on postpartum hemorrhagemdlcate (see IA 24), women dlstmgUlshbetween bleedmg and hemorrhage


BolIvIa

SSI Next to hemorrhage, sobreparto was themost frequently mentIOned problem (bysomewhat less than half the women) Aboutone-thIrd of the women who dId mentIon ItsaId It was not serIOUS, whIle the remammgSaId that It was

Report The authors report that the termsobrepw to encompasses the majorIty of thesIgns and symptoms of post-partumcomphcatJOns, mcludmg sepSIS, hemorrhage,palOS, fever, ChIlls, shakes, and foul-smellmgvagmal dIscharge

SSI Not avaIlable for BohVIa

Report Bleedmg IS not discussed, thoughpost-partum hemorrhage IS (see IA 24)

Comments/LImItatIOns

IndoneSIa (SSI) Results of askmg aboutbleedmg and hemorrhage mdlcate womendlstmgUlsh between the condItIons, thoughhow they do so IS not eVIdent



DomaIn/item


NEONATAL

IA 24-Hemorrhagepostpartum

IA 25-TWIns/Triplets

Indonesia

SSI Post-partum hemorrhage wasconsidered serious by virtually all thewomen mtervlewed

Report (FGs) Not discussed, thoughhemorrhage m relatIOn to retamedplacenta IS (see above)

SSI Only about one-quarter of thewomen Said havmg twms/trlplets was aserious problem

Report (FGs) Accordmg to theauthors, havmg twms was not viewed asrisky

Bolivia

SSI Slightly more than one-quarter of thewomen mentIOned postpartum hemorrhage asa problem Most of these considered It to beserious

Report Some focus group respondentsconsidered post-partum hemorrhage to be apart of sobrepm to

SSI Not aVailable for BoliVia (not Cited munprompted questIOns or not coded)



IndoneSia (SSI) The low number of womenstatmg that twms/trlplets were a seriousproblem IS noteworthy

IA 26-StIllblrth/neonatal death I SSI Not available for IndoneSia (notasked In closed-ended questIOns)

Report (FGs) The authors state thatstillbirth was associated with prolongedlabor


SEPTEMBER 15, 1997

SSI Only three women mentIOned stillbirths01 neonatal deaths One believed It was veryseriOUS, one said It was seriOUS, and the thirdsaid It was not serious

Report The authOl s POSit that the lownumbers of women mentlOnmg stillbirths orneonatal deaths mdlcates that compared tomaternal complicatIOns neonatal problemsare conSidered less severe, or severe mamly mrelation to the thleat they pose to the mother

BolIVia (SSI and Report) It IS strlkmg that sofew women mentIOned stillbirths or neonataldeaths ThiS findmg may mdlcate low levelsof awareness or may Indicate that neonatalproblems are not thought of or Viewed assevere when compared to maternalcomplications The findmg also may reflectproblems With questIOn wordmg, espeCiallygiven that m the SSls women were asked to listproblems With the birth of the newborn' andnot With the newborn Itself

62

Domam/Item


IA 27-Premature birth(BolIvia)

Pushmg before expected date(Indonesia)

IA 28-Cord problems

IA 29-Pushmg Every TwoMmutes

IA 30-InfectlOn of the cord

IndoneSia

SSI Half the women belIevedpremature birth was a serIOUS problemMost others did not


SSI About half the women saId that theumbIlIcal cord wrapped around thebaby's neck was a problem Mostothers saId It was not a problem


SSI About three-quarters of the womensaId pushmg every two mmutes was nota serIOUS problem, but a substantIalnumber said It was serIOUS and a fewdidn't know


SSI More than three-quarters of thewomen saId mfectIon of the cord was aserIOUS problem, most others said It wasnot and a few saId they dId not know


BolIVIa

SSI About one-sIxth of the womenmentIOned premature birth as a problemMost of these thought It was serIOUS, but afew said It was not

Report Some focus group lespondentsconsIdered premature delIvery to be a part ofmalparto

SSI Only one woman mentIOned problemswIth the umbIlIcal cord and conSidered It tobe serIOUS

Report While the authors do not mentIOnproblems with the umbilIcal cord Vis-a-VIS theneonate, they do report that focus grouppartICipants assocIate premature cuttmg of theumbIlIcal cord wIth retamed placenta (seeIA 21)

SSI Not avaIlable for BolIVIa (not Cited munprompted questIOns or not coded)


SSI Not avaIlable for BolIVia (not Cited munprompted questIOns or not coded)



IndoneSia (SSI) Many more women saidearly water break was serIOUS, compared tothose who saId pushmg before expected datewas senous ThIS may mdlcate that thequestIOns were not understood, that the Itemshad poor content valIdity, or that womenassocIate these events wIth different outcomes

Bohvla (SSI) It IS not pOSSIble to determmewhat "problems wIth the umbIlIcal cordSIgnIfies It may refer to premature cuttmg orsomethmg related to retamed placenta, ratherthan a neonatal problem



Domam/Item


IA 3 I-Baby dId not move

IA 32-Baby looks blue, notred

IA 33-Baby IS cold

1A 34-Baby IS small

IndonesIa

SSI Just over three-quarters of thewomen saId baby dId not move was asenous problem Many women statedthat It was not a problem

Report (FGs) Some women reportedthat their babies were born weak andlistless

SSI More than three-quarters of thewomen said baby looks blue, not redwas a senous problem A few said Itwas not a problem


SSI About half the women said baby IScold was a problem


SSI Only about one-third of the womenfelt that havmg a small baby was asenous problem VIrtually everyoneelse thought It was not a problem

Report (FGs) Not discussedHowever, consummg certam foods wasreported to be taboo durmg pregnancyPATH states that taboo foods arebelieved to cause the baby to grow toobig and make deltvery difficult

Boltvla

SSI Not avaIlable for BoltvJa (not cIted munprompted questIOns or not coded)


SSI Not available for Boltvla (not cIted munprompted questIOns or not coded)


SSI Not avaIlable for BoltvJa (not cited munprompted questIOns or not coded)


SSI Not avaIlable for Boltvla (not cited munprompted questions or not coded)

Report The authors report that havl11g a'low-weIght" baby was conSidered to be acompltcatlOn, but was seen as relativelyI11slg11lficant

Comments/LImItatIons

Indonesia (8SI) ThIS findmg has ImportantImpltcatlOns for nutntlOn I11terventlOn More111formatIOn IS needed on women's attitudestowards small babIes large babIes dIetarymtake, Iron, low blood pressure, etc

IndonesIa (Report-FGs) It IS not clear fromthe PATH report whether the women statedthat certal11 foods will make the baby too bigor whether thIS was the authors' mterpretatlOn


SLPTEMBER15,1997 64

Domam/Item IndonesIa BolIvIa Comments/LImItatIons


IA 35-Baby did not cry SSI Half the women mtervlewed SSI Not aVaIlable for BolIvIa (not cited mIdentIfied baby dId not cry as a senous unprompted questIOns or not coded)problem Most others felt thIS was not asenous problem Report Not dIscussed


IA 36-Baby's eye oozmg SSI More than three-quarters of the SSI Not aVaIlable for BolIvIa (not cIted mwomen SaId baby's eye oozmg was a unprompted questIOns or not coded)senous problem A substantIal portIonof the women saId they dIdn't know Report Not discussed


1B o-INDIVIDUALEXPERIENCE

IB I-Had a pregnancy SSI About one-thIrd of the women SSI Almost two-thirds of women mtervlewed General PurposIve samplmg makes ItcomplIcation mtel Viewed saId they had a pregnancy reported havmg a pregnancy complIcation mappropnate to generalIze results regardmg

complIcatIOn expenence of complIcatIOnsReport Not discussed

Report (FGs) Most women reportedthey dId not expenence complIcationsdurmg pregnancy

1B 2-Had a delIvery SSI About one-thIrd of the women SSI About one-third of the women General See 18 1complIcatIOn reported havmg a complIcatIOn durmg mtervlewed reported havmg a complIcatIOn

delIvery durmg delIvery

Report (FGs) Not discussed Report Not discussed

18 3-Had a post-partum SSt About one-third of the women SSt Less than a quarter of the women General See IB IcomplIcatIOn reported havmg a complIcatIOn after mtervlewed reported havmg a complIcation

delIvely after delIvery



SEPTEMBER J5 J997 65

Domain/item

INDIVIDUAL EXPERIENCE(contmued)

