Effective coverage of skilled birth attendance in Ethiopia: a population-based cross-sectional study Authors: Zewditu Denu, Kassahun Alemu, Della Berhanu, Atkure Defar, Theodros Getachew, Joanna Schellenberg, Tanya Marchant, Solomon Shiferaw, Amare Tariku, Tadesse Guadu, Girum Taye, Meseret Zelalem, Lars Ake Persson, Seblewengel Lemma Published: 2025 ABSTRACT AbstractBackgroundSkilled birth attendance is among the crucial maternal health interventions for maternal and neonatal mortality reduction. The availability of core components of services at healthcare facilities and the quality-of-service provision are essential for the effective utilization of skilled birth attendance. Effective coverage measures the availability of essential components at health facilities and the quality of care delivered. This study aimed to measure effective coverage of skilled birth attendance in Ethiopia.MethodsWe applied the cascade approach to measure effective coverage, focusing on four steps up to quality-adjusted coverage as a proxy for effective coverage of skilled birth attendance. To assess the structural quality of facilities, we used the PMA service delivery assessment data in 2019. Effective coverage was defined as the product of skilled birth attendance, facility readiness score, intervention coverage (proportion who received uterotonic injection), and a process quality score.ResultsThe crude coverage for skilled delivery attendance was 54% (95%CI; 52, 56). The average level of facility readiness to provide skilled delivery service was 72%, ranging from 17% for private and non-governmental facilities to 88% for hospitals. The uterotonic injection coverage was 42% and the mean process quality for skilled delivery service was 46%. Based on the effective coverage cascade, the input-adjusted coverage was 38%, and the intervention-adjusted coverage was 16%, and the quality-adjusted coverage was 7%.Low level of facility readiness and low quality of service explain the low effective coverage of skilled delivery attendance. Apart from the overall low level of readiness, there was variability between different levels of facilities. Though the skilled delivery attendance coverage is improving, the low level of readiness, intervention provision, and poor process quality resulted in low effective coverage. The minimum level of readiness needs to be ensured at all levels of facilities to provide skilled delivery services. The minimum level of service readiness and quality of care need to be ensured at all facilities to provide skilled birth attendance.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12884-025-08300-w.Keywords: Effective coverage, Skilled birth Attendance, Facility readiness, Input-adjusted, Intervention-adjusted, Quality-adjusted Background Despite low- and middle-income countries’ progress in the utilization of maternal health services such as skilled birth attendance, the rates of maternal and neonatal mortality remain high. This issue is particularly relevant in Sub-Saharan countries, emphasizing the need to prevent hemorrhage, sepsis, hypertension, and unsafe abortion. Addressing these problems can effectively reduce the number of maternal and neonatal deaths [1–4]. Skilled care during pregnancy, childbirth, and the immediate postpartum period is estimated to prevent 80% of pregnancy and childbirth-related deaths [5–9]. Such actions are highly recommended by the World Health Organization [8]. However, studies in low-income countries have shown that increased coverage of skilled birth attendance needs to be combined with good service quality to prevent maternal and neonatal mortality [10, 11]. Effective skilled birth attendance is achieved when skilled providers assist childbirth at a well-equipped health facility with quality essential interventions [12]. In many low-income countries, there is a gap between the crude and the effective coverage of maternal health services [13]. Effective coverage is a relatively new measure of service coverage, which takes into account service need, utilization, and quality of care delivered [14]. Effective coverage can be measured in different ways [15]. The Effective Coverage Think-tank group has proposed the cascade approach. It considers the target population in need of the service, the service utilization, the availability of essential inputs at health facilities to provide the required health service, the provision of essential health interventions, the quality of services delivered, the adherence of users to the recommended health intervention, and the desired health outcomes [16]. Effective coverage is a new approach of measuring health system performance that incorporates information on the population in need of the service, utilization of the service and quality of service delivered [17]. This metric not only accounts for the reach of interventions but also considers the potential health gains that the health system can deliver. By focusing on effective coverage, we can better assess and enhance the impact of health interventions on the populations they serve. Tanahashi (1978) first discussed the concept of EC, which gained momentum and developed further after 20 years. Tanahashi defined EC as the measurement of coverage based on