The Association between Women’s Perception of Community Support for and Utilization of Maternity Healthcare Services in Ethiopia

1 1 The Association between Women’s Perception of Community Support for and 2 Utilization of Maternity Healthcare Services in Ethiopia 3 4 Yongyi Lu 1*, Sally Safi1, Solomon Shiferaw2, Linnea A. Zimmerman1* 5 6 1 Department of Population Family and Reproductive Health, Johns Hopkins Bloomberg School 7 of Public Health, Baltimore, MD, USA 8 2 Department of Reproductive Health and Health Service Management, School of Public Health, 9 Addis Ababa University, Addis Ababa, Ethiopia 10 *Corresponding authors: 11 Linnea A. Zimmerman, [email protected], Department of Population Family and 12 Reproductive Health, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, USA 13 Yongyi Lu, [email protected], Department of Population Family and Reproductive Health, 14 Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, USA 15 16 17 18 19 20 21 . CC-BY 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.(which was not certified by peer review)preprint The copyright holder for thisthis version posted September 28, 2025. ; https://doi.org/10.1101/2025.09.25.25336698doi: medRxiv preprint NOTE: This preprint reports new research that has not been certified by peer review and should not be used to guide clinical practice. 2 22 Abstract 23 Ethiopia has one of the highest maternal mortality ratios in sub-Saharan Africa. Many factors 24 contribute, including limited access to and use of maternity care services. Community support 25 plays an important role in influencing women’s utilization of such services. The objective of this 26 study was to analyze the association between women’s perception of community support and their 27 utilization of maternity healthcare services in Ethiopia while exploring how this association varies 28 by urban and rural residence. Longitudinal data from the Performance Monitoring in Action 29 Ethiopia was used. We excluded women who were postpartum at baseline, did not complete the 30 six-week follow-up survey, and did not deliver a live birth. The total analytic sample for this study 31 was 1,924. We used logistic regression to analyze the relationship between a woman’s perception 32 of community support for the relevant component and the service utilization. Then, we included 33 an interaction term between community perceptions and residence for each model. The proportion 34 of women with four or more antenatal care visits, who gave birth in a health facility, and had 35 postnatal care visits within 2 days postpartum are 49%, 52%, and 41%, respectively. Women’s 36 access to comprehensive maternity care was 25%. Women who perceived their communities as 37 “fully supportive” of comprehensive maternity care were about twice as likely to receive such care 38 compared to women who perceived that the community was not fully supportive of comprehensive 39 maternity care (aOR: 1.89, 95% CI: 1.49-2.38). Regarding the full continuum of care, urban 40 women who perceive full support were significantly more likely to receive all components of 41 care. Perceived community support is an important predictor of women's utilization of maternal 42 care in Ethiopia. These findings highlight a key factor influencing care-seeking behavior and 43 variation between urban and rural residence, contributing to ongoing disparities in healthcare 44 access. . CC-BY 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.(which was not certified by peer review)preprint The copyright holder for thisthis version posted September 28, 2025. ; https://doi.org/10.1101/2025.09.25.25336698doi: medRxiv preprint 3 45 Introduction 46 Sub-Saharan Africa has the highest burden of maternal mortality in the world, with around 85% 47 of maternal deaths (1). Ethiopia, specifically, has one of the highest maternal mortality ratios in 48 sub-Saharan Africa, with approximately 401 maternal deaths per 100,000 live births, recording the 49 fourth largest number of maternal deaths in the world in 2020 (2–4). Many factors contribute to 50 maternal mortality in Ethiopia, such as lack of transportation to health facilities, shortage of life- 51 saving maternal supplies in health facilities, and lack of access to and utilization of the continuum 52 of maternity care services, which includes four or more antenatal care (ANC) visits, use of facility- 53 based delivery services, and receipt of postnatal care (PNC) for the mother within 48 hours of 54 delivery (5–7). 55 ANC is widely considered an essential service during pregnancy (8). ANC includes prevention, 56 such as iron supplementation, toxoid immunization; treatment, such as treatment for sexually 57 transmitted infections; and health education (9,10). Currently, Ethiopia has undertaken several 58 intervention programs to promote ANC access and improve the quality of ANC services such as 59 conducting health extension programs, having an organized community health structure, and 60 implementing community-based health insurance (11). Recent estimates suggest that despite these 61 efforts, ANC utilization remains low; estimates from the most recent DHS suggest that 62 approximately 74% of women received at least one ANC visit from a skilled provider and only 43% 63 received four or more visits (12). These figures also mask significant geographic variability, with 64 approximately 84% of urban women receiving at least one visit and 59% receiving four or more 65 relative to 71% and 37%, respectively, of rural women (12). 66 Delivering in a health facility, with a skilled birth attendant, is the second critical component of 67 the continuum of care. Evidence suggests that prioritizing intrapartum care, that is ensuring that . CC-BY 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.(which was not certified by peer review)preprint The copyright holder for thisthis version posted September 28, 2025. ; https://doi.org/10.1101/2025.09.25.25336698doi: medRxiv preprint 4 68 women deliver with a skilled attendant in a facility that is ready to address an obstetric emergency, 69 can significantly reduce maternal morbidity and mortality (13). Given the effectiveness of facility 70 delivery, the WHO has long recommended that pregnant women give birth in a health facility to 71 prevent maternal mortality (14). In Ethiopia, the Federal Ministry of Health introduced a policy of 72 free delivery services in public health facilities in 2013 (15). However, the use of healthcare 73 facilities for delivery is still very low in Ethiopia. According to the 2019 Ethiopia Demographic 74 and Health Survey (DHS), only 48% of women who delivered a live birth in the five years prior 75 to the survey were delivered in a health facility (12). As with ANC utilization, these numbers mask 76 significant variation with almost 80% of urban women delivering in a facility relative to 20% of 77 rural women. 78 Globally, approximately 57% of maternal deaths occur in the postpartum period within 6 weeks 79 (7). In Ethiopia, around 51% to 75% of maternal deaths occur during the postpartum period (7). 