IB 4-Had a complIcatIOn withnewborn

2A O-DECISION-MAKINGAROUND USE OF HEALTHSERVICES FORCOMPLICATIONS

IndonesIa

SSI About one-quarter of the womenreported that their baby experienced acomplIcatIOn after birth

Report (FGs) Most focus grouppartIcIpants reported havmg had a babywho died at birth, shortly after bIrth,during Infancy or m chIldhood

BolIvIa

SSI Only a few women reported having acomplIcatIOn wIth the birth of the newborn



General See I B I

BolIVIa (SSI) The questIOn posed topartIcIpants was about complIcatIOns with thebirth of the newborn, not wIth the newbornItself The wordmg IS ambIguous, so thequestIon may not yIeld accurate mformatlOnabout the experIence of newborncomplIcatIOns

IndonesIa (SSI) There were some problemswIth SSI data regardmg decIsIOn-making (seeItems below) Focus group data should beexplored m order to help elUCidate andsupplement the findmgs from the SSls

BolIvIa (SSI) WhIle there are numerousproblems wIth the Items related to declslOnmaking around use of health serVIces, It ISclear that the husbands often play an Importantrole In thiS process



Domam/Item

DECISION-MAKINGAROUND USE OF HEALTHSERVICES FORCOMPLICATIONS(contmued)

2A I-Who made/would makethe decIsIOn to seek care (or ofwhose care to seek) when acomplication arose/anses dunngpregnancy

Indonesia

SSI Over half the women mtervlewedsaid they made/would make thedecIsion Some said their husbandsmade/would make the decIsion A fewsaid they made/would make the decIsIOnwith their husbands or that their mothersmade/would make the decIsion

SSI Data also were analyzed for onlythose women who said they hadpregnancy complicatIOns and whoappeared to have sought care Amongthese women almost three-quartersreport makmg the decIsion to seek helpfrom a specific mdlvldual


BolIvia

SSI Nearly half the women who said theyhad a complication durIng pregnancy werescored' don't know/no response" ThemaJonty of the remaInmg women said thattheir husband made the decIsIOn One said adoctor deCided, and a few said they made thedecIsIOn

Report The authors report that women havea great deal of autonomy 10 decIsIOn-makIngwhen It comes to complicatIOns durIngpregnancy


Indonesia (SSI) This questIOn IS ambiguousRespondents are asked who deCided/woulddeCide to go look for whomever was or wouldbe sought out when a complicatIOn occurred ormight occur It IS Impossible to tell whetherthe question IS ask109 about the decIsIOn toseek care or the deCISIOn of whose care to seekAddltlOnaIly, hypothetical cases and real casesare merged Multiple concepts should beseparated mto different questIOns

Bohvla (8SI) This Item IS ambiguous ThequestIOn posed was "whose deCISIOn was It?'The Item Immediately precedmg this one was, who was the first person you sought helpfrom when you knew you were havmgproblems With your pregnancy? The followup question could be Interpreted as ' who madethe deCISIOn to seek care?' or as "who made thedeCISion regardIng whose care to seek? '

AdditIOnally, the high number ofmlssmg/don't know responses IS troublmgBecause the "don't know" and "no responsecategones were combIned, It IS not pOSSible toteIl whether women really do not know theanswer to thiS questIOn or whether they chosenot to answer It

F \nnachbar\comdh\report\final\appendc I

SLPTEMBER 15 1997 67

Domam/Item Indonesia BolIvIa Comments/LImItatIons

DECISION-MAKINGAROUND USE OF HEALTHSERVICES FORCOMPLICAnONS(contmued)

2A 2-Who made the decIsIOn SSI About half the women mtervlewed SSI Of the 14 women reportmg havmg a Indonesia (SSI) The same problemsto see care (or of whose care to said that they made/would make the complIcatIOn dunng delIvery, 6 said their mentIOned for 2A 1 apply to thIS questIOnseek) when a complIcatIon arose decIsIOn In about one-quarter of the husbands made the decIsIOn, 5 stated that theydurmg delIvery?) cases, women reported that theIr made the decIsIOn themselves one said a Indonesia (Report) No data are prOVided m

husbands made/would make the doctor made the decIsIOn another said her the PATH report to support or contradict thedecIsIOn Some women reported that mother made It, and there was one response authors' conclusIOnsthey make/would make the decIsIOn of "don t know/no response'wIth theIr husbands or that their mothers Bolivia (SSI) The same problems withmade/would make the decIsIOn Report The authors report that husbands ambigUIty and collapsmg of response

have pnmary authonty fOl the decIsIOn to categones that apply to 2A 1 apply hereSSI Data also were analyzed for only seek care dunng delIverythose women who saId they haddelIvery complIcatIOns and whoappeared to have sought care, but theresults were not dIscernIbly dIfferentfrom those above

Report (FGs) PATH authOls state thatwomen and their husbands rely on thetradItIOnal birth attendant (TBA) todetermme when a tramed midwifeshould be called or when the womanshould be brought to a health facilityThe decIsIOn to go or not go then restswith the husband

F \nnachbar\comdx\report\final\appendc1


Domam/ltem IndonesIa BolIvIa Comments/LImItatIOns

DECISION-MAKINGAROUND USE OF HEALTHSERVICES FORCOMPLICATIONS(contInued)

2A 3-Who made the decIsion SSI As was the case wIth decIsIOns SSI Of the 8 women reportmg havmg a Indonesia (SSI) The same problemsto seek care (or of whose care to durmg pregnancy and delIvery, about complIcatIOn after delIvery, only one mentIoned above 10 2A I apply to thIsseek) when a complIcation arose half the women Ieported that they responded to thIs Item She stated that she questIOnpost-partum? deCIded/would deCIde to seek care, made the decIsIOn to seek care

almost one-quarter reported that the BoliVia (SSI) The same problems mentIOneddecIsIOn was/would be theIr husbands' Report The authors state that women share 10 2A I apply here The results are soand some report eIther that they responsIbIlIty for the decISIon to seek care ambIguous that mterpretatlOn IS not pOSSIblemade/would make the decIsIOn WIth WIth theIr husbands durmg the post-partumtheIr husband or that theIr mother penodmade/would make the decISIOn

SSI Data also were analyzed for onlythose women who saId they hadcomplIcatIOns after delIvery and whoappeared to have sought care but thelesuIts were not dIscernIbly dIfferentfrom those above




Domam/ltem

DECISION-MAKINGAROUND USE OF HEALTHSERVICES FORCOMPLICATlONS(contmued)

2A 4-Who made the decIsIOnto seek care (or of whose care toseek) when a complicatIOn arosewith the newborn (or with thebirth of their newborn)?

2A 5-Who 10 the familygenerally decides where helpshould be sought?

Indonesia

SSI In about half the cases, womenreported that they made or would makethe decISIon to seek care Slightly lessthan one-quarter of the women said thattheir husbands made/would make thedecIsion

SSI Data also were analyzed for onlythose women who said they hadcomplicatIOns with their newborns andwho appeared to have sought care, butthe results were not discernibly dlffelentfrom those above


SSI Not available for Indonesia (notasked 10 closed-ended questIOns)


Bolivia

SSI Of the four women who reported havmga complicatIOn with the birth of theirnewborn, one said she made the deCISIOnthemselves, one said her husband made thedeCISion, and two stated that they and theirhusbands made the deCISIOn

Report The authors state that the motherplays an almost exclusive role m the deCISionto seek care when complicatIOns anse withthen newborn

SSI Of the eight women who reported thatthey did not have a complicatIOn dunngpregnancy, delivery after delivery, or withtheir newborn five stated that their husbandsmake the decIsion and one stated that theymake It Data for the remammg two womenwere mlssmg

Report Not discussed expliCitly See relateddiSCUSSions above (2A I-2A 4)

Comments/LIm ItatlOns

Indonesia (SSI) The same problemsmentIOned above 2A I apply to thiS questIOn

BolIvIa (SSI) The same problems mentIOned10 2A I apply here Additionally, the ongmalquestIOn regardmg havmg a complicatIOn Withthe newborn was worded very ambiguously,makmg mterpretatlOn difficult