the number of people who received satisfactory services, also termed as a measurement of service outputs [18]. The Tanahashi framework outlines five core components (or stages) of health service coverage, which help identify gaps and bottlenecks between service availability and actual health impact [19]. These components are availability Coverage, which measures whether the necessary resources (e.g., infrastructure, staff, and equipment) exist to deliver the service, accessibility Coverage assessing whether the population can physically and financially access the service, acceptability Coverage which Reflects whether the population is willing to use the service, influenced by cultural, social, or personal factors, Contact Coverage that measures the proportion of the target population that actually uses the service and effective Coverage assesses whether the service is delivered with sufficient quality to achieve the intended health benefit. Subsequent studies have employed various techniques to measure effective coverage. A common approach combined three elements: the target population, contact coverage, and the quality of healthcare interventions [17]. Other approaches further incorporated user adherence–adjusted and outcome-adjusted coverage [14]. In Ethiopia, facility delivery with skilled attendance has increased from 5.6% in 2000 to 49% in 2019 [20]. Despite an overall reduction in recent years, neonatal mortality has stagnated, suggesting impaired quality of skilled birth attendance [21]. Earlier studies in Ethiopia on skilled delivery service utilization only reported crude coverage [22–24]. Only one study assessed the quality of skilled birth attendance [25]. No studies have assessed effective coverage of skilled birth attendance in Ethiopia, considering facility readiness, intervention coverage, and service process quality. We aimed to analyze the effective coverage of skilled birth attendance in Ethiopia. Within a representative sample of women who recently delivered, we measured the proportion of mothers who delivered at a health facility where the necessary equipment, drugs, and infrastructure needed for delivery services were available (input-adjusted skilled attendance at delivery) and received the essential interventions (intervention-adjusted skilled attendance at delivery) according to recommended process quality (quality-adjusted skilled attendance at delivery). Methods Study setting This paper reports a secondary analysis of data from the Performance Monitoring for Action Ethiopia (PMA) project conducted from 2019 to 2020 [25] in five regions and one city administration, namely Tigray, Afar, Amhara, Oromia, Southern Nations, Nationalities, and Peoples regions, and Addis Ababa, which represent 90% of the Ethiopian population. In Afar, only rural and in Addis Ababa, only urban strata were included, while the rest had both. The Ethiopian health system encourages uncomplicated delivery at health centers and referring complicated cases to hospitals. Government health facilities offer free delivery services [26]. Study design and data This is a secondary analysis of the 2019–2020 Performance Monitoring for Action (PMA) Ethiopia women’s data collected at around six weeks after delivery. The cross-sectional service delivery data were collected in 2019. The PMA study was a collaborative effort between the School of Public Health at Addis Ababa University and the Johns Hopkins Bloomberg School of Public Health. The study involved interviewing all women aged 15–49 years in selected households in the study regions to gather information on pregnancy, childbirth, place of delivery, interventions received, and the quality of services provided. In addition to interviewing women, the study also assessed the readiness of 462 health facilities expected to provide delivery services in the women's catchment area, such as public hospitals, health centers, non-governmental such as faith based institutions and private facilities. This assessment included gathering data on the availability of services, required infrastructure, essential drugs, equipment, labor monitoring, neonatal resuscitation, and clean cord care services. Eligibility For this analysis, we included consenting postpartum women 15–49 years old who were regular members of the household. The analysis also included data on healthcare facilities located in the study areas where the eligible women were residing. Sample size and sampling procedure PMA Ethiopia panel survey was a two-stage cluster design study that used urban–rural and major regions as strata. A total of 217 enumeration areas, located within the regions described above, were selected from the master sampling frame at the Central Statistical Agency. All pregnant women and post-partum women (age 15–49) residing in households, located within the Panel survey regions, were asked to participate in the survey. For the original study, 2919 eligible women were enrolled to detect a 5% difference in maternal and child health indicators [24], and our analyses were based on data from the 2714 women who had complete information. Data collection The detailed data collection procedure and quality assurance methods were described elsewhere [26]. From the health facility