Women's reproductive health decision-making: A multi-country analysis of demographic and health surveys in sub-Saharan Africa

RESEARCH ARTICLE Women’s ep oduc i e heal h decision- making: A mul i-coun y analysis o demog aphic and heal h su eys in sub- Saha an A ica Eugene Ko uo Maa o Da ehID 1 *, Kwamena Sekyi Dickson 1 , Da id Teye Doku 1,2 1Depa men o Popula ion and Heal h, Uni e si y o Cape Coas , Cape Ghos , Ghana, 2Uni e si y o Tampe e, Facul y o Social Sciences, Heal h Sciences and PERLA (Tampe e Cen e o Childhood, You h and Family Resea ch), Uni e si y o Tampe e, Tampe e, Finland *[email p o ec ed] Abs ac In oduc ion Women’s abili y o make decisions ega ding hei ep oduc i e heal h has impo an impli- ca ions o hei heal h and well-being. We s udied he socio-demog aphic ac o s a ec ing ep oduc i e heal h decision-making among women in 27 sub-Saha a A ican coun ies. Ma e ials and me hods The s udy made use o pooled da a om cu en Demog aphic and Heal h Su ey (DHS) conduc ed om Janua y 1, 2010 and Decembe 31, 2016 in 27 coun ies in sub-Saha a A i- can. Bina y and mul i a ia e logis ic eg ession models we e used o in es iga e he associ- a ions o women’s socio-demog aphic ac o s wi h decision-making ega ding sexual ep oduc i e heal h. Resul s The p opo ion o women who can ask hei pa ne s o use a condom du ing sexual in e - cou se anged om lowes in Mali (16.6%) o highes in Namibia (93.4%). Fu he mo e, he p opo ion o women who can e use sex anged om 18.3% in Mali o 92.4% in Namibia. O e all, app oxima ely e e y i e ou o en women can ask hei pa ne s o use a condom, six ou en women could e use hei pa ne s sex and se en ou o en women could make a leas 1 decision. Women om u al a eas (OR = 0.51, CI = 0.48–0.54), hose wi h no edu- ca ion (OR = 0.11, CI = 0.10–0.12), Muslim women (OR = 0.29, CI = 0.27–0.31), women no wo king (OR = 0.53, CI = 0.51–0.56) and women whose pa ne s had no educa ion (OR = 0.17, CI = 0.16–0.19) we e less likely o make a decision on hei ep oduc i e heal h. Conclusion Residence, age, le el o educa ion, eligion, occupa ion and pa ne ’s educa ion we e ound o be associa ed wi h women’s decision-making abou sexual in e cou se, condom use and PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0209985 Janua y 9, 2019 1 / 12 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 OPEN ACCESS Ci a ion: Da eh EKM, Dickson KS, Doku DT (2019) Women’s ep oduc i e heal h decision- making: A mul i-coun y analysis o demog aphic and heal h su eys in sub-Saha an A ica. PLoS ONE 14(1): e0209985. h ps://doi.o g/10.1371/ jou nal.pone.0209985 Edi o : Mellissa H. Wi he s, USC Keck School o Medicine, Ins i u e o Global Heal h, UNITED STATES Recei ed: May 23, 2018 Accep ed: Decembe 14, 2018 Published: Janua y 9, 2019 Copy igh : ©2019 Da eh e al. This is an open access a icle dis ibu ed unde he e ms o he C ea i e Commons A ibu ion License, which pe mi s un es ic ed use, dis ibu ion, and ep oduc ion in any medium, p o ided he o iginal au ho and sou ce a e c edi ed. Da a A ailabili y S a emen : All ele an da a is a ailable o he public a h ps://dhsp og am.com/ da a/a ailable-da ase s.c m. The au ho s con i m hey had no special p i ileges o access o he da a. Funding: The au ho s ecei ed no speci ic unding o his wo k. Compe ing in e es s: The au ho s ha e decla ed ha no compe ing in e es s exis . ep oduc i e heal h decision-making index. This s udy con ibu es o he discou se on ep o- duc i e heal h decision-making in A ica. Policies and in e en ion a ge ed a imp o ing women’s au onomy and empowe ing hem o ake cha ge o hei sexual and ep oduc i e heal h issues should be ocused on hese ac o s. In oduc ion Sexual and ep oduc i e heal h may e e o indi iduals being able o eely make decisions abou hei sexual ac i i y, ha is, choose when, whe e, and wi h whom o ha e sexual in