Cultural and Social Factors Influencing Mortality Levels in Developing Countries
Post on 03-Apr-2018
215 Views
Preview:
Transcript
7/29/2019 Cultural and Social Factors Influencing Mortality Levels in Developing Countries
http://slidepdf.com/reader/full/cultural-and-social-factors-influencing-mortality-levels-in-developing-countries 1/17
http://www.jstor.org
Cultural and Social Factors Influencing Mortality Levels in Developing Countries
Author(s): John C. Caldwell
Source: Annals of the American Academy of Political and Social Science, Vol. 510, World
Population: Approaching the Year 2000, (Jul., 1990), pp. 44-59
Published by: Sage Publications, Inc. in association with the American Academy of Political
and Social Science
Stable URL: http://www.jstor.org/stable/1046793
Accessed: 14/08/2008 22:37
Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at
http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless
you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you
may use content in the JSTOR archive only for your personal, non-commercial use.
Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained athttp://www.jstor.org/action/showPublisher?publisherCode=sage.
Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed
page of such transmission.
JSTOR is a not-for-profit organization founded in 1995 to build trusted digital archives for scholarship. We work with the
scholarly community to preserve their work and the materials they rely upon, and to build a common research platform that
promotes the discovery and use of these resources. For more information about JSTOR, please contact support@jstor.org.
7/29/2019 Cultural and Social Factors Influencing Mortality Levels in Developing Countries
http://slidepdf.com/reader/full/cultural-and-social-factors-influencing-mortality-levels-in-developing-countries 2/17
ANNALS, AAPSS,510, July 1990
Culturaland Social Factors
InfluencingMortalityLevels in
DevelopingCountries
ByJOHNC. CALDWELL
ABSTRACT:ecentanalysesof ThirdWorlddata,bothat the level of national
or otherlarge aggregatesandat that of individualsstudied n sample surveys,have revealedthe surprising act thatsocial characteristics,uch as the level of
schooling or fertilitycontrol,or culturalcharacteristics,uch as ethnic group,are usually more influentialin determiningmortality evels than is access tomedicalservices, income,or nutritionalevels. Evidence from theUnitedStates
at thebeginningof thecenturysuggeststhat thiswas not the case earlier n the
West.Thisarticleexamines theevidence,showswhy developingcountriesare
currentlynan unusual ituation,andpresentsanthropologicalvidence on how
cultural, ocial, andbehavioral actorsachievetheir mpact.An attempts madeto begin the constructionof a moregeneral theoryof mortality ransition.
John Caldwell receivedhis Ph.D. degreeindemographyat theAustralianNationalUniversity,Canberra.He and his wife,Pat Caldwell,haveresearchedpopulation changevia anthropologicalanddemographic ield-research echniquesn Thailand,Malaysia,sub-SaharanAfrica,India,andSri Lanka.From1970 to 1988he was headof theAustralianNationalUniversity'sDepartmentofDemographyand now isdirectorof its Health TransitionCentre.He is authorof Theoryof FertilityDecline; and coauthorof The Causes of DemographicChange:ExperimentalResearch n SouthIndia.
44
7/29/2019 Cultural and Social Factors Influencing Mortality Levels in Developing Countries
http://slidepdf.com/reader/full/cultural-and-social-factors-influencing-mortality-levels-in-developing-countries 3/17
MORTALITYLEVELSIN DEVELOPINGCOUNTRIES
OVER the last 100 years,the historyof
mortality n the West has given all
the appearancesof supportinga common-sense and economic-deterministnterpre-tationof healthchange.During hatperiod,life expectancy in the most developedcountries increasedby more than 50 per-cent, from under 50 years to around75
years.Nothinglike this hadhappenedbe-fore inhumanhistory.But noone wasverysurprised,becauseotherchangesof funda-
mental mportancehadalsooccurred:withthe full floweringof the IndustrialRevolu-
tion, real incomes in the most economi-
cally advanced countries had multipliedalmost tenfold over that period. This al-
lowed people to be better fed and clad,
permitted the construction of improved
hospitalsthat the populacecould increas-
ingly afford to use, and providedthe re-
sources to treatdrinkingwaterandsewage.At thesametime,theinterrelated cientificrevolutionfirst madesafermedicalproce-dures possible and ultimately producedsulfa drugs,antibiotics,new vaccines,and
powerfulinsecticides.
Material improvement and scientific
advance seemed to have been the main
engines drivingdown mortality,an inter-
pretationthat this article will arguewasnot, for theWest,farwrong.Furthermore,for most of this period, the interpretationwas largely confined to the West, partlybecause most analysts lived there but
largelybecausethere was little in the wayof healthstatisticselsewhere.Thisfact wasto cloud our interpretation f the forcesbehind heglobalmortality ransitionwhen
itbegantogathermomentum nthepresentcentury.
Certainly, t was known thatthese ad-vanceswere notequallyshared.As earlyas1852WilliamFarrhaddemonstratedmajor
mortalitydifferentials nEnglandby socio-
economicclass,'butthesewereassumed o
reflect real differences in the means tobring minimum comfort and to pay fortreatment and little else. Not everyoneagreedthatindividualbehaviorplayedno
significantrole in determining helevel of
mortality,as was shown by the aims andactivitiesof the InfantWelfareMovementin English-speaking countries before
WorldWarI or by the MaternalandChild
Welfare Movements of the 1920s and1930s.2A revisionistapproach o the his-
tory of medicine has been developed inrecentdecadesby ThomasMcKeownand
colleagues,3 but the thrust of this workhas been to emphasizethe impactof eco-nomic changeat the expense of scientificmedicine.
NEWEVIDENCEFROMDEVELOPINGCOUNTRIES
A major hiftin ourinterpretationf the
mortalityransition asbeenmadepossibleby a change of focus to the developmentproblemsof the ThirdWorld,where life
expectancy in sub-SaharanAfrica is stillbelow50 yearsand nSouthAsiabelow 55
years.These are
populations argelylack-
1. William Farr,Vital Statistics: A MemorialVolume fSelectionsfromtheReportsandWritings fWilliamFarr(Metuchen,NJ:ScarecrowPress,1975).
2. EllenRoss,"Mothers nd theState nBritain,1904-1914" Paperdeliveredat theConferenceon the
HistoricalContext and Consequencesof Declining
Fertility n Europe,CambridgeMeeting,July 1989);Philippa Mein Smith, "InfantSurvival, the Infant
Welfare Movement and Mothers'Behaviour,Aus-
tralia with Referenceto New Zealand, 1900-1945"(Ph.D. thesis, AustralianNational University,Can-
berra,1989).3. Thomas McKeown, The Role of Medicine:
Dream,MirageorNemesis (London:NuffieldHospi-talsTrust,1967).
45
7/29/2019 Cultural and Social Factors Influencing Mortality Levels in Developing Countries
http://slidepdf.com/reader/full/cultural-and-social-factors-influencing-mortality-levels-in-developing-countries 4/17
THEANNALS OF THE AMERICANACADEMY
ing adequate death registrationsystemsand medical identificationof the cause of
death,so adequatedatabankson mortalityand social and economic conditionshave
been amassedby the United Nations and
the WorldBankonly in recentyears.This evidence shows that levels of in-
come andhealth servicesareweakpredic-torsof mortality evels and that social de-
terminantsapparentlyplay a majorrole in
determiningmortality.Thisarticlesumma-
rizes thatevidence, attempts o employ itto explain global mortality ransition,and
analyzesthe available nformation n how
social factors affect death rates.