BolIVIa (SSI) The assumptIOn IS made that thedeCISion occurs 10 the family so there IS nomformatlon on the mfluence of others (e g ,parteras)

ThiS questIOn refers to help-seekmg behavIOr,but does not speCify whether It IS related tomaternal complicatIOns Women may havemterpreted It more broadly



II Domam/Item


2A 6-Where IS help firstsought (general comments)

2A 7-Where dId you first seekhelp for problems m pregnancy?(IndonesIa)

Where IS help first sought whenproblems occur durmgpregnancy? (BolIVIa)

IndonesIa

SSI For both open- and closed-endedquestIons on where help IS first sought(for al1 stages), large numbers of womenreport usmg bldans and bldan dl desas


SSI ThIs questIon was both open-andclosed ended Nearly half the womenreported seekmg help from bldans andalmost a quarter saId they sought helpfrom bldan dl desas Some womenwent to the doctor or TBA Only a fewsaId they went to the pustu orpuskesmas There were no discernIbledIfferences m responses betweenwomen from the three study sites


BolIvIa

SSI Data are mlssmg for nearly threequarters of the women who said they had apregnancy complIcatIOn Of those whoresponded, only one saId they first soughthelp from theIr mother or father a few saIdthey did not seek help from anyone/werealone, and a handful saId "both

Report Whtle this question IS not dIscussedfor women who expenenced pI oblems per se,the authors report that women tend to use themstitutlonal health system for hemorrhagedurmg pregnancy and use pm teras for fetaldystOCia For symptoms of preeclampSia theauthors state that women use neIther systembut WIll go to health faCIlItIes for attacks andfamtmg ,


IndoneSIa (SSI) The data IS somewhatsuspect smce al1 women report that theywould seek out someone

IndoneSIa It IS not possIble to tell from theway the questIon IS wOlded, whether or not thewomen went through the mstItutlOnal healthsystem when they sought out the bldans andbldan dl desas Focus group data may helpexplIcate the SSI findmgs but none werereported

BoliVIa (SSI) The very high number ofmlssmg data decreases the value of the ItemAddItIOnally there were problems With thewordmg of the questIOns and responsecategones Specifically, the response category

no one/was alone" encompasses two Ideasthat should be separated Also, It IS notpossible to discern who IS referred to m theresponse category both" Fmally,mother/father are combmed Into one responsecategory Given the difference m theIr roles,separate categones should be used

F \nnachbal \comdx\report\final\appendc I

SEPTLAfBER 15, 1997 71

Domam/ltem

DECISION-MAKINGAROUND USE OF HEALTHSERVICES FORCOMPLICAnONS(contInued)

2A 8-Where did you/wouldyou first seek help for problemsm delivery? (Indonesia)

Where IS help first sought whenproblems occur durmg delivery?(Bolivia)

2A 9-Where did you/wouldyou first seek help for problemsm delivery? (Indonesia)

Where IS help first sought whenproblems occur afterdelivery?/Was the care of a postor center doctor or nurse soughtwhen problems occuned afterdelively? (BolIVIa)

Indonesia

SSI Results are Virtually Identical tothose fOl pregnancy (see 2A 7)


SSI Results are Virtually Identical tothose for pregnancy (see 2A 7)


Bolivia

SSI All women who said they had a problemdurmg delivery responded to this questIOnMost said they either went to a doctor/centerdoctor/hospital or that they went to thepat tel a One respondent said she went firstto her husband another said she went first toher mother/father, another said she went toother family members and one said she wentto a nurse

Report While this questIOn IS not discussedfor women who expenenced problems per se,the authors report that women go to themstitutlOnal health system for hemorrhagmgdUl mg delivery

SSI Half the women who reported problemsafter delivery said they sought the care of aphysIcian or nurse at the center or post Theother half said they did not seel-. such care

Report While not discussed for women whoexpenenced problems per se, the authors statethat women do not genelally use themstltutlOnal health system for post-partumcomplicatIOns, regardless of whether theycome from commUnities with high or low useof mstltutlOnal health services Data flom thefocus groups mdlcate that women expect theywill be scolded by health prOViders for havmgsoblepatto


Indonesia The same Issues that apply to theabove questIOn on pregnancy (see 2A 7) applyhere

BoliVia (SSI) It appears women often seekhealth servIces when they have problemsdurmg pregnancy Because doctor, centerdoctor, and hospital were combmed as oneresponse category It IS not pOSSIble todlstmgUlsh among these The same problemwith mother/father discussed III 2A 7 applieshere


BoliVia (SSI) It IS not pOSSible to tell whetherthe women who said they did not seek care (ata post of a center doctor or of a nurse) did notseek any care or sought cale elsewhere (e gparents, hospital)

ThiS question IS different from those designedto get at the same general mformatlOn forpregnancy delivery, and the newbornQuestIOns on the same tOpiC should bestandardIzed



DomaIn/Item

DECISION-MAKINGAROUND USE OF HEALTHSERVICES FORCOMPLICAnONS(contmued)

2A IO-Where did you/wouldyou first seek help for problemswith the newborn? (Indonesia)

2A II-When women haveproblems In pregnancy,dehvery, post-partum, or withthe newborn/upon the birth oftheir baby, where do they firstseek help?

Indonesia

SSI Results ale virtually Identical tothose for pregnancy (see 2A 7)


SSI Not avatlable for Indonesia


BolIVia

SSI Not avatlable for BolIvia

Report While not discussed for women whoexpenenced problems, per se, the authorsstate that the InstitutIOnal health systembecomes an Important altematlve to thetraditIOnal system once a newborncomplIcatIOn IS Identified No focus groupdata are reported

SSI Eight women reported they had noproblems In pregnancy, dehvery post-partum,or with the newborn Data are missIng fortwo of the women Three respondents saidwomen first seek help from a midwife Of theremaInIng three respondents, one each saidwomen seek help from other relatives, fromthe post/center/hospltal/doctor, or from"other"

Report See 2A 7-2A 10

Comments/LlmltatlOns


Bohvia (SSI) Whtle more women said(unprompted) that they normally seek helpfrom the partera, the small number ofrespondents means that httle weight should beascnbed to thiS Item Further, while womenseem to seek out the partera for complIcatIOnsIn delIvery, none reported dOIng so forpregnancy problems Thus, the extent towhich the partera or others are sought out forhelp IS not clear from the responses to theSSIs


SEPTEMBER J5, J997 73

Domain/item


2A 12-Blrth place preference(past and future)

3A O-ACCESS TO CAREROLE OF OTHERS

3A I-Who generally goes withthe woman when she goes to thepuskesma:, pembantu,pusk/..,smas or hospital(IndoneSia)

Indonesia

SSI More than three-quarters of thewomen interViewed said their last birthoccurred at home About three-quartersof the women also said they wanted togive birth at home In the future Morethan three-quarters of the women whohad complIcatIOns dUring pregnancy,delivery, In the post-partum penod, orwith the newborn also said they wantedto give birth at home

Report (FGs) Data from the focusgroup participants indicate that about70 80 percent of all births occur athome and most are attended by TBAsAccording to PATH the decIsion todeliver at home IS based onconvenience, practicalIty andaffordabllIty

SSI By far the most common response(given by about two-thirds of thewomen) was that they wereaccompanied by neighbors Somereported gomg with their husbands, andsome with then children


BolIVia

SSI About half the women mtervlewedreported that they wanted future births tooccur In their homes and half expressedpreference for health centers

Report The authors state that the homeenvironment and family support m thatsetting are Viewed as a cntlcal part of thebirth process Focus group data support thiScontention

SSI Not available fOJ BoliVia (IndonesIaspeCific)

Report Not discussed (IndoneSia-specific)

Comments/LimitatIOns

IndoneSia (Report) While the PATH reportindicates that the deCISIOn to give birth at homeIS based on convenience, practicalIty andaffordabllIty, the authors do not prOVide anydata to support thiS claim Thus, It IS difficultto determme whether thiS conclusIOn IS basedon focus group diSCUSSIOns or IS the authors'mterpretatlon

IndoneSia (SSI) ThiS questIOn may have beenmterpreted very broadly, smce It IS not askmgspeCifically about complIcatIOns It ISnoteworthy, however, that the vast maJonty ofwomen report gomg to the health center withneighbors Also, the question was "If you goto the puskesmas pembantu, puskesmas, orhospital you are usually accompanied bywhom? Or do you go alone?' Yet no womenreported gomg by themselves

F \nnachbar\comdx\1 eport\final\appendc 1


Domam/Item

3A O-ACCESS TO CAREROLE OF OTHERS

3A 2-Who goes with thewoman to get help whencomplicatIons anse mpregnancy?