inventory, we used data on 22 tracer items for delivery service readiness (Table 1). The selection of tracer items was based on national guideline and WHO recommendation and also availability of the items in the PMA data [27, 28]. Information on the availability of the tracer items to provide delivery services was collected by interviewing health care providers using a predefined checklist [26]. Quality of delivery service was assessed by interviewing the postpartum women at around six weeks [26]. Data included place of delivery, provision of a uterotonic injection and 13 tracer items for delivery process quality (Table 2). Information was also collected on wealth status based on household assets, geographic region, urban or rural residence, education, age, marital status, and family size. Table 1Items used to measure facility readiness to provide skilled delivery care, PMA Ethiopia 2019Availability of items based on observation and interviews with providersInfrastructure available1Electricity2Water source3Functional toilet4Delivery room5Emergency transportInfection prevention items available6Soap7Surgical gloves8Chlorhexidine9Chlorine solution10Chemical or electric sterilizerLabor monitoring items available11Blood pressure manometer12Thermometer13Urine dipstick14Fetoscope or fetal DopplerEssential drugs available15Oral or parenteral oxytocic drugs16Antibiotics17Diazepam18Magnesium sulphateNeonatal resuscitation19Neonatal Ambu bag20Neonatal maskCord care items available21Sterile cord cutter22Sterile cord tie Items used to measure facility readiness to provide skilled delivery care, PMA Ethiopia 2019 Table 2Items of quality of care based on interviews with the mother at around 6 weeks post delivery1Treated respectfully2Treated friendly3Allowed to ask questions4Explained the procedure during the examination5Allowed family or friends to accompany6Able to be in a preferred position7Involved in decision-making8Allowed to express feelings9Covered or screened during examination10Healthcare providers used their preferred name11Explained about medications given12Healthcare providers paid attention13Provider put the baby on the breast Items of quality of care based on interviews with the mother at around 6 weeks post delivery Outcome The outcome for this study was quality-adjusted skilled birth attendance as a proxy measure of effective skilled birth attendance. It was defined as delivery attended by a doctor, nurse, midwife, and health officer in a health facility where necessary inputs were available, and necessary intervention delivered according to the recommended process quality (Table 3). Table 3Definition of outcome measurementsOutcome measuresDefinitionSkilled birth attendance (contact) coverageThe proportion of women who gave birth to their recent baby at a health facilityThe input-adjusted skilled birth attendance coverageThe proportion of women who gave birth at a health facility with the necessary skilled professionals, infrastructure, equipment, drugs, supplies, and emergency transport, as defined in Box 1Intervention-adjusted skilled birth attendance coverageThe proportion of women who gave birth at a health facility where the necessary inputs were available and received a uterotonic injection immediately after their delivery as reported by the womenQuality-adjusted skilled delivery attendance coverageThe proportion of women who delivered at a health facility where the necessary inputs were available, received uterotonic injection after delivery and were attended according to the recommended process quality*Uterotonic injection immediately after delivery was considered as a compulsory intervention Definition of outcome measurements *Uterotonic injection immediately after delivery was considered as a compulsory intervention Calculating readiness and process quality score The average readiness and process quality scores received by women at these facilities are combined throughout the study area. The availability of tracer items was measured with yes or no questions. To calculate the readiness index, we aggregated all responses with a "yes" value for each item within the facilities included in the study using STATA. Subsequently, we determined the average level of readiness by dividing the readiness index by the total number of items. Similarly, we calculated the process quality index and average level of quality using the same methodology [29]. Intervention coverage We calculated the intervention coverage as the proportion of women who received uterotonic injection (compulsory intervention) immediately after delivery. We considered the administration of uterotonic injections as a compulsory intervention, as hemorrhage stands as the primary cause of maternal mortality. The World Health Organization (WHO) recommends the use of uterotonic to prevent postpartum hemorrhage [30]. Calculating the effective coverage cascade Effective coverage was calculated using cascade approach. In order to calculate the estimates for the cascade, each additional measure outlined in the box below was multiplied by the preceding measure [15]. Quality adjusted skilled birth attendance coverage was used as a proxy for effective coverage of skilled birth attendance. Analyses The analysis was weighted