80 To reduce maternal and infant deaths, the WHO suggests that both mother and infant should 81 receive a PNC visit within one day (24 hours) of birth, no matter where the baby is born, and have 82 a minimum of four PNC visits within the first six weeks after birth (16). Visits should include 83 screening for obstetric complications, prevention of infectious diseases, general well-being 84 management, breastfeeding promotion, and discussion of postpartum family planning options (17). 85 PNC visits are also important opportunities to check mother and baby for danger signs such as low 86 body temperature, shortness of breath, and fever (18). According to the 2019 Ethiopia DHS, only 87 35% of newborns and 34% of women received PNC services within two days of birth, again with 88 significant geographic disparity (48% of urban women received a PNC check for their health 89 versus 29% of rural women) (12). There is evidence that utilization of PNC services is increasing, . CC-BY 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.(which was not certified by peer review)preprint The copyright holder for thisthis version posted September 28, 2025. ; https://doi.org/10.1101/2025.09.25.25336698doi: medRxiv preprint 5 90 however, overall use remains low; the utilization of PNC services among women increased from 91 17% in 2016 to 34% in 2019 (19). 92 Due to the generally low utilization of each component service, utilization of the entire continuum 93 of maternal health services – four or more ANC visits, facility delivery, and receipt of a PNC visit 94 within 48 hours after birth – remains well below optimal levels. Estimates in Ethiopia range from 95 9% to 42 %, with urban women having approximately 2.8 times higher odds of completing the 96 continuum than their rural counterparts (20). 97 A number of barriers have been identified that impact use of each component service, in addition 98 to the entire continuum of care. ANC utilization is negatively impacted by both supply-side factors, 99 such as shortage of required materials and disrespectful care by providers, and demand-side factors, 100 such as transport costs and partner approval (21). Reasons for the low rate of use of health facilities 101 include a perceived high cost of delivery, lack of transportation, shortage of supplies, low level of 102 awareness about complications of pregnancy, and cultural barriers (15). Some barriers to accessing 103 PNC include lack of funding for PNC, lack of PNC priorities from obstetricians, and lack of 104 supplies for PNC provision (22). 105 Qualitative evidence suggests that community and cultural factors also impact receipt of maternal 106 care services (21,23). One study in the Amhara region of Ethiopia found that communities 107 supported facility delivery through taking care of children, managing household chores, and 108 supporting transport costs and logistics (24). The study did not, however, explore whether or how 109 normative support for different maternal health services affected use. Another study in areas 110 surrounding Addis Ababa, the capital, found that community views of quality and utility of 111 antenatal services were impactful on both women’s initial and return attendance (21). Community . CC-BY 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.(which was not certified by peer review)preprint The copyright holder for thisthis version posted September 28, 2025. ; https://doi.org/10.1101/2025.09.25.25336698doi: medRxiv preprint 6 112 norms that facility delivery should only be used in case of emergency have been found to support 113 high rates of home delivery in pastoralist communities (25). Conversely, as utilization of services 114 has increased, more women may have had positive experiences, leading to higher community 115 support for services (26). 116 This qualitative evidence identifies ways that community factors can both inhibit and support 117 utilization of maternity care services, but few quantitative studies in Ethiopia have explored this 118 question, and none in relation to the effect of community support across the entire continuum of 119 care. Teferi and colleagues found that if a woman perceived more community support for facility 120 delivery, she had higher odds of expressing a preference for facility delivery, but they did not 121 assess whether perceived community support was associated with facility delivery itself (27). 122 Mamo and colleagues found that women who perceived greater support from partners, family 123 members, and friends to deliver in a facility were significantly more likely to do so (28). Significant 124 gaps remain, however, as few studies have explored the effect of community support on postnatal 125 care, on the continuum of care more broadly, or how the influence of community support may 126 differ by residence and contribute to the differentials in utilization by urban and rural residence 127 identified above. 128 Due to persistent maternal mortality and poor utilization of ANC, facility-based delivery, and PNC 129 in Ethiopia, many previous studies have analyzed other factors affecting the utilization of these 130 maternity healthcare services such as education, mass media, and family size (29,30). However, 131 to our knowledge, no study has examined how women’s perception of community support 132 influences the utilization of each component and the entirety of the continuum of care in Ethiopia, 133 nor whether these differ by residence. It is important to understand community-level support of 134 ANC, PNC, and facility-based delivery and how this may affect women’s utilization of these . CC-BY 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.(which was not certified by peer review)preprint The copyright holder for thisthis version posted September 28, 2025. ; https://doi.org/10.1101/2025.09.25.25336698doi: medRxiv preprint 7 135 healthcare services. Our objective is thus to 1) identify the relationship between perceived 136 community support for each service (ANC, facility delivery, and PNC) with receipt of the service; 137 2) to identify whether greater community support is associated with completing the entire 138 continuum of care; and 3) whether the influence of each differs by urban versus rural residence. 139 Materials and Methods 140 Ethical Clearance 141 All procedures were reviewed and approved by Addis Ababa University, College of Health 142 Sciences (AAU/CHS) (Ref: AAUMF 01-008) and the Johns Hopkins University Bloomberg 143 School of Public Health (JHSPH) Institutional Review Board (FWA00000287) and carried out in 144 accordance with relevant guidelines and regulations. Verbal informed consent was obtained from 145 all of the participants. The IRB approves verbal consent procedures (without a need for written 146 consent) for simple surveys without any invasive procedures in an environment where literacy is 147 low. Women under the age of 18 who are married are considered emancipated minors and are able 148 to provide informed consent. No unemancipated minors were included in this survey. 