It has become increasinglyclear that
ThirdWorld nationalmortality evels ex-
hibit a very differentpatternfrom what
their income levels would imply. Some
poor developing countries have largely
escaped the Malthusianshackles. A 1985Rockefeller Foundation Conference4se-
lected for investigatory study four Third
Worldsocieties thathadachievedlow lev-
els of mortality"at low cost": certainly,when comparedwith developedcountries
with similar mortalitylevels, at absolute
low cost, but not always at relativelylow
cost if the measure is the proportionof
national ncome spenton health andsocialservices likelyto assistthemaintenance f
health. The societies chosen-Sri Lanka;Kerala State, in southwest India;China;andCostaRica-all had life expectanciesin the 66- to 70-yearrange.In the case ofthe firstthreesocieties, thiswas at least 15
yearshigherthantheaverageforcountrieswith similarincomes, about3 yearslower
than Eastern Europe, and only 7 yearsbelowWesternEuropeandNorthAmerica.This level of mortalityhadbeen achieved
4. Scott B. Halstead,Julia A. Walsh,and Ken-nethS. Warren,GoodHealthat LowCost(New York:
RockefellerFoundation,1985).
by these three societies with per capitaincomes in the $300-400 range, or one-
fortieththat of WesternEuropeand one-fiftieth of North America.5 Considerable
attentionwas paid to the nature of theirhealth and social services.
Partof theexplanation or the high life
expectancies probablylies in the social-
service net that Sri Lanka, China, and
Costa Rica provide. It should be noted,however that Sri Lankaspends only 1.2
percent of its gross nationalproductonhealth, slightly above the averagefor its
income level, comparedwith 3.7 percentforWestern ndustrializedountries,which
spend, in absoluteterms,about140 times
as muchper person.6 ntermsof the num-
berof inhabitants er physician,SriLanka,with 7500 personsperdoctor,or 15 times
as many as in the West, is typical of its
income level;7Keralaprobablypresentsasimilarpicture,although he identification
for statisticalpurposesof doctors in both
India and China includes many personswhom othercountrieswould exclude andso makes comparisons impossible. Theconclusion is inescapablethatneitherin-
come northe levels of healthservicesandinterventionsare the primaryexplanation
for the remarkable ealthachievementsofthese societies.
The new data banksprovidedvaluableclues. Multivariateanalysis allowed thelevel of a rangeof socioeconomicfactorsand healthinputsin developingcountriesto be related to mortality levels. Flegg
5. Cf. WorldBank,WorldDevelopmentReport,
1988 (New York:OxfordUniversityPress,1988), pp.222-23, tab. 1.
6. Ibid.,pp.266-67, tab.23. Therearenofiguresfor Kerala, which probablyspends less, or China,while Costa Rica spends a proportionof the grossnationalproduct n the rangeof the proportion pentby the industrialized ountries.
7. Ibid.,pp.278-79, tab.29.
46
7/29/2019 Cultural and Social Factors Influencing Mortality Levels in Developing Countries
http://slidepdf.com/reader/full/cultural-and-social-factors-influencing-mortality-levels-in-developing-countries 5/17
MORTALITYLEVELS IN DEVELOPINGCOUNTRIES
showed that the level of literacywas the
best indicatorof low infant mortality,al-
though the degree of equality in incomeand the level of medical care also played
importantroles.8 Caldwell demonstrated
that ow nationalmortalitywas mosthighlycorrelatedwith the proportionof femalesin school a generationearlier,and that the
levels of familyplanningpracticeandmale
school attendancewere also importantn-
dicators of low mortality.9Lowercorrela-
tions were found with the ratio of doctorsto populationandnutritionalevels, andastill lower correlationwith income levels.
Recently,Rogers and Wofford have con-
firmed the prime role of literacy and the
proportionof the populationworkingout-
side agricultureand,of lesser importance,the safety of the water supply.'?Health
inputs,as measuredby the ratio of physi-
cians to population,showed a lower levelof correlation, while nutrition was notfoundto be significant.The importanceof
schooling a generationearlierlies in the
fact that it determinesthe current evel of
parental ducation,especially importantn
the case of mothers.The markeddifferen-tial in thesurvivalof childrenaccording othe level of mother'seducation had been
noted nGhana nthe1960sbyGaisie" andin a rangeof LatinAmericancountries n
8. A. T. Flegg, "Inequality f Income,IlliteracyandMedical Careas Determinants f InfantMortalityin UnderdevelopedCountries,"PopulationStudies,
36(3):441-58 (Nov. 1982).9. John C. Caldwell,"Routes o Low Mortality
in PoorCountries," opulationandDevelopmentRe-
view, 12(2):179, tab. 3 (June 1986).
10. RichardG.RogersandSharonWofford,"LifeExpectancy n Less DevelopedCountries:Socioeco-nomic Development or Public Health?"Journal ofBiosocial Science, 21:245-52 (1989).
11. S. K.Gaisie,DynamicsofPopulationGrowthin Ghana, GhanaPopulationStudies no. 1 (Legon,Accra: University of Ghana, Demographic Unit,1969).
the 1970s by Behm.Y2 he importanceof
child survival for determining mortality
levels lies inthefactthat, n high-mortalitycountries,typicallyat least one-quarter fall birthsresultin deaths before 5 yearsof
age, andbecause of theage structure f the
population,halfof all deaths n thesocietyoccurto personsunder5 years.
Meanwhile, these macro observationswere being increasingly supported byindividual-leveldata collectedby national
and subnationalrepresentative urveys inthe ThirdWorld.A majoropportunitywas
presentedby theorganization f surveysof
good scientific quality in 45 developingcountries by the World Fertility Surveyprogramin the decade after 1975. Twodifferentanalysesl3argelysupported achotherin theirfindingthatparentaleduca-tion is the most important nfluence on
child survival, with mother's schoolingusually havingthegreater mpact.Income,evidenced by father'soccupation,is also
important.Child mortalitydeclines with
every additionalyear of mother'seduca-tion with no lower threshold,so that even
one or two years of schooling in a ruralschool has some impact.Furthermore,o-cial influences are of greater mportance,
as evidencedby widerdifferentialmortal-ity betweengroups,for childrenaged 1-4
yearsthanforinfants,presumablybecause
12. Hugo Behm, Final Reporton the Research
Project on Infant and ChildhoodMortality in theThird World Paris:Comit6international e cooper-ationdans les recherchesnationalesen demographie,
1983).13. John C. Caldwell and Peter F. McDonald,
"Influence f MaternalEducation n Infantand ChildMortality:Levels andCauses," nInternationalPop-ulation Conference,Manila, 1981, (Liege: Interna-tional Union for the Scientific Studyof Population,1981), 2:79-96;J. N. Hobcraft,J. W.McDonald,andS. O. Rutstein,"SocioeconomicFactors n InfantandChild Mortality: A Cross-National Comparison,"PopulationStudies,38(2):193-223 (July 1984).
47
7/29/2019 Cultural and Social Factors Influencing Mortality Levels in Developing Countries
http://slidepdf.com/reader/full/cultural-and-social-factors-influencing-mortality-levels-in-developing-countries 6/17
THEANNALS OF THEAMERICANACADEMY
of theleveling effect on thelatterof nearlyuniversalbreast-feeding.