e - cou se [1]. Rep oduc i e heal h and igh s o women became o global in e es a e he In e - na ional Con e ence on Popula ion De elopmen (ICPD) P og am o Ac ion in Cai o in 1994. A o al o 179 coun ies a he ICPD in Cai o ad oca ed ha women should ha e he igh s o eely decide on hei ep oduc i e heal h de oid o any disc imina ion [1]. This was u he a i med by he ou h women con e ence in Beijing in 1995 [2]. Gende inequali ies ha e e ec s on ep oduc i e heal h decision-making and i is a p inci- pal componen o he social con ex o ep oduc i e heal h [3,4]. Women’s abili y o make decision ega ding hei ep oduc i e heal h had been cu bed due o hei lack o au onomy and powe ela ions wi h men being mo e in luen ial in decision-making [4]. Women’s au on- omy is c i ical o he social, economic and sus ainable de elopmen o any coun y. In iew o his, he sus ainable de elopmen goal (SDG) emphasizes gende equali y. I inc eases hei ep oduc i e con ol, a i udes and abili y o nego ia e o sa e sex [5,6]. In he global comba agains HIV and AIDS, s udies had highligh ed he essence o commu- nica ion be ween sexual pa ne s conce ning sa e sex [7,8]. Nego ia ion be ween sexual pa - ne s abou condom use and sa e sex is associa ed wi h inc eased use o condoms [9]. S udies ha e shown ha women’s au onomy is epea edly cons ained by educa ional, economic, cul- u al and social condi ions [10,11,12,13]. In low- and middle-income coun ies, pa icula ly sub-Saha an A ica and Sou he n Asia, he e a e se e al socio-economic and cul u al ac o s which a ec women’s abili y o make decision ega ding hei own heal h and hose o hei dependen s, especially child en [14,15]. Ea lie s udies ha e ocused on some aspec s o ep oduc i e heal h, women au onomy and decision-making, using sub-popula ions wi hin some coun ies in sub-Saha an A ica [5,10,16,17,18,19,20]. As a as we know, no s udy has in es iga ed he subjec in mul iple coun ies in he egion. We aim o p o ide na ional and egional es ima es o women’s au on- omy in ep oduc i e heal h decision-making in 27 sub-Saha an A ican coun ies and we u - he examine he ac o s associa ed wi h ep oduc i e heal h decision-making in hese coun ies. The concep o decision-making can be ca ego ized in h ee main ideas, hus, decision mak- ing as a igh [21], as a choice [22] and as a p ocess [23]. Decision making is a p ocess ha s a s wi h p oblem iden i ica ion, da a collec ion and ga he ing, analysis o da a, indings, selec ing an app op ia e and mos sui able solu ion om he al e na i e solu ions and inally e alua ion o he p ocess [23]. Socio-demog aphic cha ac e is ics such as le el o educa ion [5], pa ne educa ional le el [5,10] and weal h s a us [10,16] ha e he po en ial o in luencing an indi idual in making in o med decision. Exposu e o in o ma ion and knowledge and he abili y o p o ide o one’s needs is a posi i e in luence on decision-making. By ex ension, a woman’s decision making abou he sexual and ep oduc i e heal h will be e ec i ely accom- plished based on he abili y o p o ide o he needs and he exposu e o knowledge. Women’s ep oduc i e heal h decision making in sub-Saha an A ica PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0209985 Janua y 9, 2019 2 / 12 None heless, he e a e possibili ies o o he backg ound cha ac e is ics such as eligious a ilia- ion [10,16], place o esidence [23] and cul u al leanings [18,19,20] ha can ha e impac on he decision-making p ocess [15]. In sub-Saha an A ica in pa icula , eligious a ilia ion, cul- u al and social and place o esidence se ings also a ec women’s ep oduc i e heal h deci- sion-making [10]. This a gumen is consis en wi h he P ed’s Beha iou al Ma ix [23]. Ma e ials and me hods The s udy made use o pooled da a om cu en Demog aphic and Heal h Su ey (DHS) con- duc ed om Janua y 1, 2010 and Decembe 31, 2016 in 27 coun ies in sub- Saha