https://orcid.org/0000-0002-5719-4294
Figures
Abstract
Background
Quality Antenatal Care (ANC) is considered if pregnant women have access to essential services that align with the best evidence-based practice. Although several studies have been conducted on ANC uptake in Ethiopia, they have focused on the timing and number of visits and the level of complete uptake of care contents according to the WHO recommendation remains scarce. Hence, this study aimed to assess the magnitude of missing care content during ANC visits, its spatial variations, and individual- and community-level determinants in Ethiopia.
Methods
The study was conducted using the 2016 Ethiopian Demographic and Health Survey and included a total weighted sample of 4,771 women who gave birth within five years before the survey. Spatial analysis was carried out using Arc-GIS version 10.7 and SaTScan version 9.6 statistical software. Spatial autocorrelation (Moran’s I) was checked to determine the non-randomness of the spatial variation in the missing contents of care. Multilevel multivariable logistic regression analysis was performed using STATA version 16. The adjusted odds ratio (aOR) with its corresponding 95% CI was used as a measure of association.
Results
The prevalence of missing full contents of ANC in Ethiopia was 88.2% (95% CI: 87.2, 89.0), with significant spatial variations observed across regions. Missing essential contents of care was higher among women who live in rural areas (aOR = 1.68, 95% CI: 1.47, 2.71), not completed formal education (aOR = 1.94, 95% CI:1.24, 3.02), late initiation of ANC (aOR = 3.05, 95% CI:1.59, 6.54), attended only one ANC (aOR = 4.13, 95% CI: 1.95, 8.74), and not having a mobile phone (aOR = 1.44, 95% CI: 1.07, 1.95).
Conclusion
The level of missing care content during prenatal visits was high in Ethiopia, with significant spatial variation across regions. Health systems and policymakers should promote early initiation and encourage multiple visits to provide optimal care to pregnant women. In addition, it is vital to focus on enhancing education and healthcare infrastructure in rural parts of the country.
Citation: Hailegebireal AH, Bizuayehu HM, Asgedom YS, Feyisa JW (2024) Spatial patterns and predictors of missing key contents of care during prenatal visits in Ethiopia: Spatial and multilevel analyses. PLoS ONE 19(11): e0313893. https://doi.org/10.1371/journal.pone.0313893
Editor: Laura Hannah Kelly, Public Library of Science, UNITED KINGDOM OF GREAT BRITAIN AND NORTHERN IRELAND
Received: November 8, 2023; Accepted: November 1, 2024; Published: November 18, 2024
Copyright: © 2024 Hailegebireal et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: The data for this study was obtained from the DHS office upon a reasonable request, with the authorization to use it. The authors were granted the right to use the data solely for study purposes, with the condition that the data not be shared with any other third parties. Thus, the one who needs the data supporting the findings of this study can get it in anonymised form from the DHS website at https://dhsprogram.com/data/dataset/Ethiopia_Standard-DHS_2016.cfm?flag=1 upon reasonable request.
Funding: The author(s) received no specific funding for this work.
Competing interests: The authors have declared that no competing interests exist.
Abbreviations: ANC, Antenatal Care; aOR, Adjusted Odds Ratio; CSA, Central Statistical Agency; EDHS, Ethiopian Demographic and Health Survey; FANC, Focused Antenatal Care; WHO, World Health Organization
Introduction
In 2020, an estimated 287,000 women died from maternal causes globally, equivalent to 800 maternal deaths per day, or one every two minutes [1]. Sub-Saharan Africa has the highest Maternal Mortality Ratio(MMR) (545 per 100,000 live births), which is more than two times the global estimate (223 per 100,000 live births) [1]. The region alone accounted for almost 70% of the global maternal mortality in 2020 [1, 2]. According to a World Bank modelled prediction in 2020, Ethiopia’s MMR was 267 per 100,000 live births, which was higher than the global average of 223 deaths per 100,000 live births [3].