Mensch, Lentzner, and Preston ana-lyzed 15 surveysinAfrica, Asia, andLatin
America, only 6 of which were from theWorldFertility Survey program.They ex-
ploredthe impactof 12 groupsof social,
economic, andhealth-carevariables,bothuncontrolled and then controlledfor the
influence of theothervariables.Theycon-cluded that the majorinfluences on child
survivalwere mother'seducation,ethnic-ity, and, largely in urbanareas, father's
education.14Theextent towhichmaternal ducation
has beenidentifiedas amajor- or eventhe
major factor ndetermining hildmortal-
ity is astonishing,althougheven this find-
ing merely providesclues to the forces at
work ratherthan a simple answer.When
two socioeconomically similar areas inNigeria's Ekiti districtwere compared norder to discover the mortality mpactofdifferent levels of health services, it wasdiscoveredthat,even where therewas noaccess to modemhealthservices,mother's
schoolingwas animportant eterminant fchildsurvival.'5Research n southwestNi-
geria,especially in Ibadan, or the Chang-
ing African Family Project, showed thatthe importanceof maternaleducationre-mained after controllingfor the occupa-tions of bothmotherandfather, he urban-ruraldivision and the residential ocationwithin urbanareas,whetherthe marriage
14. United Nations, Socio-Economic Differen-tials in ChildMortality n Developing Countries,by
BarbaraMensch,HaroldLentzner,ndSamuelPreston,ST/ESA/SER.A/97NewYork:UnitedNations,1985).
15. I. O. Orubuloyeand JohnC. Caldwell,"The
Impact of Public Health Services on Mortality:A
Study of MortalityDifferentialsin a RuralArea of
Nigeria," Population Studies, 29(2):259-72 (July1975); Caldwell,"Routes o Low Mortality."
was monogamous or polygynous, and
whethertheparentspracticed amily plan-
ningor not.'6Mensch,Lentzner,andPres-tonconcluded:
When xaminedyitself,anadditionalearofmother'schooling educes hildmortality yanaverage crossour15countries f 6.8per-cent,with hemajorityfcountriesallingnthe
range f 5.0to9.0percent.Afterallother ari-ablesareenteredntotheestimationquation,theeffect s stilla reduction f 3.4 percentn
mortalityperyearof schooling.This latters the"direct" ffectof schoolingand s biaseddown-ward as an estimateof the "total" ffect by the
inclusion of variables whose value is partlydeterminedby mother'sschoolingitself.17
Maternal education and child survivalwere the focus of two papers,the first byClelandandvanGinneken'8 ndthesecond
by Clelandalone.The latterconcluded:
The mostimportanteatures f thematernaleducation-childhoodortalityssociationmaybe summarized hus:thereis no threshold; heassociation is found in all major developingregions; the linkage is strongerin childhood
than in infancy;only about half of the grossassociationcan be accountedfor by material
advantagesassociated with education;repro-ductive risk factorsplay a minorintermediaterole in the relationship; reaterequityof treat-mentbetweensons anddaughterss no partofthe explanation; the association betweenmother's education and child mortality is
slightly greater han for father'seducationand
mortality.19
16. John C. Caldwell,"Education s a Factor n
Mortality Decline: An Examination of NigerianData,"PopulationStudies,33(3):395-413 Nov.1979).
17. United Nations, Socio-Economic Differen-
tials, p. 287.18. JohnClelandandJeroenvanGinneken,"Ma-
ternalEducationand Child Survival in DevelopingCountries:The Search for Pathwaysof Influence,"SocialScienceandMedicine,27(12):1357-68(1988).
19. JohnCleland,"MaternalEducation ndChildSurvival: FurtherEvidence and Explanations," n
48
7/29/2019 Cultural and Social Factors Influencing Mortality Levels in Developing Countries
http://slidepdf.com/reader/full/cultural-and-social-factors-influencing-mortality-levels-in-developing-countries 7/17
MORTALITYLEVELSIN DEVELOPINGCOUNTRIES
The pointthatmay not havebeen suffi-
ciently stressed is that education has two
separatebutmultiplicative mpacts,oneon
individualswhose behavior s changedrel-ative to theirsociety andone thatchangesthe whole society. An examinationof theWorldFertilitySurveysconcluded:
There s as close a correlation etweenchildsurvivalandgeneral evels of [female]edu-cation in a communityas thereis betweenchildsurvival ndmaternalducation. nedu-catedwoman
mayeelmore
deprivednacoun-
trywheremostotherwomenareeducatedhanin one wheretheyarenot; nevertheless, erchildren tanda muchgreater hanceof sur-vival.Ifwe take hese wo factorsogether,hecontrasts reenormous.nLatinAmerica,hedeathrateamong hechildren f uneducatedPeruvian omen s almost timesgreaterhan
amongVenezuelanwomenwithsevenyearsofeducation.nAsia, hemortalitymong hil-
drenof uneducatedepalesewomens almost15timesgreaterhan t is amonghoseof Ma-
laysianwomenwithseven or moreyearsof
schooling.20
One persistent, but underresearched,
finding is that there are major ethnic orcultural differentials in mortality, espe-cially child mortality,even in the same
countryandwiththe sameaccess to health
services- differences hatsurvivecontrol-ling for income and education.This hasbeen shown between Chinese andMalaysin Malaysia2'and between the different
What We Know about Health Transition:The Pro-
ceedings of an InternationalWorkshop,Canberra,
May 1989, ed. John C. Caldwell et al. (Canberra:Australian National University, Health Transition
Centre,1990).20. JohnC.
Caldwell,"MassEducationas a De-terminant of Mortality Decline" (CASID Lecture,
MichiganState University,25 Oct. 1988), reprintedin Selected Readings in Cultural,Social and Bhav-iouralDeterminantsof Health, ed. JohnC. CaldwellandGigi Santow(Canberra: ustralianNational Uni-
versity,HealthTransitionCentre,1989), pp.103-11.21. Julie DaVanzo,William P. Butz, and Jean-
Pierre Habicht, "How Biological and Behavioural
peoples of the West African savanna inruralMali.22Mensch,Lentzner,andPres-
tonexamined 60 ethnicgroups n 11coun-
triesof Africa,Asia,andLatinAmericaandalmost invariably ound significantethnicdifferentialsnchildsurvival neachcoun-
try,with the mortality evel in one groupsometimes being twice or more that ofanother.They noted thatChinesepopula-tionsinSoutheastAsiaarecharacterized yunusually ow mortality.23
Oneaspect
of theimpact
on child mor-
tality of different cultural attitudes and
practices s thatof preference orsonsover
daughters. This preference, where it is
found,almostcertainlykeepschildmortal-
ity higher han t mightotherwisebe inthatthe preferentialtreatmentis unlikely toforce male mortalitydown by as greatanadditional incrementas it unnecessarily
keeps female mortalityhigh. The WorldFertility Surveys provided for the first
time,by theuse of a life-historyapproach,substantially ccuratedataon childmortal-
ity by age and sex. This showed that inthe sensitive age rangeof 1-4 yearsthereis excess female over male mortalitythroughoutnearly all North Africa, theMiddle East, South Asia, and East Asia,
withgreaterdiversity nSoutheastAsiaandLatinAmerica,and little or no additional
danger for females only in sub-SaharanAfrica.24Data on differential ex mortalityfrom the IndianSampleRegistrationSur-
vey and the InternationalDiarrhoealDis-
Influences on Mortality n MalaysiaVaryduring he
FirstYearof Life,"PopulationStudies,37(3):381-402(Nov. 1983).
22. Allan G. Hill, ed., Population Health andNutritionin the Sahel: Issues in the Welfareof Se-lected WestAfrican Communities London: KeganPaulInternational, 985).
23. United Nations, Socio-Economic Differen-tials, pp. 77-111.
24. Shea O. Rutstein,"Infantand ChildMortal-
ity: Levels, TrendsandDemographicDifferentials,"Comparative tudies:Cross-National ummaries,no.