3A 3-Who goes with thewoman to get help whencomplicatIOns anse durmgdelivery?

3A 4-Who goes with thewoman to get help whencomplications anse afterdelivery?

Indonesia

SSI Pregnancy-specific questIOn notavailable for Indonesia (see 3A 1)


SSI Delivery-specific questIOn notavailable for Indonesia See generalquestIOn (see 3A I)


SSI Post-partum-speclfic questIOn notaVailable for Indonesia See generalquestIOn (see 3A I)


Bolivia

SSI Data are aVailable for only two-thirds ofthe women who reported havmg a pregnancycomplicatIOn Most women leported thattheir husbands went with them A couple saidthey went alone, and a few others Said theperson provldmg help came to their house


SSI Less than half the women reportmgcomplicatIOns durmg delivery said theirhusbands came with them to get help Acouple said the partera went with them, acouple said a parent went with them, and acouple said they received help m their home,but did not state from whom Two responseswere 'don t know/no response"


SSI Data for four of the eight women whosaid they had problems after delivery aremlssmg Of the lemammg four whoresponded, all said then husbands went withthem


Comments/LImitations

Bolivia (SSI) The large number of miss109

data makes the results difficult to mterpretHowever, husbands do appear to play anImportant role

BoIvla That women "go somewhere" whencomplicatIOns anse IS Implicit m the questIOnwordmg yet this may not be the case

Bolivia (88I) While the absolute numbers aresmall, husbands appear to play an Importantrole m help-seekmg durmg delivery This Itemsuffers from the same problems with theresponse categones (mother/father, don'tknow/no response) as discussed 10 2A 8

BoIvla That women "go somewhere" whencomplicatIOns anse IS Implicit 10 the questIOnwordmg, yet this may not be the case

Bolivia (88I) While the number of responsesare very low, husbands stili appear to play anImportant role m accompanymg the womanwhen help IS sought for problems afterdelivery

Bolvla That women' go somewhere whencomplicatIOns anse IS Implicit 10 the questIOnwordmg, yet this may not be the case



Domam/Item Indonesia Bohvla Comments/LimitatIons I3A 5-Who goes with the SSI Newborn-specific questIOn not SSI Of the four women who said they had BolIvia (SSI) Agam, the number of responseswoman to get help when aVailable for Indonesia See general problems with their newborn, three said their are very low but husbands playa role Thecomphcattons arIse with the questIOn (see 3A I) husbands accompanied them and one said response category "mother/father" should benewborn? their mother/father did divided mto two

Report (FGs) Not discussedReport Not discussed Bolvla That women "go somewhere" when

compltcatlons arIse IS Impltclt m the questionwordmg, yet this may not be the case



Domain/Item Indonesia BolivIa Comments/LImItatIons

3B O-ACCESS TO CARE Indonesl3 (SSI) Only one questIon was askedPERCEPTION OF CARE about servIce quality I e "How was the

servIce quality/response you got from X whenyou had a problem dunng Y?" ThIS doesproVIde a general pIcture of how women mIghtfeel about the servIce they receIved, but It doesnot allow any distinctIOn of the elements ofquahty (e g , courtesy, tImeliness, c1eanhness,etc) These data were entered as a strIngvanable and could not be analyzed

Data from questIOn C5 of the IndoneSIan SafeMotherhood mstrument, m whIch women whosaId the servIce they receIved was not goodstated theIr reasons, would be more useful toanalyze However thiS IS In text form andwould need to be translated

Future questIOnnaIres should ask all women toexplam their answers, not Just those who werenot happy wIth the servIce they receIved

BoliVIa From the SSls, women appearoverwhelmmgly satIsfied wIth the quality ofcare they received ThIS could mdlcate one ofseveral thmgs (I) genume satisfactIOn WIthcare, (2) lack of content validity of the Items,(3) response bIas In general, Items related tothIS domam yielded questIOnable resultsFocus group mformatlOn and open-endedresponses probably pamt a more accuratepicture



Domain/item

ACCESS TO CAREPERCEPTION OF CARE(contmued)

38 I-Were the problemsoccurring dunng pregnancyproperly resolved at thepost/center/hospital?

38 2-Were the problemsoccurring during delIveryproperly resolved at thepost/center/hospital?

38 3-Were the problemsoccurring after delIvery propellyresolved at thepost/center/hospital?

Indonesia

SSI Not available for Indonesia (notasked In closed-ended questions)


SSI Not available for Indonesia (notasked In closed-ended questIOns)



80hvla

SSI Of the 12 women who said they went tothe post/center/hospital for problems dunngpregnancy, about three-quarters said that theirproblems were resolved well and one-quartersaid they were not resolved well

Report Not discussed for pregnancy per se,but focus group data are used to support theauthors' statement that the institutIOnal healthsystem IS Viewed as efficaCIOUs In solvingobstetnc and neonatal plOblems

SSI All ten of the women who reportedgOing to the post/centerlhospltal for problemsdunng delIvery said their problems wereresolved "well

Report 38 I

SSI Of the four women who reported gOingto a health center/post/hospital for problemsafter delIvery, three said their problems wereresolved "well" One said her problems wereresolved "not well

Report See 38 I


Bolivia (SSI) The high number of satisfiedresponses IS somewhat suspect The questIOnalso IS very vague and does not necessanlyassess perceived qualIty of care There may bemany reasons why a problem IS not 'properlyresolved" More targeted questIOns would bemore hkely to yield good informatIOn

Another problem IS that It IS difficult todetermine whether women were answennghonestly, or whether there was bias In theirresponses ThiS Item IS not particularly useful

BoliVia (SSI) There may be response biashere Also, the question IS vague Moretargeted questIOns would be more lIkely toyield good information (see 38 I)

BoliVIa (SSI) The comments above (see 38 Iand 38 2) apply here as well


SEPTEMBER J5, J997 78

Domam/ltem


38 4-When you went to thethe post/center/hospital forproblems m pregnancy, wereyou seen In a timely manner?

38 5-When you went to thepost/centerlhospltal forproblems dunng delivery, wereyou seen m a timely manner?

38 6-When you went to thepost/center/hospital forproblems after delivery, wereyou seen In a timely manner?

38 7-When you went to thepost/center/hospital forproblems With your newborn,were you seen m a timelymanner?

Indonesia




SSI Not available for IndoneSia (notasked In closed-ended questIOns)

80llvla

SSI About one-quarter of the womeninterviewed reported gOing to thepost/center/hospital fOi problems Inpregnancy Of these all but two said theywere seen m a timely manner

Report Not discussed vis-a-vIswomen/families The authors state that healthpersonnel, particularly those workmg mhospitals, viewed timely attentIOn to the clientas synonymous with quality care

SSI Eight of the 10 women who reported thatthey went to the post/center/hospital forproblems m delivery said they were seen In atimely manner

Report See above

SSI Three of the four women who said theywent to a health center for problems afterdelivery said they were seen In a timelymanner One responded don t know/noresponse"

Report See above

SSI Three of the four women who Said theywent to a post/center/hospital fOi ploblemsWith their newborn said they were seen In atimely manner The fourth responded"other"

Report See above

Comments/LimitatIOns

Bohvia (8SI) The relatively large number ofwomen reporting that they were seen In atimely manner could indicate (I) that thewomen did not walt an Inordmately long time,(2) response biaS, (3) that women have lowexpectatIOns regarding waltmg time DirectobservatIOns of patient flow might reveal moremformatlOn regardmg time flOm arrival toconsultatIOn, as would askmg respondents tomdlcate their waltmg time on average, at theirlast VISit, and m an emergency SituatIOn

BoliVia (8SI) See above (38 4)

Bohvla (8SI) See above (38 4)

Bohvla (8SI) See above (38 4) Also the"other category IS not defined and IS notuseful



DomaIn/item


3B 8-Are you convmced aboutthe village midwives' (bldan dldum, ') skill?