149 Data sources 150 This longitudinal study used data from the Performance Monitoring for Action (PMA) Ethiopia 151 project. PMA Ethiopia is a collaboration between the Johns Hopkins Bloomberg School of Public 152 Health, the Ethiopian Federal Ministry of Health (FMoH), and Addis Ababa University (AAU). 153 PMA Ethiopia implemented a longitudinal survey of pregnant and postpartum women, which 154 focused on maternal and newborn health and reproductive health topics (31). Women were 155 enrolled during pregnancy or within six weeks postpartum and completed baseline interviews at 156 enrollment and follow-up interviews at six weeks, six months and one year postpartum. . CC-BY 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.(which was not certified by peer review)preprint The copyright holder for thisthis version posted September 28, 2025. ; https://doi.org/10.1101/2025.09.25.25336698doi: medRxiv preprint 8 157 Two datasets from the PMA Ethiopia panel survey were used for this analysis: the baseline survey, 158 conducted from September 15, 2019 to December 30, 2019, and the 6-week follow-up survey, 159 conducted throughout September 22, 2019 to June 26, 2020. The panel study was conducted using 160 a multi-stage design in six regions - Addis Ababa, Afar, Amhara, Oromia, Tigray, and Southern 161 Nations, Nationalities, and People’s region. 1 Urban and rural stratification was applied within 162 Amhara, Oromia, Tigray and SNNP. In the first stage of sampling, a total of 216 enumeration areas 163 (EAs), were randomly selected using probability-proportional-to-size. In the second stage of 164 sampling, all households in the selected EAs were listed and census of all household members was 165 conducted. All women between the ages of 15-49 were screened for eligibility (self-reported 166 pregnancy or less than six weeks postpartum) and, if eligible, were invited to participate in the 167 survey. At the baseline interview, resident enumerators explained the study's purpose and read the 168 approved consent language to the women approached for consent (31). In accordance with the 169 National Research Ethics Review Guidelines in Ethiopia, oral consent was granted due to 170 widespread illiteracy (32). During each follow-up, women were asked if they still agreed to 171 participate and if they had any questions (31). Women under age 18 are considered to be 172 emancipated minors within Ethiopia if they are married. No unemancipated minors were included 173 in this study. 174 A total of 2,855 women participated in the baseline interview and 2,669 women participated in the 175 6-week follow-up survey. The exclusion criteria in this study were women who were postpartum 176 at baseline (n=616), did not complete the six-week follow-up survey (n=175), and who did not 177 deliver a live birth (n=140). The total analytic sample for this study was 1,924. 1 Initial study design and sampling took place prior to the creation of three new regions from within SNNPR and thus represents estimates within the previous administrative borders of SNNP. . CC-BY 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.(which was not certified by peer review)preprint The copyright holder for thisthis version posted September 28, 2025. ; https://doi.org/10.1101/2025.09.25.25336698doi: medRxiv preprint 9 178 Measures 179 Outcome measures 180 Our dependent variables (outcomes) were based on survey questions that assessed women’s 181 utilization of maternity care. Using the 6-week follow-up survey, we assessed four measures: 182 (1) receipt of four or more ANC visits from a trained provider, including a health extension 183 worker (HEW), 184 (2) delivered in a health facility, and 185 (3) received PNC within 2 days of giving birth. Women were classified as receiving PNC 186 if a trained provider visited them in their home and discussed their health, they visited a trained 187 provider at a facility and discussed their health, or in the case of women who delivered in a facility, 188 if someone checked on their health prior to discharge. 189 (4) receipt of all three of the above components comprehensive maternity care. 190 Key independent variables 191 Our independent variables (predictors) were based on categorical measures from the baseline 192 survey assessing women's perception of community support for specific maternity healthcare 193 services. Specifically, we assessed three measures using the following questions from the baseline 194 survey: 195 (1) Do most, some, few, or no people in your community encourage women to deliver at 196 a facility? 197 (2) Do most, some, few, or no people in your community encourage going to antenatal 198 care? . CC-BY 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.(which was not certified by peer review)preprint The copyright holder for thisthis version posted September 28, 2025. ; https://doi.org/10.1101/2025.09.25.25336698doi: medRxiv preprint 10 199 (3) Do most, some, few, or no people in your community encourage women to seek 200 postnatal care? 201 The response options for the above three questions included: no people, few people, some people, 202 and most people. For our analysis of each specific service, we collapsed the first two categories 203 into “no or few people.” “Do Not Know” was included as a potential option, however, the number 204 of "do not know" responses ranged from 1.46% (n=28) for community support of antenatal care 205 to 2.08% (n=40) for community support of postnatal care. The small sample sizes did not allow 206 for sufficient power to detect differences in this group and thus, were treated as missing. 207 We also created a dichotomous measure for perceived community support for all types of 208 maternity care. Respondents who answered “most people” for each of the above three questions 209 were classified into a “High support” category and all other respondents were classified in a 210 “Lower support” category. 211 Adjustment variables 212 We identified sociodemographic variables that we hypothesized were likely to confound the 213 relationship of perceived community support and receipt of each service, including maternal age 214 (categorical variable in five-year age groups), education (categorical variable indicating none, 215 primary, secondary and above), household wealth quintile, and region. Due to almost universal 216 marriage in this population of currently pregnant women, we did not adjust by marital status. 217 Analysis 218 Exploratory analyses were used to describe the study sample and summarize missingness. For each 219 component of the continuum of care and for the entire continuum, we used logistic regression to 220 analyze the relationship between a woman’s perception of community support for the relevant . CC-BY 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.