49
7/29/2019 Cultural and Social Factors Influencing Mortality Levels in Developing Countries
http://slidepdf.com/reader/full/cultural-and-social-factors-influencing-mortality-levels-in-developing-countries 8/17
THEANNALS OF THE AMERICANACADEMY
eases Research Centre'sBangladeshpop-ulation laboratory n the Matlab district
demonstrate how culturally specific be-haviorcan affectmortality.Female mortal-
ityis not above thatof malesduring he first
yearof life, when breast-feedingprovides
equalnourishment ndprotectiveantibod-
ies; it is relativelyhigh in the 1- to 4-year
age range,when thatprotectionhas dwin-
dled but childrenare stillhighlydependenton others; t falls towardparitybetween 5
years and marriageas girls become morecapableof fending for themselves;and it
rises above that of males again during he
reproductive ears, argelybecause of highmaternalmortality in the poor obstetric
conditionsof much of SouthAsia. There-
after t falls below male levels.5
There is convincing evidence that the
achievementof a smallfamily,or even the
intentionof havingoneby employingbirthcontrol, s associatedwith declinesinchild
mortality.There is a correlationbetweennational evels of child mortalityandfam-
ily planning practice that compares onlywith that between maternaleducationand
ethnicity.26The Nigerian segment of the
ChangingAfricanFamily Project oundin
Ibadan ity child-mortalityevels thatwere
far lower among those women who hadachieved relatively small families than
amongthose who hadnot.27The One-per-
43, rev. ed. (London:WorldFertility Survey, Dec.
1984); Pat Caldwell and John C. Caldwell,"Where
There Is a NarrowerGapbetween Female and Male
Situations:Lessons from South Indiaand Sri Lanka"
(Paper delivered at the Social Science Research
CouncilWorkshop nGenderDifferentialsnMortal-
ity in SouthAsia, Dhaka,Bangladesh,Jan.1987).25. Caldwell and Caldwell, "Where There Is a
NarrowerGap."26. Caldwell,"Routes o Low Mortality," . 179,
tab. 3.
27. John C. Caldwell and Pat Caldwell, "TheAchieved Small Family: EarlyFertilityTransition n
anAfricanCity,"Studies nFamilyPlanning, 9(1):2-18, app.B (Jan.1978).
ThousandSurveyof Chinafoundextraor-
dinarily ow mortality mongonlychildren
whose parentshad completed the docu-mentationoptingfor thatstatus.28The in-
terrelationshere arecomplex and aredis-
cussed in the next section.I continued furtherwith the approach
adopted n the"GoodHealthat Low Cost"
conference.29A comparisonof the mortal-
ity andper capitaincome rankingsof the
99 ThirdWorldcountriesreportedully by
the WorldBankbecausetheirpopulationsexceed 1 million showed that,in termsof
their income, 11 did exceptionally well
withregard o health,being25 to62 places
higher in their health rankingsthan their
incomes would have predicted,while an-
other11didexceptionallybadly, alling25
to 70 placesbelowprediction.Armedwith
this informationand thatfromcorrelation
analysis, together with anthropological,
sociological,and historical nformation n
the societies thathad beenmost successfulin drivingdown mortalitywithin their in-
come constraints,hestudycameto a num-
ber of conclusions.Parental ducation s of
greatimportance, specially that of moth-
ers.So is the controlof fertilityor even the
attempt o controlit. Femaleautonomy
s
important,and its relative lack was the
mainreason that 9 of the 11 countries esssuccessful inconverting heir ncomesintolow mortalitywere found in the westernbranchof Islamstretching romSenegaltoIran. Grass-rootsradicalism,egalitarian-ism,anddemocracywereimportantnboth
creatinga successful populardemand for
health andeducational ervices andensur-ing that they worked. Neitherfemale au-
tonomy norradicalismhas as yet been as
successfully researched as education,
28. John C. Caldwell and K. Srinivasan,"NewData onNuptialityandFertility nChina,"PopulationandDevelopmentReview,10(1):71-79(Mar.1984).
29. Caldwell,"Routes o Low Mortality."
50
7/29/2019 Cultural and Social Factors Influencing Mortality Levels in Developing Countries
http://slidepdf.com/reader/full/cultural-and-social-factors-influencing-mortality-levels-in-developing-countries 9/17
MORTALITYLEVELSIN DEVELOPINGCOUNTRIES
partlybecause there are no simple mea-
sures equivalent to years of schooling.30
Clearly,as is discussedin the next section,these characteristicsare interrelated.The
studywent far toward uggestingthatsoci-
eties arelargely prisonersof theircultures
and historiesand thatthe rootsof contem-
poraryhealthsuccesseslie farback nthose
histories. The exceptions were the suc-
cesses achievedbyCommunist evolutionsin Chinaand Vietnamand, ess certainly
because it earlierwas markedby some ofthe othercharacteristics-by Cuba.
Where thegreatestsuccesses over mor-
tality have been gained, this achievement
has been the productof an interactionbe-
tween certainculturaland social character-
istics on the one handand the easy acces-
sibilityof basic modem health services on
the other.In spite of the fact thatparental
educationandthe practiceof fertilitycon-trol correlate so much more highly with
mortality evels than do medicalinterven-tionsin thecontemporaryThirdWorld, heevidencestrongly suggeststhat alone theycannot make dramaticreductions n mor-
tality levels. They may, in fact, correlate
morehighlybecause health nvestmenthas
beenrunningaheadof social investment n
termsof the optimummix.Sri Lankahadexperiencedmassive so-
cial change by the 1920s.The 1921 census
hadfound 56 percentof males and21 per-cent of females to be literate,a level thatPakistanwas not to reach for anotherhalf
century.Yetlife expectancywas little over30 years.3'It was the provisionof health
30. John C. Caldwelland Pat Caldwell,"Wom-en's Position and Child Mortalityand Morbidity n
LDCs," n Conferenceon Women's ositionand De-
mographic Change in the Course of Development,Asker(Oslo) 1988 (Liege:InternationalUnionfor theScientific Studyof Population,1988), pp. 213-36.
31. Caldwell,"Routes o Low Mortality";T.Na-
darajah,"Trendsand Differentials in Mortality,"n
Population of Sri Lanka (Bangkok:Economic and
services,first n urbanareasandthen,from
1945 onward,rapidlyin ruralareas, that
allowed the subsequentdramaticfall indeathrates.Yetearlier hecountryhadbeen
highlysensitiveto the need to combatsick-
ness andpossessed one of the mostexten-
sive and developedsystems of traditional
medicine n the world. Traditionalmedical
systems may provide solace and reduce
pain and even symptoms in chronic or
other conditions,but the evidence seems
clear that modem medicine is needed todrive downmortality ates.
Francehad reduced ts fertility evel to
the equivalentof 3.5 birthsperwomanby1850, but its life expectancywas only 39
years.32Three societies that had experi-enced a great deal of social change-SriLanka,Kerala,and Costa Rica-enjoyed
periodsof intensive activity when health
serviceswerespreadmuch morewidely tothe ruralpopulationsand the urbanpoor
during the years 1946-53, 1956-71, and
1970-80,respectively;neach case mortal-
ity fell muchmorerapidly han nearlieror
lateryears.33 herearealso societieswherelack of specific types of social change,often female autonomyor female educa-
tion,means thatmajor nfusionsof modem
healthservices do not achieve theirantici-
pated impact. By 1980 Libya employedmore doctorspercapitathanJapanor Ire-
land and was reaching the levels of theUnitedKingdomandNewZealand,but ife
expectancytherewas 16 yearsshorterandthe infant mortalityrate seven times as
high.Thereis, then,some kind of
symbiosisbetween social changeand modernmedi-
Social Commission for Asia and the Pacific, 1976),
p. 148.