The primary causes of maternal death in Ethiopia, are obstructed labor, uterine rupture, hypertensive disorders of pregnancy, infection, and hemorrhage [4, 5]. Such causes can be prevented by adequate utilization of prenatal, intrapartum, and postpartum care [6]. For instance, high-quality antenatal care (ANC), defined as getting an adequate amount of visits and items of care on time, is one of the cost-effective approaches to improving MMR [7, 8].
ANC primarily aids women and newborns in maintaining normal pregnancies by identifying pre-existing problems and preventing complications during pregnancy and childbirth [9]. The extent to which pregnant women have access to essential services that fit with evidence-based professional knowledge as part of Antenatal Care (ANC) is an indicator of the quality of care during pregnancy [9]. To be effective, important ANC components must be offered to the ANC plan [10].
The contents of care are focused on the prevention of pregnancy complications (counseling on danger signs), prevention of iron deficiency anemia(iron supplementation), prevention of neonatal tetanus (via at least two doses of Tetanus Toxoid injections), diagnosis of hypertension (blood pressure measurement), and diagnosis of asymptomatic infections (urine testing, blood tests, and HIV test) [11]. The services provided to pregnant women with ANC can reduce the chances of poor pregnancy outcomes such as stillbirths, premature deliveries, and low birth weight [12, 13]. When compared to any single element/component of ANC, compliance with recommended contents of care provided higher protection against preterm birth and low birth weight [14, 15]. Poor quality and inadequate utilization of ANC services throughout pregnancy, on the other hand, may raise the risk of preventable detrimental pregnancy outcomes [14].
Ethiopia implemented the WHO organization (WHO) focused ANC (FANC) model in all health institutions until February 2022, which was a goal-oriented approach to delivering evidence-based interventions at four crucial points throughout pregnancy [16]. However, in 2019, only 62.9% and 43% of pregnant women had at least one and four ANC contacts, respectively [17]. However, in Ethiopia, relatively little attention has been paid to the content of the care provided.
Although numerous studies have been conducted on the uptake of ANC in Ethiopia, practically all have focused on the timing and number of visits [18–20], and the level of adequacy of services received during visits with their spatial variability has not been thoroughly assessed. Hence, the purpose of this study was to uncover the magnitude, geographic variations, and individual- and community-level determinants of the missing vital elements of ANC in Ethiopia. By identifying specific geographic places with a high magnitude of missing contents of care, resources, and interventions could be prioritized and allocated for those locations needing the most. Furthermore, the identified individual- and community-level barriers to compliance with recommended items of care would be used to improve efforts by various stakeholders such as program planners, policymakers, and healthcare providers to meet Sustainable Development Goal 3(SDG3) [21]. Working on safer pregnancies and timely access to healthcare services could significantly improve maternal and newborn health outcomes.
Methods
Data source and population
This study was conducted using the 2016 Ethiopian Demographic and Health Survey (EDHS 2016), which was conducted from January 18, 2016, to June 27, 2016 [22]. Study populations were all women aged 15–49 years who gave birth within the five years before the survey. Women who had complete information on key WHO-recommended components of care during their last ANC visits were included in the study unless excluded.
Sampling procedure and sample size
A stratified two-stage cluster sampling approach was used to select study participants. The first stage involved selecting 645 enumeration areas (EAs) with a probability proportional to EA size. The second stage involved selecting a fixed number of 28 households per cluster using a systematic equal probability selection of eligible women. Finally, the analysis was based on the records of 4712(weighted sample size of 4771) (Fig 1). Detailed methodologies are addressed in the 2016 EDHS report [22]. The position data of each surveyed cluster (geographic coordinates) were obtained using Global Positioning System (GPS) data. The GPS readings for each cluster were obtained from the center. To preserve respondents’ privacy, all survey groups’ GPS latitude/longitude positions were randomly displaced. The maximum displacement was two kilometres (km) for urban clusters and five kilometres (km) for 99% of rural clusters. The remaining 1% of rural clusters shifted up to a ten-kilometer distance [23].
