Maternal health care in vietnam demand for antenatal care and choice of delivery care services

Problem statement

Thereisagrowingconcernaboutthematernalhealthcareglobally,especiallyinlowi n c o m e countries.WorldHealthOrganization(WHO2014)reportedthattheglobalmaternalm o r t a l i t y ratio(MMR)in2013was210maternaldeathsper100000livebirths,decreasingfrom3 80maternaldeathsper100000livebirthsin1990.However,theratioindevelopingregionsw as14timeshigherthanindevelopedregions.Eventhoughmaternaldeathisgenerallydecreasingworldwi de,ithasyettoachievethetargetofMillenniumDevelopmentGoal5byreducingthe

Thematernaldeathhasdirectcausesandindirectcauses.Thedirectcauseresultsfromarisingc omplicationsduringpregnancy,deliveryandpostpartum,orimpropertreatmentsuchashemorrhage, infection,obstructedlabor,unsafeabortion,ectopicpregnancyandanesthesia- relateddeathswhiletheindirectcauseresultsfromthediseasewhichpreviouslyexistsorben o t duetoindirectobstetric causeslike hepatitisanemia,malaria,heartdiseaseandtetanus( W H

O 2005).Itwasreportedthatdirectcausesmade upthehighernumberofmaternaldeatht h a n indirectcauseswith 80% ofthe total MMR (WHO2005).

Preventable complications during pregnancy can be significantly reduced through effective healthcare interventions, such as antenatal and delivery care, as highlighted by the WHO's Safe Motherhood package introduced in 1994 Antenatal care provides essential information to pregnant women and their families regarding maternal health and the growth of the unborn baby, allowing for dietary improvements that can prevent low birth weights Regular check-ups during this period help identify danger signs and risks associated with pregnancy and delivery, enabling timely interventions, such as tetanus immunization, which is crucial for protecting both maternal and infant lives Proper management of conditions like high blood pressure during pregnancy is vital for ensuring maternal health and improving infant survival rates Additionally, delivery care is critical in reducing maternal deaths, with WHO recommending childbirth in health facilities attended by skilled health staff to ensure safe deliveries and healthy babies Access to good hygiene practices and adequate medical equipment in these facilities further minimizes complications such as hemorrhage and obstructed labor, while skilled health professionals provide necessary emergency management.

InpursuitofMillenniumDevelopmentGoal5“Improvingmaternalhealth”,Vietnamalso ismakingprogressinimprovingthematernalhealthwiththedropofmaternalmortalityratio.The WorldBankshowsthatMMRinVietnamhasremarkableimprovementsinlast15yearsindecrease from81deathsper100,000livebirthsin2000to54per100,000in2015.T h e accesstoantenatalc are,animportantperiodforhealthofpregnantwomenandtheirbabyanddeliveryservicehasalsoinc reased.Multiple indicatorclustersurveyin2014(MICS5)s h o w s thatthepercentageofwo menaged15-

49withalivebirthinthelasttwoyearswhoreceivedantenatalcareatleastonceis95.8percentna tionwide.However,thereareconsiderabledisparityinmaternalmortalityratioandutilizationofmat ernalhealthcareamonge t h n i c i t y group,placeofresidenceandtheregionswherethepregnant women areliving.MPI2 0 1 5 reportedthatmaternalmortalityinmountainous areasismorethanth reetimeshighert ha n i n l o w l a n d areas.Furthermore,MICS5indicatest h a t t h e t i m e s o f pren atalc a r e v i s i t s differsamongthewomenlivingruralandurbanarea,especiallyregardinghavingmor ethan

4v i s i t s Inaddition,t h e ethnicm i n o r i t y groupsgetm o r e disadvantageo f accesst o maternalh ealthcarew i t h 7 9 % o f t h o s e having1 v i s i t a n d 3 2 7 % o f t h o s e h a v i n g atleast4 v i s i t s comparedto 99.2%and82.1 %of theKinhgroupasshown.Therefore,growingdisparities inhealthoutcomesandhealthcareutilizationhaveposed agreatchallenginginrecentyears.

Theabovechallengesleadtoseveralstudiesintheutilizationofmaternalhealthcarei n Vie tnam.Mostofthemfocusedontheinfluenceofdemographicandsocioeconomicfactors(Sepherietal.2

2012,Malqvistetal.2013).Demographicfactorswhichwereshowntoincreasetheprobabilityofthehealth servicesareyoungerage,l o w b i r t h orderw h i l e t h e f a c t o r s reportedt o decreaset h e p r o b a b i l i t y a r e separatedorunmarriedstatus,unintendedpregnancy.Inaddition,socio- economicfactorsmakegreaterinfluenceontheuseofthematernalhealthcareservices.Highereduc ationlevelofawomanisthemostimportantdeterminantreportedinthepreviousstudies(Sepherietal.2

008,Tranetal.2011,Golandetal.2012,Malqvistetal.2012,Malqvistetal.2013,Wakabayashi2014)

.Lowerhouseholdincomeisalsoshowntobeastrongfactorinthelikelihoodofusingmaternalhealt hcare(Sepherietal.2008,Golandetal.2012).Somestudiesemphasizedthed i s p a r i t y inth e maternalhealthcareutilization amongethnicmajority andminority groups(Malqvistetal.2012,Malqvistetal.2013).Ontheotherhand,majorequityinruralandurban areasalsowasidentifiedbyTranetal.(2011)andSepherietal.

(2008)estimatingtheimpactsofpovertyrate.Theomissionofthefactormaybiastheinfluenceo nm a t e r n i t y health c a r e utilizat ion, especiallyforth e disadvantagedw o m e n T h e o mi ss io n o f c o m m u n i t y effectsmayresultinbiasedestimatesoftherolesoffactors(Singhetal

2014).C o m m u n i t y b e l i e f s andn o r m s encourageorpreventhealthc a r e s e e k i n g beha viorso f t h e women.Ina d d i t i o n , community economicde ve lo pm en t may influence the accesst o health servicesandindirectlyincreasedecisionmakingpowerofwomenandpositiveattitudestow ardhealthserviceuse.

Community-level factors significantly influence maternal healthcare utilization Key indicators include the poverty rate, the percentage of women with higher education, and the rate of facility-based childbirth Higher poverty rates are linked to lower likelihoods of accessing antenatal care and facility deliveries, while a greater proportion of educated women correlates positively with the use of maternal health services.

Therefore,thereisaneedtoproperlyinvestigatethedeterminantsonhealthcareserviceutilizationi n c l u d i n g individuallevel,householdlevelandc o m m u n i t y l e v e l characteristics.Usingthelate stdatasetfromVietnamMultipleindicatorclustersurveyin2014(MICS5),thiss t u d y appliesthePo issonModeltoestimatetheimpactofsocialdeterminantsonthedemandf o r prenatalcarevisits,andt heMultinomialLogisticModeltomeasuretheassociationbetweensocialfactorsandchoiceon deliverycareproviders.

Researchobjectives

Theoverallobjectiveofthisthesisresearchistwofold.First,weanalyzethedemandofprenatalhealthcare.Particularly,weexaminethedeterminantsofthenumberofantenatalcarevisitsofwomentakingdurin gthelasttwoyearsusingdatafromMICS5.Second,weinvestigatethechoiceonservicefacilityfor deliveryofpregnantwomeninVietnam.

Researchquestions

Toinvestigatetheaboveobjectives,the followingquestionsneedto be answeredthoroughly:

Question 2: What arethedeterminantsof the choiceon deliverycareprovider?

Structure

Thep a p e r i s o r g a n i z e d asf o l l o w s InC h a p t e r 2,t h e generaltheoriesrelatedt o t h e d emandofhealthcareandthechoiceofhealthcareproviderarediscussedandthepreviousstudiesont heimpactofsocialdeterminantsonutilizationofmaternalhealthcarearereviewed.Chapter2presentsth econceptualframeworkandthemethodology withtwoappliedmethodsu s i n g thedatasetnamelyMultipleIndicatorClusterSurvey2013-2014(MICS5).Inthesamechapter,methodsinvestigatingthedeterminantsofthedemandofprenatal healthcareandthechoiceofservicefacilityfordeliverywillbepresented.Theresultsanddiscussionarepresent edChapter4.Chapter5concludes thepaperanddiscusses policyimplications.

Thischapterpresentsthetheoreticalreviewandempiricalreviewregardingthedemandf o r p renatalcarevisitsandthechoiceoffacilityfordelivery.Thefirstpartistoprovidether o l e ofm aternalhealthcareandtheoverviewofmaternalhealthcareinVietnam.Thenextparti s topresentthetheo reticalbackgroundforthedemandforhealthcareservices,andthechoiceo f healthcaref a c i l i t y a n d t h e i r determinants.T h e finals e c t i o n reviewst h e determinantsaffectingthedemandofpre natalcarevisitsandthechoiceofdeliverycareprovidersreportedi n thepreviousstudiesindeveloped countriesanddevelopingcountries,especiallyinVietnam.

Theroleofmaternityhealthcare

Motherhoodisapositiveexperiencewhichawomenencounters;however,thereares o m e healthproblemshappeningduringpregnancy,childbirth,andthepostpartumperiod.Theconsequ encesimpactseriouslynotonlyonthewomen’shealthbutalsoonthebabies.Threequarterofmater naldeathsisreportedtooccurduringchildbirthandthepostpartumperiod.However, antenatalcareanddeliverycarecanpreventthesecomplications.

Antenatal care (ANC) began in the early 1900s to support the health of pregnant women and their unborn children while detecting adverse conditions for timely interventions This care enhances women's understanding of fetal growth and their health status, helping to prevent issues like low birth weight through improved nutrition Women are also educated about the risks associated with pregnancy and delivery The World Health Organization (WHO) recommends at least one visit to a skilled health provider or a minimum of four ANC visits WHO guidelines for ANC include assessments of both mother and fetus, such as body weight, height, blood pressure, and various tests, alongside medical provisions like tetanus vaccination and iron and folate supplementation, as well as health education.

Thedeliveryathealthfacilitiesalsoplaysanimportantroletoensure womentogothrou ghchildbirthsafelyanddeliverhealthyinfants.Infact,propermedicaltechnologyandhygieni cconditionsduringd e l i v e r y c a n preventcomplicationsandinfectionsleadingt o m o r b i d i t y andm o r t a l i t y o f motherandh e r child.Inaddition,s k i l l e d b i r t h attendantsa r e avail ableinmosthealthfacilities.AccordingtoWHO,askilledbirthattendantisdefinedasamidwife,docto rorn urs e, who is wellskilled toensurenormal childbirthandthepostnatal

O recommendedthatincountrieswithveryhighMMR,atleast60%ofchilddeliveriess h o u l d beassistedbyskilledbirthattendantsby2015.During2005-

Overviewofmaternalhealth andhealthcareinVietnam

Theculture

VietnamcultureishighlyinfluencedbyConfucianism,especiallyinthenorthregiono f thecountry.AccordingtothetraditionofConfucianism,thesonofthefamilywillinheritt h e fa milyresourcesandworshiptheancestors.Inaddition,hehasresponsibility fortakingcareofth efamilymembersandmaintainsthecontinuityofthefamilyline.Therefore,givingb i r t h t o a sonbringsaproudto thefamilyandimproves the status of thewomenin thefamily.However,itwillputhighpressureofhavingasononthewomeninthefamily,especiallyfo rt h e womenwithdaughters.Thestrongpreferenceforsonsisthemainreasonforincreasingsexratioatbir th.

In many families, male members are viewed as the primary breadwinners and decision-makers, while female members are often seen as vulnerable, with their lives largely influenced by their parents After marriage, women typically reside with their husband's family, where financial control is often exerted by their in-laws and husbands The strong influence of Confucian values and existing hierarchies restrict their autonomy and independent decision-making, particularly regarding their health For instance, the childbirth experiences of mothers and mothers-in-law significantly impact the maternity care of younger women, potentially deterring them from seeking essential maternal healthcare services.