32. NathanKeyfitz, WorldPopulation:An Anal-
ysis of Vital Data (Chicago:Universityof ChicagoPress,1968).
33. Caldwell,"Routes o Low Mortality," . 181.
51
7/29/2019 Cultural and Social Factors Influencing Mortality Levels in Developing Countries
http://slidepdf.com/reader/full/cultural-and-social-factors-influencing-mortality-levels-in-developing-countries 10/17
THEANNALS OFTHE AMERICANACADEMY
cine, the lattermeasuredmoreby itsacces-
sibility to a wide populationthan by its
level of technology.A comparisonof twosocioeconomically similarpopulationsin
Nigeria,one with access to a hospitaland
doctors and the other isolated from such
interventions, uggeststhatthegain in life
expectancy equivalent o the moreeasilymeasured hangesin childmortality- was
20 percentwhen the sole interventionwas
easy accessto adequatehealth acilities for
illiteratemothers,33 percentwhen it waseducationwithout healthfacilities, and87
percentwith both.34
Clelandand vanGinneken ummarized
datafroma wide rangeof countriesshow-
ing that the use of modem healthservicesincreased with duration of education.35
They believed that mostevidenceshowedthe interaction between education and
healthservices to be less spectacular hanthat found in Nigeria.36They reachedtheconclusion thataboutone-half of the verygreat differentialsfound across the ThirdWorld in child survival by educationofmother are probably explained by "eco-
nomic advantagesassociatedwith educa-tion (income, water and latrinefacilities,
clothing, housing quality, etc.)."37They
were morecautious abouthow the "pure"impact of mother's education was to bedivided between interactionwith modemmedicine and behavioral and care factorsthatpreventchildren frombecomingsickorhavingan accident n thefirstplace,but
theyemphasized hatbothwerelikelyto be
important. Income also interacts withhealth-serviceprovision,and this interac-
34. Ibid., p. 204; cf. Orubuloyeand Caldwell,
"Impactof PublicHealthServices."35. ClelandandvanGinneken,"MaternalEduca-
tion,"pp. 1361-62.
36. Ibid.,pp. 1362-63.37. Ibid.,p. 1360.
tion is especially strongwhenthere s little
attempt o providea free healthservice.
A rangeof researchers ttributemost ofthepureeffect of maternal ducation othebetteruse of modernhealthservices,buta
significantnumberalso attribute nimpor-tant role to family healthmanagementn-
dependent of curative services.38 In a
Nigerianvillage thatwas so farfrommod-em health services thatvery few childrenhad ever been takento doctors or nurses,
motherswith some schoolingexperiencedonly one-thirdthe child loss of motherswith no schooling. Only some of this canbe explained by greater use of modem
pharmaceuticals such as the malaria
suppressantsbroughtby a cyclist who ranan itinerantpharmacyservice.39 t mightalso be notedthat theskills in health man-
agement hatcanpreventchildren rombe-
comingsickordying nthe absenceof mod-em healthservices arethe same ones thatallow healthservices to be exploitedmore
successfully.Further onvincingevidenceof culturaland social differentials nchild
mortality n the pre-moder-medicine erahas been providedby researchamongso-cieties in ruralMaliwhere the modemerahas not yet begun.The substantialdiffer-
ence in child mortalitybetween adjacentculturalgroupswas explainedby different
styles of child care.40n contemporaryo-
cieties,somefamilies aremuchmoreproneto experiencesicknessand to lose childrenthanothers,as was shown fortyyearsago
38. For detailed references, see Caldwell and
Caldwell,"Women'sPosition,"pp. 222-23.
39. Orubuloyeand Caldwell,"Impactof PublicHealthServices,"p. 268.
40. KatherineHilderbrand t al., "Child Mortal-
ity and Careof Children n RuralMali"(Paperdeliv-eredat the National Institute or ResearchAdvance-ment and IUSSP Seminaron Social and BiologicalCorrelates f Mortality,Tokyo,24-27 Nov. 1984).
52
7/29/2019 Cultural and Social Factors Influencing Mortality Levels in Developing Countries
http://slidepdf.com/reader/full/cultural-and-social-factors-influencing-mortality-levels-in-developing-countries 11/17
MORTALITYLEVELSIN DEVELOPINGCOUNTRIES
for Newcastle-upon-Tyne,England,41nd
recentlyin India.42
INTERPRETING HE EVIDENCE
TOFORMULATEA THEORYOF
HEALTHTRANSITION
The first proposition s that there have
alwaysbeen socioeconomicdifferentialsn
mortality evels andthattheypredated he
impactof modernmedicine.This situation
was partlya function of income and the
abilityto eat betterandenjoyothermaterial
comforts,as Malthusnoted.43But the evi-
dence on social differentials by ethnic
group,andof greaterpropensity or child
loss in some householdsthanothers,even
in ratherhomogeneouscontemporaryEn-
glish urban areas or Indianvillages, sug-
gests thatsocial differentialswere also im-
portant. t is unlikelythatresearchwill ever
identify pre-modern-medicine opulationswith no social differentialsin mortality,
especially child mortality,but it is highly
probable that the differentials will be
smaller than those in the era of modem
medicine. It should be noted thatthis era
did notsuddenlybegin.Moreover,modem
41. J. Spence et al., A Thousand Families in
Newcastle-upon-Tyne:An Approachto the Study ofHealth and Illness in Children(New York:Oxford
UniversityPress,1954).42. Monica Das Gupta,"DeathClustering,Ma-
ternalEducationand theDeterminants f ChildMor-
tality nRuralPunjab, ndia," nWhatWeKnowabout
Health Transition, d. Caldwell et al.
43. Cf. JohnC. Caldwell, "Family Changeand
DemographicChange:TheReversalof theVeneration
Flow," in Dynamics of Populationand Family Wel-
fare 1987,ed. K. Srinivasan ndS. Mukerji Bombay:
Himalaya,1988), pp.71-96;JohnC. CaldwellandPat
Caldwell,"FamilySystems:TheirViabilityand Vul-
nerability:A Studyof Intergenerational ransactions
and Their DemographicImplications" Paperdeliv-
ered at IUSSP Seminaron ChangingFamily Struc-
turesandLife Coursesin LDC's,East-WestPopula-tion Institute,Honolulu,HI, 5-7 Jan.1987).
medicinehas become ever moreeffective,so thatsocial differentialsnmortalityaris-
ing out of interactionwith modernmedi-cine are likely to have increased in the
presentcentury.The second propositionis that a sub-
stantialpart,probably he majority,of the
explanation orsocial differentialsn mor-
talityinthecontemporaryThirdWorld ies
in the interactionwith modernmedicine.
Evidence for the mechanicsof thisinterac-
tion is presentedat the end of this article.The interfacebetweensociety and modem
medicine is broader hantheproponents f
scientific medicineusually ike to admit.It
includes not only doctors, nurses, mid-
wives, and pharmacistsbut also pharma-ceuticals distributed through traditional
markets, by wandering untrained sales-
men,and,on a massivescale,throughboth
traditionalmedical practitionersand non-traditional untrained practitioners or
quacks. This informal system helps to
changebeliefs andpracticeswithregard o
illness and its treatmentand increasinglyactsas a referral ystemto the moreformal
health sector. It is also probablethat this
uncontrolledspreadof modern medicine
saves more lives rather hancauses addi-
tional deaths, althoughthe whole matterhas hardlybeen researchedat all. This in-
formalsector s theonlychannelof modem
medicine to muchof ruralSouthAsia and
sub-SaharanAfrica and almost certainly
plays a role in the continuingdecline in
mortalityin both regions. The impact of
modem medicinein the formal sector is a
function ess of its scientific levels thanof
its accessibility throughrural clinics andnationalhealth schemes reducingthe costto the patient.The breakthroughperiodsin reducing mortalitylevels in differentThird Worldcountries have been associ-ated with the democratization f services,
53
7/29/2019 Cultural and Social Factors Influencing Mortality Levels in Developing Countries
http://slidepdf.com/reader/full/cultural-and-social-factors-influencing-mortality-levels-in-developing-countries 12/17
THE ANNALS OFTHEAMERICANACADEMY
not with an increase n the qualityof med-
ical technology.