3C o-ACCESS TO CAREDISTANCE/TIME TOFACILITY AND COST OFTRANSPORTATION

3C I-How far IS the sub-healthcenter (pu:.ke:.mas pembantu)from your house? (IndonesIa)

Is the closestpost/center/hospital vary faraway somewhat far away, orclose by the community?(BoliVia)

Indonesia

SSI More than three-quarters of thewomen Interviewed reported that theywere "convInced" of the bldan dl desas'skill

Report (FGs) PATH authors reportthat women expressed concernregardmg the skills and experIence ofthe bldan dl desas but had confidence Inthe TBAs

SSI Most women from DistrIcts I and 2reported that the puskesmas pembantuwas close A substantial number didn'tknow whether It was close or notWomen from Dlstnct 3 were diVided Intheir responses About one-quarterthought It was close, one-quarter said Itwas fa! 01 very far, and about onequarter said they didn't know Nearlyone quarter of the responses wei emissIng


Bohvla

SSI Not available for BohvIa (Indoneslaspecific)

Report Not available for BolivIa (IndonesIaspecific)

SSI About three-quarters of the women fromlow service use areas and about half thewomen from high service use aleas stated thatthey considered the health center to be closeto the community

Report No focus group data are reported

Comments/LimItatlOns

Indonesia (SSIs) While some clarIty mayhave been lost m the translation the questIOnIS very vague Also a comparIson between themidWives' skill and that of the TBAs IS placedas the middle lesponse m thiS Likert-typescale ThiS IS not approprIate FInally thiS IS a"yes/no' questIOn, yet the responses are multilevel

IndoneSia (Report) AccordIng to the PATHauthors focus group participants doubted theskills and expenence of the bldan dl desaswhile expressmg confidence m the TBAsThese results confhct With the SSls butprobably are a more accurate reflectIOn ofperceptions



Domam/Item

ACCESS TO CARED1STANCE/TIME TOFACILITY AND COST OFTRANSPORTAnON(contInued)

IndoneSIa Bohvla Comments/LImItatIons

3C 2-How far IS the healthcenter (puskesmas) from yourhouse?

See 3C 1 for Bohvla

SSI Most women from DIstricts 1 and 3 I SSI See 3C 1conSIdered the puskesmas to be very faror far from theIr homes Very few I Report See 3C 1women from dIstrict one Said that thepuskesmas was close About half thewomen lo DIstrict 2 saId the puskesmaswas close and about half saId It was farOnly a few women from any dIstrictsaId they dldn t know whether thepuskesmas was close or far


3C 3-How far IS the hospItalfrom your house? (IndonesIa)

See 3C 1 for Bohvla

3C 4-Type of transportatIonused to go to the pustu (pubhcand private, IndonesIa)

SSI Nearly all the women fromDIstricts 2 and 3 Said the hospItal wasfar from theIr house Almost half thewomen from DIstrict I saId the hospItalwas close by Very few women saIdthey dIdn't know whether the hospItalwas close or far from theIr homesThere were several mlsslOg responses


SSI Nearly all the women respondIngto thIS questIOn saId they go to the pustuby foot Responses were very SImilaracross the thl ee dIstricts


SSI See 3C I

Report See 3C I

SSI See 3C 7 and 3C 12 (for BohvIa,transportatIon IS dIVIded lOto pubhc andprivate)

Report See 3C 7 and 3C 12

IndoneSIa (SSI) About half the responses falllo the category "not apphcable' It IS dIfficultto tell whether these were non-responders, orwhether the women lo the category notapphcable" dId not go to the pustu If thewomen were non-Iesponders, the large numberof mIss109 data makes lOterpretatIon dIfficult


SEPTElI1BER 15 1997 81

Domain/item

ACCESS TO CARED1STANCE/TIME TOFACILITY AND COST OFTRANSPORTATION(contmued)

3C 5-Type of transportatIOnused to go to the puskesmas(Indonesia)

3C 6-Type of transportationused to go to the hospital (Pubhcand private Indonesia)

Indonesia

SSI About one-quarter of the womenreport gomg to the puskesmas on footA substantial portIOn of women reportgOing by motorboat (In Dlstncts 2 and3) Some report gOing by nckshawVery few report uSing other forms oftransportatIOn (e g , bicycle, motorcycle,mlklo) Many women responded "notapphcable


SSI Most responses were "notapphcable" In Dlstnct I, most of thewomen for whom the questIOn wasconSidered apphcable reported gOing byrickshaw In Dlstnct 2 most reportedgOing by mlkro In Dlstnct 3 mostreported gOing by motorboat


Bolivia

SSI See 3C 7 and 3C 12 (for Bohvla,transportatIOn IS divided Into pubhc andpnvate)


SSI See 3C 7 and 3C 12 (for Bohvla,transportatIOn IS divided Into pubhc andpllvate)



IndoneSia (SSI) See above (3C 4)

IndoneSia (SSI) In all three dlstncts thenumber of women who reported then methodof transportatIOn to the hospital was verysmall Results should be mterpreted withcautIOn


SEP fEMBER 15, 1997 82

Domam/Item IndonesIa Bohvla Comments/LImItatIOns

ACCESS TO CAREDlSTANCEITIME TOFACILITY AND COST OFTRANSPORTAnON(contlDued)

3C 7-Type of transportatIOn SSI See 3C 4-3C 5 SSI Data for nearly half the women who BoliVia (SSI) WhIle It IS noteworthy thatused to go to post/health reported that they sought help at the among the women who responded, mostcenterlhospltal(pubhc) (BolIvIa) Report See 3C 4-3C 5 post/center or hospItal for a comphcatlOn are sought care on foot, too much data are

eIther mlssmg or fall under the "don't mlssmg to draw any conclUSIOns Also, thereknow/no response" category Over one- IS no way to dlstmgUlsh what transportatIOnquarter ofthe total sample reported gomg to women used to go to whereby foot A few reported gomg by mInIbus,mICro, or by bus As mentIOned preVIOusly, the "don't know/no

response' category should be separatedReport The authors state that women fromrural areas generally use pubhc buses to go tohealth centers, whIle urban women usemInIbuses or go by foot



Domam/Item

ACCESS TO CAREDISTANCEITIME TOFACILITY AND COST OFTRANSPORTATION(contmued)

3C 8-How long did It take (toget to the pu!>ke!>mas pembantu,pu.~kesmas, or hospItal)? (Publtcand pnvate transportatIonIndonesia)

How long did It take? (publtctransportation-BoliVia)

Indonesia

SSI (for the Pustu) Almost half of thewomen mtervlewed (regardless ofdIstrict) reported that It took 15 mmutesor less to get to the pustu

SSI (for the puskesmas) More than onequarter of the women said It took them15 mmutes or less to reach thepuskesmas and about the same numbersaid It took them more than 15 mmutes,but less than 30 mmutes Some womensaId It took between 30 and 180mmutes Over one-quarter of the datawere coded as mlssmg All reports thatIt took more than 30 mmutes to reachthe puskesmas were from District 3

SSI (for the hospital) More than halfthe data were mlssmg About onequarter of the women said they couldreach the hospital m 30 mmutes or lessA few women said It took them between30 and 60 mmutes to reach the hospital,and a few women said It took thembetween two and four hours AII thewomen who reported that It took morethan one hour to reach the hospital welefrom Dlstnct 3


Boltvla

SSI Over 50 percent of the data are missingIt took longer for the fOUl women from lowuse areas to reach their destmatlon than It didfor any of the 7 women from high use areas

Report The authors report that most womencan reach a health center Within 10-15mmutes No data source IS given


Indonesia (881) A large portIOn of the datawas coded as "missing Responses should beInterpreted With caution

Bolivia (8SI) Too much data are mlssmg tolend credence to thiS Item Additionally, Itappears data may have been entered m bothhours and mmutes, making It difficult toInterpret

Again, the "don't know/no response" categoryshould be separated


StPTEMBER15,1997 84

Domam/ltem

ACCESS TO CAREDISTANCE/TIME TOFACILITY AND COST OFTRANSPORTATION(continued)

3C 9-How much did It cost?(PublIc and privatetransportatIOn-Indonesia)

How much did It cost? (PublIctransportatIOn-BolIvia)