Thetwo-childpolicy

Thetwo- childp o l i c y i n Vietnamwasissuedi n l a t e 1 9 8 0 s w i t h restrictiono f t h e m a x i m u m n u m b e r o f childrenperhousehold.Vietnamgovernmentpracticeds o m e f a m i l y p l a n n i n g measures to reach thegoalsuchasprovidingthefreebirth controldevicesaswellasp r o s i n g thefacilitiesfor thosewhowasallowedforabortions.Inaddition,thefamilywhodidn o t complywiththetwo-

The child policy in Vietnam has led to various penalties for non-compliance, including high fees for families and delayed salary raises or demotions for government staff This fear of repercussions has caused some women to hide their pregnancies and avoid proper maternal healthcare Additionally, having more children places a significant burden on women, creating time and financial constraints that limit their access to maternal health services Following several revisions, the 2009 Population Ordinance now allows couples to decide when to have children, permitting one or two children per family As a result, Vietnam's total fertility rate decreased from 2.55 in 2001 to 1.99 in 2011, indicating the policy's effectiveness in stabilizing population growth However, challenges remain, such as ineffective contraceptive methods, with IUDs being popular despite their side effects, leading to hesitation among women Furthermore, Vietnam experiences a high abortion rate, particularly among youth, largely due to a lack of knowledge about contraceptive options, societal preferences, and financial or health-related issues.

MaternalmortalityratioandmaternalhealthcareinVietnam

InVietnam,thegovernmentrecommendspregnantwomenshouldhaveatleastthreepren atalvisitsduring thepregnancytodetectandpreventtherisksnegativelyaffectingthehealthof motherand baby.Thecontentofprenatalcareincludesbloodpressuremeasurement,u r i n e testing,b l o o d t e s t i n g andm e a s u r e ofweightandheight.Ina d d i t i o n , t h e nationalguidelinessuggestt h a t pregnantwomens h o u l d d e l i v e r th e b a b y ath e a l t h facilities.Propermedicalattentionandhyg ienicconditionathealthfacilitiesreducethecomplicationsoccurringi n andafterthechildbirth.Inad dition,forthecomplications,Caesareansectionisrequiredbuts h o u l d b e performedbyt h e s k i l l e d obstetricd o c t o r s t o e n s u r e s a f e childbirth.D u r i n g t h e postpartumperiodtheguideline srecommendatleasttwohealthcheckupsforbothmotherandchild.

The maternal mortality ratio (MMR) measures the number of women who die from pregnancy-related causes within 42 days after childbirth, per 100,000 live births during a specific period According to World Bank data, Vietnam has made significant progress in reducing its MMR over the past 15 years, decreasing from 81 per 100,000 in 2000 to 54 per 100,000 in 2015 This achievement has surpassed the Millennium Development Goal 5 target of 58.3 per 100,000 live births by 2015 However, despite this improvement, Vietnam still lags behind more developed Asian countries like Singapore and Malaysia in terms of maternal health outcomes.

Maternal mortality ratio (per 100,000 live births)

500 Lao PDR Malaysia Indonesia Thailand Cambodia

Brunei Darussalam Singapore Myanmar Philippines Vietnam

0 andThailand.Therefore,VietnamshouldmakegreaterefforttoreducetheMMRandensuret h e pers istentpopulationgrowth.

At least 4 times by any providers At least 1 visit by skilled health worker

To reduce maternal mortality rates (MMR) and infant mortality rates (IMR), the World Health Organization (WHO) recommends that each pregnant woman should have at least four prenatal care visits or at least one visit attended by professional health staff Antenatal care is crucial for providing essential information to women and their families about potential risks during pregnancy and childbirth Vietnam has made significant strides in antenatal care coverage, with 95.8% of pregnant women having at least one prenatal care visit in 2014, a notable increase from 1997 However, only 73.7% of women receive more than four visits, highlighting an area for improvement Additionally, there are considerable disparities in maternal healthcare utilization based on ethnicity, place of residence, and region Women in rural areas experience fewer prenatal care visits compared to those in urban settings, particularly regarding the number of visits exceeding four Ethnic minority groups face greater challenges in accessing maternal healthcare, with only 79% having one visit and 32.7% having at least four visits, compared to 99.2% and 82.1% of the Kinh group, respectively.

Figure3 : Percentageofwomenh a v i n g a t least1 v i s i t anda t least4 v i s i t s duringpr egnancy

Antenatal care visits by residence

Antenatal care visits by ethinicity

Thedemandforhealthcare

Theoreticalbackground

Grossman(1972)arguedthatwhatindividualdemandwhenpurchasinghealthcareisn o t healthcarebutgoodhealth.Firstly,hesetupthemodelofdemandforhealth,inwhichthe individualhaspositiveutilityofconsumptiongoodsandnegativeutilityofsicktime� 𝑠 (�). (Zweifeletal.2009)

Inaddition,thehealthstockHchangesovertime.Therefore,thehealthcapital decreaseattherate�.However,theindividualcanincreasehealthcapitalbyinvestinginI.In atwo-periodmodel,thecurrenthealthcouldbespecifiedas

Where,Uisutility,Hisstockofhealthcapital,�istherateofthedepreciationofthe heathstock,wiswagerate,Xisconsumptiongoods,Mistheconsumedamountofmedical services,Iisinvestmentinhealth,� 𝑠is sicktimeand� �is timeinvestedinfavorofh e a l t h

Grossman(1972)arguedt h a t t h e i n d i v i d u a l demandf o r h e a l t h f o r t w o reasons:first,as investmentcommodityandsecond,consumptioncommodity.

Basingonthefunction�(�,� � )and� 𝑠 (� 1 )from(1.1)and(1.2),Grossmanconstructed thedemandfunctionformedicalservicesin investmentmodel: ln�=𝑐����−(1+� � (�−1))���+(1+� � (�−1))���−(1−�)� 𝐸 𝐸 (1.3) Where,� �is theproductionelasticityofmedicalservicesand� 𝐸is theeffectivenessofedu cationE;�isthemarginalefficiencyofhealthcapital� 1

Fromtheutility function(1.1)including sicktime andconsumptiongood,heconstru ctedthedemand functionformedicalservices in consumptionmodel: ln�=𝑐����.−(1+� � (�−1))���+(1−�)(1−� � )���−(1−�)� 𝐸 𝐸−

Fromtheinvestmentmodel(1.3)andconsumptionmodel(1.4),thedemandofhealthservic esdependsonthepriceofmedicalservices,thewagerate,educationandwealth.Itcould beseenthatthe priceofmedicalservicesdecreasetheoptimumquantity of� 1whereas t h e w agerateincreasethequantityof� 1 Inthemultipleperiodmodel,bothofdemandfunctionsalsoareadd edtheagebecausethedepreciationrate�ispositivelycorrelatedwithage.(Zweifel etal.2009)

In addition to the Grossman Model, various studies have explored individual health-seeking behavior, notably through Rosenstock's Health Belief Model introduced in 1974 This model posits that an individual's beliefs about health problems, perceptions of benefits and barriers, and cues to action collectively influence health-promoting behavior Modifying variables, including demographic and psychosocial characteristics, indirectly affect these perceptions The model emphasizes that the perception of serious consequences and risks associated with health problems significantly increases the likelihood of engaging in health-promoting behaviors Following this, the perception of benefits from taking action versus barriers, such as inconvenience or side effects, plays a crucial role; if perceived benefits outweigh barriers, health-promoting behavior is likely to occur Cues to action, both internal (e.g., pain, symptoms) and external (e.g., information from friends or mass media), are essential in this process The model has been effectively applied to design interventions aimed at changing health-related behaviors by focusing on these key components.

ThaddeusandMaine(1994)developedthethreedelaytheorytoidentifythebarrierstot h e tim elyandadequatelyutilizationofmaternalhealthcare.Theyviewedthatthefirstphasewasdelayinsee kingcare.Thefactorseffectingthephaseincludeddecisionsofindividualandfamily,thestatusofwom en,previousexperienceregardinghealthcaresystem,financialando p p o r t u n i t y cost.Then extphasewasdelayingettingaccesstothehealthfacilityduetothea v a i l a b i l i t y o f facilitie s,distancet o facilities,transportationcostandtransportationinfrastructure.Thelastphasewasdelayi ntakingadequatecare.Therelevantcauseswerelacko f equipmentandskilledhealthstaffs,qualificat ion ofhealthstaffs.

ThebehavioralModelofHealthServicesUtilizationbyAndersen(1995)hasbeenusedi n t h e s tudieso n utilizationo f healthservicesi n b o t h developedandd e v e l o p i n g c o u n t r i e s (Thindet al.2008).IntheAndersen’sModel,oneindividualgettingaccesstoandusinghealthcareserviceswasafun ctionofthreefactors,namelypredisposing,enablingandneedfactors.T h e predisposingfactorsw ereclassifiedintothreegroupssuchasdemographiccharacteristics,

13 socialstructureandh e a l t h beliefs.Demographiccharacteristicsrepresentt h e t e n d e n c y o f i n d i v i d u a l s touseservices,includingage,gender,maritalstatuswhilesocialstructurere flectt h e a b i l i t y o f i n d i v i d u a l s t o seekhealthc a r e s e r v i c e s , i n c l u d i n g e d u c a t i o n , o ccupationandethnicity.Healthbeliefswereperceptionandattitudesregardingthehealthcaresyste msthatinfluencet h e u t i l i z a t i o n o f t h e services.T h e n e x t wase n a b l i n g factors,b o t h pers onalandorganizational,whichmeasuretheactualabilitytoobtainhealthservices.Personalenab lingfactorsincludedi n c o m e , healthinsurance,travelandw a i t i n g t i m e s w h i l e organiz ationalenablingfactorsweretheavailabilityofhealthfacilities.Thelastoneisneedfactors,thedirectcau sestoutilizationofhealthservices.Theneedwastheself- assessmentofhealthstatusandevaluationfromhealthstaffs.

EmpiricalLiterature Review

Theoretically,prenatalcareisconsideredavitalfactorinmaternalandinfantmortality.M e a s u r i n g andc o u n s e l i n g d u r i n g prenatalcarehelpwomenu n d e r s t a n d i n g h e a l t h s t a t u s o f herselfandherbaby,detectingrisksinordertoensuresafepregnancy,childbirthaswellasd ecreasepost- partumproblems.Whilesomeresearcheswereconductedtoestimatetheassociationbetweenmater nalhealthcareandhealthoutcome,somestudiesfocusondeterminantseffectingutilizationofmat ernalhealthcare.Thebelowistopresentthefactorsinfluencingthedemand o f prenatalcare visi tsandthe c h o i c e ofdelivery locationfromthepreviousstudies.

Thedeterminantsarecategorizedintothreecharacteristicsincludingindividuallevel,hou seholdlevelandcommunitylevel.Theindividualvariablesincludeeducationattainment,maternal age,maritalstatus,religion,ethnicitywhilehouseholdcharacteristicsarehouseholds i z e andho useholdwealth.Further,community levelconsistsofp la ce ofresidence,regiond u m m i e s , povertyrate andilliteracyrate.

(Arthur2 0 1 2 , Bbaale2 0 1 1 , Navaneetham& Dharmalingam2002).Itwase x p l a i n e d t h a t educatedwomenhadmoredecision-makingpoweronhealth-

2002) However, therewasno significantdifferenceintheutilizationofmaternalhealthcareamongeducatedwomenbylevelo f educa tion,e s p e c i a l l y betweenp r i m a r y l e v e l andsecondl e v e l

(Navaneetham& Dharmalingam2002).Inaddition,whenmeasuringtheeffectoftheintroduction ofNationalHealthInsurance(NHI)inTaiwan,Chenetal.(2003)foundthateducationattainmentdid nothaveinsignificantimpactonantenatalcarebeforeNHIbutwassignificantlypositiveafterNHI.T h e reasonforthisdifferencewasambiguous.Ingeneral,educatedwomenarewellinformedo f theimpo rtanceofmaternalcareandwillincreasemorefrequenciesofantenatalcarevisitst h a n less- educatedmothers.