The thirdproposition s thatthe varioussocial mechanisms dentifiedas playinga
role in reducingmortalityarereallydiffer-
entfacetsof thesamephenomenon,which
might be called social modernization,or
the rise of individualism rWesternization.It is really somethingbroaderandin many
ways is the socialcounterpartf thetransi-
tion from subsistence productionto the
marketeconomy.It is the move toward asystemwhere individualshaveoptionsand
canexercise choices- andrealizethat heycan do so and act on that realization.The
findingswithregardo maternal ducation,female autonomy,andgrass-rootsradical-
ism are all partof this picture. It is also
why sudden social shocks can accelerate
demographic rocesses,asthe FrenchRev-
olution and the Japanesedefeat in WorldWarIIdidin the case of thefertilitydeclineand as the Chinese Revolution did with
regardto mortalityeven if its stated aim
was farfrom the promotionof individual-ism. It is thedismantling f thesubsistence-
production organizationand the controland belief systems thatwas necessarytoensuresurvival.
Underlying these changes were pro-foundeconomicchanges.Itwas economic
growththatproduced he full market con-
omy andultimatelyallowed individualsa
degree of independencefrom the unified
familyeconomy.Nevertheless, n termsof
demographicbehavior,the shifts in belief
systemswereveryimportant.ntheareaofhealththe important hangeswere toward
a belief that sickness and death were theresult of nondivine andnonmagical orcesof thisworld,that herewas something hatcould be done about them in the form ofeithercarefulbehavioror seekingthebest
help,andeventuallythatmodernmedicinewas usually the most effective help that
couldbe obtained.In a studyareain rural
India,we called theprocesstheseculariza-
tion of healthbehavior.44 hatseculariza-tiondoes notnecessarily nvolveadiminu-
tion of religion, but it does involve its
retreat from intervention in causing the
everyday disasters of this world. In the
Indianvillage the decline of the so-called
little tradition nd tsvillage goddessesand
profusionof evil spirits,in the face of the
great tradition of mainstreamHinduism,
associatedwith literacy,courts,andcities,is an example of this, as was the Puritan
movement in England,which eventuallymoved ordinarypeople towardthe view
that mostearthlyphenomena n theirday-
to-day iveswere a matter f material ause
andeffect andthat o thinkotherwisecould
be blasphemous.It might be noted thatearlierbehavior was not irrational n that
much less could at that time be done toavertsickness and death.
The West experienced these changesfirst because of unparalleled economic
growthfromat least the sixteenthcentury.It was eventually to export some of this
growthas theworldmoved towardaglobaleconomy,but in theprocessit exported ts
behavioralbeliefs andsocial attitudesand
ultimately ts medical technologyand ac-
companyinghealth philosophy.This ex-
portwas achievedby colonialadministra-
tors, missionaries,the media, and, most
powerfully,by the moderneducationsys-tems thatareladen with Western,market,so-calledrationalvalues withregard obe-havior and family relationshipsand sys-tems. Theseconduitswere so effective be-cause they were hardlyconsciousof their
proselytizingrole butusuallybelievedthey44. John C. Caldwell,P.H. Reddy,andPat Cald-
well, "TheSocial Componentof MortalityDecline:An Investigation n SouthIndiaEmployingAlterna-tive Methodologies,"PopulationStudies,37(2):185-205 (July 1983).
54
7/29/2019 Cultural and Social Factors Influencing Mortality Levels in Developing Countries
http://slidepdf.com/reader/full/cultural-and-social-factors-influencing-mortality-levels-in-developing-countries 13/17
MORTALITYLEVELSIN DEVELOPINGCOUNTRIES
were propagatingeitherobjectivetruthor
objectivelydesirablebehavior.Theimpact
of such ideological exportshas probablyplayed a central role in the near-global
fertilitytransition hatis occurring.45There is compelling evidence that the
impactof maternal ducationon child sur-
vival is notmerelya case of learningmore
abouthealth.Themostimportant vidence
is that it occurs everywhere: in goodschools with good teachers who do teach
about health and in poor schools with un-derqualified eacherswho devote no time
to thesubject,as well as ineverypartof the
Third World. Even strongerevidence is
providedby the linear mpactof education
so that even a little elementaryschoolinghas a proportionalmpact.Clearly,we are
witnessingthegeneral mpactof ideas,ide-
ologies, and behavioral models. In rural
Bangladesh,Lindenbaum oundthatmoth-ers who had been to school were cleaner
andraised heirchildrenmorehygienicallyandcarefully,not becausetheyhad earned
that hiswouldsavethe children's ives but
becausethey assumed hat thosewith edu-
cation behavedin such a superiorway.46n
a south Indianruralarea,we found that
motherswith schooling associated them-
selves and their schooling much moreclosely with "modern" nstitutions- inde-
pendence and five-year plans as well as
health centers and the case for using
45. JohnC.Caldwell,Theoryof FertilityDecline
(London:Academic Press, 1982), esp. chap.9, "The
Failure of Theoriesof Social and EconomicChangeto ExplainDemographicChange:Puzzlesof Modern-
ization orWesternization," p.
269-300.46. Shirley Lindenbaum,ManishaChakraborty,
and MohammedElias, "The Influence of Maternal
Educationon Infant and Child Mortalityin Bang-ladesh"(Reportfor the InternationalCentrefor Di-
arrhoealDisease Research,Bangladesh, 1983), re-
printed n SelectedReadings,ed. CaldwellandSan-
tow, pp. 112-31.
them-than did illiteratemotherswho felt
thatthey were not partof this new world.
The educatedalso felt thisabouttheuned-ucated,thusreinforcing helatter'smental
set.47
A corollaryof thisarguments thatma-
ternaleducation s likely to producemuch
greaterdifferentials n child mortalityin
the contemporaryThirdWorldthanin the
West,eventhehistoricalWest,becausethe
market-attunedehavioralsystem had al-
readyevolved in the West. This baldstate-ment might be modified by noting that
educationin the West did acceleratethe
workingclasses' adoptionof middle-class
values and that "rational ndividualistic"
behaviorhas continuedto develop in the
West. Prestonhasproducedevidencefrom
the U.S. census of 1900 to show thatthe
gapbetween educatedprofessionalclasses
andthe rest of the society in childsurvivalwas much smaller than in the contempo-
raryThirdWorldand evidence fromBalti-
morein 1915 to show only small differen-
tials between literateand illiteratemothers
once father's ncome hadbeen controlled.48
He arguedthat this was because the level
of healthignoranceof the middle class in
America of thattime was high and closer
to that of theworkingclass than s the casein thecontemporaryThirdWorld.I subse-
quentlycontested this view, partlyon the
basis of a social-historical tudyof health
behaviorin nineteenth-centuryAustralia,
47. JohnC.Caldwell,P. H. Reddy,and Pat Cald-
well, The Causes of DemographicChange:Experi-mentalResearch nSouthIndia(Madison:Universityof Wisconsin
Press, 1988), esp. chap. 6, pp.132-60
andchap.7, pp. 161-86.