by enumeration area and household to ensure the sample’s representativeness. To account for the clustered nature of PMA data, we used the svy command in STATA. Skilled birth attendance was stratified by region (Tigray, Afar, Amhara, Oromia, the Southern Nations, Nationalities, and Peoples’ region, and Addis Ababa). The mean availability of the tracer items that should be available to provide delivery services at the health facilities was calculated as the facility readiness score. This calculation was done for the overall availability and each domain (infrastructure, labor monitoring, infection prevention, essential drugs, neonatal resuscitation, and clean cord care). Input-adjusted skilled birth attendance coverage was calculated by multiplying the proportion of women who delivered at health facilities by the facility readiness score. Intervention-adjusted coverage was computed by multiplying the proportion of women who delivered at health facilities by the proportion who received uterotonic injections immediately after delivery and then multiplying the product by the average facility readiness score. The quality-adjusted skilled birth attendance coverage was computed by multiplying the proportion of those who delivered at a health facility by the average readiness score, by the proportion of those who received uterotonic injection, and by the average level of process quality score. Results Participation Out of 2691 eligible women, information on the place of delivery was retrieved from 2605 participants, i.e., a response rate of 97% (Supplemental Fig. 1). We did a complete case analysis. One-fifth of the postpartum women lived in urban areas. Approximately three-quarters were in the age range of 19–34 years with a mean age of 27 years. Nearly four in ten (42%) did not attend formal education (Table 4). Table 4Background characteristics of women who gave birth at around six weeks before interview, Performance Monitoring for Action Ethiopia 2019–2020Characteristics N = 2605Unweighted n (%)Weighted % (95% CI)Region* Tigray441 (s17)7 (6,7) Afar216 (8)2 (2, 3) Amhara452 (17)20 (20,21) Oromia644 (25)44 (43,44) Southern Nations, Nationalities, and Peoples603 (23)22 (22,23) Addis Ababa249 (10)4 (5, 5)Residence Urban1019 (38)21 (21, 22) Rural1672 (62)79 (78, 79)Age 15–24 years893 (33)33 (32,34) 25–34 years1383 (51)1187 (1169,1192) 35–48 years38 (16)416 412,419)Educational level* No formal education994(38)42(40,44) Primary education944(36)40(38,42) Secondary and above667(26)18(16,19)Household wealth quintiles* 1 Lowest466 (18)21 (19,23) 2396 (15)20 (18,22) 3409 (16)20 (18,22) 4482 (19)20 (18,22) 5 Highest852 (32)19 (18,20)*Data missing on 86 women Background characteristics of women who gave birth at around six weeks before interview, Performance Monitoring for Action Ethiopia 2019–2020 *Data missing on 86 women Skilled birth attendance coverage Of the 2605 women who recently delivered, 54% (95% CI: 52, 56) gave birth in a health facility, ranging from 18% in Afar to 95% in Addis Ababa (Supplemental Fig. 2). In this study 44% of rural women utilized skilled birth attendance compared to 92% of urban women. Among the health facilities where women delivered, 35% were government hospitals, 52% were government health centers and 13% were non-governmental and private health facilities. Input-adjusted skilled birth attendance We assessed the facility readiness score to provide skilled delivery care based on 22 tracer items under six domains (Table 1). Four percent of the health facilities had all items available (Fig. 1). The average level of health facility readiness score was 71%. The average level of readiness for hospitals was 88%, for health centers 44%, and for non-governmental and private health facilities 4% (Supplemental Table 2). Multiplying the crude skilled birth attendance of 54% by the mean facility readiness score of 71%, the input-adjusted skilled birth attendance coverage was 38%. Availability of tracer items necessary to provide skilled delivery care in Ethiopia, Performance Monitoring for Action Ethiopia, 2019–2020 Intervention-adjusted skilled birth attendance Forty-two percent of the participating women reported that they received a uterotonic injection immediately after delivery. When stratified by facility type, it was found that 77% of women who delivered in hospitals received uterotonic injections immediately after delivery. In comparison, only 44% of women who delivered at health centers received uterotonic injections post-delivery. Of women delivering in non-governmental and private health facilities, 63% reported receiving uterotonic injections immediately after delivery at these facilities. When multiplying the contact coverage (54%) by the average facility readiness level (71%) and by the intervention coverage (42%), the intervention-adjusted skilled birth attendance was 16%. Quality-adjusted skilled birth attendance We computed the process quality score using 13 tracer items (Fig. 2) and the mean process quality score was 46%. Proportion of women who received quality skilled birth attendance with process quality components in Ethiopia, Performance Monitoring for Action Ethiopia 2019–2020 Quality-adjusted skilled birth attendance was used as a proxy measure to