(2008)suggestedthatthemaritalstatusimpactedtheuseofanyprenatalcaregreaterthanthen u m b e r of vi si ts buthasnosignificantinfluenceo n decisiono f deliveryplace.Itcouldb e e x p l a i n e d thatsinglemothersunwillinglygetaccesstothehealthcareduetofaceofstigmatizationbecauseth echildbirthiswidelyconsideredtheresponsibilityofmotheraswellasherhusbandinVietnam(Sepeh rietal.2008).Researchesinothercountrieswereconsistentw i t h thestudyofSepehrietal. (2008).InthestudyinTaiwan,Chenetal.

Theageofexpectantmotherisotherfactorintheutilizationofmaternalhealthservices.T h e highe ragethemothersget,thelowertheprobabilitytheywillseekhealthcare(Arthur2012).Itc o u l d b e e x p l a i n e d t h a t experiencea n d knowledgerelatedmaternalh e a l t h mayinfluenceonthema ternalhealthseekingbehavior(Chenetal.2003).Tsawe&Susuman(2014)showedthatthewomeninth eageof15-39morelikelytotakehealthcheck- upsfrequentlyt h a n thoseintheageofabove40.H o w e v e r , thereweresomeresearchesshowingthat women’sagewasinsignificantlyassociatedwiththeattendanceofANCforexample,inTurkey(C elik

Ast h e sameast h e ageo f expectantmother,mo st studiesf o u n d th at t h e increasingn u m b e r ofchildreneverbornhadthesamenegativeeffectontheuseofmaternalhealthcareservices.T hehigherorderbirthswilldecreasethelikelihoodofusingtheservicesforsomereasons.Itm aybed u e tot i m e andresourcec o n s t r a i n t s f o r womenhavinglargerf a m i l i e s

First-time mothers are more likely to utilize antenatal care due to their lack of experience (Navaneetham & Dharmalingam, 2002) However, previous negative experiences with childbirth can lead to fewer antenatal care visits, as argued by Arthur (2012) This sentiment is echoed by Tsawe & Susuman (2014), who noted that maternal healthcare usage is significantly influenced by prior experiences with the services provided Women who receive better care are more inclined to return for additional services Conversely, policies such as the two-child policy and fears of penalties may discourage those with multiple children from seeking adequate maternal healthcare (Sepehri et al., 2008).

(2013)f o u n d thatpregnancyintentionwassignificantlyrelatedtotheutilizationofantenatalcare buti n s i g n i f i c a n t l y tothedeliverycare.Thereasonwhytheunwantedpregnantwomenlessli kelyt o receiveantenatalcareadequatelyisambiguous.Wadoetal.

(2013)hypothesizedthatwomenw i t h unwantedpregnanciesd i d n o t preparewelli n t h e respecto f emotionandfinancef o r childbirthandchildbearingsotheylesstakecareoftheirhealthandtheirun born.Oneoftheargumentisthewomenwithunintendedpregnanciesrecognizedtheirpregnancyl atesotheym i s s t h e firstt i m i n g antenatalc a r e v i s i t s W a d oeta l

( 2 0 1 3 )observedt h a t womenw i t h unwantedpregnancydetectthepregnancyaroundonem onthlatercomparedwithoneswithpregnancyintention.Ingeneral,t h e s t u d y impliedt h a t t h e pr egnancyintention wash i g h l y associatedwiththeutilizationofmaternalcare;however,itsassocia tionwiththedeliverycarewasstillunclear.

Somestudiesresultedthatfrequenciesofusingmassmedia positivelyrelatedtotheutil izationofANC.Thehigheristhelevelofexposuretomassmedialikeradioandtelevisiont h e higherist helikelihoodofusingANC.Navaneetham&Dharmalingam(2002)foundthatwomenw a t c h i n g t e l e v i s i o n andl i s t e n i n g radior e g u l a r l y increasedt h e o p p o r t u n i t i e s o f herseekingANC.Itcanbeexplainedthatthemassmediawillprovidetheinformationregarding tothematernalhealthandenhancetheawarenessoftheavailablematernal healthservices.Tsa we&Susuman(2014)reportedthatthosewhohadmoreknowledgeofthematernalhealthservicemore likelytoreceivetheservicetimelyandadequately.Infact,lackofinformationposesthechallengest o theutilization ofmaternalhealth.

The significance of working status in maternal healthcare utilization is well-documented in previous studies Navaneetham and Dharmalingam (2002) found that non-working women are more likely to seek maternal healthcare services compared to their employed counterparts, potentially due to their relatively higher socioeconomic status In developing countries, women's employment often comes with low pay, which decreases the likelihood of accessing maternal healthcare Additionally, Baale (2011) highlighted that the type of job held by pregnant women influences their use of antenatal care, as job conditions and income significantly affect healthcare affordability Women employed in factories tend to utilize antenatal care more than housewives, and those working in government are more likely to seek antenatal care when covered by national health insurance (Chen et al., 2003).

Research indicates that women from ethnic majority groups in Vietnam are more likely to give birth in health facilities compared to those from ethnic minority groups, primarily due to communication challenges stemming from language barriers Additionally, the lower rates of maternal healthcare access among ethnic minority women can be linked to their lower levels of education and poorer socio-economic conditions This pattern is reflected in other countries as well; for instance, women from "Scheduled" castes and tribes in India have been found to have fewer antenatal care (ANC) check-ups, while Kudish women in Turkey are similarly less likely to utilize ANC services.

SimilartoEthnicity,religiousdifferenceisoneofthemostimportantdeterminantsofhealths erviceutilizationbecausereligionaffectsthewaypeopleliveandbehaveaswellast h e i r belief (Bbaale2011).Ifwomenbelievinginthetraditionalbirthattendantsmorelikelygivebirthathom eratherthangoingtohospitalorclinic(Tsawe&Susuman2014).Studyingi n India,Singhetal.

Muslimrestrictwomenamongthisreligiont o enjoyhealthcarebenefits.Similarly,Muslimwom eninKeralatendtogivebirthathomeandrefusetobeassistedbyskilledhealth providers;however,theprobabilityofantennalcarev i s i t s i n t h e firsttrimesterishigheramongthem(Navaneetham&Dharmalingam2002).

Previous research indicates that household wealth positively influences the utilization of antenatal care (ANC) services (Bbaale, 2011; Tsawe & Susuman, 2014) For instance, Navaneetham and Dharmalingam (2002) found that women with higher living standards are three times more likely to receive adequate prenatal care compared to those with lower living standards Additionally, owning a car has been associated with increased access to maternal healthcare (Celik & Hotchkiss, 2000) Despite some countries offering free healthcare services, barriers remain, including direct and indirect costs such as transportation (Arthur, 2012) Women who struggle to afford travel expenses to hospitals are less likely to seek ANC compared to those who do not face financial barriers (Tsawe & Susuman, 2014) Overall, financial constraints significantly hinder the utilization of maternal healthcare services.

PlaceofresidenceiskeyconcerninANCutilization.Thepreviousstudiesresultedthatwomeni nruralareaslesslikelyusematernalhealthservicesthanurbanwomenduetolongdistancetotheh ealthfacilities.Inaddition,financialproblemsalsowereabarrierforthemtocompleteuseofmaternal healthcarebecausetheycouldnotaffordthetransportationcostaswellasANCcost(Tsawe&Susum an2014).Theshortageofskilledattendantalsoisthereasonw h y ruralwomenlessfrequentlyundertak eantenatalcarethanurbanarea.Infact,thereislacko f qualifiedhealthproviderinruralareaincompari sonwiththecities(WHO,2006).

Understandingthesematters,somecountriesappliedtheprogramtominimizetheproblems.Fo rexample,whenstudyingfourstatesinthesouthIndianin2002,Navaneetham&Dharmalingam(2 002)foundthat thereisnosignificantdifferencebetweenruralandurbanareaintermsofthede mandforantenatalcare.Theunderlyingreasonwasthatmultipurposehealthworkersprovideant enatalcaretopregnantwomenbyvisitingtheirhome.Incontrast,Celik&Hotchkiss(2000)foundthat urbanandruraldifferencedidnotsignificantlyimpacttheprenatalc a r e afterh o l d i n g t h e constantr egionals t a t u s ando t h e r variablesi n t h e s t u d y i n Turkey.

Thedifferenceinutilizationofmaternalhealthca re isduetotheimplementationofhealth c a r e program,t h e a v a i l a b i l i t y anda c c e s s i b i l i t y ofth e h e a l t h careserviceamongt h e regi ons.Studyingi n Vietnam,Sepehrieta l

(2008)pointedo u t t h a t womenl i v i n g i n t h e disadvantagedregionslikeCentralHighlandsles slikelyundertakeanymaternalhealthcaret h a n thoseliving inRedRiverDelta.However,there wasnosignificantregionaldifferenceint h e frequencyofprenatalvisits(Sepehrietal.,2008).Simila rly,inTurkey,Celik&Hotchkiss(2000)alsofound thatlivingindevelopedregionsofthecountr yhadsignificant ly positiveassociationwithattendanceofprenatalcare service.

Besidei n d i v i d u a l levelandhouseholdlevelcharacteristics,c o m m u n i t y levelcha racteristicshavebeenpaidgreaterattentioninthepreviousstudiesinrecentyears(Gage2 0 0 7 , S epherietal.2008,Ononokponoetal.2013andSinghetal.2014).Stephensonetal.

(2006)arguedt h a t c o m m u n i t y couldi n f l u e n c e healthoutcomesthroughseveralways.Forexa mple,communitybeliefsandnormsmayhavegreateffectonpositiveornegativeattitudeso n healthc a r e services,whichi n t u r n l e a d t o healthc a r e s e e k i n g b e h a v i o r Inaddition,economic developmentofacommunitywillattracttheinvestmentofhealthcareinfrastructure,socials u p p o r t a swella s improvemento f k n o w l e d g e i n b e n e f i t s o f h e a l t h careservices.Studyinginru ralMali,Gage(2007)suggestedthatcommunitywithhigherportionofwell- educatedwomenwas highlyassociatedto likelihood ofgettingaccesstomaternalhealthcare.InVietnam,Sepehrietal.

(2008)pointedoutthatwomenlivingincommunity withhigherp o v e r t y ratehavefewerte ndencytodelivertheirbabyinhealthfacilitiesincomparisonwitht h o s e livingincommunitywit hlowerpovertyrate.Oneofthecommunitylevelcharacteristics,

19 whichwerealsofoundtobepositivelyassociatedwiththedemandofprenatalcarevisitsistheproportiono fwomeninsamecommunitygivingbirthathealthfacility (Ononokponoetal.2 0 1 3 ) Th eresearchshowed that thewomenlivingin thecommunitywiththehighproportionm o r e l i k e l y 5 8 t i m e s t o t a k e adequateprenatalc h e c k - u p comparedt o t h o s e i n t h e lowerc o m m u n i t y levelproportion.

Thechoiceofhealthcare provider

Theoreticalbackground

Int h e l at e1 98 0s , Gertler,LocayandS a n d e r s o n (1987)f i r s t l y developedt h e healt hseekingbehaviorinthechoice ofhealthcareproviders.Theyarguedthatatthefirststage,thei n d i v i d u a l decideswhethertosee khealthcareandatthesecondstage,heorsheselectsthehealthcareproviderswiththemaximumuti lity.Theutilityfunctionofindividualireceivinghealthcare fromproviderj ispresentedasfollows:

Inwhich,� ��i s t h e u t i l i t yofthe individual iafter receivinghealthcarefromproviderj,ℎ ��i s expectedhealthstatusoftheindividualafterreceivinghealt hcarefromproviderj and

Thehealthstatusafterreceivinghealthcarefromproviderjforanindividualidepends on the qualityofproviderj’shealthcare

Therefore,thequalityofhealthcaredependsonthecharacteristicsofhealthcareproviders� � andcharacteristicsofindividual� �as follows:

Theotherconsumptionexpenditure𝐶 �a fterreceivingthehealthcareistheremainingofthei ncome� �a fterpayinghealthcareproviderj.Theprice� ��o f thealternativejisthe paymentforthehealthcareincludingdirectcostsuchasconsultationcost,medicinecostandindirectco st suchastransportationcostandwaitingtime.