Indonesia

SSI (for the pustu) There were onlyfour women who said they neitherwalked, nor took a bicycle, or for whomdata were not mlssmg All four womensaid It cost between 450 and 1,000 rp toget to the pu5tu Via their transportatIOnmethod

SSI (for the Puskesmas) More than 50percent of the respondents saId It cost500 rp or less to get to the puskesmasMore than one-quarter said It costbetween 500-1000 rp Only a few saidIt cost more than 1000 rp

SSI (for the hospital) Almost half thewomen said It cost them 500 rp or lessto get to the hospital Most otherwomen said It cost between 500 and1500 rp A few women said It costmore than 1500 rp, wIth one statmg thatIt cost 40 000 rp


Bohvla

SSI About half the data are mlssmg (womenwho went on foot were excluded from theanalysIs) It IS mterestmg to note, however,that of the four women from low service useareas, two reported costs of less than 5 OBs(less than $1 00) and two reported costs of20 OBs or greater ($4 00 or more) Of thewomen from high use areas no one reportedspendmg more than 10 OBs ($2 00) ontransportatIOn

Report The authors state that transportatIOnm urban areas costs about IBand m ruralareas, It costs between 3-5Bs No datasources are given

Comments/LimitatIons

IndoneSia (SSI) (for the Pustu) Given thatso httle data are avaIlable for women who didnot walk or use bicycles conclUSIOns cannotbe made

Bohvia (SSI) There IS too much mlssmg datato draw definitive conclUSIOns However datafrom other sources (focus groups) should beused to determme the extent to whIchtransportatIOn cost plays a role m use of healthservices


SEPTEMBER J5 J997 85

Domam/ltem IndonesIa Boltvla Comments/LImItatIons

ACCESS TO CAREDISTANCE/TIME TOFACILITY AND COST OFTRANSPORTATION(contmued)

3C IO-How regular IS this SSI Not avaIlable for Indonesia (not SSI Well over half the data are mlssmg Bohvla There IS too much mlssmg data totransportatIon (public)? asked m closed-ended questIOns) Only two women saId theIr transportatIOn was draw definItive conclusIOns from the SSls

megular, both were from low servIce use Focus group data are more mformativeReport (FGs) Not dIscussed areas

Report Usmg focus group data the authorsstate that publtc transportatIon generally runslegularly durmg the day but transportatIOndunng emergencIes and at nIght IS dIfficult tofind

3C I I-How comfortable IS thiS SSI Not avaIlable for Indonesia (not SSI Agam, nearly half the data are mlssmgtransportatIOn (public)? asked m closed-ended questIOns) Of the nIne women respondmg, about half

said It then transportation was comfortableReport (FGs) Not discussed and about half said It was uncomfortable

Report Women partlclpatmg m focus groupsreport that transportatIOn generally IScomfortable, but not durmg pregnancy

3C 12-Type of transportatIOn SSI Data not divIded mto public and S8I Too much data are mlssmg to analyze BolIVIa (SSI) There IS too much mlssmg dataused (pnvate) prIvate for IndonesIa results to warrant analySIS The large numbers of

mlssmg responses may reflect problems WithReport (FGs) Not discussed Report Not dIscussed the questIOnnaire deSign 01 WIth mtervlewmg

SSI Data not divided mto public andtechnIques

3C 13-Cost of pnvate SSI Too much data are mlssmg to analyzetl ansportatlOn pllvate for Indonesia results


3C 14-Rellablltty of SSI Not available for Indonesia (not SSI Too much data are mlssmg to analyzetransportatIOn asked m closed-ended questIOns) results

Report (FGs) Not dIscussed Report Not discussed



II DomaInlltem

ACCESS TO CAREDlSTANCEITIME TOFACILITY AND COST OFTRANSPORTAnON(contlDued)

3C IS-Comfort oftransportatIOn

3D O-COST OF SERVICES

3D I-How much do ANCservices cost? (Indonesia)

Old you have enough money for(staft) to take care of you at thepost/center/hospital? (Bolivia)

3D 2-How much IS the averagecost of medicIne? (Indonesia)

Old you have enough money tobuy medicIne? (Bolivia)

Indonesia



SSI Almost one-quarter of the womensaid the services cost nothIng and aboutthree-quarters said they cost between 01500 rp


SSI Almost half the women saidmedicInes were free


Bolivia

SSI Too much data are mlssmg to analyzeresults


S51 Regardless of whether women withcomplicatIOns were from high or low serviceuse areas, responses were almost evenly splitbetween those who felt they had enoughmoney and those who felt they didn't

Report The authors state that families fromhigh service-use areas perceive the cost oflegular care (not emergency care) to bereasonable, whereas families from lowservice-use areas perceive the cost to be highor moderate (but stili prohibitive) Focusgroup data support the authors' contentionthat for all families cost becomes a majorlimitation to care In emergency situatIOns

SSI Women expenencIng complicatIOns whowere from high and low service use areaswere about evenly split between those whoreported havmg enough money and those whoreported they did not

Report See 3D I


Indonesia (551) ThiS Item while useful forobtammg mformatlOn on the cost of prenatalcare says nothIng about the cost or perceivedcost of services for pregnancy-relatedcomplicatIOns

BoliVia (5SI) SInce thiS questIOn was askedonly of women who went to a health center,post, or hospital for complicatIOns It IS notuseful for determmmg the degree to whichservice cost acts as a ban ler to care Also thenumber of respondents was low

IndoneSia (SSI) About one-quarter of the datawere mlssmg for thiS vanable AdditIOnally,since the prevIOus questIOn pertaIned toantenatal care, women may have responded tothiS Item as If It pertaIned only to prenatal caleand not to emergency situatIOns

BoliVia (SSI) See comments for 3D I

•



Domam/Item

3E o-ACCESS TO CARECHILD CAREARRANGEMENTS

3E I-Who did you leave yourchildren with when you soughthelp for problems In pregnancy,delivery after delivery, or withthe newborn? (Bolivia)

When women go to thepuskesmas pembantu,puskesmas, or hospital, whousually cares for your childrenor do you take them along?(Indonesia)

Indonesia

SSI Almost half the women respondingstated that their children generallyaccompany them when they go for careIn some cases, the mother's mother orthe mother s eldest child was thereported caregiver In others thechlld(ren) stayed alone Only a fewwomen said that their husbands or anyother family member took care of theirchildren when they went to get care


BoliVia

SSI For all stages of pregnancy, responseswere pretty consistent With about equalfrequency, women said they left their childrenWith their husbands, parents, or other familymembers Other responses given were "otherpeople," "other," and "don't know/noresponse"

Report Focus group data support theauthors' statement that the extended familyplays an Important role 10 car10g for childrenwhen the mother goes to a health faCIlity


Indonesia (SSI) The question IS very generalWomen may have 10terpreted It very broadly(I e , not complicatIOn-specific or even speCificto their own health needs) However, the factthat most women report takmg their children ISnoteworthy Smce they also often report gOingWith a neighbor, It IS pOSSible that neighborsact as child care proViders Focus group datamight help elUCidate the role of the family andneighbors but none are reported

BoliVia As IS the case for other questIOns, the"no response/don t know category should beseparated Even though there IS very littlemiSSing data the small sample size precludesdraWing definitive conclUSIOns

Focus groups revealed more about the role ofthe family mcludIng caretakmg of childrenthan did the SSIs



Domain/Item

3F o-ACCESS TO CAREHELP FROM THECOMMUNITY

3F I-Old women received helpfrom a neighbor or theircommunIty when they hadproblems and did they find thishelp helpful?" (Indonesia)

Old women receive help frompeople In their communIty whenthey had problems? (Bolivia)

3F 2-Who did women receivehelp from If they got assistancefrom someone In theircommunIty?