48. Samuel H. Preston,"Resources,Knowledgeand Child Mortality:A Comparisonof the U.S. in
the Late NineteenthCenturyand Developing Coun-
triesToday,"nInternationalPopulationConference,Florence 1985 (Liege: InternationalUnion for the
Scientific Studyof Population,1985), 4:373-86.
55
7/29/2019 Cultural and Social Factors Influencing Mortality Levels in Developing Countries
http://slidepdf.com/reader/full/cultural-and-social-factors-influencing-mortality-levels-in-developing-countries 14/17
THEANNALS OF THEAMERICANACADEMY
and maintained hat the lesser differential
was due to the fact thatmodern medicine
was much less developed and had rela-tively littleto offer.49 now believe thatmy
interpretation as onlypartof theexplana-tion and that the other partwas that, as
Western ducation'smessagewas so much
closer to the Westernbehavioralpatternof
the time, its impacton changingattitudes
and behaviorwith health mplicationswas
much less than in the contemporaryThird
World.It mightbe noted that PrestonandEwbank have produceda study showingU.S. child-mortalityrates by social class
wideningbetween 1895 and1925,"consis-
tent,"they argue"withthe fasteradoptionof behavioral innovations by the upperclass groups."50Amongthebehavioral n-
novations thatthey document s the grow-
ing resort to modern and increasinglyef-
fective medicineby professionalclasses.One furtherpointshould be made with
regard to education. All contemporaryThirdWorld data show a significant im-
pact on child mortality rom fathers'edu-
cation as well as thatof mothersandthat
much of this effect survivescontrolling or
income. Discussionhascenteredundulyon
the maternaleffect, even thoughthe exis-
tence of such an effect is good evidencethat education probably affects both afather'sattitudeand behaviorwith regardto his children's health and also his rela-
tionshipwith hiswife, with aresultant ec-
ondary mpacton his children'shealthandtreatment.
There is a related but distinct matterwith regard o children'shealth.That s the
49. Caldwell, "Routes,"p. 206.
50. Douglas C. EwbankandSamuel H. Preston,"PersonalHealthBehaviourand theDecline inInfantandChildMortality:The UnitedStates, 1900-1930,"in WhatWeKnow aboutHealth Transition, d. Cald-
well et al.
matterof the intrafamilialemotional and
resource-allocation priorities. These
change as the marketdevelops and withWesternizationbut are better treated as a
separate trand nsocialchange ikelytobeaccelerated or retardedaccordingto thenature of family structures. have calledthe intrafamilial low of resourceswealthflows and the changethatdirects more of
them toward childrenthanparentsor fa-thers the reversal of the intergenerational
wealth flows to a downwarddirection.51Thefourthpropositions thatchildmortal-
ity will fall more rapidly as the inter-
generationalwealth flow turnsdownward.
This almostinevitablyhappensas fertilitydeclines. Indeed, not only does parentalconcern for child survival ncrease,as has
happened n contemporaryChina,but sodoes communityand nationalinterest in
encouraging parentsto care for these in-creasingly rare and precious commodi-
ties, as happened n the case of the infant-welfaremovement ntheWest romaround1900 as the full extentof therecent ertilitydeclinebecame clear.Families arewillingto spendmoreeffort and a greaterpropor-tion of income on child care andsurvival.The situationis even morecomplex than
this because there is a correlation n theThird Worldbetween the level of familyplanningpracticeandchild survival evenbefore fertility decisively declines. Thereasonappears o be thatthe wealth flowhasbegunto turn; he familiesarealreadyplacingmoreemphasisonchildrenrelativeto theold, areplanning or theirfuture,are
findingthat the adequateallocation of re-
sources to each child for the successes ofthoseplanscan be attainedonlywithfewer
children,and are trying harderto ensuretheirsurvival.
51. Caldwell,Theoryof FertilityDecline.
56
7/29/2019 Cultural and Social Factors Influencing Mortality Levels in Developing Countries
http://slidepdf.com/reader/full/cultural-and-social-factors-influencing-mortality-levels-in-developing-countries 15/17
MORTALITYLEVELSIN DEVELOPINGCOUNTRIES
ELABORATIONS
Two modifications need to be intro-
duced to this pictureof social and family
change.The first is that the rate of family
changedepends o a considerable xtenton
its preexisting structure.In sub-Saharan
Africa, partlybecause of the lineage sys-tem andpartlybecauseof widespreadpo-
lygyny, wives usually have separatebud-
gets fromthose of their husbandsandare
themselves responsible for many of the
resources neededby theirchildren.52 his
gives mothers a great deal of autonomywith regardto health decisions affectingboththeirown andtheir children'shealth,but it often severely limits the resources
available. In these circumstances, a
strengthening f thespousalemotionaland
economic bond,as well as any movement
towardmonogamy,
islikely
to accelerate
child-mortalitydecline.Similarly n South
Asia and elsewhere, nuclear-familyresi-
dence in contrastto extended-familyresi-
dence is likely to give the young mother
greatercontrolover her children's healthtreatment.In south India, education can
producea degreeof emotionalnucleation
even within the extendedfamily and can
givea mother
greatercontrol over health
decisions affectingherchildren.Thesecondmodification s thata strong
cultural radition imitingwomen's auton-
omy, especiallywhen reinforcedby a reli-
gion thatregardsthe seclusion of women
as a prime moral objective, can have a
deleteriouseffect both on female healthin
generaland on all child health because of
52. John C. Caldwell and Pat Caldwell, "The
CulturalContext of High Fertility in Sub-Saharan
Africa,"PopulationandDevelopmentReview,13(3):409-37 (Sept. 1987);JohnC. Caldwell,PatCaldwell,and Pat Quiggin, "The Social Context of AIDS in
Africa,"PopulationandDevelopmentReview,15(2)
(June1989).
the limitation n mothers' akingquickand
effective action. This is the majorreason
why the Arab world does conspicuouslybadly relative to income in attaining ow
mortality.Suchtraditions an limit theed-
ucation of women and can limit the health
effectiveness of thateducation.This iswhyin the WorldFertilitySurveyprogram he
differentials n child survivalby mother's
educationwere so smallinBangladeshand
why in Syria and Jordan,although they
were considerable etweenmotherswho hadneverbeen to school andthosewho had-
possibly a cultural or ethnic effect-theywere very smallby durationof schooling.
The finalpropositions thatcultural, o-
cial, andbehavioral actorshave animpactbothonan individual'smortalityandonthe
mortalityof anindividual'sdependents.So
muchanalysishasbeen carriedout on child
survival because demographers' tech-
niquesforestimatingmortalityevels from
most Third Worlddataare muchbetter at
the youngest ages. This has also allowed
the specific study of the impact of the
mother'ssocial characteristics, ecause of
the particularly mportant ole she usually
plays in the treatment f youngchildren.It
would beunwise, however,
to believe that
parentsplayedthe sole role inensuring he
survival of children.In muchof theThird
World,grandparentsndsiblings play im-
portantroles. The improvedrelative sur-
vival chance of girlsafter5 yearsof age in
South Asia shows that increasinglychil-drenplay a role in their own survival and
presumably,then, that their own social
characteristics re increasingly mportant.There are data now for Europeshowingthat mortality rates for adult males aremuch lower among the more educated,53
53. T.Valkonen,"SocialInequality nthe Faceof
Death," nEuropeanPopulationConference:Plenar-
ies, ed. International nionfortheScientificStudyof
57
7/29/2019 Cultural and Social Factors Influencing Mortality Levels in Developing Countries
http://slidepdf.com/reader/full/cultural-and-social-factors-influencing-mortality-levels-in-developing-countries 16/17
THE ANNALS OF THE AMERICANACADEMY
evidence probablyof a firmer decision to
controltheirown lives andfate.Thedeter-
minantsofdependent
children'smortalityare a point of prime importance n high-
mortalitysocieties, for where life expec-
tancyis below 50 yearsand thepopulationis growing at 3 percentper annum,as in
sub-SaharanAfrica,the majorityof deaths
arelikely to be to personsunder5 yearsof
age.Witha life expectancyof 60 yearsand
a growthrateof 2 percent,however,as is
now found in SoutheastAsia,
thatpro-
portion drops to aroundone-quarter; nd
with a life expectancyof 75 years and a
growthrate of 0.5 percent,as now charac-
terizes the West, the proportionfalls to
one-fortieth.Hence, as mortality alls, the
emphasis on health behavior shifts from
parenthood o how the middle-agedlook
afterthemselves.