Researcherscouldnotmeasuretheutilityoftheindividualbutinvestigatethe characteristicso f t h e alternativesandt h e characteristicso f t h e i n d i v i d u a l (Train,2 0 0 9 ) Therefore,the utilityfunction oftheindividualisexpressedasfollows:

Theobservedcharacteristicso f t h e individualaregender,a g e , education,income, insurancewhereastheunobservedcharacteristicsareperceptionofqualityofthehealthcareprov ider,thepreferentialmedicaladministration.The observedcharacteristicsofthehealthcareprovid eraretheprice,thedistancefromhouseofthepatienttothehealthcareproviderwhereastheunob servedcharacteristicsarethe fame,prestigeof thehealthcareprovider.

Empiricalliteraturereview

Itiswidelyacknowledgedthatchildbirthathealthfacilitiesisassociatedwithlowerrateo fmaternalmortalityandmorbidity.However,therearestillwomenchoosingthechildb i r t h at home,especiallyindevelopingcountries.Therefore,somepreviousstudieshavefurtherconductedtoinve stigatethedeterminantsofthechoiceofdeliverycareprovider.Theyhavefocusedthecharacteristi csofindividualofthewomen,householdandcommunitywheretheyareliving.

(2006)arguedthattheprenatalcareduringpregnancywashighlyassociatedwiththechoiceofdeliverya thealthfacilitiesbyinformingthebenefitofinstitutionaldeliverya n d appropriateservices.Sepehr ieta l

(2008)agreedt h i s p o i n t t h a t t i m i n g andadequatep r e n a t a l carev i s i t s couldraiset h e awarenesso f t h e needo f c a r e f o r delivery.

Educationattainmentalsoisfoundtoplaytheimportantroleinthechoiceofchildbirthplace(Sep ehrietal.2008).Forexample,Celik&Hotchkiss(2000)pointedoutthatthewomenw i t h highereduc ationm o r e l i k e l y t o choosef a c i l i t y d e l i v e r y r a t h e r t h a n traditionalh o m e delivery.Iti s l i k e l y t h a t t h e h i g h l y educatedwomenhavet h e s t r o n g e r andindependentdecisionstot hebesthealthcareservices.Furthermore,theeducatedwomenhavemuchmorechanceto bewellinformed of thebenefit offacility-baseddelivery.

Withrespecttothebirthorder,thepreviousstudiesreportedthatithasstrongassociationw i t h t h e c h o i c e o f d e l i v e r y ath e a l t h facilities.Navaneetham& Dharmalingam(2002)f o u n d t h a t t h e womenw i t h firstb i r t h orderm o r e l i k e l y to giveb i r t h ath e a l t h carei n s t i t u t i o n s ratherthanthosewithsecondbirthorder.Therearesomeargumentsregardingthematter.Someresearc herspointedoutthatthewomenwithhigherparityfacedtimeandresourceconstraintsto utilizethefacilitiesservicesdueto largefamilies.On theotherhand,itcouldbet h e badexperiencesfrompreviouschildbirth,whichmakethemtounderes timatethedemandf o r facility-baseddelivery.

Concerningtheageofthepregnancywomen,therearemixedresultregardingitseffecti n thec hoiceofbirthdelivery alternatives.Celik&Hotchkiss(2000)reportedthattheageofwomenatlastchildbirthwasnot significantlyassociatedwith thechoiceofdeliverylocation.O n contrary,S t e p h e n s o neta l

( 2 0 0 6 )arguedt h a t t h e a g e oft h e i n t e r v i e w e d w o m e n hadsignificantassociationwitht hechoiceoffacilitydeliveryinthestudyofsixAfricacountries.H e foundthatthewomenwith a geof40-49and30-

Inaddition,maritals t a t u s showedm i x e d r e s u l t i n t h e d e c i s i o n i n t h e b i r t h d e l i v e r y location.Stephensonetal.

Thechoiceoftheplaceofdelivery isalsohighlyinfluencedbythehouseholdlevelc haracteristics,n a m e l y h o u s e h o l d wealthi n d e x , e t h n i c i t y andreligion.W i t h respectt o t h e householdwealthindex,Stephensonetal.

(2006)foundthatwomenwithhigherhouseholdi n c o m e indexmorelikelytochoosethe deliveryathealthinstitution ratherthanthosewithlowerincomeindex.Thereasonisthatthec ostrelatedbirthdeliverycareaswellastransportationrestrictst h e p o o r w o m e n fromu t i l i z i n g t h e healthservices.Somepreviousstudiesidentifiedthatethnicityhasgreatinfluenceinthechoiceo fthefacilitydelivery.Celik

&Hotchkiss(2000)agreedthattheeffectistrueinurbanwomenandruralwomen.Hesaidt h a t i tcouldbeduetotheculturalandeconomicbarriersorthepoorofhealthcareservice.Similart o t h e ethnicity,religionalsoi s consideredast h e k e y determinanti n t h e women’sdecisionintheplaceof childbirth.WhenstudyingsixAfricacountries,Stephensonetal.

(2006)f o u n d t h a t M u s l i m womeni n Ghanaarel e s s l i k e l y t o delivert h e i r b a b y i n healt hf a c i l i t y comparedtoCatholicwomen,whileProtestantwomenmorelikelytoutilizetheservic ethanCatholicwomen.

&Hotchkiss(2000)foundthaturbanwomenweremorelikelytoseekthefacilitydelivery ratherthanruralwomen.Furthermore,Celik&

Hotchkiss(2000)saidthattheEasternregionsi n TurkeyislessadvantagedthanWesternandNo rthernregions,whichcausedmoredifficultiesforthe womeninEasternregionstochoosefa ci li t y deliverycomparedtothose l i v i n g inotherregions.Gage(2007)arguedthatthe regionaldi fferencespartlyindicatetheunequalaccessibilityandavailabilityofhealthcarefacilities.

Inaddition,somepreviousstudiesinvestigatedtheimpactofcommunitylevelcharacteri sticsinthedecisiontodeliveryathealthfacilities.Thereareseveralwaysthoughwhichthecomm unityaffectthedeliveryfacilitychoiceofwomen.Stephensonetal.

(2006)reportedthatthe community ofhigher concentrationratioofwomenw i t h highereduca tioni m p r o v e thegreaterautonomylevelandprovidemoreopportunityforthewomentoseekhealth

23 careduringpregnancyanddelivery.Furthermore,healsofoundthathigherconcentrationratioo f wom eninthecommunitydeliveringtheirbabyathealthfacilitieshadasignificantlypositiveimpactonthedecisi oninthedeliverylocation.Itsuggeststhatthepracticesofothersinthec o m m u n i t y couldaff ectthehealthseekingbehavioroftheindividualinthesamecommunity.O n e o f theimportantcommunitylevelfactorsasreportedinthestudyofSepehrietal.

(2008)i n Vietnami s communitylevelp o v e r t y r a t e Hepointedt h a t t h e c o m m u n i t y pover ty rateaffectednegativelytheprobabilityofgivingbirthathealthfacilities.Womeninthecommunityw i t h higherpovertyrateabove50%lesslikelytodelivertheirbabyathealthinstitutionsincomp arisonto thosein thecommunitywith povertyrate ofequalandlessthan10%.

Thischapterpresentst h e m e t h o d o l o g y anddatadescriptionusedt o investigatet h e deter minantsoftheutilizationofmaternalhealthcare.Thefirstpartistointroduce conceptual frameworkdescribingtheassociationsbetweenthedeterminantsandthedemandofprenatalcarevi sitsandthechoiceofdeliverycareprovidersrespectively.Next,twomethodsrelatedtot h e objectiveoft hestudy,namelytheNegativeBinomialmodelandtheMultinomialLogisticsm o d e l willbediscuss edtoanalyzetheassociationsasmentionedinthepreviouspart.Thefinalsectionpresentsthe datausedin thestudy.

Conceptualframework

Prenatal care visits and the selection of delivery care providers are influenced by various factors at the individual, household, and community levels Individual characteristics such as education, access to mass media, employment status, marital status, pregnancy intentions, and birth order play a significant role Household factors, including wealth index, household size, ethnicity, and the religion of the household head, also contribute to these decisions Additionally, community-level characteristics, such as the place of residence, poverty rate, illiteracy rate, and the proportion of women giving birth in health facilities, impact maternal health-seeking behaviors This multifaceted framework highlights the interconnectedness of these influences on prenatal care utilization.

EducationExp ose tomassWo rking status

Figure6Theassociationbetweenindividuallevel,householdlevelandcommunitylevelcharact eristicswith theutilizationofmaternalhealthcare services

Empiricalframework

DemandforPrenatalcare

PoissonModelisappliedfordependentvariableswhicharenon- negativeinteger.Int h e model,Poissondistributionisusedtodescribethe likelihood ofeventsoccurringk timesi n a givenperiod oftime.Thedistributionfunctionis

Withtheconditionthat�i snon-negativeandmeanequaltovariance𝐸(�)=���(�)�.ThemeaningisthatwhenXchanges,howtheexpectedvalueofychanges

However,thelimitationofPoissonModelisthatthevarianceequaltothemean.The countdataisobservedthattheequalityisunreasonable,whichmeansthatthevariancemaybedifferentfr omt he mean Inorder tosolve this matter, thealternativeis NegativeBinomialregression,w hichallowsthevariancetodifferfromthemean.Ingeneral,NegativeBinomialregressionhasthesa memeanstructureasPoissonregressionandithasanextraparametertom o d e l t h e over- dispersion.T h e commandnbregi nStatai s implementedt o estimatet h e NegativeBinomialregr essionandprovidethetestofover-dispersion.

Inthestudy,thedependentvariableisnumberofmaternalcheck- upsthatapregnantwomantakes.T h e independentvariablesa r e d i v i d e d i n t o t h r e e group s:i n d i v i d u a l level,householdlevelandcommunitylevelcharacteristics.Differentfromthet heorybackground,t h e pricesofhealthcareandtheincomearenotincludedduetolimitationofthedata.T hedetailwill beintroducedin nextsection.

Choiceofbirthdeliveryfacility

MultinomialLogitModelisappliedtomeasuretherelationshipbetweencategoricalvariable sando t h e r e x p l a n a t o r y variables.M o t h e r s a r e assumedt o c h o o s e t h e f a c i l i t y t h a t maximeutility.Andthechoiceoffacilityjofindividualicodedas� �is 1,2,3…

Inthestudy,thedependentvariableisthelogoddsthatindividualiwillchosedelivery alternativej(j=2,3)relativetothereferencealternative.T h e alternative1isthedeliveryath o m e whilethealternative2and3 are respectivelydeliveryatpublichospitalsand deliveryatprivatehospitalsorclinics.Thealternative2issetupasthebaseoutcome.Inthestudy,onl yt h e characteristicsrelatedtothechooserwillbeinvestigatedbecausethedataMICS4doesnotcoverth echaracteristicsofthehealthcareproviders.Similartotheanalysisofthedemandofprenatalcarevisi ts,theindependentvariablesaredividedbythreegroups:individual,householdandcommunity.

ThecommandmlogitinStataisemployedtoestimatethemultinomiallogisticmodel.T h e i ndependentvariablesaredividedinto threegroups:individuallevel,householdlevelandc o m m u n i t y levelcharacteristics.Thedeta ils will beintroducedin next section.