Variables of the study
Dependent variable.
Seven nationally recommended contents of care during ANC in Ethiopia were assessed to define non-compliance with or missing WHO-recommended components of care. The items included receipt of counselling on danger signs(pregnancy complications), iron supplements, at least two doses of Tetanus Toxoid injections, blood pressure measurement, urine testing, blood tests, and HIV tests as part of ANC [9, 20, 22, 24]. The answers were coded as (1 = Yes, and 0 = No). Based on the responses, a composite index was created with minimum and maximum values of 0 and 7, respectively. Finally, this count variable was dichotomized as missed care (= 1) and full care (= 0).
Explanatory variables.
After reviewing relevant and current literature, both individual- and community-level factors were considered. Individual-level factors include sociodemographic, obstetric, and healthcare-related characteristics. Community-level factors, were features shared by all women living in the same community (cluster), such as place of residence and region (Table 1).
Data management and statistical analysis
STATA version 16.1 was used for data cleaning, coding, and analysis, whereas ArcGIS 10.7 and SaTScan were employed for geographical analysis. All data were weighted using a weighting factor to minimize under or over-representation. The weighted proportion of the outcome variable computed in STATA was then generated in Microsoft Excel 2016 (CSV format) and imported into ArcGIS 10.7 for spatial analysis.
Spatial analysis.
The CSV file containing the outcome variable was first imported into ArcGIS version 10.7 and merged with the GPS data (shape file) to obtain latitude and longitude data. The event data were then transformed into a shape file and displayed on an XY plane. Before initiating the study, geographically coordinated data were projected using the UTS system.
Spatial autocorrelation (Global Moran’s I).
Spatial autocorrelation (Moran’s index) was used to determine whether the missing contents of care in Ethiopia were dispersed, clustered, or randomly distributed. This analysis produced a single output value, which ranged from −1 to 1. A Moran’s I value close to 1 suggests a significant positive spatial autocorrelation (the outcome of interest is clustered), whereas a value close to -1 indicates a large negative spatial autocorrelation (the outcome is dispersed) across clusters. A Moran’s I value near zero implies that the spatial distribution is random [25]. Furthermore, a statistically significant Moran I value (p<0.05) led to rejection of the null hypothesis (missing content care was randomly distributed) [25, 26].
Hot spot analysis (Getis-Ord Gi* statistics).
Getis-Ord Gi* statistics were run to detect significant hot spots and cold spots for missing key elements of the ANC. To determine the statistical significance of clustering, the Z-score was obtained and the p-value was determined. A statistical output with a high GI suggests a "hot spot," whereas one with a low GI indicates a "cold spot" [27, 28].
Spatial interpolation.
Using data from the sampled EAs, spatial interpolation was performed using the Ordinary Kriging technique to forecast the status of missing care contents in unsampled areas. The ordinary Kriging spatial interpolation technique was chosen because it has numerous advantages over other techniques. First, it considers the spatial dependence or autocorrelation in the data, which is vital to effectively predicting unobserved values based on values observed at nearby places [29]. It assumes that the statistical features of the data, such as the mean and variance, are constant across the research area, which is known as the assumption of stationarity, which simplifies the modelling process and often works well in practice [30]. In addition, it reduces prediction variance by providing more weight to nearby sample points with comparable values, which aids in producing more reliable estimates, particularly in areas where data is sparse [31]. Furthermore, because ordinary Kriging is based on geostatistical models that account for both spatial trend and spatial variability in data, it provides a more accurate depiction of the underlying spatial patterns in ANC content [30].
Spatial scan statistical analysis.