Indonesia

SSI Somewhat less than half the datawere missing The "missing" categorymay Include women who said they didnot have a problem during any stage butIt IS not possible to tell About one thirdof the women who apparently hadproblems In pregnancy, delivery, thepost-partum penod, or with the newbornsaid they got help from a neighbor orsomeone In the communIty A littlemore than one-quarter said they did notV Irtually all the women saId this helpwas "helpful


S81 Not available for IndonesIa (datawere coded as a string vanable and forthe purposes of this report, could not beanalyzed)


Bolivia

881 About two-thirds of the women reportingproblems during pregnancy, delivery, afterdelivery, or with the newborn stated that theydid not receive help from anyone In theircommumty

Report Not discussed outSide the context ofthe family

S81 Of the nIne women who reportedreceIVIng help from people In theircommunIty, most said they got assistanceeither from a membel of the health profession(pwtew doctor or nurse) or flom a familymember (but not their spouse) Two womenreported getting help from theIrneighbors/friends

Report Not discussed outSide the context ofthe family

Comments/LimItatlons

Indonesia (881) While the questIOn wasapparently asked to only those women whoreported haVing a problem In any of the fourstages, 37 of the 90 women were reported as"missing" These probably were women whosaid they did not have a complicatIOn Codingthem as ' missing" gives misleading resultswhen frequencies are tabulated

Information was collected on the kind of helpwomen received and from whom, but the dataare not In numencal form and could not beanalyzed herein These data and focus groupdata should be explOied to further eluCidatethis tOPiC

Bolivia (SSI) The term communIty" IS vagueand lacks content validity (see 3F 2) Also, theterm help" needs clanficatlon Those womenwho said they did get help speCified thecontent of that assistance but since the data aretextual, they could not be analyzed for thisreport

Bolivia (881) The fact that most women whosaid they got help from people In theircommunIty mentIOned their family ormembers of the health professIOn indicates thatthis questIOn may be lacking In contentvalidity Women may have received somecommunIty assistance but problems withcontent validIty make interpretatIOn difficult

F \nnachbar\comdx\1 eport\final\appendc I


REFERENCES

1 Koblmsky M (1995) On the pathway to maternal health-results from IndonesiaMothelCare Matters 5(1) 1-2 (Koblmsky M and A Hamson, eds) Arlmgton, VAMotherCare/John Snow, Inc

2 MotherCare (1996) QuestIOnnaIre of Mothers of ChIldren Under FIve/Pregnant WomenSafe Motherhood (IndoneSIa) Jakarta Indonesia MotherCare/John Snow, Inc

3 Seoane GS, Kaune VM and JV Cordova (1996) Dzagnostlco Barreras y Vlabllzzadoresen la AtenclOn de Compllcaczones Obstetncas y Neonatales (Estudzo Cualztatlvo enComumdades y Servlczos de SaIud de Cznco D'StrztOS de Salud en La Paz yCochabamba) La Paz, BohvIa MotherCare/John Snow, Inc, and Marketmg S R L

4 Marsaban J, PATH-IndonesIa, and MotherCare-IndonesIa (1996) Report CommumtyDzagnosls MotherCare Safe Motherhood Project South Kalzmantan 1996 Jakarta,IndoneSia PATH and MotherCare/John Snow, Inc

5 Marsaban J, PATH-IndoneSIa, and MotherCare-IndonesIa (1996) Report CommumtyDzagnosls MotherCare Safe Motherhood Project, South Kalzmantan, 1996 Jakarta,IndoneSia PATH and MotherCare/John Snow, Inc

6 Marsaban J, PATH-IndoneSia, and MotherCare-IndonesIa (1996) Report CommumtyDzagnosls MotherCare Safe Motherhood Project, South Kalzmantan, 1996 Jakarta,IndoneSia PATH and MotherCare/John Snow, Inc

7 Seoane GS, Kaune VM, and JV Cordova (1996) Dzagnostlco Barreras y Vzabllzzadoresen fa Atenczon de ComplzcaclOnes Obstetrzcas y Neonatales (EstudlO Cualztatlvo enComumdades y SerVlCZOS de Salud de Cznco D,stntos de Salud en La Paz yCochabamba) La Paz, BohvIa MotherCare/John Snow, Inc, and Marketmg S R L

8 Marsaban J, PATH-IndoneSIa, and MotherCare-IndonesIa (1996) Report CommumtyDwgnosis MotherCare Safe Motherhood Project South Kalzmantan, 1996 Jakarta,IndoneSia PATH and MotherCare/John Snow, Inc

9 Johnson BB (1996) Control Program ofAnemw zn Pregnant Women The Results oftheQualztative InvestIgatIOn Washmgton, DC OMNI/John Snow, Inc

10 Johnson BB (1996) Control Program ofAnemw m Pregnant Women The Results oftheQualztatlve Investigatwn Washmgton DC OMNI/John Snow, Inc

F \nnachbar\comdx\report\\fina14 wpdSEPTE\fBER 15 1997

11 Seoane GS, Kaune VM, and JV Cordova (1996) Dzagnostzco Barreras y Vzabzlzzadoresen la AtenclOn de ComplzcaclOnes Obstetricas y Neonatales (EstudlO Cualztatzvo enComumdades y ServlclOs de Salud de Cmco Distntos de Salud en La Paz y ,Cochabamba) La Paz, BohvIa MotherCare/John Snow, Inc, and Marketmg S R L

12 Marsaban J, PATH-Indonesia, and MotherCare-IndonesIa (1996) Report CommumtyDzagnosls MotherCare Safe Motherhood Project, South Kalzmantan, 1996 Jakarta,IndoneSIa PATH and MotherCare/John Snow, Inc (get p #)

13 Marsaban J, PATH-Indonesia, and MotherCare-IndoneSia (1996) Report CommumtyDzagnoszs MotherCare Safe Motherhood Project, South Kalzmantan, 1996 Jakarta,IndoneSIa PATH and MotherCare/John Snow, Inc

14 Marsaban J, PATH-Indonesia, and MotherCare-IndonesIa (1996) Report CommumtyDzagnosls MotherCare Safe Motherhood Project, South Kalzmantan, 1996 Jakarta,IndoneSIa PATH and MotherCare/John Snow, Inc

15 Johnson BB (1996) Control Program ofAnemza m Pregnant Women The Results oftheQualztative InvestlgatlOn Washmgton, DC OMNI/John Snow, Inc

16 MotherCare (1996) QuestlOnnazre ofMothers ofChIldren Under FIve/PregnantWomen Safe Motherhood (lndonesza) Jakarta, IndoneSia MotherCare/John Snow, Inc

17 Marsaban J, PATH-Indonesia, and MotherCare-IndonesIa (1996) Report CommumtyDzagnosls MotherCare Safe Motherhood Project South Kalzmantan 1996 Jakarta,IndoneSIa PATH and MotherCare/John Snow, Inc

18 Marsaban J, PATH-IndonesIa, and MotherCare-Indonesia (1996) Report CommumtyDzagnoszs MotherCare Safe Motherhood Project South Kalzmantan 1996 Jakarta,IndoneSIa PATH and MotherCare/John Snow, Inc

19 Seoane GS, Kaune VM, and JV Cordova (1996) Dzagnostico Barreras y Vlabllz=adoresen la AtenclOn de ComplzcaclOnes Obstetncas y Neonatales (E'itudlO Cualztatzvo enComumdades y SerV1ClOS de Salud de Cmco Dzstrztos de Salud en La Paz yCochabamba) La Paz, BolIVIa MotherCarelJohn Snow, Inc, and Marketmg S R L

20 PareJa R and Galloway R (1996) TllP Reportto lndonesza May 27-June 19, 1996Washmgton, DC MotherCare/John Snow, Inc

21 Seoane GS, Kaune VM, and JV Cordova (1996) Dwgnostlco Barreras y Vwbllzzadoresen la AtenclOn de ComplzcaclOnes Obstetncas y Neonatales (EstudlO Cualztatlvo enComumdades y SerV1ClOS de Salud de Cmco Dlstrztos de Salud en La Paz yCochabamba) La Paz, BolIVia MotherCare/John Snow, Inc, and Marketmg S R L (get



p#)

22 Seoane GS, Kaune VM, and N Cordova (1996) Dwgnostlco Barreras y Vwblllzadoresen la Atencwn de Compllcacwnes Obstetncas y Neonatales (Estudw Cualltatlvo enComumdades y Servlcws de Salud de Cmco Dlstntos de Salud en La Paz yCochabamba) La Paz, BolIvIa MotherCare/John Snow, Inc, and MarketmgS R L (quote from partIcIpant), p 68 (old)