Data

MICSwascarriedoutinVietNambyVietNamGeneralStatisticsOfficeincollaborationwi thUNICEF.ThesampleforVietNamMICSwasdesignedtoprovideestimatesforalargenumberofindi catorsonthenationallevelsituationofchildrenandwomeni n urbanandruralareasaswellassixgeogra phicregionsRedRiverDelta,NorthernMidlandsandMountainareas,NorthCentralareaandCentral Coastalarea,CentralHighlands,SouthEastandM e k o n g R i v e r Delta.T h e s u r v e y c o n d u c t s threes e t o f questionnaires:first,t h e questionnaireonhouseholdsto collectinformation r egardinghousehold membersandeconomicstatus;second,thequestionnairesonfemalehouseho ldmembersintheproductiveage(15-

49ages)andthelastoneisadministeredtochildrenunder5yearsandtheircaretakers.MICS5isbasedon asampleof10,018interviewedhouseholds,with9,827womenand3,316children.Forthepurposeoft hestudy,thesampleisnarroweddownintothewomenwhogaveb i r t h t o a l i v e childw i t h i n t w o yearsbeforet h e s u r v e y i n o r d e r t o reducer e c a l l errors.Therefore,thereare1,479women in theproductiveageappliedin thestudy.

Variablesdefinition

Dependentvariables

The study evaluates two key indicators of maternal healthcare service utilization: the amount of antenatal care coverage and the place of delivery Antenatal care coverage, as defined by MICS, is the percentage of women aged 15-49 who had an alive birth within the two years prior to the survey and received care from skilled health personnel at least once The World Health Organization recommends that pregnant women have a minimum of four antenatal care visits during their pregnancy to prevent and identify potential health issues for both the mother and the baby It is essential for pregnant women to attend these visits early and consistently throughout their pregnancy "Skilled personnel" refers to accredited health professionals such as midwives, physicians, and nurses, excluding traditional birth attendants, making antenatal care visits critical variables in assessing maternal health.

Placeofdeliveryisdefinedasdeliveryinahealthfacility(privateorpublic)oroutsidet h e healt hsystem( h o m e delivery).T h e increasedproportionofb i r t h s deliveredath e a l t h facilitiesi s a nimportantfactort o reduceh e a l t h r i s k s t o motherandbaby.P r o p e r m e d i c a l attentionandhy gienicconditionsd u r i n g deliveryc a n reducer i s k s o f complicationsand

Independentvariables

Theindividual-householdlevel,c o m m u n i t y - l e v e l e x p l a n a t o r y variableswereu s e d basedonthetheoreticalandempiricalliterature,c onsideringtheuseofmaternalhealthcareservicesandt h e i r a v a i l a b i l i t y int h e dataset.Becau set w o regressionsf o r t h e tworesearchobjectiveswillbeanalyzed,theywillsharethesamesetofinde pendentvariables.Thedetaileddescriptionofselectedvariablesisgivenbelow.

The study examines various individual-level factors influencing maternal health, including the mother's age at childbirth, birth order, education, material status, exposure to mass media, and pregnancy intention Maternal age is treated as a continuous variable ranging from 15 to 49 years Education is categorized into five dummy variables: no education, primary, lower secondary, upper secondary, and tertiary Birth order is also a continuous variable reflecting the total number of children a woman has given birth to Material status is represented by dummy variables indicating whether a woman was previously married but is no longer in a union, never married, or otherwise Exposure to mass media is assessed through four dummy variables based on access frequency to mobile phones, newspapers, radio, and television Lastly, pregnancy intention is measured as a binary variable, indicating whether the woman did not wish to become pregnant.

Householdlevelfactorsinclude ethnicityandreligionofhouseholdhead,householdw ealths t a t u s E t h n i c i t y wasdefinedo n eitherK i n h o r H o a ( = 1 ) o r non-Kinh/

Hoa( = 0).Althoughtheyarethe sixthlargestminoritygroupin VietNam,theirlivingstandardsareon aparwiththoseoftheKinhmajority.Thereligionvariableisusedasadummyvariable:1ifthehousehold hasn o r e l i g i o n and0 otherwise.T h e h o u s e h o l d wealths t a t u s alsoi s dummyvariable:

1 if thehousehold is in thepoorestor poorquintilesand0otherwise.

Inadditiontoindividualandhouseholdlevelfactors,communitylevelfactorsareaddedt o meas ureanyeffectinc o m m u n i t y level.First,p l a c e o f residencei s dividedi n t o t w o categories:urbancodedas1 andruralas0 Moreover,si xregionald u m m i e s represents i x socioeconomicregi onssuchasRedRiverDelta,NorthernMidlandsandMountainareas,North

CentralareaandCentralCoastalarea,CentralHighlands,SouthEastandMekongRiverDelta.Second,i lliter acy rateis basedonpercentageof illiteratewomeninthe community.Third, p o v e r t y r ateiscalculatedaspercentageofwomeninthecommunenthelowestwealthquintile.Finally,theperce ntageofthewomeninthecommunitydeliveringtheirbabyathospitalsisapplied.

ANC Number of antenatal carevisits visits

NOEDU Dummyvariableindicatingtheindividual hasnotfinished primaryschool Yes=1,No=0

PRIMARY Dummyvariableindicatingtheindividual finished primaryschool Yes=1,No=0

LOWSECOND Dummyvariableindicatingtheindividual finishedlowersecondaryschool Yes=1,No=0 UPSECOND

Dummyvariableindicatingtheindividual finished upper secondaryschoolDummyvariableindicatingtheindividual finishedthecollegeabove

Yes=1,No=0 Yes=1,No=0 Yes=1,No=0

CEB Number of childreneverborn children

NEWSPAPER WhetherthewomanreadsNewspaperor Magazineeveryday Yes=1,No=0

RADIO Whetherthewomanlistenstoradio everyday Yes=1,No=0

TV WhetherthewomanwatchesTV everyday Yes=1,No=0

UNWANTED Whetherthewomanwanted thelast child No=1,Yes=0

POOR Whetherwomenbelongtothepoorestand poorquintilesgroup Yes=1,No=0

ETHNIC Whetherthehouseholdheadbelong totheethnicminoritygroup Yes=1,No=0

RURAL Whetherthewomenliveinrural area Yes=1,No=0

MN NorthernMidlandsand MountainousArea Yes=1,No=0

CH Central Highlands Yes=1,No=0

SE South East Yes=1,No=0

MD MekongRiver Delta Yes=1,No=0

POVERTY Percentageofwomenwithpoorestandthe2nd quintileinthecommune percentage(%)

HOSPDELIRATIO Percentageofwomeninthecommunegiving birththelast child at hospitals percentage(%)

Thec h a p t e r presentst h e resultso f s t u d y t o reportt h e relationshipsbetweendetermina ntsandthedemandofprenatalcarevisitsandthechoiceofdeliverycareprovidersrespectively.T h e firstp a r t i s t o presentd e s c r i p t i v e statisticso f dependentvariablesandindependentvaria bles.Thefollowingpartistoanalyzethebivariateassociationsbetweeneachdependentvariableandinde pendentv a r i a b l e s T h e finali s t o demonstratet h e regression analysisregardingthedemandofprenatalcarevisitsandthechoiceofdeliverycareaswellast h e compa risonbetweentheresultwith theonesofpreviousstudies.

DescriptiveResults

ThesamplesummarystatisticsareshownintheTable2andTable3.Inthefirsttable,i t canbese enthatthereare1479womenreportedthatundergochildbirthinthelast2years.T h e averageof prenatal carevisits is

The World Health Organization (WHO) recommends that pregnant women have at least four antenatal care visits to ensure a safe pregnancy and fetal development The average age of women in the study is 28, with ages ranging from 15 to 47 On average, women have two children, aligning with Vietnam's two-child policy, although some women in rural areas have as many as 11 children Alarmingly, 100% of women from the poorest households live in communities that lack basic necessities Additionally, many women in these areas remain illiterate, with a significant portion not having completed even primary education.

Table 3 presents descriptive statistics for dummy variables related to childbirth Most women prefer to deliver in government hospitals or community health centers to ensure safe childbirth, although 136 cases still report home deliveries Among the interviewed women, 35% completed lower secondary education, followed by upper secondary, tertiary, primary, and 6% with no education Additionally, 3% of the women reported being previously married or never married Nearly 20% experienced unintended pregnancies, and many were not employed In terms of media exposure, women predominantly watch television and read SMS messages more frequently than they read newspapers or listen to the radio Television and mobile phones are widely used in Vietnam, as they provide timely updates on both international and domestic affairs and facilitate quick communication.

Therearemorethan40%ofwomenlivinginthepoorestandpoorhouseholdwealthquinti lesandapproximately24%ofthosebelongingtotheethnicgroup.Mostofhouseholds inwhichthewomenlivedonotfollowanyreligion.ItisobviouslycommonthatVietnamesefamilie stend to worship theirancestors.However,consideringthereligion,Buddhismmakesu p thehighestratio,followed byChristianCatholic.Residentiallocationisoneofthemostimportantfactorsintheutilizationo fmaternalhealthcare.Mostofinterviewedwomenarel i v i n g intheruralareas,whichdonotha veheathcarecenteradequatelyandlessdevelopedeconomicconditionthanurbanareas.Howeve r,thereisnoobviousdifferenceinthenumbero f womenlivingin sixsocio-economicregions.

Variable Obs Mean Std.Dev Min Max

Analysis ofDemandforprenatalcare

Bivariateanalysis

TheTable4providesthebivariateresultsregardingtheassociationbetweenthenumbero f prenat alcarevisitswitheachsocialdeterminant.Concerningthehighestlevelofeducationwhatthewomenc ompleted,itcanbeseenthatthereislargedifferenceamongthosewhohasn o educationalcertificatea ndlittlevariationamongthehighereducatedwomen.Thosewithouteducationalcertificatehavethenumber ofvisitslowerby3prenatalcarevisitsthanthosehavehighereducationcertificates.Inaddition,thediffer encesamongthoseinrural,poorhouseholdwealthquintileandethnicgroupappeartoberemarkable.The disadvantagedwomenlesslikelygetaccesst o t h e p r e n a t a l care.Further,in t h e d e v e l o p e d regio nssuchasRedR i v e r Delta,

MekongRiverDeltaandSouthEast,theprobabilityofusingtheprenatalcareisslightlyhighert h a n theles sdevelopedregionssuchasCentralHighlandaswellasNorthernMidlandsandMountainousArea.Inc ontrast,thereislittlevarianceinthedemandofmaternalhealthcaref o r w o r k i n g womenand religiouswomen.Lookingatt h e Figure9 i t i s illustratedt h a t t h e likelihoodofutilizingmaternalhe althcaredecreaseswiththeageofwomen,thenumberofchildreneverbornandtheilliteracyrateo f the communityin which thewomen live.

ANC Observation Mean Std.Dev Min Max

Mobilephone re ca l at an et na d ve i c e re s me i T

Figure7:T h e associationbetweent h e demando f maternalc a r e v i s i t s a n d numericalinde pendentvariables

Analysis ofNegative BinomialModel

Firsto f all,t h e r e g r e s s i o n t e s t s t h e likelihoodratio,estimatingwhetherdispersionp arameteralphai s e q u a l t o z e r o T he te st statisticw it h p - v a l u e n e a r z er o suggestst h a t t h e responsevariableisover- dispersedsotheNegativeBinomialmodelispreferredtoPoisson m o d e l asshowninAppen dix6.Next,theregressionisrunwithrobustinordertosolvewithth e heteroscedasticity.Theresults ofNegative Binomialmodelaredescribed inTable5.