(which was not certified by peer review)preprint The copyright holder for thisthis version posted September 28, 2025. ; https://doi.org/10.1101/2025.09.25.25336698doi: medRxiv preprint 11 221 component and the utilization of that service. To determine if the influence of community support 222 differed by urban versus rural residence, we then included an interaction term between community 223 perceptions and residence for each model. All analyses accounted for complex survey design 224 through the application of survey weights and accounting for clustering within EAs. All analyses 225 were conducted using Stata 18 SE (33). 226 Results 227 Table 1 shows the distribution of characteristics of the sample, with weighted percentages. About 228 half of the women (55%) were aged 20-29 and 42% had never attended school. In terms of 229 perceptions of community support for maternity care, forty-nine percent (49%) of respondents 230 indicated that they felt that most people in their community encouraged pregnant women to seek 231 ANC. Fifty-two percent (52%) of respondents felt that most people in their community encouraged 232 pregnant women to give birth in a health facility. Forty-one percent (41%) felt that most people in 233 the community encouraged women who had given birth to seek PNC. In terms of use of maternity 234 care, 44% of respondents indicated that they had 4 or more ANC visits, 55% indicated that they 235 delivered in a health facility, 39% reported they had a PNC visit within 2 days and 25% reported 236 they had received all these services. 237 238 239 240 241 242 243 244 245 246 247 248 . CC-BY 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.(which was not certified by peer review)preprint The copyright holder for thisthis version posted September 28, 2025. ; https://doi.org/10.1101/2025.09.25.25336698doi: medRxiv preprint 12 249 250 251 252 253 Table 1 Sociodemographic characteristics of respondents (n=1,924) Sociodemographic characteristics Unweighted n Weighted n % Region Tigray 336 138 7.2 Afar 157 36 1.9 Amhara 333 398 20.7 Oromia 472 841 43.7 SNNP 454 443 23.0 Addis 172 69 3.6 Women’s age (in years) 15-19 166 193 10.0 20-24 470 476 24.8 25-29 616 572 29.8 30-34 366 368 19.1 ≥35 306 315 16.4 Household wealth index Lowest quintile 350 377 19.6 Lower quintile 293 385 20.0 Middle quintile 306 398 20.7 Higher quintile 354 383 19.9 Highest quintile 621 381 19.8 Education level Never attended 754 808 42.0 Primary 683 751 39.0 Secondary, technical or higher 487 366 19.0 Marital status Currently married and living with a man 1877 1885 98.0 Not currently partnered 47 39 2.0 Perceptions of community support for maternal care Unweighted n Weighted n % Woman’s perception of community support for ANC 1,896 1,896 No or few people 492 533 28.1 Some people 391 432 22.8 Most people 1023 931 49.1 Woman’s perception of community support for delivery at a facility 1891 1891 No or few people 487 531 28.0 Some people 314 376 19.9 Most people 1090 984 52.0 Woman’s perception of community support for PNC 1884 1884 No or few people 637 681 36.2 Some people 408 436 23.1 Most people 839 767 40.7 Woman’s perception of community support for comprehensive maternity carea 1867 1867 Not fully supportive community 1177 1274 68.2 Fully supportive communityb 690 593 31.8 Receipt of maternity care Unweighted n Weighted n % Received 4+ ANC visits No 990 1,088 56.5 Yes 934 836 43.5 Delivered in a health facility No 749 875 45.5 Yes 1175 1,049 54.5 Received PNC within 2 days postpartum No 1069 1,170 60.8 Yes 855 754 39.2 Received 4 + ANC visits, delivered at health facility, and received PNC No 1338 1,445 75.1 . CC-BY 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.(which was not certified by peer review)preprint The copyright holder for thisthis version posted September 28, 2025. ; https://doi.org/10.1101/2025.09.25.25336698doi: medRxiv preprint 13 Yes 586 479 24.9 254 a Comprehensive maternity care = 4+ ANC visits + facility delivery + PNC 255 b Fully supportive community = women who indicated that “most people” in community encourage receipt 256 of ANC, facility delivery, and receipt of PNC 257 Results from the logistic regression analysis examining the association of community support for 258 maternity services with women’s utilization of such services are presented in Table 2. Across all 259 components of the continuum of care, greater perceived support was associated with higher odds 260 of completing each component. For example, women who felt that some people in the community 261 encouraged the use of ANC had 2.35 times higher odds (95% CI: 1.70-3.26) of completing 4+ 262 ANC visits and women who felt that most people encouraged utilization of ANC had more than 263 three times higher odds compared to women who felt there was less community support for ANC 264 (aOR: 3.15, 95% CI: 2.37-4.17). These findings were similar for both facility delivery and 265 postnatal care, with progressively higher odds associated with more perceived support. Lastly, 266 women living in communities that they perceived to be “fully supportive” of encouraging 267 comprehensive maternity care were about twice as likely to receive comprehensive maternity care 268 than women who perceived that the community was not fully supportive of perinatal care (aOR: 269 1.89, 95% CI: 1.49-2.38). 270 271 272 273 274 . CC-BY 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.(which was not certified by peer review)preprint The copyright holder for thisthis version posted September 28, 2025. ; https://doi.org/10.1101/2025.09.25.25336698doi: medRxiv preprint 14 275 276 Table 2. Association between perceived community support for maternity health services with 277 women’s utilization of such services Crude OR (95% CI) Adjusted OR (95% CI) Variables (n=1,924) OR 95% CI P-value aOR 95% CI P-value Outcome: Receipt of 4+ ANC visits Portion of community that encourages utilization of ANC No/few people ref ref Some people 3.19 (2.37, 4.30) 0.000 2.35 (1.70, 3.26) 0.000 Most people 6.66 (5.17, 8.58) 0.000 3.15 (2.37, 4.17) 0.000 Outcome: Facility delivery Portion of community that encourages facility delivery No/few people ref ref Some people 2.30 (1.72, 3.09) 0.000 1.96 (1.39, 2.77) 0.000 Most people 7.79 (6.13, 9.90) 0.000 3.04 (2.27, 4.08) 0.000 Outcome: Receipt of PNC visit within 2 days Portion of community that encourages utilization of PNC No/few people ref ref Some people 1.85 (1.42, 2.41) 0.000 1.45 (1.07, 1.95) 0.015 Most people 3.64 (2.91, 4.54) 0.000 1.96 (1.52, 2.54) 0.000 Outcome: Receipt of comprehensive maternity care Portion of community that encourages comprehensive maternity care a Not fully supportive community ref ref Fully supportive community b 3.39 (2.76, 4.16) 0.000 1.89 (1.49, 2.38) 0.000 a Comprehensive maternity care = 4+ ANC visits + facility delivery + PNC b Fully supportive community = women who indicated that “most people” in community encourage receipt of ANC, facility delivery, and receipt of PNC 278 279 Table 3 demonstrates if and how these effects differ based on urban and rural residence. For ANC, 280 there are significant differences in utilization based on perceived support (aOR some:3.84, 95%CI: 281 1.98-7.43 and aOR most: 6.18, 95% CI: 3.55-10.78) and between urban and rural women 282 (aOR rural:0.42: 95% CI: (0.24-0.75), however, there is no evidence that the effect of perceived 283 support differs by residence based on the interaction terms (aOR some*rural 0.67, 95% CI: 0.32-1.42 . CC-BY 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.