23 PareJa R Bollvw Tnp Report Apnl21-May 5, 1996 Washmgton, DCMotherCare/John Snow, Inc

24 Seoane GS, Kaune VM, and N Cordova (1996) Dwgnostlco Barreras y Vwblllzadoresen fa Atencwn de Compllcacwnes Obstetncas y Neonatales (EstudlO Cualltatlvo enComumdades y Servlclos de Salud de Cmco DIStrltOS de Salud en La Paz yCochabamba) La Paz, BolIvIa MotherCare/John Snow, Inc, and Marketmg S R L

25 PATH-IndonesIa (1996) Key Fmdmgs Focus Group DISCUSSIOns Among Women,Husbands and Key Informants on Safe Motherhood and on Anemw, m Three Dlstncts mSouth Kallmantan Jakarta, IndonesIa PATH-IndoneSIa

26 Marsaban J, PATH-IndonesIa, and MotherCare-IndonesIa (1996) Report CommumtyDzagnosls MotherCare Safe Motherhood Project South Kallmantan, 1996 Jakarta,IndonesIa PATH and MotherCarelJohn Snow, Inc

27 Seoane GS, Kaune VM, and N Cordova (1996) Dwgnostlco Barreras y Vwblllzadoresen la Atencwn de Compllcacwnes Obstetncas y Neonatales (Estudw Cuahtatlvo enComumdades y Servlcws de Salud de Cmco DIStrltOS de Salud en La Paz yCochabamba) La Paz BolIvIa MotherCare/John Snow, Inc, and Marketmg S R L

28 Marsaban J, PATH-IndoneSia, and MotherCare-IndonesIa (1996) Report CommumtyDwgnosls MotherCare Safe Motherhood Project, South Kahmantan 1996 Jakarta,IndonesIa PATH and MotherCarelJohn Snow, Inc

29 Marsaban J PATH-IndoneSia, and MotherCare-IndonesIa (1996) Report CommumtyDzagnosls MotherCare Safe Motherhood P, oject, South Kallmantan, 1996 Jakarta,IndonesIa PATH and MotherCare/John Snow, Inc

30 Seoane GS, Kaune VM, and JV Cordova (1996) Dwgnostlco Barreras y Vwblllzadoresen fa Atencwn de Compllcaclones Obstetncas y Neonatafes (Estudw Cualltatlvo enComumdades y Servlcws de Salud de Cmco Dlstrztos de Salud en La Paz yCochabamba) La Paz, BolIvIa MotherCare/John Snow, Inc, and Marketmg S R L


SEPTEMBER 15 199791

31 Seoane GS, Kaune VM and N Cordova (1996) Dwgnostlco Barreras y Vwblb=adoresen fa AtenclOn de CompbcaclOnes Obstetrlcas y Neonatafes (EstudlO Cuabtatlvo enComumdades y ServlclOs de Safud de Cmco Dlstntos de Safud en La Pa= yCochabamba) La Paz, BolIvia MotherCare/John Snow, Inc, and MarketIng S R L

32 Seoane GS, Kaune VM, and N Cordova (1996) Dwgnostlco Barreras y Vwblbzadoresen fa AtenclOn de CompbcaclOnes Obstetncas y Neonatafes (EstudlO Cuabtatlvo enComumdades y ServlclOs de Safud de Cmco DlStrltOS de Salud en La Paz y

Cochabamba) La Paz, BolIvia MotherCare/John Snow, Inc, and MarketIng S R L (p126)

33 Marsaban J, PATH-Indonesia, and MotherCare-IndonesIa (1996) Report CommumtyDwgnosls MotherCare Safe Motherhood Project, South Kabmantan, 1996 Jakarta,Indonesia PATH and MotherCare/John Snow, Inc

34 Seoane GS, Kaune VM and N Cordova (1996) Dwgnostlco Barreras y Vwblbzadoresen fa AtenclOn de CompbcaclOnes Obstetrlcas y Neonatafes (EstudlO Cuabtatlvo enComumdades y ServlclOs de Safud de Cmco Dlstnto5 de Safud en La Paz yCochabamba) La Paz, BolIvia MotherCare/John Snow, Inc, and MarketIng S R L (p96)

35 Seoane GS, Kaune VM, and JV Cordova (1996) Dwgnostlco Barreras y Vwblbzadoresen fa AtenclOn de CompbcaclOnes Obstetncas y Neonatales (EstudlO Cuabtatlvo enComumdades y ServlclOs de Safud de Cmco Dlstntos de Salud en La Paz yCochabamba) La Paz, BolIvIa MotherCare/John Snow, Inc and MarketIng S R L (p97)

36 Seoane GS, Kaune VM, and JV Cordova (1996) Dwgnostlco Barreras y Vwblbzadoresen la AtenclOn de CompbcaclOnes Obstetncas y Neonatales (EstudlO Cuabtatlvo enComumdades y ServlclOs de Salud de Cmco Dlstnto') de Safud en La Paz yCochabamba) La Paz, BolIvIa MotherCare/John Snow, Inc, and MarketIng S R L (p128)

37 Marsaban J, PATH-IndonesIa, and MotherCare-Indonesia (1996) Report CommumtyDwgnosls MotherCare Safe Motherhood Project, South Kalzmantan 1996 Jakarta,IndoneSIa PATH and MotherCare/John Snow, Inc

38 Marsaban J, PATH-Indonesia and MotherCare-IndonesIa (1996) Report CommumtyDwgnosls MotherCare Safe Motherhood Project South Kabmantan 1996 Jakarta,IndoneSia PATH and MotherCare/John Snow, Inc

F \nnachbar\comd"\.\report\\fina14 wpdSEPTEMBER 15 1997

92

39 Seoane OS, Kaune VM, and N Cordova (1996) DlGgnostlco Barreras y Vlabllz=adoresen la AtencIOn de ComphcacIOnes Obstetncas y Neonatales (EstudIO Cuahtatlvo enComunzdades y SerVlcIOs de Salud de Cznco Dlstntos de Salud en La Paz yCochabamba) La Paz, BolIvia MotherCare/John Snow, Inc, and Marketmg S R L (p91)

40 Seoane OS, Kaune VM, and N Cordova (1996) DlGgnostlco Barreras y VlGblhzadoresen la AtencIOn de ComphcacIOnes Obstetrlcas y Neonatales (EstudIO Cuahtatlvo enComunzdades y ServlcIOs de Salud de Cznco D1StrztOS de Salud en La Paz yCochabamba) La Paz, BohvIa MotherCare/John Snow, Inc, and Marketmg S R L (p88)

41 Bernard HR (1994) Research Methods zn Anthropology Quahtatlve and QuantltatlveApproaches (2nd EdItIOn) Thousand Oaks, CA Sage PublIcatIOns (p 209)

42 Bernard HR (1994) Research Methods zn Anthropology Qualztatlve and QuantztatlveApproaches (2nd EdItIOn) Thousand Oaks, CA Sage PublIcatIOns (pp 211-212)

43 Crabtree, p 16 (Get complete reference)

44 Patton MQ (1990) Quahtatlve EvaluatIOn and Research Methods (2nd EdItIOn)Newbury Park, CA Sage PublIcatIOns (p 289)

45 Patton MQ (1990) Quahtatlve EvaluatIOn and Research Methods (2nd EdItIOn)Newbury Park, CA Sage PubhcatIOns (p 169)

46 Peterson RA (1984) Askzng the Age QuestIOn Pubhc OpmIOn Quarterly, 48 379-383

47 Bernard HR (1994) Research Methods zn Anthropology Qualztatlve and QuantztatlveApproaches (2nd EdItIOn) Thousand Oaks, CA Sage PubhcatIOns (p 233)

48 Hehtzer Allen D, Makhambera M, and Wangel AM 1994 Obtammg sensItIvemformatIOn the need for more than focus groups Reproductlve Health Matters, 3 75-82

49 Bentley, P Personal commUnICatIOn, 1997

50 Bentley, P Personal commUnICatIOn, 1997

51 Patton MQ (1990) Qualztatlve EvaluatIOn and Research Methods (2nd EdItIOn)Newbury Park CA Sage PublIcatIOns (p 294-295)

F \nnachbar\comdx\report\\fmaI4 wpd

SEPTEMBER 15,199793

Documents

REpORTON THE USE OF THE COMMUNITY DIAGNOSIS TO …