THE TRANSLATIONOF
BEHAVIOR NTOSURVIVAL
This section documents he fact thatbe-
havior,especially mother'sbehavior,can,inThirdWorldsocieties,be translated nto
lower child mortality.It focuses on rural
southIndia,with a life expectancyaround
50 years,and SriLanka,nearing70 years,areaswhere I have undertakenanthropo-
logical studiesof demographicbehavior.54
The term "health management" de-
scribes behavior that prevents sickness
Populationand theEuropeanAssociation forPopula-tionStudies,for theFinnishCentralStatisticalOffice
(Helsinki:Central tatisticalOffice,1987), pp.201-61.
54. Caldwell, Reddy, and Caldwell, Causes of
DemographicChange;JohnC. Caldwellet
al.,"Sen-
sitization to Illness and the Risk of Death:An Expla-nation for Sri Lanka'sApproach o Good Health for
All," Social Science and Medicine, 28(4):365-79
(1989);John C. Caldwell et al., "Cultural, ocial andBehaviouralDeterminants f HealthandTheirMech-
anisms:A Reporton Related ResearchPrograms,"n
WhatWeKnowaboutHealthTransition, d. Caldwellet al.
from occurringor limits the damageonce
it does occur.Greater emaleautonomyor
education ncreasesa woman'scapacity
n
healthmanagement n two ways: first,by
giving hergreaterdetermination ndself-
confidence and, second, by reducingthe
family and other constraintsplaced uponher. In traditionalsociety, child care is
oftena diffusedresponsibility.Galalel Din
showed how in aSudanesevillagechildren
were rathercasually looked after by the
wholevillage,
as well asby
theirsiblings,but, as mothers became more educated,
theytookgreater ontrolandresponsibilitythemselves.55A research program in a
north Indianvillage showed how women
in semiseclusion had little confidence in
theirabilityto identifysickness or to take
theappropriateteps.56n rural outhIndia,we foundthatmoreeducatedmothersgave
greateremphasisto
cleanliness, hygiene,nutrition,and the need for rest and sleepwhen childrenwere sick. Theywere more
effective in demanding from their hus-
bandsagreater hareof availableresources
for their children rather than for their
husband's relatives. When sickness did
occur, they were more likely to adoptef-
fective home action. This is an important
point,for home care is
reportedo consti-
tuteat east halfof alltreatmentntheThird
World.57
55. Mohamedel Awad Galal el Din, "TheEco-
nomic Value of Childrenin Rural Sudan,"in The
PersistenceofHighFertility:PopulationProspects n
the ThirdWorld,ed. John C. Caldwell (Canberra:Australian National University, 1977), 2:617-32;
idem, "The Rationalityof High Fertility in Urban
Sudan," n ibid.,2:633-58.
56. M. E. Khan etal., Inequalities
betweenMen
and WomennNutritionandFamily Welfare ervices:
AnIn-DepthEnquiry nan IndianVillage,PopulationandLabourPolicies ProgramWorkingPaperno. 158
(Geneva:International abourOffice, 1987).57. N. A. Christakis nd A. M. Kleinman, llness
Behavior and Health Transition n the DevelopingWorld,mimeograph Cambridge,MA: HarvardUni-
versity,School of PublicHealth,1989).
58
7/29/2019 Cultural and Social Factors Influencing Mortality Levels in Developing Countries
http://slidepdf.com/reader/full/cultural-and-social-factors-influencing-mortality-levels-in-developing-countries 17/17
MORTALITYLEVELSIN DEVELOPINGCOUNTRIES
In southIndia,we studiedin detailthe
various interrelationsbetween sick chil-
drenandtheirfamilies on the one handandthe medical service on the other n a situa-
tion where one village in a ruralareahada
governmenthealth center with a resident
doctor.Thepersonwho first detectedchild
sickness was in 80 percentof the cases the
mother; however, illiterate mothers were
unlikely to take action or even draw at-
tention to the sickness, waiting for their
mothers-in-lawor husbands to take noteand action. One reason thatmortalitywas
higher in the south Indian research areathan ntheSriLankanonewas thatonly 10
percentof mothers n theformer ook treat-
ment action on their own responsibility,
comparedwith 50 percent n the latter.As
a mother's education increased,she was
more likely to be the chief proponentof
action when her children were sick andmorelikelyto ensure hat heyweretreated
by the doctor.
In the same study,one of the steepestdifferentialsby maternaleducationwas in
the time spentby themotherwith the doc-tor. Given the absence of backuplabora-
tory testing, diagnosis depends to a verylargeextenton case historiesas presented
by mothers. Doctors think that illiteratewomen cannotadequatelypresentsuch ev-
idence and make relativelylittle efforttolisten to them.Partlybecause of theirlackof educationandpartlybecausedoctors ayless tothem, lliteratewomenare ess likelyto carryout the doctor'sinstructionsprop-erlyand less likely topersistwith thetreat-ment.Avery steepandsignificantdifferen-
tialbyeducation s foundwithregard othemother'sreactionwhen the child's condi-tion does not improve.With moreschool-
inga mother s increasingly ikelyto return
to the healthcenter oreport heproblem o
the doctor, while an uneducatedmother
frequently fails to do so partly on thegroundsthat the doctorhas alreadydone
his best andpartlyon the grounds hatshe
cannottell animportantman he hasfailed.
When we contrasted his situationwith
Sri Lanka,with its much higherlevels of
femaleeducation,we found n theSriLan-kanhouseholdan almostcompetitiveatti-
tude to thequickdetectionof sicknessand
the seeking of treatment. The strongestcontrast, however,was in the Sri Lankan
impatiencewith treatmenthatwas notre-
sulting n improvement ndtheconsequent
changing of doctors or from doctors to
hospitalsafter only a few days. The low
mortality evels suggest that this is an ef-fective treatment trategy.
THEDIRECTIONOF CHANGE
In most of the Third World,with its
limitedandonly slowly spreadingmodem
healthservices,mortalityevels canbe dra-
maticallyreducedby behavioralchanges.Those changesare not easily achieved,as
theyaffect notonlymortalityevels but thestructure f societyand all social relations.
Nevertheless,there is a potential or rapidchange that did not exist in the Westbe-
cause Westernsocial patternsare spread,
largelywithoutthataim in mind,by educa-
tion, the media, and religious proselytiz-
ing.Educationhas had a major mpact,andthisis now being supplementedby the wo-men's movement.Underlyingit all is the
development f themarket conomyandac-
companyingmovements way rom hefam-ily controlpatternsandresourceprioritiescharacteristicf subsistence griculture.
59
top related