an A ica. DHS is a na ionwide su ey collec ed e e y i e-yea pe iod ac oss low and middle-income coun ies. DHS ocuses on ma e nal and child heal h by in e iewing women o ep oduc i e age (15–49 yea s). DHS su eys ollow he same s anda d p ocedu es—sampling, ques ion- nai es, da a collec ion, cleaning, coding and analysis—which allows o c oss–coun y compa - ison. The su ey employs a s a i ied wo-s age sampling echnique. The i s s age in ol ed he selec ing o poin s o clus e s (enume a ion a eas [EAs]). The second s age is he sys ema ic sampling o households lis ed in each clus e o EA. All women in hei ep oduc i e age (15– 49) who we e usual o selec ed households o isi o s who slep in he households on he nigh be o e he su ey we e in e iewed. The esponse a e a ied om 86.2% o 100.0%. Fo he pu pose o his, only women who had in o ma ion on ep oduc ion heal h decision-making we e used (N = 210,536), hus, women who we e ei he ma ied o li ing wi h a pa ne . Women ga e o al and w i en consen . E hical app o al o DHS is usually ob ained om he e hics egula o y boa ds o he coun ies o which he s udies a e conduc ed and by ICF In e - na ional’s ins i u ional e iew boa d. Pe mission o use he da a se was so om MEASURE DHS. Da a se is a ailable o he public a h ps://dhsp og am.com/da a/a ailable-da ase s.c m (da a was no collec ed o owned by he au ho s; po en ial use s would be gi en access once a concep no e is sen o MEASURE DHS) De ini ion o a iables Ou come a iables. The h ee main ou come a iables used we e: (1) decision-making on sexual in e cou se, (2) decision-making on condom use, and (3) ep oduc i e heal h decision- making index. Fo he i s a iable, women we e asked i hey can e use hei pa ne sex. Fo he second a iable (i.e. decision-making on condom use), women we e asked i he can ask hei pa ne s o use condoms. The esponse ca ego y o hese a iables we e: 1 = “yes”, 2 = “no” and 3 = “don’ know/ no su e”. This esponse was ca ego ized as 0 = “no and don’ know” and 1 = “yes” (see Da eh e al. [10]). The hi d ou come a iable, ep oduc i e heal h decision-making index, is gene a ed om he combina ion o he decision-making on sexual in e cou se and he decision-making on condom use a iables. This was ca ego ized as 0 = no decision and 1 = a leas 1 decision. Explana o y a iables. The explana o y a iables consis o : esidence, age, weal h s a us, educa ion, eligion, occupa ion and pa ne ’s educa ion. Residence was ca ego ized as u ban and u al. Age was g ouped in 5 –yea in e al: 15–19, 20–24, 25–29, 30–34, 35–39, 40–44, 45– 49. Weal h s a us was de i ed om he owne ship o a a ie y o household asse s and ca ego- ized as poo es , poo e , middle, iche and iches . Le el o educa ion and pa ne ’s educa ion was cap u ed as no educa ion, p ima y, seconda y and highe educa ion. Religion was eco ded as Ch is ian, Muslims and O he s. Religion was no a ailable o Nige . Occupa ion was ca ego ized as no wo king, wo king. Women’s ep oduc i e heal h decision making in sub-Saha an A ica PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0209985 Janua y 9, 2019 3 / 12 S a is ical analysis All da a se s om he 29 coun ies downloaded om MEASURE DHS we e me ged and appended as one da a se be o e he analysis was done. Desc ip i e and in e en ial s a is ics we e used. Desc ip i e igu es a e epo ed in pe cen ages by coun ies. Bina y and mul i a i- a e logis ic eg ession models we e used o in es iga e he ela ionship be ween he explana o y a iables and he ou come a iables. Two models we e used o assess he p edic o s o wom- en’s decision-making on sexual in e cou se, decision-making on condom use, and ep oduc- i e heal h decision-making index. Model I looked a a bi a ia e analysis be ween each o he p edic o a iables and he ou come a iable. Model II was i ed o in es iga e he associa ion be ween he independen a