(which was not certified by peer review)preprint The copyright holder for thisthis version posted September 28, 2025. ; https://doi.org/10.1101/2025.09.25.25336698doi: medRxiv preprint 15 284 and aOR most*rural 0.68, 95% CI: 0.36-1.27). There is a similar relationship for facility delivery, 285 though the statistical significance among women who perceive some support is attenuated. 286 There is evidence of interaction between community support and residence for both receipt of PNC 287 and the full continuum of care, however. When accounting for interaction, the main effect of 288 community support on PNC utilization is no longer significant - among urban women, there is no 289 difference in the odds of receiving PNC by level of perceived community support. The main effect 290 of residence remains significant; among women who perceive little support in the community, 291 those who live in rural areas have significantly lower odds of utilizing PNC than women who live 292 in urban areas (aOR:0.10, 95% CI: 0.06-0.15). The interaction, however, demonstrates that the 293 effect of community support is significantly stronger among rural women, with perceived levels 294 of community support being strongly associated with increased odds of PNC utilization 295 (aOR some*rural 2.60, 95% CI: 1.39-4.84 and aOR most*rural 2.27, 95% CI: 1.34-3.85). Regarding the 296 full continuum of care, the main effects of both perceived support (aOR full support 1.81, 95% CI: 297 1.33-2.45) and residence (aOR rural 0.15, 95% CI: 0.13-2.45) retain statistical significance; among 298 urban women, women who perceive full support are significantly more likely to receive all 299 components of care while among women who perceive low support, rural women are significantly 300 less likely to receive all components relative to urban women. As with PNC utilization, the effect 301 of perceiving full support is stronger among rural women than among urban women (aOR full 302 support*rural 1.57, 95%CI: 1.00-2.45). 303 304 305 306 . CC-BY 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.(which was not certified by peer review)preprint The copyright holder for thisthis version posted September 28, 2025. ; https://doi.org/10.1101/2025.09.25.25336698doi: medRxiv preprint 16 307 Table 3. Interaction effects of community perceptions and residence on maternity care utilization Variable OR 95% CI P-value Outcome: Receipt of 4+ ANC visits Portion of community that encourages utilization of ANC No/few people reference Some people 3.84 (1.98, 7.43) 0.000 Most people 6.18 (3.55, 10.78) 0.000 Residence Urban reference Rural 0.42 (0.24, 0.75) 0.003 Interaction term Some people * rural 0.67 (0.32, 1.42) 0.298 Most people * rural 0.68 (0.36, 1.27) 0.227 Outcome: Receipt facility delivery Portion of community that encourages utilization of facility delivery No/few people reference Some people 3.23 (0.97, 10.74) 0.056 Most people 5.54 (2.53, 12.12) 0.000 Residence Urban reference Rural 0.06 (0.03, 0.13) 0.000 Interaction term Some people * rural 0.67 (0.19, 2.34) 0.529 Most people * rural 0.80 (0.35, 1.83) 0.593 Outcome: Receipt PNC within 2 days Portion of community that encourages utilization of PNC within 2 days of delivery No/few people reference Some people 0.83 (0.49, 1.40) 0.477 Most people 1.37 (0.89, 2.11) 0.150 Residence Urban reference Rural 0.10 (0.06, 0.15) 0.000 Interaction term Some people * rural 2.60 (1.39, 4.84) 0.003 Most people * rural 2.27 (1.34, 3.85) 0.002 Outcome: Receipt of comprehensive maternity care Portion of community that encourages comprehensive maternity care* Not fully supportive community reference Fully supportive community** 1.81 (1.33, 2.45) 0.000 Residence urban reference rural 0.15 (0.132, 2.45) 0.000 Interaction term Some people * rural reference Most people * rural 1.57 (1.00, 2.45) 0.049 . CC-BY 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.(which was not certified by peer review)preprint The copyright holder for thisthis version posted September 28, 2025. ; https://doi.org/10.1101/2025.09.25.25336698doi: medRxiv preprint 17 308 Discussion 309 To our knowledge, this is the first study to quantitatively examine the association between 310 women’s perception of community support for and utilization of maternity healthcare services 311 across the continuum of care in Ethiopia. We find that perceived support from the community 312 significantly influences the likelihood of receiving care across each component of the continuum 313 and across the full continuum, with the effect being stronger amongst rural women for completion 314 of PNC and the full continuum. 315 Our findings suggest that when women feel that at least some people in their community are 316 supportive of maternal health services, they are more likely to receive these services and that this 317 applies across the entire continuum of care. This is consistent with previous qualitative studies 318 indicating that communities have considerable influence on the receipt of services. For example, 319 a study in Addis Ababa highlighted that community members felt there was little utility to 320 receiving ANC services and discouraged women from accessing this care (21). Similarly, women 321 who received poor quality delivery services discouraged others from attending services, a finding 322 which was echoed in a similar study in Kenya (34). These studies, which focus on only one 323 component of the continuum, demonstrate the critical role that community support plays in 324 women’s utilization of specific maternity care services. No studies that we could find examined 325 the influence of community support across all components of the continuum of care, limiting our 326 ability to compare this finding; nonetheless, that each component was consistently associated with 327 higher uptake of services points to the importance of building community support for each service 328 in order to increase utilization of the entire continuum. We also found that, while women generally 329 considered that at least some people in their community encouraged care seeking, between one in 330 four and one in three, depending on the service, reported that either no or few people supported . CC-BY 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.(which was not certified by peer review)preprint The copyright holder for thisthis version posted September 28, 2025. ; https://doi.org/10.1101/2025.09.25.25336698doi: medRxiv preprint 18 331 them. This points to the continued need to address community opinions on maternal health services 332 and improve perceptions of the need for and quality of services. Understanding the specific beliefs 333 associated with each service and barriers that prevent engagement is crucial for improving 334 women's utilization of services. 