ThestudyfindsthatthevariableAGEhasacoefficientof0.016,whichisstatisticallysignific antwithp-valuenearzero.Concerning marginaleffect,itcouldbeinterpretedthattheincreaseoftheageby1yearwillleadtotheincreaseofpre natalcarevisitsby0.08.Itmeanst h a t thehighertheageofthewomanis, themore frequentlytheytakematernalcarevisit.TheresultisinconsistentwiththestudyofArthur(2012)and

Tsawe&Susuman(2014).Itcouldb e e x p l a i n e d t h a t t h e p r e g n a n c y women arewell- informedaboutt h e p o t e n t i a l r i s k s d u r i n g pregnancyatthehigherage.

Regardingtheeducationattainment,similartothestudies(Arthur2012,Bbaale2011,Nava neetham&Dharmalingam2002),thesignificantcoefficientsofnoeducation,primaryandlowers e c o n d certificatesuggestst h a t t h e r e areo b v i o u s differencesbetweent h e lower educationholdersandhighereducationholdersi n t h e demando f prenatalcarev i s i t s T h e margin aleffectc o e f f i c i e n t s h o w s t h a t t h o s e w i t h o u t n o certificatecomparedtot h o s e w i t h t e r t i a r y certificate,whileholdingtheothervariableconstantinthemodel,areexpectedtohavea rateof1. 67timeslowerinnumberofprenatalcarevisits.Comparedtothetertiarycertificateholder,thosewithpri maryandthosewithlowersecondareexpectedtohavearateof0.76and

4.98timesrespectivelylowerinthenumberofprenatalcarevisits.Incontrast,thereisnogreatdifference betweentheuppersecondcertificateholderandthe tertiaryholder.

Concerningtheexposuretomassmedia,theresultshowsthatonlythecoefficientoff r e q u e n c y ofreadingnewspaperandmagazinesisstatisticallysignificant.Withmarginaleffecto f 0.39 ,thedifferenceinnumber ofprenatalcare visits is higher0.39 times forthosewhoreadingn ewspaperandmagazineseverydaycomparedtothosewholessreadnewspaperornotreadatall.T h e coe fficientso f f r e q u e n c y ofr e a d i n g S M S messages,l i s t e n i n g t o radioandw a t c h i n g TVar enot statisticallysignificant.

(2013),thecoefficientofunwantedpregnancyisnots t a t i s t i c a l l y significant.Similarly,t h e coefficiento f w o r k i n g s t a t u s alsoi s n o t s t a t i s t i c a l l y significant.

Thecoefficientofnon- unioninmaritalstatusishighlysignificantwiththenegativeparameter,whichconfirms theresultsfromthestudiesofSepehrietal.(2008)andChenetal.

(2003).Thenegativesignmeansthatthewomenwholivewiththeirhusbandnolongerorbeneverma rriedlesslikely1.16timestakeprenatalcarevisitcomparedtothosewhoarelivingw i t h theirhusb and.Itcouldbeindicatedthatsingleexpectantmotherfacesthedisgracewhent a k i n g maternalch eckupatpublicbecausethechildbirthisconsideredastheresultofmaritalrelationship.

Inaddition,thecoefficientofbirthorderishighlystatisticallysignificantwithp- valuealmostzero.Thenegativecoefficientshowsthatwomenhavingonemorechildtendtohavet h e numberofvisitslowerby0.52.Thisisattributedtoburdenofchildrenthatthewoment a k i n g careofsoshedoestakeprenatalcheck- upfrequently.Theresultissimilartothestudieso f Navaneetham&Dharmalingam(2002),Sepehrie tal.(2008),Arthur(2012)andTsawe&Susuman(2014).

Theresultindicatesthatthecoefficientsoflivinginpoorhousehold,belongingtoethnicm i n o r i t y groupandnotfollowinganyreligionarestatisticallysignificantwhereastheonesofhousehol dsizeisnotsignificant.Itisremarkablethatthewomenfromthepoorandpoorestquintilesappear tohavethenumberofvisitslowerby1.1comparedtothoseinotherquintiles.D u e tofinancialproblems ,thewomenfinddifficulttoaffordthecostofprenatalcareaswellastransportationcosts.Thestudysu pportsthehypothesisofthestudies(Bbaale2011,Tsawe

SimilartothesestudiesofCelik&Hotchkiss(2000),Navaneetham&Dharmalingam(2002) andSepehrieta l

(2008),t h e womanw h o s e h o u s e h o l d h e a d belongst o t h e ethnicm i n o r i t y grouphas0.6 visitslessthanthoseinKinhorHoagroup.Languagebarrierandculturedifferencesprobablypreventtheet hnicgrouptomakeuseofprenatalcarevisits.Similarly,thewomanw h o s e householdheaddoesn o t f o l l o w a n y religionhasp o s i t i v e influenceo n t h e demandofprenatal carevisits with themarginaleffectof0.3.Itindicatesthatreligiousnormsandpracticesofthehouseholdheadpreven tthewomeninthereligiousgroupfromutilizingt h e prenatalc a r e v i s i t s T h e s t u d y reconfirm st h e resultfromBbaale( 2 0 1 1 ) andTsawe& Susuman(2014).

In 2008, significant community-level variables were identified, particularly the proportion of women with no education and the proportion of women delivering in hospitals Consistent with Gage's (2007) findings, expectant mothers in communities with higher literacy rates had 0.05 more prenatal care visits compared to those in lower literacy areas Conversely, a higher hospital delivery ratio positively influenced the demand for antenatal care, with women in communities with a greater ratio of hospital deliveries having 0.03 more visits than those in areas with lower ratios This indicates that community practices significantly affect women's attitudes and healthcare-seeking behavior, supporting the conclusions of Ononokpono et al (2013).

Furthermore,thereisgreatregionaldisparityinutilizationofprenatalcarevisits,whichi s thesa meresultasCelik& Hotchkiss(2000)andSepherietal.

(2008).ResidenceintheunderdevelopedareassuchasCentralCoast,CentralHighlandaswellasNo rthernMidlandsandMountainousAreafarlesslikely1.05,0.76and0.71timesrespectivelytota keprenatalcarev i s i t s f r e q u e n t l y comparedt o residencei n MekongDeltaareas.However,t h e r e i s n o s t a t i s t i c a l l y significantdifferencebetweenlivinginSouthEastareasandMekong RiverDelta.Itimpliesthatthereareremarkeddifferencesintheimplementationofhealthcareprogram,t hea v a i l a b i l i t y andaccessibilityofthehealthcare servicesamongtheregions.

MOBIPHONE(using mobile phone every day=1) 0.047 (0.030) 0.239 (0.157)

Analysis ofChoiceinthedeliverycareproviders

Bivariateanalysis

Table6presentstheassociationbetweenthechoiceof deliveryfacilityandcontinuousindependentvariables.Itcanbeseenthatthewomenwithhigherbir thordermorelikelytochoose childbirthathomethanthosewithlowerbirthorder.Inaddition,thehigheraverageofp o v e r t y ra teandilliteracyratefocusonthewomengivingbirthathomewhereasthewomendelivering childathomelivesin thecommunitywithlowerratioof facilitydelivery.

Obs Mean Std.Dev Min

Table 7 illustrates the relationship between the choice of delivery care providers and various independent variables Delivery locations are categorized into three types: home delivery, which is the most affordable but often lacks hygiene and medical equipment; public hospitals, which are government-supported and low-cost but frequently overcrowded; and private hospitals or clinics, which are more expensive but offer superior services and advanced technology Notably, women opting for home births have less exposure to mass media, such as mobile phones, television, and newspapers, compared to those delivering at health facilities Additionally, the likelihood of home childbirth increases among women residing in disadvantaged areas with higher poverty and illiteracy rates, like the Central Highlands and North Mountainous regions, where access to healthcare is hindered by poor infrastructure Interestingly, many women delivering at home had been employed in low-paying farm work for two years prior to the interview, which may restrict their ability to afford healthcare costs Furthermore, women with higher living standards and those from ethnic majority groups are more likely to give birth in private hospitals, whereas rural women are less likely to access these facilities due to their concentration in urban areas and the high cost of services.

Analysis ofMultinomialLogisticModel

Theresult of estimatedmultinomiallogisticregressioncoefficientsf o r themodelis pres entedinTable8andmarginaleffectshowninTable9.Thereferencegroupisthegroupw h o choset odeliverinpublichospital.Thesignofcoefficientsshowsthepositiveornegativerelationshipbetweeni ndependentvariablesandthelikelihoodofchoosingchildbirthathomeo r atp r i v a t e h o s p i t a l s r e l a t i v e t o d e l i v e r y atp u b l i c h o s p i t a l w h i l e t h e marginaleffecti s interpretedas howanincreaseintheindependentvariableaffectstheprobabilityofchoosingd e l i v e r y ath o m e o r d e l i v e r y atprivateclinicso r h o s p i t a l s r e l a t i v e l y t o d e l i v e r y atp u b l i c h o s p i t a l , givent h e o t h e r variableconstant.T h e variablehospdeliratioi s omittedbecausecollinearit y

Itcanbeseenthathousehold- levelcharacteristics,namelylivinginpoorhouseholdsandinethnicminoritygrouprarehighlystatisti callysignificantwithp- valuenearlyzero,livingi n thepoorandinethnicgroupisfoundtoincreasethelikelihoodofchildbirthat homerathert h a n atpublichospitals.Itindicatesthatfinancialdifficultiesaswellaslanguageandc ulturebarrierconstraintsthepregnantwomentowillinglygivebirthathealthfacilities.Theresult saffirmthepreviousstudiesofCelik&Hotchkiss(2000),Stephensonetal.(2006)andWadoeta l

(2013).Unexpectedly,t h e householdheadf o l l o w i n g n o religioni s n o t significantlyassociate dwiththechoiceofdeliveryathome,whichisdifferentfromStephensonetal.

(2006).Inrespecttothemarginaleffect,livinginthepoorandinethnicminoritygroupincreasethe p r o b a b i l i t y ofgivingb i r t h ath o m e by0 0 8 % a n d 0 0 7 % r e s p e c t i v e l y ; however,th eya r e s t a t i s t i c a l l y insignificant.

Inaddition,someindividuallevelcharacteristicsalsoaffectconsiderablythechoiceofd e l i v e r y l o c a t i o n Similart o Navaneetham& Dharmalingam(2002),t h e negativesignso f l i s t e n i n g toradioorreadingnewspaperregularlycould bereportedtobenegativelyassociatedw i t h theprobabilityforchildbirthathome.Itcouldbeexplainedthat frequentexposuretomass

45 mediaincreasetheknowledgeregardingthematernalhealthservicesandenhancetheawarenesso fneedforcareofdelivery.Inrespecttomarginaleffect,listeningtoradioandreadingnewspaper decreasethe probabilityofgivingbirthathomebyaround0.03percentagep o i n t s Inaddition,themarginaleffe cto fnumber ofprenatalcarevisits is -

0.0002,whichmeansthatthenumberofprenatalcaredecreasetheprobabilityfordeliveryathom eby0.02percentagep o i n t s , h o l d i n g allo t h e r variablesc o n s t a n t andt h e effecti s h i g h l y s t a t i s t i c a l l y significant.ManypreviousstudiessuchasStephensonetal.(2006),Sepehrietal.

(2013)alsoshowedthesameresult.I n contract,themarginaleffectofthenumbero f childreneverborn is0.0001couldbeinterpretedthatforthewomenhas onemorechildren,t h e probabilityofdeliveryathomeincreaseby0.01percentagepoints.Thear gumentisthatfrompreviousexperience,womenwithhighbirthorderunderestimatetheneedofdeli veryathealthfacilitiesortheyhavegreaterconstraintstousetheservices(Navaneetham&Dharm alingam2002,Stephensonetal.2006andSepehrietal.2008).Itissurprisingthatageandeducationatt ainmentaren o t significantlyassociatedw i t h t h e choiceo f deliveryp l a c e alternatives.