determine effective skilled birth attendance coverage. When multiplying the contact coverage (54%) by the average facility readiness level (71%), the intervention coverage (42%), and the mean process quality score (46%), the effective skilled birth attendance coverage was 7% (95%CI: 5.8) (Fig. 3). Figure 3 illustrates women’s utilization of maternal health services, with each bar representing the corresponding percentage. The cascade of effective skilled birth attendance coverage in Ethiopia, proportions with 95% confidence interval, PMA 2019–2020 Discussion This study revealed a significant gap between the proportion of Ethiopian women giving birth with a skilled attendant present and the effective coverage of skilled birth attendance at health facilities being ready for necessary input, offering the needed intervention, and having the recommended quality procedures. While slightly more than half received skilled birth attendance, the effective coverage dropped to approximately one in ten women when considering factors such as facility readiness, availability of uterotonic injection, and process quality score. Our study has revealed that in the 2019–2020 period, 54% of Ethiopian women gave birth at health facilities. The finding showed a slight increase in skilled birth attendance compared to the 2019 mini-Demographic and Health Survey results for 2015–2019, which reported a coverage of 48% [21]. However, compared to the average skilled birth attendance coverage in 39 sub-Saharan countries, which stands at 75%, Ethiopia’s proportion remains low [31]. Our study revealed regional disparities in skilled birth attendance coverage. Only a tiny percentage of women in rural Afar received skilled birth attendance compared to nearly all women in Addis Ababa. In many low-income countries, rural women are less likely to have access to skilled birth attendance [32]. The average level of facility readiness to provide skilled birth attendance was 71%, which is nearly identical to the service availability readiness assessment conducted in 2014, where the readiness level was 68% [29]. These figures indicate that there has been no improvement in facility readiness over the past few years. There was a considerable drop in percentage between the contact and input adjusted (54% versus 38%) coverage. Such disparity highlights the need for improvements in facility readiness to ensure that all women have access to quality skilled birth attendance. This finding is consistent with previous research in Uganda in 2018, which also identified a substantial drop in percentage between the crude coverage of skilled birth attendance (55%) and input-adjusted skilled birth attendance coverage (10%) [33]. In the current study, hospitals had the highest facility readiness score (88%), while health centers managing more deliveries had a much lower score (44%). These results underscore the need to enhance facilities with essential infrastructure, infection prevention protocols, monitoring tools, vital medications, and newborn care supplies. If delivery facilities are adequately equipped, the probability is enhanced that mothers giving birth in these settings receive the comprehensive care they require. Our study showed that 42% of women received a uterotonic injection, which is an essential intervention to control postpartum hemorrhage. This percentage was significantly lower than in a previous Ethiopian study, where 87% of women were administered uterotonic injections as part of actively managing the third stage of labor [34]. However, the mentioned study included only hospital deliveries, excluding lower-level facilities. The lack of complete coverage of uterotonic injections implies that many women may not receive the necessary intervention to prevent and manage postpartum hemorrhage, a leading cause of maternal mortality [35]. Additionally, this result is clinically important since anemia affects one in four pregnant women, and additional bleeding can pose a problem [36, 37]. A considerable drop along the effective coverage cascade was seen between the target women and the crude coverage, showing nearly half of women did not visit health facilities for delivery service. The second significant drop was observed between input-adjusted coverage and intervention-adjusted coverage, implying that among those who delivered at health facilities, a considerable number did not receive the essential medicine to control bleeding, with an implication of potential exposure to risk of bleeding. Lastly, the perceived poor quality of providers in providing respectful maternity care further reduced the effective coverage of skilled birth attendance. Thus, there is a need for targeted interventions and investments in the healthcare system to address this low coverage of skilled birth attendance both from the demand and supply sides. Strengths and limitations This paper reports the first Ethiopian study that analyzed skilled birth attendance considering need, utilization, and quality of care, using most components of the recommended effective coverage cascade, except for the adherence-adjusted and the outcome-adjusted coverage [16]. Our study sample represented 90% of the country's population, and findings can be generalized to