335 Additionally, we found that across the entire continuum of services, urban women were 336 consistently more likely to utilize services than rural women, but that the effect of perceived 337 community support was significantly stronger among rural women for receipt of PNC services and 338 for completing the entire continuum. Disparities in use of maternal health services between urban 339 and rural women in Ethiopia have been well documented in other studies (35–37). There are a 340 number of reasons why these disparities exist, including access-related challenges (38,39) and 341 socioeconomic differences such as education and wealth (40,41), but few studies have examined 342 whether and how specific factors, and particularly community support, differ in their effect based 343 on residence. A study in Morocco demonstrated that restrictive gender norms impacted sexual and 344 reproductive healthcare-seeking behavior more strongly for rural than urban women (42), while in 345 Ethiopia, community-level gender norms were significantly related to experience of childhood 346 violence in rural, but not urban, areas (43). While there are no studies that we could find that 347 include a comparative perspective of the influence of community support on maternal health care- 348 seeking by residence, our results indicate that difference do exist and are important to consider, at 349 least for services that are less commonly used such as PNC. Perceptions of community support for 350 postnatal care may be particularly impactful in rural areas, where women traditionally have less 351 autonomy to make individual decisions due to less access to economic resources, and persistence 352 of conservative patriarchal norms, prioritizing husband and extended family decision-making (44). 353 Conversely, in urban areas, women may have greater access to individual financial resources and . CC-BY 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.(which was not certified by peer review)preprint The copyright holder for thisthis version posted September 28, 2025. ; https://doi.org/10.1101/2025.09.25.25336698doi: medRxiv preprint 19 354 traditional norms are frequently weaker (43,45), thus the effect of perceived community support 355 may have less influence on individual decisions around care. Understanding community 356 perceptions of the need for PNC care, particularly in rural areas, in order to design effective 357 community-based interventions may serve as a means to reduce disparities in urban and rural 358 utilization. 359 Our study has several strengths. First, this study used longitudinal data collected amongst a 360 representative sample of currently pregnant women who were asked to report on their perceptions 361 of community support prior to utilization of services (with the potential exception of ANC), which 362 reduces concerns about temporality. Additionally, we utilized a novel measure of perceived 363 community support, asking women to report on the community in general, rather than rely on 364 aggregate measures of individual data, which are generally used to measure community influences 365 (46–48). As these data are generally collected only among women age 15-49, they fail to capture 366 the opinions of other individuals in the community (46), who are likely impactful in determining 367 care-seeking behavior. Finally, we explore the modifying effect of residence on the effect of 368 perceived support, providing more evidence into how mechanisms that influence care seeking may 369 differ between urban and rural women and explain disparities in care. We do note limitations in 370 our data, including that our measure was non-specific in is wording. “Support” can entail any 371 number of actions or behaviors, but we were intentionally vague to allow the respondent to 372 interpret support as she saw fit. Additional research is needed to understand the components of 373 community support and its specific impact on women's utilization of maternal health services. 374 Additionally, while we collection information from women on their perceptions of community 375 support, we did not collect information from male partners and other community members who . CC-BY 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.(which was not certified by peer review)preprint The copyright holder for thisthis version posted September 28, 2025. ; https://doi.org/10.1101/2025.09.25.25336698doi: medRxiv preprint 20 376 influence decisions. Continued research to understand beliefs amongst a more diverse population 377 than women of reproductive age would help to identify barriers to care. 378 Conclusions 379 Perceived community support is an important predictor of women's utilization of maternal care in 380 Ethiopia. These findings shed light on an important mechanism to care-seeking, and how it differs 381 by urban and rural residence, contributing to ongoing disparities in care between urban and rural 382 women. We note that while support is generally high across the continuum, at least one in four 383 women report that few or no people support care-seeking, highlighting a lack of awareness of 384 services that are important to address. Additional research is needed to better understand the 385 reasons for this lack of support, particularly towards services in the postpartum period when the 386 majority of maternal and newborn mortality occur. 387 388 Declarations 389 Abbreviations 390 ANC: Antenatal Care 391 PNC: Postnatal Care 392 AAU/CHS: Addis Ababa University, College of Health Sciences 393 JHSPH: Johns Hopkins University Bloomberg School of Public Health 394 PMA: Performance Monitoring for Action 395 FMoH: Ethiopian Federal Ministry of Health 396 EAs: Enumeration Areas . CC-BY 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.(which was not certified by peer review)preprint The copyright holder for thisthis version posted September 28, 2025. ; https://doi.org/10.1101/2025.09.25.25336698doi: medRxiv preprint 21 397 Ethical declarations 398 Ethics approval and consent to participate 399 PMA Ethiopia received ethical approval from Addis Ababa University, College of Health Sciences 400 (AAU/CHS) (Ref: AAUMF 01-008) and the Johns Hopkins University Bloomberg School of 401 Public Health (JHSPH) Institutional Review Board (FWA00000287). Verbal informed consent 402 was obtained from all of the participants. The IRB approves verbal consent procedures (without a 403 need for written consent) for simple surveys without any invasive procedures in an environment 404 where literacy is low. Women under the age of 18 who are married are considered emancipated 405 minors and are able to provide informed consent. No unemancipated minors were included in this 406 survey. Detailed information about the ethical guidelines can be found in the National Research 407 Ethics Review Guidelines (http://www.ccghr.ca/wp-content/uploads/2013/). 408 Consent for publication 409 Not applicable. 410 Availability of data and materials 411 The datasets generated during the study are publicly available from the PMA website 412 (https://www.pmadata.org/data/available-datasets). 413 Competing interests 414 The authors declare that they have no competing interests. 