iables and he ou come a iables (decision-making on sexual in e cou se, decision making on condom use, and ep oduc i e heal h decision-making index). All equency dis ibu ions we e weigh ed whils he su ey command in S a a was used o adjus o he complex sampling s uc u e o he da a in he eg ession analyses. All esul s o he logis ic analyses we e p esen ed as odds a ios (ORs) wi h 95% con idence in e - als (CIs). Resul s Table 1 shows selec ed in o ma ion om women om 27 coun ies in sub-Saha an A ica. The p opo ion o women who could ask hei pa ne o use a condom du ing sexual in e - cou se anged om 16.6% in Mali o 93.4% in Namibia. P opo ion o women who could e use sex anged om 18.3% in Mali o 92.4% in Namibia. The p opo ion o women who could make a leas 1 o he ep oduc i e heal h decisions anged om 25.5% in Mali o 97.2% in Namibia. O e all, app oxima ely e e y i e ou o en women could ask a pa ne o use a condom, six ou en women could e use hei pa ne s sex and se en ou o women could make a leas 1 decision. Table 2 p esen s he odds o making a decision on sexual in e cou se, condom use and ep oduc i e heal h among women in 29 sub–Saha an A ican coun ies. Residence, age, weal h s a us, educa ion, eligion, occupa ion and pa ne ’s educa ion s ongly p edic ed deci- sion-making on sexual in e cou se and condom use among women in he 29 sub–Saha an A ican coun ies. Women om u al a eas we e less likely o decide on sexual in e cou se (OR = 0.58, CI = 0.55–0.61), condom use (OR = 0.54, CI = 0.51–0.57) and o e all decisions on ep oduc i e heal h (OR = 0.51, CI = 0.48–0.54), compa ed o hose om u ban a eas. The e ec s o women’s age on hei abili y o ake decision a ied and we e mainly posi i e. Fo ins ance, women aged 20–24 yea s we e 1.07 imes mo e likely o decide o ha e sex o no (CI = 1.02–1.12), 1.54 imes mo e likely o decide o use condom o no (CI = 1.47–1.62) and 1.20 imes mo e likely o decide on hei ep oduc i e heal h (CI = 1.14–1.26), compa ed o women aged 15–19 yea s. The ela ionship be ween weal h s a us and women’s abili y o make decision was mainly nega i e. Women wi h poo es weal h s a us we e 0.44 imes less likely o decide o ha e sex o no , 0.32 imes less likely o decide o use condom o no and 0.33 imes less likely o decide on hei ep oduc i e heal h, compa ed o women wi h iches weal h s a us. The ela ionship be ween eligion and women’s abili y o make decision was mainly nega- i e. Muslim women we e 0.39 imes less likely o decide o ha e sex o no , 0.30 imes less likely o decide o use condom o no and 0.29 imes less likely o decide on on hei ep oduc- i e heal h, compa ed o Ch is ian women. Women who we e no wo king we e 0.58 imes less likely o decide o ha e sex o no , 0.69 imes less likely o decide o use condom o no and 0.54 imes less likely o decide on hei ep oduc i e heal h, compa ed o women wo king ou side hei home. On he o he hand, women who wo k a home we e 1.33 imes mo e like Women’s ep oduc i e heal h decision making in sub-Saha an A ica PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0209985 Janua y 9, 2019 4 / 12 o decide o ha e sex o no , 2.56 imes mo e likely o decide o use condom o no and 1.89 imes mo e likely o decide on hei ep oduc i e heal h, compa ed o women wo king ou side hei home. Women wi h no educa ion we e 0.19 imes less likely o decide o ha e sex o no , 0.12 imes less likely o decide o use condom o no and 0.11 imes less likely o decide on hei ep oduc i e heal h, compa ed o women wi h highe educa ion. In he inal mul i a ia e model, he s a is ical associa ions be ween all he exposu es and he h ee ou comes we e e ained, al hough mos o he odds we e a enua ed (Table 3). The only excep ion is ha he associa ions o decision-making on sexual in e cou se wi h age and weal h s a us los hei s a is ical signi icance. Discussion This pape examined he p e alence