Concerningthecommunitylevelcharacteristics,livinginruralareashassignificantlyp o s i t i v e influenceinthedecisiontotheplaceofdeliveryathome,whichissimilartothestudieso f Celi k&Hotchkiss(2000)andGage(2007).Themarginaleffectof0.0003indicatesthatl i v i n g i nruralareasincreasetheprobabilityofgivebirthathomeby0.03percentagepoints.Moreover,the concentrationratioofthewomenwithnoeducationcertificatearefoundtobes i g n i f i c a n t l y positivelyassociatedwiththechoiceofchildbirthathometochildbirthatpublich o s p i t a l

Theresultshowsthatonlytheestimatedcoefficientsofworkingstatus,residenceinrura landpoorhouseholdarestatisticallysignificantwithp- valuenearzero.Itcouldimplythatthereisnosomuchgreatdifferencedeterminedbydemographic factorsin thechoicebetweenp u b l i c hospitalandprivatehospital.Indetail,themarginaleffectoflivin ginpoorhouseholdi s -

0.0012,livinginruralareadecreasetheprobabilityofdeliveryatprivateh o s p i t a l s by0.12pe rcentagepoints.Inaddition,thereareregionaldifferences in thechoiceofd e l i v e r y atprivatehospital.ComparedtoMekongRiverDelta,livinginthetwore gionssuch asNorthCentralandSouthEastdecreasestheprobabilityforgivingbirthatprivatehospitalsby0.12 and0.1percentagepointsrespectively.Itisobviouslyseenthattheprivatehospitalsareonlyavail ableintheurbanareasandthedevelopedregionsandthecostofdeliverycareinprivatehospitalsishigher thanoneof thepublichospitals.

CEB(numberofchildren) Household- levelCharacteristics HHSIZE(numbero fmembers)

NORELI(followingnoreligion=1) Com munity- levelCharacteristics RURAL(livinginru ralarea=1)

Marginal effect Marginal effect Marginal effect

Thislastchaptersummarizesallthemainfindingsintheresearch.Basedontheseconclusions,i t w i l l d i s c u s s andrecommendrelevantpolicies.Finally,limitationsof the studyanddirectionf o r furtherstudiesarediscussed.

Mainfindings

2014andtwomethods,i n c l u d i n g t h e NegativeBinomialModelandM u l t i n o m i a l Logis ticModel,t h e s t u d y h a s investigatedthedeterminantsofthedemandofprenatalcareandthec hoiceofbirthdeliveryfacility,namelyindividuallevelcharacteristics,householdlevelcharacteristic sandcommunitylevelcharacteristics.

Prenatal care demand is influenced by various individual, household, and community factors Key individual determinants include women's age, education level, marital status, and the number of children they have Studies have shown that higher education significantly increases the likelihood of utilizing prenatal care, as educated women tend to have greater autonomy and awareness of healthcare benefits Additionally, community illiteracy rates correlate with lower maternal healthcare utilization Women from lower household wealth and ethnic minority backgrounds are also less likely to seek adequate health check-ups compared to those from more affluent and majority groups Family support plays a crucial role, as unmarried or separated women, as well as those with more children, often face challenges in accessing prenatal care due to increased responsibilities and lack of support Furthermore, there are notable regional disparities in prenatal care utilization, with residents in disadvantaged areas like the Central Highlands and North Mountainous regions significantly less likely to access maternal health check-ups compared to those in more prosperous areas.

The choice of birth delivery place is significantly influenced by individual, household, and community characteristics Individual factors such as the number of prenatal care visits and exposure to mass media negatively affect the preference for home delivery compared to public facilities, while having more children increases the likelihood of choosing home delivery However, individual characteristics do not significantly differentiate between public and private hospital deliveries At the household level, women from poorer households and ethnic minority groups tend to prefer home delivery over public hospitals, whereas poorer women are more likely to give birth in public hospitals than in private ones These findings align with previous studies by Navaneetham & Dharmalingam (2002), Stephenson et al (2006), and Sepehri et al (2008).

(2013).Finally,thereisconsiderabledifferenceint h e availabilityandaccessibilityofthehea lthfacilitiesbetweenruralandurbanareas.Ruralwomenmorelikelytogivebirthathomecomparedto publichospitalsandatpublichospitalscomparedtoprivatehospitals.Thematterhadbeenalsore portedinVietnam(Sepehrietal.2 0 0 8 )andotherdevelopingcountries (Celik & Hotchkiss2000andGage2007).

PolicyRecommendation

Theresultsp r o v i d e s o m e p o l i c y recommendation.Firstly,t h e importanceo f educations h o u l d behighlightedinindividualcharacteristics,especiallytheilliteracyrateinthetwotypeso f mat ernalhealthcare.Itmeansthatthepracticesofthewomenwithhigheducationalsoaffectt h e behavioro fotherwomeninthesamecommunity.Therefore,educationreformsshouldbeconductedextensivel ytoreducepotentialbarrierandensurethefulluniversalofprimaryandsecondeducationforallpeop le,especiallywomen.Thehigheducationattainmentenhancest h e women’sautonomy andt heirrolesinthe family inorder toseekthebestcarefortheir health.

Secondly,t h e maternalhealthcareinterventionss h o u l d reduced i s p a r i t i e s i n healthcareuti lizationinethnicsgroup.TheethnicminoritygroupsliveinremoteandmountainousareassuchasCen tralHighland andNorthernMidlandsandMountainousArea.Therefore,livingfara w a y fromt h e h o s p i t a l al ongw i t h underdevelopedinfrastructure,theyfeeldifficultt o getaccesstothehealthfacilitiesforpre natalcareandchildbirth.Inaddition,duetolanguageand

50 culturebarrier,theethnicsgroupcannotcommunicatewellwithhealthproviderandevenmayb e discri minated(Malqvistetal.2012)

Thirdly,theprenatalcarevisitisfoundtobeoneofthemostimportantdeterminantsinthechoiceofd e l i v e r y l o c a t i o n Therefore,t h e p r e n a t a l c a r e i n t h e h e a l t h facilitiess h o u l d b e e f f e c t i v e l y conductedtoobtainthetrustofthepregnantwomenandtheirfamily.Furthermore,appropri atecounselingshouldbeofferednotonlyforwomenbutalsootherfamilymembers.Becausetheap propriateadviceinprenatalcarescouldraiseawarenessofbenefitsofsubsequentserviceu s e aswel last h e p o t e n t i a l r i s k s d u r i n g p r e g n a n c y andchildbirth,t h e womanandherfamilycoul dbeencouragedtochoosethedeliveryathealthfacilitiesinsteado f d e l i v e r y athome.

Fourthly,theinfrastructuredevelopmentispaidgreaterattentionincommunitylevel.Thegover nmentshouldintensifythematernalhealthprogramandprovidethehealthfacilitiesinremotean dmountainousareas.Byapplyingthevariousmeansofcommunicationsuchasradio,televisionandnews papertheresidentsintheseareascouldeasilygetinformationregardingthebenefitsofprenatalcareandfacilit ydeliveryaswellasgetaccesstohealthfacilities.Moreover,t h e roadandp u b l i c t r a n s p o r t a t i o n me anss h o u l d b e upgradedt o removet h e difficultieso f travelingt o t h e healthfacilitiesandencoura get h e residentst o utilizethematernalhealthservice.

LimitationandFurtherResearch

This research has several limitations Firstly, the data on antenatal care utilization and delivery location may be affected by recall errors, as women might not accurately remember the number of prenatal visits or when they began care To mitigate this, the study focused on women who delivered their babies within two years prior to the interview Secondly, some factors influencing prenatal care visits and delivery location were not observed due to data limitations, such as pregnancy complications and community practices that affect maternal healthcare-seeking behavior Additionally, community-level variables may differ from individual-level data due to aggregation, potentially skewing results Lastly, including prenatal care visits in the facility choice regression may introduce endogeneity issues that have not been addressed Therefore, further research is necessary to address these limitations.

(2000).Factorsinfluencingchoiceofdeliverysitesi n Rakaidistrict ofUganda.SocialScienceand Medicine,50(2),203–213.

BabalolaS.&FatusiA.(2009).DeterminantsofuseofmaternalhealthservicesinNigeria– l o o k i n g beyondindividualandhouseholdfactors.BMCPregnancyChildbirth, 9:43.

ChenC.S.,LiuT.C.&ChenL.M.(2003)Nationalhealthinsuranceand theantenatalcare use:a casein

GageAJ&CalixteMG.(2006).Effectsofthephysicalaccessibilityofmaternalhealthserviceso n t h e i r useinruralHaiti.Population Studies,60,271–88.

(2007)BarrierstotheU t i l i z a t i o n ofMaternalHealthCarein RuralMali.Social Scie nce&Medicine, 65,1666–1682

GeneralS t a t i s t i c s Office(2011).VietnamMultipleIndicatorClusterS u r v e y 2 0 1 1 , F i n a l Report,2011 HaNoi, Vietnam.

GeneralS t a t i s t i c s Office(2013).T h e 1 / 4 / 2 0 1 3 Time- pointP o p u l a t i o n changea n d Familyp l a n n i n g Survey:MajorFindings HaNoi,

(2006).Effectofmothers’exposuretoelectronicmassmediaonknowledgeanduseo f prenata lc a r e s e r v i c e s : comparativeanalysiso f Indians t a t e s T h e P r o f e s s i o n a l Geographers,58, 278–93.

(2012).I n e q u i t y i n maternalh e a l t h careu t i l i z a t i o n i n Vietnam.InternationalJo urnalEquityHealth2012, 11: 24.

Liu.T.C.,ChenB.,ChanYS.&ChenCS.(2015).Doesprenatalcarebenefitmaternalhealth?

Malqvist,M.,LincettoO.,NguyenD.,BurgessC.,DinhH.

(2013).Maternalhealthutilizationi n Vietnam:increasingethnicinequality.BullWorldHea lthOrgan, 91,254-261.

MinistryofPlanningandInvestment(2015).Countryreport15yearsachievingtheVietNamMillenn iumDevelopmentGoals.Hanoi

(2005).Urbanp o v e r t y andh e a l t h i n developingc o u n t r i e s : householdandneighborh oodeffects.Demography, 42, 397–425.

(2002)Utilization ofmaternalhealthcareservicesinsouthernIndia.SocialScience&Me dicine, 55(10), 1849–1869

ReichmanN.E.,CormanH.,NoonanK.&Schwartz-SoicherO.

Rosenstock,IM(1974).Historicaloriginso f t h e healthbeliefm o d e l HealthEducation& Behavi or2(4):328–335.

(2008).Howimportantareindividual,householdandcommunecharacteristicsi n explainin gu t i l i z a t i o n o f maternalhealthservicesin Vietnam?SocialScience&Medicine, 67, 1009–1017.

(2008)Factorsaffectingt h e utilizationofantenatalcareindevelopingcountries:systematicre viewoftheliterature.JournalofAdvancedNursing,61(3),244–260.

(2014).Factorsassociatedwithmaternalhealthcareservicesutilization in ninehighfocusstatesinIndia:amultilevelanalysisbasedon143 8 5 communitiesin

StephensonR,BaschieriA,ClementsS,HenninkM&MadiseN.

(1994).Toofarto walk:Maternalmortali ty inc o n t e x t SocialScience&Medicine38

ThindA.,MohaniA.,BanerjeeK.andHagigiF.(2008).Wheretodeliver?Analysisofchoiceo f d e l i v e r y locationfromanationalsurveyinIndia.BMCPublicHealth2008, 8:29

TranT.K.,NguyenC.T.,NguyenH.D.,ErikssonB.,BondjersG.,GottvallK.,HenryA.& MaxP

(2011).U r b a n – ruraldisparitiesi n antenatalc a r e u t i l i z a t i o n : a s t u d y oft w o cohortsofpregnantwome n inVietnam.BMCHealthServicesResearches,11: 120.

(2012).A n t e n a t a l a n d deliverycareu t i l i z a t i o n i n u r b a n a n d r u r a l context si n Vietnam:astudyintwohealthanddemographicsurveillancesites.NordicSchoolofP u b l i c HealthNHV,Sweden.