415 Funding 416 This work was supported, in whole, by the Gates Foundation [INV 009466] 417 https://www.gatesfoundation.org/. Under the grant conditions of the Foundation, a Creative 418 Commons Attribution 4.0 Generic License has already been assigned to the Author Accepted . CC-BY 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.(which was not certified by peer review)preprint The copyright holder for thisthis version posted September 28, 2025. ; https://doi.org/10.1101/2025.09.25.25336698doi: medRxiv preprint 22 419 Manuscript version that might arise from this submission. LAZ received the award. Sponsors 420 played no role in the study design, data collection and analysis, decision to publish, or preparation 421 of the manuscript. 422 Author information 423 Authors and Affiliations 424 Department of Population Family and Reproductive Health, Johns Hopkins Bloomberg School of 425 Public Health, Baltimore, MD, USA 426 Yongyi Lu & Sally Safi & Linnea A. Zimmerman 427 Department of Reproductive Health and Health Service Management, School of Public Health, 428 Addis Ababa University, Addis Ababa, Ethiopia 429 Solomon Shiferaw 430 Contributions 431 YL, S. Safi, and LAZ conceived of the study design. YL analyzed the data and wrote the draft. 432 YL, S. Safi, and LAZ reviewed results and contributed to interpretation. LAZ, S. Safi, and 433 S.Shiferaw critically reviewed the manuscript. YL, S. Safi, S.Shiferaw, and LAZ revised the final 434 version of the manuscript. All authors contributed to the article and approved the submitted version. 435 Corresponding author 436 Correspondence to Linnea A. Zimmerman and Yongyi Lu. 437 Acknowledgments 438 The PMA Ethiopia project relies on the work of many individuals, both in the United States and 439 in survey countries. Thanks to the Ethiopia country team and resident enumerators who are 440 ultimately responsible for the success of PMA Ethiopia. 441 . CC-BY 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.(which was not certified by peer review)preprint The copyright holder for thisthis version posted September 28, 2025. ; https://doi.org/10.1101/2025.09.25.25336698doi: medRxiv preprint 23 442 443 References: 444 1. Onambele L, Ortega-Leon W, Guillen-Aguinaga S, Forjaz MJ, Yoseph A, Guillen-Aguinaga 445 L, et al. Maternal Mortality in Africa: Regional Trends (2000–2017). Int J Environ Res Public 446 Health. 2022 Oct 12;19(20):13146. 447 2. Ayele AA, Tefera YG, East L. Ethiopia’s commitment towards achieving sustainable 448 development goal on reduction of maternal mortality: There is a long way to go. Womens 449 Health. 2021 Dec 16;17:17455065211067073. 450 3. Kassebaum NJ, Bertozzi-Villa A, Coggeshall MS, Shackelford KA, Steiner C, Heuton KR, et 451 al. Global, regional, and national levels and causes of maternal mortality during 1990–2013: a 452 systematic analysis for the Global Burden of Disease Study 2013. The Lancet. 2014 Sept 453 13;384(9947):980–1004. 454 4. Oladapo OT, Nihlén Å. Maternal health in a dramatically different world: tailoring actions to 455 achieve targets for 2030 and beyond. Lancet Glob Health. 2024 Feb 1;12(2):e185–7. 456 5. General (OSG) O of the S. Strategies And Actions: Improving Maternal Health And Reducing 457 Maternal Mortality And Morbidity. In: The Surgeon General’s Call to Action to Improve 458 Maternal Health [Internet] [Internet]. US Department of Health and Human Services; 2020 459 [cited 2024 Feb 12]. Available from: https://www.ncbi.nlm.nih.gov/books/NBK568218/ 460 6. Berhan Y, Berhan A. Antenatal Care as a Means of Increasing Birth in the Health Facility and 461 Reducing Maternal Mortality: A Systematic Review. Ethiop J Health Sci. 2014 Sept;24(0 462 Suppl):93–104. 463 7. Tesfay N, Tariku R, Zenebe A, Woldeyohannes F. Critical factors associated with postpartum 464 maternal death in Ethiopia. PLoS ONE. 2022 June 24;17(6):e0270495. 465 8. Kuhnt J, Vollmer S. Antenatal care services and its implications for vital and health outcomes 466 of children: evidence from 193 surveys in 69 low-income and middle-income countries. BMJ 467 Open. 2017 Nov 15;7(11):e017122. 468 9. van Pelt S, van der Pijl M, A.C. Ruiter R, Ndaki PM, Kilimba R, Shields-Zeeman L, et al. 469 Pregnant women’s perceptions of antenatal care and utilisation of digital health tools in Magu 470 District, Tanzania: a qualitative study. Sex Reprod Health Matters. 31(1):2236782. 471 10. Tesfay N, Hailu G, Woldeyohannes F. Effect of optimal antenatal care on maternal and 472 perinatal health in Ethiopia. Front Pediatr [Internet]. 2023 [cited 2024 Feb 12];11. Available 473 from: https://www.frontiersin.org/articles/10.3389/fped.2023.1120979 474 11. Tesfay N, Kebede M, Asamene N, Tadesse M, Begna D, Woldeyohannes F. Factors 475 determining antenatal care utilization among mothers of deceased perinates in Ethiopia. Front 476 Med [Internet]. 2023 [cited 2024 Feb 12];10. Available from: 477 https://www.frontiersin.org/articles/10.3389/fmed.2023.1203758 . CC-BY 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.(which was not certified by peer review)preprint The copyright holder for thisthis version posted September 28, 2025. ; https://doi.org/10.1101/2025.09.25.25336698doi: medRxiv preprint 24 478 12. Ethiopian Public Health Institute (EPHI) [Ethiopia] and ICF. Ethiopia Mini Demographic 479 and Health Survey 2019: Final Report. 2021. Available from: 480 https://dhsprogram.com/pubs/pdf/FR363/FR363.pdf 481 13. Campbell OM, Graham WJ. Strategies for reducing maternal mortality: getting on with 482 what works. The Lancet. 2006 Oct 7;368(9543):1284–99. 483 14. Organization WH. Trends in maternal mortality 2000 to 2017: estimates by WHO, 484 UNICEF, UNFPA, World Bank Group and the United Nations Population Division: executive 485 summary. 2019 [cited 2024 July 31]; Available from: 486 https://iris.who.int/handle/10665/327596 487 15. Zewdu Amdie F, Landers T, Woo K. Institutional delivery in Ethiopia: Alternative 488 options for improvement. Int J Afr Nurs Sci. 2022 Jan 1;17:100436. 489 16. Organization WH. WHO recommendations on postnatal care of the mother and newborn 490 [Internet]. World Health Organization; 2014 [cited 2024 Aug 1]. Available from: 491 https://iris.who.int/handle/10665/97603 492 17. McCauley H, Lowe K, Furtado N, Mangiaterra V, van den Broek N. Essential 493 components of postnatal care – a systematic literature review and development of signal 494 functions to guide monitoring and evaluation. BMC Pregnancy Childbirth. 2022 May 495 28;22:448. 496 18. Kebede SA, Weldesenbet AB, Tusa BS. Determinants of Postnatal Care and Timing of 497 the First Postnatal Care for Newborns in Ethiopia: Further Analysis of 2019 Ethiopian 498 Demographic and Health Survey. Front Pediatr. 2022 Mar 24;10:809643. 499 19. Ethiopian Minstry of Health. Health Sector Transformation Plan: 2020/21-2024/25 500 [Internet]. 2015. Available from: 501 https://www.globalfinancingfacility.org/sites/default/files/Ethiopia-HSTP-II.pdf 502 20. Mose A, Haile K, Timerga A. Prevalence of completion of maternity continuum of care 503 and its associated factors in Ethiopia: a systematic review and meta-analysis. BMJ Open. 2022 504 Nov 1;12(11):e062461. 