o ep oduc i e heal h decision-making and he ac o s ha de e mine women’s decision-making on ep oduc i e heal h issues in 27 sub-Saha an A ican coun ies. The s udy ound ou ha he e we e wide a ia ions in he p opo ion o Table 1. Backg ound cha ac e is ics o sampled women in 27 sub–Saha an A ican coun ies. Coun y Weigh ed N�% can ask pa ne o use a condom % can e use sex Rep oduc i e heal h decision index No Yes No Yes No decision A leas 1 Benin, 2011–2012 11,045 61.6 38.4 47.2 52.8 41.9 58.1 Bu undi, 2011 5,424 33.7 66.3 39.4 60.6 19.6 80.4 Came oon, 2011 4,682 42.7 57.3 25.4 74.6 19.9 80.1 Chad, 2014–2015 4,553 82.6 17.4 55.6 44.4 51.8 48.2 Como os, 2012 3,087 54.4 45.6 55.7 44.3 43.5 56.5 Congo, 2011–2012 6,259 34.4 65.6 29.4 70.6 14.9 85.1 Congo DR, 2013–2014 12,119 60.1 39.9 33.8 66.2 25.8 74.2 Co e d’ oi e, 2011–2014 6,290 58.0 42.0 42.9 57.1 35.3 64.7 E hiopia, 2011 10,225 63.2 36.8 47.0 53.0 38.0 62.0 Gabon, 2012 4,383 18.2 81.8 18.4 81.6 8.3 91.7 Gambia, 2013 6,763 55.7 44.3 47.7 52.3 39.6 60.4 Ghana, 2014 5,382 33.3 66.7 26.1 73.9 17.8 82.3 Guinea, 2012 6,727 75.8 24.2 59.9 40.1 39.6 60.4 Kenya, 2014 8,679 25.6 74.4 25.7 74.3 14.3 85.7 Leso ho, 2014–2015 1,662 7.9 92.1 24.6 75.4 3.2 96.8 Libe ia, 2013 5,382 43.7 56.3 15.4 84.6 12.7 87.3 Malawi, 2010 15,577 22.4 77.6 25.1 74.9 12.3 87.7 Mali, 2012–2013 7,509 83.4 16.6 81.7 18.3 74.5 25.5 Mozambique, 2011 9,337 61.1 38.9 51.2 48.8 44.9 55.1 Namibia, 2015 3,090 6.6 93.4 7.7 92.3 2.8 97.2 Nige ia, 2013 27,749 61.9 38.1 38.7 61.3 33.7 66.3 Rwanda, 2014–2015 6,962 16.7 83.3 17.3 82.7 6.9 93.1 Sie a Leone, 2013 10,800 67.4 32.6 31.4 68.6 28.8 71.2 Senegal, 2010–2011 10,390 69.9 30.1 71.0 29.0 58.2 41.8 Togo, 2013–2014 6,298 39.8 60.2 28.1 71.9 21.7 78.3 Zambia, 2013–2014 9,837 24.2 75.8 30.4 69.6 15.4 84.6 Zimbabwe, 2015 6,155 29.2 70.8 28.2 71.8 14.1 85.9 All Coun ies ( o al) 210,536 50.2 49.8 39.3 60.7 30.7 69.3 �Numbe o women wi h da a on ep oduc i e heal h decision h ps://doi.o g/10.1371/jou nal.pone.0209985. 001 Women’s ep oduc i e heal h decision making in sub-Saha an A ica PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0209985 Janua y 9, 2019 5 / 12 Table 2. Logis ic eg ession model showing he ela ionship (Odds a io) be ween backg ound cha ac e is ics and women’s decision-making on sexual in e cou se, condom use and ep oduc i e heal h decision-making index in 27 sub-Saha an A ican coun ies, 2010–2016. Explana o y ac o s P opo ions N = 210,536 Decision-making on sexual in e cou se Decision-making on condom use Rep oduc ion heal h decision-making index Model I Odds Ra io (OR) 95% CI Model I Odds Ra io (OR) 95% CI Model Odds Ra io (OR) 95% CI Residence U ban 73,097 Re Re Re Ru al 137,439 0.58���(0.55–0.61) 0.54���(0.51–0.57) 0.51���(0.48–0.54) Age 15–19 16,259 0.75���(0.70–0.79) 1.05���(0.98–1.11) 0.79���(0.74–0.84) 20–24 39,968 1.07��(1.02–1.12) 1.54���(1.47–1.62) 1.20���(1.14–1.26) 25–29 51,524 1.11���(1.06–1.16) 1.54���(1.47–1.62) 1.23���(1.17–1.29) 30–34 43,210 1.14���(1.09–1.19) 1.49���(1.43–1.56) 1.25���(1.19–1.31) 35–39 36,717 1.10���(1.05–1.16) 1.39���(1.33–1.46) 1.19���(1.32–1.25) 40–44 25,567 1.04(0.99–1.09) 1.21���(1.15–1.28) 1.08��(1.03–1.14) 45–49 20,075 Re Re Re Weal h s a us Poo es 42,002 0.44���(0.42–0.47) 0.32���(0.30–0.34) 0.33���(0.31–0.35) Poo e 42,981 0.53���(0.50–0.56) 0.41���(0.39–0.44) 0.42���(0.39–0.45) Middle 41,850 0.58���(0.55–0.62) 0.49���(0.46–0.51) 0.47���(0.45–0.51) Riche 42,003 0.72���(0.68–0.75) 0.64���(0.60–0.67) 0.63���(0.59–0.67) Riches 41,680 Re Re Re Educa ion No educa ion 88,230 0.19���(0.17–0.21) 0.12���(0.11–0.13) 0.11���(0.10–0.12) P ima y 66,581 0.45���(0.41–0.49) 0.40���(0.36–0.43) 0.34���(0.30–0.38) Seconda y 47,764 0.65���(0.60–0.72) 0.56���(0.52–0.61) 0.56���(0.50–0.63) Highe 7,961 Re Re Re Religion Muslims 76,621 0.39���(0.36–0.40) 0.30���(0.29–0.32) 