TsaweM.&SusumanA.S.(2014).Determinantsofaccesstoanduseofmaternalhealthcareservicesin theEasternCape,SouthAfrica:aquantitativeandqualitativeinvestigation.BMCResearchNo tes, 7:723.

(2013).Unintendedpregnanciesandt h e u s e of maternalh e a l t h servicesi n southwester nE t h i o p i a BMCInternationalHealtha n d HumanRights, 13:36.

WehbyGL,MurrayJC,CastillaEE,Lopez-CameloJ.S.&OhsfeldtR.L.

(2009).Prenatalcareeffectivenessand utilization inBrazil.HealthPolicyand Planning,

WorldHealthOrganization(2003).Antenatalcareindevelopingcountries:Promises,achieveme ntsandmissedopportunities:Ananalysisoftrends,levelsanddifferentials,1 9 9 0 - 2001.Geneva:WHO.

2001:Anindeptha n a l y s i s o f causesand determinants.Manila:WHO,WesternRegionalRegion.

WorldHealthOrganization(2014).TrendsinMaternalMortality: 1990 to2013.Zweifel,P.,Breyer,F.,&Kifmann,M.(2009).Healtheconomics.Springer.

Variable Obs Mean Std.Dev Min Max

Variable Obs Mean Std.Dev

The data analysis reveals significant insights across various metrics For age distribution, delivery 1 shows an average age of 26.43, while delivery 2 averages at 27.68, and delivery 3 at 28.08 In terms of the CEB (Children Ever Born), delivery 1 has a mean of 2.78, delivery 2 has 1.75, and delivery 3 has 1.76 Household size (HHSIZE) averages 6.61 for delivery 1, 5.67 for delivery 2, and 5.65 for delivery 3 Poverty levels are notably high in delivery 1 at 87.40, with delivery 2 at 35.44 and delivery 3 at 29.13 Illiteracy rates present a stark contrast, with delivery 1 at 39.73, while both delivery 2 and delivery 3 show significantly lower rates of 4.35 and 4.36, respectively Finally, the hospital delivery ratio (HOSPDELIRATIO) indicates a striking difference, with delivery 1 at 29.81, delivery 2 at 96.93, and delivery 3 at 98.12, highlighting the variability in healthcare access.

Variable Obs Mean Std.Dev Min Max

Variable Obs Mean Std.De v Min Max

Appendix5:T h e associationbetweenthedemandofmaternalcare vi si ts and numericalinde pendentvariables

>E POORETHNICNORELIRURALPOVERTYILLITERACYhospdeliratioRRDNMNCCHSEMD note:TERTIARYomittedbecauseofcollinearitynote:MD omittedbecauseofcollinearity

ANC Coef Std.Err z P>|z| [95%Conf.Interval]

Likelihood-ratiotestofalpha=0:chibar2(01)= 87.67Prob>=chibar2= 0.000

=5.0515799 variable dy/dx Std.Err z P>|z| [ 95%C.I ] X

AGE 0808842 01791 4.52 0.000 045774 115994 27.5842NOEDU* -1.670329 43445 -3.84 0.000 -2.52184 -.818817 060176PRIMARY* -.7639974 24053 -3.18 0.001 -1.23543 -.292565 13455LOWSEC~D* -.490159 20498 -2.39 0.017 -.891903 -.088415 344828UPSECOND* -.2702806 17973 -1.50 0.133 -.622539 081978 233942MOBIPH~E* 2393492 15673 1.53 0.127 -.067832 546531 273158NEWSPA~R* 3853626 18374 2.10 0.036 025243 745482 192698RADIO* -.307939 1898 -1.62 0.105 -.679945 064067 121704TV* -.0706393 23625 -0.30 0.765 -.533684 392405 834348MARITAL* -1.165153 41725 -2.79 0.005 -1.98295 -.347358 027045UNWANTED* -.3054536 18475 -1.65 0.098 -.667562 056655 177823WORKING* -.111988 19738 -0.57 0.570 -.49885 274874 824882CEB -.526979 10892 -4.84 0.000 -.740463 -.313495 1.84314HHSIZE -.0123077 03265 -0.38 0.706 -.076309 051694 5.75321POOR* -1.080706 23102 -4.68 0.000 -1.5335 -.62791 421231ETHNIC* -.5715099 2478 -2.31 0.021 -1.05719 -.085831 236646NORELI* 3050419 16018 1.90 0.057 -.008902 618986 745098RURAL* -.2449753 16931 -1.45 0.148 -.576808 086857 62407POVERTY -.0035125 0041 -0.86 0.392 -.011548 004523 39.9499ILLITE~Y -.0501067 01199 -4.18 0.000 -.073606 -.026608 7.60748 hospde~o 0303695 00729 4.16 0.000 016073 044666 90.8046RRD* -.5180725 25484 -2.03 0.042 -1.01755 -.018598 152806NM* -.7165916 26996 -2.65 0.008 -1.2457 -.187479 189317NC* -1.059297 21455 -4.94 0.000 -1.47981 -.638784 149425CH* -.7608472 24568 -3.10 0.002 -1.24237 -.279329 208249SE* 2804522 26002 1.08 0.281 -.229187 790091 163624(*)dy/dxisfordiscretechangeofdummyvariablefrom0to1

>CEBHHSIZEPOORETHNICNORELIRURALPOVERTYILLITERACYRRDNMNCCHSEMD,robust note:TERTIARYomittedbecauseofcollinearitynote:MDo mittedbecauseofcollinearityIteration0: logpseudolikelihood=-708.35443

ANC 0435657 0329205 1.32 0.186 -.0209573 1080886 AGE -.032046 0336321 -0.95 0.341 -.0979637 0338716 NOEDU 5487294 8578957 0.64 0.522 -1.132715 2.230174 PRIMARY -.5984313 5340849 -1.12 0.263 -1.645218 4483558 LOWSECOND -.7018067 3990452 -1.76 0.079 -1.483921 0803074 UPSECOND -.269463 358309 -0.75 0.452 -.9717358 4328098

TV -.3248382 4126861 -0.79 0.431 -1.133688 4840116 MARITAL -.474224 1.093957 -0.43 0.665 -2.61834 1.669892 UNWANTED -.0265002 3924757 -0.07 0.946 -.7957384 742738 WORKING -.5619982 3171955 -1.77 0.076 -1.18369 0596934 CEB 2349459 2036672 1.15 0.249 -.1642344 6341262 HHSIZE -.0421632 0719171 -0.59 0.558 -.1831181 0987916 POOR -.9170955 4792335 -1.91 0.056 -1.856376 022185 ETHNIC -.1205223 4615013 -0.26 0.794 -1.025048 7840037 NORELI 4736594 3281441 1.44 0.149 -.1694912 1.11681 RURAL -.6398772 2979514 -2.15 0.032 -1.223851 -.0559031 POVERTY 007296 0083815 0.87 0.384 -.0091314 0237235 ILLITERACY 0046362 0187234 0.25 0.804 -.0320609 0413334 RRD -2.097738 5949775 -3.53 0.000 -3.263873 -.9316039

Marginaleffectsaftermlogit y= Pr(DELIVERY==1)(predict,poutcome(1))

=.00031878 variable dy/dx Std.Err z P>|z| [ 95%C.I ] X

ANC -.0001634 00006 -2.61 0.009 -.000286 -.000041 5.71941 AGE -4.25e-06 00001 -0.41 0.684 -.000025 000016 27.5842 NOEDU* 0004249 00062 0.68 0.494 -.000791 001641 060176 PRIMARY* 0001634 00035 0.47 0.637 -.000514 000841 13455 LOWSEC~D* -.0001505 00023 -0.65 0.514 -.000602 000301 344828 UPSECOND* -.0001815 00021 -0.86 0.388 -.000594 000231 233942 MOBIPH~E* -.0003431 00022 -1.54 0.123 -.000779 000093 273158 NEWSPA~R* -.0033406 00174 -1.92 0.054 -.006742 000061 192698 RADIO* -.0002605 00014 -1.81 0.070 -.000542 000021 121704 TV* -.0001644 00015 -1.10 0.271 -.000457 000128 834348 MARITAL* -.0001417 00014 -1.01 0.311 -.000416 000133 027045 UNWANTED* 0000416 00014 0.29 0.769 -.000236 000319 177823 WORKING* 0001389 00011 1.26 0.208 -.000077 000355 824882 CEB 000119 00008 1.57 0.117 -.00003 000268 1.84314 HHSIZE -.0000252 00002 -1.20 0.230 -.000066 000016 5.75321 POOR* 0008473 00058 1.47 0.141 -.000281 001975 421231 ETHNIC* 0006663 00042 1.58 0.114 -.00016 001492 236646 NORELI* 0000906 0001 0.88 0.378 -.000111 000292 745098 RURAL* 0002581 00018 1.45 0.146 -.00009 000606 62407 POVERTY -4.77e-06 00000 -1.03 0.304 -.000014 4.3e-06 39.9499 ILLITE~Y 8.30e-06 00001 1.65 0.098 -1.5e-06 000018 7.60748 RRD* 0009084 00113 0.80 0.423 -.001316 003132 152806 NM* 001187 00124 0.96 0.337 -.001237 003611 189317 NC* 0013201 0015 0.88 0.378 -.001612 004253 149425 CH* 0012689 00129 0.98 0.327 -.001268 003806 208249 SE* 0008003 00096 0.84 0.402 -.001072 002672 163624 (*)dy/dxisfordiscretechangeofdummyvariablefrom0to1

Marginaleffectsaftermlogit y= Pr(DELIVERY==2)(predict,poutcome(2))

=.99800811 variable dy/dx Std.Err z P>|z| [ 95%C.I ] X

ANC 0000904 00008 1.10 0.272 -.000071 000251 5.71941AGE 0000578 00006 1.01 0.313 -.000054 00017 27.5842NOEDU* -.0016039 00244 -0.66 0.510 -.006377 00317 060176PRIMARY* 0006525 00069 0.94 0.347 -.000708 002013 13455LOWSEC~D* 0012234 00063 1.94 0.052 -.000013 00246 344828UPSECOND* 0006015 00057 1.06 0.287 -.000507 00171 233942MOBIPH~E* -2.26e-06 00063 -0.00 0.997 -.001242 001238 273158NEWSPA~R* 0032791 00183 1.79 0.073 -.000306 006864 192698RADIO* 000098 0007 0.14 0.889 -.00128 001476 121704TV* 0007713 00088 0.88 0.381 -.000956 002498 834348MARITAL* 0007809 00118 0.66 0.508 -.001532 003094 027045UNWANTED* 2.32e-06 00067 0.00 0.997 -.001303 001308 177823WORKING* 0010022 00079 1.27 0.202 -.000539 002543 824882CEB -.0005112 00036 -1.42 0.156 -.001217 000194 1.84314HHSIZE 0000956 00012 0.78 0.433 -.000144 000335 5.75321POOR* 0006292 00095 0.66 0.510 -.001242 0025 421231ETHNIC* -.0004701 00085 -0.56 0.578 -.002128 001188 236646NORELI* -.0008017 00047 -1.71 0.087 -.001721 000117 745098RURAL* 0009188 00062 1.47 0.141 -.000304 002142 62407POVERTY -7.42e-06 00001 -0.50 0.614 -.000036 000021 39.9499ILLITE~Y -.000016 00003 -0.50 0.614 -.000078 000046 7.60748RRD* 0011125 0012 0.93 0.355 -.001244 003469 152806NM* 0450928 00761 5.92 0.000 030173 060013 189317NC* -.0001213 00155 -0.08 0.938 -.003159 002916 149425CH* -.0006039 0014 -0.43 0.666 -.003343 002135 208249SE* 000242 00107 0.23 0.821 -.001854 002338 163624(*)dy/dxisfordiscretechangeofdummyvariablefrom0to1