Ethiopia. We included governmental, non-governmental, and private health facilities in the women’s residence area in the facility inventory. The recall period of these self-reported data was short, minimizing recall bias. However, women might fail to recall the timing and specific interventions received, which could under- or overestimate the effective coverage. The availability of uterotonic injections was reported by 79% of health facilities, indicating that more may have received the drug than reported by the participating women. Such an error could be reduced if data were collected from the women as well as from a chart review [38]. Our analysis did not account for the mode of delivery, such as assisted or operative deliveries, when calculating facility readiness to provide skilled birth attendance. As a result, the readiness estimates may not fully capture the resources and capacity required for different types of deliveries. In addition, we did not consider the variations in staff distribution between rural and urban health facilities, which could influence the availability of skilled personnel and, consequently, the effective coverage of skilled birth attendance. These limitations may have introduced some bias in our estimates and should be taken into account when interpreting the findings. We linked women's data with the ecologically linked service delivery data. Such linking has been shown to give acceptable estimates [39] and assumes that women visited health facilities in their catchment areas, while some might bypass the facilities in their locality. Conclusion This study showed that Ethiopia’s effective coverage of skilled birth attendance was low. This low level was mainly due to low service utilization and poor quality of care. The health centerswith the highest number of births were mostly not ready to provide these services. Enhancing the quality of delivery services is essential to attract more women to utilize these services, ultimately leading to improved pregnancy outcomes. To provide effective skilled birth services, all facility levels must ensure minimum readiness and availability of the recommended interventions. Continuous capacity-building initiatives are required to enhance the skills necessary for providing high-quality services. Supplementary Information Supplementary Material 1. S Figure 1. Study flow diagram, Performance Monitoring for Action Ethiopia, 2019-2021 Supplementary Material 2. S Figure 2: Distribution of facility delivery proportions by region, Performance Monitoring for Action Ethiopia, 2019-2020. SNNP= Southern Nations, Nationalities, and Peoples region Supplementary Material 3. S Table 1: Readiness level to provide skilled birth attendance by type of health facility, Performance Monitoring for Action Ethiopia, 2019-2021 Acknowledgements The authors thank the PMA data managers for assisting in data access and management. Author’s contributors ZA, KA, LP, and SL designed, conceptualized, and drafted the manuscript. KA, LP, SL, DB, AD, TM, JS, SS, TG, AT, TG. MZ and GT contributed to the editing and revision of the manuscript. Funding The analysis of the PMA data was funded by grants from the Bill and Melinda Gates Foundation to the London School of Hygiene & Tropical Medicine (INV-009691 and INV-010320). The funder did not play a role in data collection, analysis, or interpretation of results. Data availability The data presented in this manuscript were collected by Addis Ababa University and Johns Hopkins Bloomberg School of Public Health to gather information on key indicators of reproductive, maternal, and newborn health in Ethiopia. This dataset has been made publicly available and can be accessed at [https://www.pmadata.org/countries/ethiopia] (https://www.pmadata.org/countries/ethiopia). Declarations Ethics approval and consent to participate This paper presents findings from a secondary analysis of data collected as part of the PMA Ethiopia project. The original research protocol was reviewed and approved by the Ethical Review Board at Addis Ababa University, College of Health Sciences (Ref: AAUMF 01–008), as well as the Johns Hopkins University Bloomberg School of Public Health (FWA00000287). Informed consent was obtained from all women to participate in the study. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Supplementary Materials Supplementary Material 1. S Figure 1. Study flow diagram, Performance Monitoring for Action Ethiopia, 2019-2021 Supplementary Material 2. S Figure 2: Distribution of facility delivery proportions by region, Performance Monitoring for Action Ethiopia, 2019-2020. SNNP= Southern Nations, Nationalities, and Peoples region Supplementary Material 3. S Table 1: Readiness level to provide skilled birth attendance by type of health facility, Performance Monitoring for Action Ethiopia, 2019-2021 Data Availability Statement The data presented in this manuscript were collected by Addis Ababa University and Johns Hopkins Bloomberg School of Public Health to gather information on key indicators of reproductive, maternal, and newborn health in Ethiopia. This dataset has been made publicly available and can be accessed at [https://www.pmadata.org/countries/ethiopia] (https://www.pmadata.org/countries/ethiopia).