0.29���(0.27–0.31) Ch is ians 121,473 Re Re Re O he s 12,442 0.54���(0.50–0.59) 0.41���(0.38–0.45) 0.40���(0.37–0.44) Occupa ion No wo king 64,520 0.58���(0.56–0.60) 0.67���(0.65–0.70) 0.53���(0.51–0.56) Wo king 146,016 Re Re Re Pa ne ’s educa ion No educa ion 72,118 0.25���(0.23–0.27) 0.18���(0.17–0.19) 0.17���(0.16–0.19) P ima y 59,380 0.56���(0.53–0.61) 0.58���(0.55–0.62) 0.51���(0.47–0.55) Seconda y 62,269 0.80���(0.75–0.86) 0.73���(0.69–0.78) 0.75���(0.70–0.81) Highe 16,769 Re Re Re Coun y Benin 11,045 Re Re Re Bu undi 5,424 1.55���(1.44–1.65) 3.40���(3.18–3.65) 3.24���(3.00–3.50) Came oon 4,682 2.73���(2.53–9.94) 2.40��(2.24–2.57) 3.08���(2.84–3.34) Chad 4,553 0.65���(0.61–0.70) 0.29���(0.27–0.32) 0.62���(0.57–0.66) Como os 3,087 0.76���(0.70–0.82) 1.56���(1.44–1.69) 1.05���(0.97–1.14) Congo 6,259 1.80���(1.70–1.89) 1.00(0.94–1.05) 2.05���(1.95–2.17) Congo DR 12,119 2.47���(2.31–2.63) 2.83���(2.66–3.02) 4.52���(4.18–4.89) Co e d’ oi e 6,290 1.09��(1.03–1.16) 1.10��(1.04–1.18) 1.24���(1.16–1.32) E hiopia 10,225 1.06�(1.00–1.11) 0.98(0.92–1.03) 1.18���(1.11–1.24) Gabon 4,383 4.55���(4.18–4.94) 6.55���(6.04–7.10) 7.74���(6.96–8.61) (Con inued) Women’s ep oduc i e heal h decision making in sub-Saha an A ica PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0209985 Janua y 9, 2019 6 / 12 women making decisions on sexual in e cou se (18%-92%), condom use (16%-93%) and ep oduc i e heal h decision-making index (25%-97%) ac oss he egion. The s udy ound ou ha abou 1 in 2 women could no eques hei pa ne s o use con- dom whils wo ou o i e could no e use hei pa ne s sexual in e cou se when hey eques . Th ee ou o en women could no make decision ega ding hei ep oduc i e heal h. Simila esul s we e seen by Da eh and collegues in 2014 [10] They epo ed ha one- i h o sampled Ghanaian women could no e use hei pa ne s’ eques o sexual in e cou se whils one ou o ou could no demand he use o condoms by hei pa ne s. Residence, women’s educa ion, eligion, occupa ion and pa ne s’ educa ion we e independen ly associa ed wi h bo h decision-making on sexual in e cou se, condom use as well as he ep oduc i e heal h decision-making index while age and weal h we e independen ly associa ed wi h decision- making ega ding condom use and he ep oduc i e heal h index. This is e iden in ea lie wo ks o Exa e y e al. [5]; Rahman e al. [13], Hammed e al.[24] and Suja h and Reddy[6]. Olde women we e seen o be mo e likely o make decision on sexual in e cou se, condom use and ep oduc i e heal h decision-making index, compa ed o he younge ones. This ind- ing a i ms he s udy by Hameed e al. [24] and Da eh e al. [10] who ound ha women wi h inc eased age end o ge highe decision-making powe . In mos A ican coun ies, younge women a e no expec ed o a gue wi h olde pe sons and a e equi ed o espec hei opinions a all ime. The e a e also aces o in e gene a ional sexual ela ionships and age mixing be ween younge women and olde men in a sexual ela ionship. The e is also e idence o ea ly ma iage by younge women which is being used as a s a egy o economic su i al [25]. Table 2. (Con inued) Explana o y ac o s P opo ions N = 210,536 Decision-making on sexual in e cou se Decision-making on condom use Rep oduc ion heal h decision-making index Model I Odds Ra io (OR) 95% CI Model I Odds Ra io (OR) 95% CI Model Odds Ra io (OR) 95% CI Gambia 6,763 0.86���(0.81–0.91) 1.19���(1.12–1.26) 1.00(0.94–1.07) Ghana 5,382 2.64���(2.46–2.84) 3.29���(3.08–3.53) 3.35���(3.10–3.62) Guinea 6,727 0.66���(0.62–0.70) 0.58���(0.54–0.62) 0.69���(0.65–0.74) Kenya 8,679 2.31���(2.18–2.45) 3.61���(3.40–3.83) 3.38���(3.17–3.61) Leso ho 1,662 2.69���(2.38–3.04) 26.14���(21.07–32.43) 20.06���(15.29–26.33) Libe ia 5,382 4.79���(4.42–5.18) 2.12���(1.99–2.26) 4.90���(4.61–5.33) Malawi 15,577 2.92���(2.78–3.08) 6.45���(6.11–6.81) 