505 21. Tsegaye ZT, Abawollo HS, Desta BF, Mamo TT, Heyi AF, Mesele MG, et al. 506 Contributing barriers to loss to follow up from antenatal care services in villages around Addis 507 Ababa: a qualitative study. BMC Womens Health. 2021 Apr 7;21(1):140. 508 22. Beňová L, Semaan A, Portela A, Bonet M, van den Akker T, Pembe AB, et al. 509 Facilitators and barriers of implementation of routine postnatal care guidelines for women: A 510 systematic scoping review using critical interpretive synthesis. J Glob Health. 13:04176. 511 23. Weldegiorgis SK, Feyisa M. Why Women in Ethiopia Give Birth at Home? A Systematic 512 Review of Literature. Int J Womens Health. 2021 Nov 9;13:1065–79. . CC-BY 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.(which was not certified by peer review)preprint The copyright holder for thisthis version posted September 28, 2025. ; https://doi.org/10.1101/2025.09.25.25336698doi: medRxiv preprint 25 513 24. Community’s experience and perceptions of maternal health services across the 514 continuum of care in Ethiopia: A qualitative study | PLOS ONE [Internet]. [cited 2024 Sept 515 7]. Available from: https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0255404 516 25. Kebede K, Tadesse AH, Bekele B. A qualitative study of factors influencing the 517 utilization of institutional delivery: Insights from pastoral communities, Southwest Ethiopia. J 518 Midwifery Reprod Health. 2020 July 1;8(3):2284–95. 519 26. Hill Z, Amare Y, Scheelbeek P, Schellenberg J. ‘People have started to deliver in the 520 facility these days’: a qualitative exploration of factors affecting facility delivery in Ethiopia. 521 BMJ Open. 2019 June 1;9(6):e025516. 522 27. Teferi HM, San Sebastian M, Baroudi M. Factors associated with home delivery 523 preference among pregnant women in Ethiopia: a cross-sectional study. Glob Health Action. 524 2022 Dec 31;15(1):2080934. 525 28. Mamo A, Abera M, Abebe L, Bergen N, Asfaw S, Bulcha G, et al. Maternal social 526 support and health facility delivery in Southwest Ethiopia. Arch Public Health. 2022 May 527 11;80:135. 528 29. Factors affecting utilization of antenatal care in Ethiopia: A systematic review and meta- 529 analysis - PMC [Internet]. [cited 2024 Mar 20]. Available from: 530 https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6459485/ 531 30. The factors associated with antenatal care utilization in Ethiopia - PMC [Internet]. [cited 532 2024 Mar 20]. Available from: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9289506/ 533 31. Zimmerman L, Desta S, Yihdego M, Rogers A, Amogne A, Karp C, et al. Protocol for 534 PMA-Ethiopia: A new data source for cross-sectional and longitudinal data of reproductive, 535 maternal, and newborn health. Gates Open Res. 2020 Sept 9;4:126. 536 32. National Research Ethics Review Guideline. Fift Edition. 537 33. StataCorp 2023. Stata Statistical Software: Release 18. College Station, TX: StataCorp 538 LLC. 539 34. Odiase O, Akinyi B, Kinyua J, Afulani P. Community Perceptions of Person-Centered 540 Maternity Care in Migori County, Kenya. Front Glob Womens Health. 2021 Oct 8;2:668405. 541 35. Mekonnen ZA, Lerebo WT, Gebrehiwot TG, Abadura SA. Multilevel analysis of 542 individual and community level factors associated with institutional delivery in Ethiopia. 543 BMC Res Notes. 2015 Aug 26;8(1):376. 544 36. Bobo FT, Yesuf EA, Woldie M. Inequities in utilization of reproductive and maternal 545 health services in Ethiopia. Int J Equity Health. 2017 June 19;16(1):105. . CC-BY 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.(which was not certified by peer review)preprint The copyright holder for thisthis version posted September 28, 2025. ; https://doi.org/10.1101/2025.09.25.25336698doi: medRxiv preprint 26 546 37. Memirie ST, Verguet S, Norheim OF, Levin C, Johansson KA. Inequalities in utilization 547 of maternal and child health services in Ethiopia: the role of primary health care. BMC Health 548 Serv Res. 2016 Feb 12;16(1):51. 549 38. Hendrix N, Warkaye S, Tesfaye L, Woldekidan MA, Arja A, Sato R, et al. Estimated 550 travel time and staffing constraints to accessing the Ethiopian health care system: A two-step 551 floating catchment area analysis. J Glob Health. 13:04008. 552 39. Hierink F, Oladeji O, Robins A, Muñiz MF, Ayalew Y, Ray N. A geospatial analysis of 553 accessibility and availability to implement the primary healthcare roadmap in Ethiopia. 554 Commun Med. 2023 Oct 7;3(1):1–10. 555 40. Yesuf EA, Calderon-Margalit R. Disparities in the use of antenatal care service in 556 Ethiopia over a period of fifteen years. BMC Pregnancy Childbirth. 2013 June 15;13(1):131. 557 41. Gebre E, Worku A, Bukola F. Inequities in maternal health services utilization in 558 Ethiopia 2000–2016: magnitude, trends, and determinants. Reprod Health. 2018 July 559 4;15(1):119. 560 42. Ouahid H, Mansouri A, Sebbani M, Nouari N, Khachay FE, Cherkaoui M, et al. Gender 561 norms and access to sexual and reproductive health services among women in the Marrakech- 562 Safi region of Morocco: a qualitative study. BMC Pregnancy Childbirth. 2023 June 563 2;23(1):407. 564 43. Murphy M, Jones N, Yadete W, Baird S. Gender-norms, violence and adolescence: 565 Exploring how gender norms are associated with experiences of childhood violence among 566 young adolescents in Ethiopia. Glob Public Health. 2021 June 3;16(6):842–55. 567 44. Geleta D, Birhanu Z, Kaufman M, Temesgen B. Gender Norms and Family Planning 568 Decision-Making Among Married Men and Women, Rural Ethiopia: A Qualitative Study. Sci 569 J Public Health. 2015 Mar;3(2):242–50. 570 45. Williams TJ. Women’s Attitudes toward Gender Equality and Culture: The Influence of 571 Urbanisation and Education [Internet]. Stellenbosch : Stellenbosch University; 2021 [cited 572 2024 Sept 10]. Available from: http://hdl.handle.net/10019.1/110068 573 46. Stephenson R, Baschieri A, Clements S, Hennink M, Madise N. Contextual Influences on 574 the Use of Health Facilities for Childbirth in Africa. Am J Public Health. 2006 Jan;96(1):84– 575 93. 576 47. Gueye A, Speizer IS, Corroon M, Okigbo CC. Belief in Family Planning Myths at the 577 Individual And Community Levels and Modern Contraceptive Use in Urban Africa. Int 578 Perspect Sex Reprod Health. 2015 Dec;41(4):191–9. 579 48. Ononokpono DN, Odimegwu CO, Imasiku E, Adedini S. Contextual Determinants of 580 Maternal Health Care Service Utilization in Nigeria. Women Health. 2013 Oct 1;53(7):647– 581 68. . CC-BY 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.(which was not certified by peer review)preprint The copyright holder for thisthis version posted September 28, 2025. ; https://doi.org/10.1101/2025.09.25.25336698doi: medRxiv preprint 27 582 583 584 585 586 587 . CC-BY 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.(which was not certified by peer review)preprint The copyright holder for thisthis version posted September 28, 2025. ; https://doi.org/10.1101/2025.09.25.25336698doi: medRxiv preprint