5.96���(5.60–6.33) Mali 7,509 0.22���(0.21–0.23) 0.37���(0.34–0.39) 0.29���(0.27–0.31) Mozambique 9,337 1.13���(1.06–1.19) 1.41���(1.33–1.49) 1.58���(1.51–1.65) Namibia 3,090 9.99���(8.80–11.34) 21.60���(18.84–24.75) 22.30���(18.39–27.04) Nige ia 27,749 1.55���(1.48–1.62) 1.06�(1.01–1.11) 1.58���(1.51–1.65) Nige 9,866 0.44���(1.48–1.62) 0.55���(0.52–0.58) 0.54���(0.52–0.57) Rwanda 6,962 4.46���(4.15–4.79) 8.45���(7.85–9.10) 10.36���(9.27–11.46) Sie a Leone 10,800 2.11���(1.99–2.23) 0.93�(0.88–0.98) 1.94���(1.83–2.05) Senegal 10,390 0.34���(0.32–0.37) 0.64���(0.60–0.67) 0.49���(0.46–0.51) Togo 6,298 2,28���(2.01–2.60) 2.10���(1.97–2.23) 2.28���(2.13–2.44) Uganda 7,125 3.31���(3.10–3.54) 2.92���(2.74–3.10) 4.20���(3.90–4.52) Zambia 9,837 2.28���(2.16–2.42) 5.72���(5.38–6.08) 4.58��(4.28–4.90) Zimbabwe 6,155 2.50���(2.34–2.67) 4.37���(4.09–4.68) 4.78���(4.40–5.19) �p<0.05 �� p<0.01 ���p<0.001 h ps://doi.o g/10.1371/jou nal.pone.0209985. 002 Women’s ep oduc i e heal h decision making in sub-Saha an A ica PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0209985 Janua y 9, 2019 7 / 12 Table 3. Mul i a ia e logis ic eg ession model showing he ela ionship (Adjus ed odds a io) be ween backg ound cha ac e is ics and women’s decision-making on sexual in e cou se, condom use and ep oduc i e heal h decision-making index in 27 sub–Saha an A ican coun ies, 2010–2016. Explana o y ac o s P opo ions N = 210,536 Decision-making on sexual in e cou se Decision-making on condom use Rep oduc ion heal h decision-making index Model II Adjus ed Odds Ra io (AOR) 95% CI Model II Adjus ed Odds Ra io (AOR) 95% CI Model II Adjus ed Odds Ra io (AOR) 95% CI Residence U ban 73,097 Re Re Re Ru al 137,439 0.89���(0.86–0.91) 0.83���(0.81–0.85) 0.81���(0.81–0.86) Age 15–19 16,259 0.93���(0.88–0.97) 1.35���(1.28–1.41) 0.99(0.94–1.05) 20–24 39,968 1.04�(0.99–1.07) 1.53���(1.46–1.59) 1.14���(1.10–1.20) 25–29 51,524 1.04��(1.00–1.08) 1.49���(1.44–1.66) 1.16���(1.11–1.20) 30–34 43,210 1.06��(1.02–1.10) 1.41���(1.36–1.47) 1.15���(1.03–1.20) 35–39 36,717 1.02(0.98–1.06) 1.32���(1.27–1.38) 1.10���(1.05–1.14) 40–44 25,567 0.99(0.95–1.03) 1.16���(1.11–1.21) 1.02(0.97–1.07) 45–49 20,075 Re Re Re Weal h s a us Poo es 42,002 0.74���(0.71–0.77) 0.59���(0.57–0.62) 0.61���(0.58–0.64) Poo e 42,981 0.80���(0.76–0.83) 0.69���(0.66–0.72) 0.69���(0.66–0.72) Middle 41,850 0.82���(0.78–0.85) 0.74���(0.71–0.77) 0.72���(0.69–0.75) Riche 42,003 0.88���(0.95–1.03) 0.81���(0.78–0.84) 0.81��(0.77–0.84) Riches 41,680 Re Re Re Educa ion No educa ion 88,230 0.49���(0.45–0.53) 0.34���(0.32–0.37) 0.37���(0.34–0.41) P ima y 66,581 0.66���(0.61–0.71) 0.52���(0.48–0.55) 0.54���(0.50–0.51) Seconda y 47,764 0.83���(0.77–0.88) 0.72���(0.67–0.77) 0.78���(0.71–0.85) Highe 7,961 Re Re Re Religion Muslims 76,621 0.60���(0.59–0.62) 0.76���(0.74–0.78) 0.60���(0.58–0.62) Ch is ians 121,473 Re Re Re O he s 12,442 0.89���(0.85–0.93) 0.82���(079–0.85) 0.84���(0.81–0.88) Occupa ion No wo king 64,520 0.78���(0.76–0.80) 0.82���(0.80–0.83) 0.72���(0.70–0.74) Wo king 146,016 Re Re Re Pa ne ’s educa ion No educa ion 72,118 0.74���(0.70–0.78) 0.60���(0.57–0.63) 0.65���(0.61–0.68) P ima y 59,380 0.92��(0.88–0.97) 0.75���(0.71–0.79) 0.84���(0.79–0.89) Seconda y 62,269 0.99(0.95–1.04) 0.91���(0.87–0.95) 0.95(0.90–1.01) Highe 16,769 Re Re Re Coun y Benin 11,045 Re Re Re Bu undi 5,424 1.20���(1.12–1.30) 2.88���(2.68–3.10) 2.48���(2.28–2.69) Came oon 4,682 2.01���(1.85–2.18) 1.52���(1.40–1.64) 2.10���(1.92–2.29) Chad 4,553 0.86���(0.80–0.93) 0.33���(0.30–0.35) 0.82���(0.76–0.88) Como os 3,087 0.86��(0.79–0.94) 1.26���(1.15–1.39) 1.06(0.97–1.16) Congo 6,259 1.10��(1.04–1.16) 0.54���(0.51–0.57) 1.10��(1.04–1.17) Congo DR 12,119 1.45���(1.35–1.56) 1.68���(1.56–1.80) 2.42���(2.22–2.64) Co e d’ oi e 6,290 1.18���(1.11–1.26) 1.10��(1.03–1.18) 1.32���(1.23–1.41) E hiopia 10,225 1.15���(1.09–1.22) 0.96(0.90–1.02) 1.30���(1.23–1.38) (Con inued) Women’s ep oduc i e heal h decision making in sub-Saha an A ica PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0209985 Janua y 9, 2019 8 / 12 Women wi h poo es , poo e , middle and iche weal h s a us we e less likely o make deci- sion on sexual in e cou se, condom use and ep oduc i e heal h decision-making index, com- pa ed o women wi h iches weal h s a us simila o hose ound by [5,10,18,21]. Weal h may lead o au onomy and independence and is ela ed o occupa ion and educa ion. Weal h could also be associa ed wi h sel -es eem and con idence and his may ha e e ec s on he abili y o make decision ega ding ep oduc i e heal h issues. Women wi h seconda y and lesse educa ion we e less likely o make decision on sexual in e cou se, condom use and ep oduc i e heal h decision-making index, compa ed o women wi h highe educa ion. Also, women whose pa ne s had no educa ion, p ima y and seconda y educa ion we e less likely o make decision on sexual in e cou se, condom use and ep oduc i e heal h decision-making index, compa ed o hose wi h highe educa ion. This is consis en wi h p e ious s udies [5,10,16,20]. Educa ion empowe s women o be independen and equip hem wi h he essen ial in o ma ion ega ding sexual in e cou se and condom use ha may be impo an o making decisions ega ding hei ep oduc i e heal h isssues. How well a pe son is in o med a ec s decsion making a e e y s age [26]. Ealie s udies had con i med ha one’s eligious a ilia ion may ha e some e ec s on he decision-making wi h ega ds o ep oduc i e heal h. The s udy also ound ha Muslim women and women om o he eligions we e less likely o make decision on sexual in e - cou se, condom use and ep oduc i e heal h decision-making index, compa ed o Ch is ian women. This inding is consis en wi h Da eh e al. [10] bu goes con a y o ha o Exa e y Table 3. (Con inued) Explana o y ac o s P opo ions N = 210,536 Decision-making on sexual in e cou se Decision-making on condom use Rep oduc ion heal h decision-making index Model II Adjus ed Odds Ra io (AOR) 95% CI Model II Adjus ed Odds Ra io (AOR) 95% CI Model II Adjus ed Odds Ra io (AOR) 95% CI Gabon 4,383 2.79���(2.54–3.06) 3.64���(3.33–3.98) 4.12���(3.67–4.63) Gambia 6,763 1.19���(1.11–1.27) 1.30���(1.21–1.39) 1.35���(1.26–1.45) Ghana 5,382 1.94���(1.80–2.09) 2.25���(2.09–2.42) 2.29���(2.11–2.49) Guinea 6,727 0.92�(0.86–0.99) 0.68���(0.63–0.73) 0.95(0.87–1.01) Kenya 8,679 1.60���(1.50–1.71) 2.34���(2.20–2.50) 2.22���(2.06–2.49) Leso ho 1,662 1.29���(1.14–1.47) 12.28���(9.83–15.32) 8.18���(6.19–10.81) Libe ia 5,382 4.06���(3.74–4.41) 1.73���(1.61–1.85) 4.05���(3.71–4.42) Malawi 15,577 2.09���(2.65–3.09) 4.58���(4.31–4.87) 4.00���(3.74–4.28) Mali 7,509 0.34���(0.32–0.37) 0.46���(0.43–0.50) 0.44���(0.41–0.47) Mozambique 9,337 2.09���(1.97–2.21) 1.02(0.95–1.09) 0.89���(0.84–0.95) Namibia 3,090 5.98���(5.23–6.83) 12.40���(9.83–15.32) 12.32���(10.05–15.10) Nige ia 27,749 1.52���(1.45–1.60) 0.78���(0.74–0.82) 1.49���(1.41–1.57) Rwanda 9,866 2.86���(2.65–3.09) 5.73���(5.29–6.20) 6.12���(5.51–6.80) Sie a Leone 6,962 2.73���(2.557–2.90) 0.99(0.93–1.05) 2.45���(2.30–2.61) Senegal 10,800 0.55���(0.52–0.59) 0.84���(0.79–0.89) 0.78���(0.73–0.83) Togo 10,390 1.67���(1.56–1.79) 1.70���(1.59–1.82) 1.83���(1.70–1.96) Zambia 6,298 1.38���(1.30–1.47) 3.33���(3.12–3.56) 2.49���(2.31–2.68) Zimbabwe 7,125 1.31���(1.22–1.41) 1.94���(1.80–2.09) 2.08���(1.90–2.27) �p<0.05 �� p<0.01 ���p<0.001 NB: Rep oduc i e heal h decision-making index is he combina ion o decision-making on sexual in e cou se and he decision-making on condom use h ps://doi.o g/10.1371/jou nal.pone.0209985. 003 Women’s ep oduc i e heal h decision making in sub-Saha an A ica PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0209985 Janua y 9, 2019 9 / 12