Sat Scan analysis was performed to identify substantial hot spots of missing ANC services using Kuldorff’s SaTScan version 9.6 statistical software. The Bernoulli-based approach was used to find statistically significant spatial clusters with a high number of women who missed the content of care [32]. The main reasons for using Bernoulli model is because it accommodates binary outcomes; the occurrence of women missing the content of antenatal care (coded as 1 = cases), and the non-occurrence would be women receiving the recommended antenatal care contents (coded as 0 = controls). In addition, the results of the Bernoulli-based spatial scan analysis tend to be flexible and easy to interpret, such as revealing clusters with high or lower risk factors, making it easier for policymakers and public health experts to understand and focus interventions [33, 34]. The SaTScan statistics gradually scanned the area to determine the number of observed and expected events inside the window at every point. The default maximum spatial cluster size of 50% of the population was set as an upper limit, allowing both small and large clusters to be recognized. The scanning window with the highest likelihood was selected as the most likely and high-performing cluster to represent a case, and the level of significance was set at p-value<0.05 [32, 35].
Multilevel mixed-effect logistic regression.
Given the hierarchical nature of the EDHS data, multilevel modelling was employed. Because the outcome variable was in binary form, we used multilevel multivariable regression analysis.
Model building and selection.
Fixed effects (measures of association). A multilevel bivariable logistic regression analysis was first carried out to examine the influence of each predictor on the response variable, and variables with a p-value< 0.25 were accounted for in the multilevel multivariable logistic regression. The Variance Inflation Factor (VIF) was used to look for multicollinearity across covariates at a cut-off point of 10 and none was found (the VIF ranged from 1.02 to 2.33 with a mean of 1.44). Finally, multilevel multivariable logistic regression analysis was performed to identify significant predictors of missing contents of care, and statistically significant variables (p <0.05) were reported with their corresponding 95% confidence intervals.
Random effects. Four different models were fitted by using multilevel methods. Model one (null model) had no explanatory variable. The second and third models contained only individual- and community-level factors, respectively. The fourth (complete) model incorporated all individual- and community-level characteristics. The intraclass correlation coefficient (ICC) and proportional change in variance (PCV) were used to quantify the random effects in each model (variability in the level of missing care contents between and across clusters).
where Var(b) is the variance at the group level and Var(w) is a predicted individual variance component, which is π2/3 ≈3.29.
Proportional Change in Variance (PCV) was estimated as
where Va is the variance of the initial model (null model) and Vb = variance of the subsequent models (models 2, 3, 4, and 5).
Goodness of fit. Deviance = (-2 * (Log Likelihood (LL) of each model), Schwarz’s Bayesian Information Criterion (BIC), and Akaike’s information criterion (AIC) were used to assess goodness of fit. After comparison, the fourth model with the lowest deviation, AIC, and BIC values was selected as the best-fit model for this study (Table 6).
Ethical consideration and consent to participate
All methods and procedures were performed according to the relevant guidelines and regulations of the Declaration of Helsinki. The DHS office provided us with written permission to use both the DHS and GPS datasets following registration. Furthermore, as this is secondary data, the ethics committee of Wachemo University College of Medicine and Health Sciences declared that no formal ethics approval was required with a written letter at the reference number (WCU/327/2023).
Results
Sociodemographic characteristics of the respondents
The current study examined a total weighted sample of 4771 women, with the mean (±SD) age 28.79(±6.61) years. Most participants (90.4%) were from the central region (Oromia, Amhara, and Tigrai). Rural residents accounted for 81.7% of the study participants. Orthodox and Muslim religious followers accounted for 42.5% and 33.0% of respondents, respectively. The vast majority (88.2%) of women missed at least one key element of care during their prenatal visit (Table 2). Nine in every ten women from the poorest wealth quintile (93.9%) and without formal education (91.9%) did not receive full content of care. The highest mean number of contents of care was observed among women who lived in urban areas (5.46) or with the richest wealth quintile (5.71) (Table 2).
Obstetric and health service-related- characteristics
Nearly half (47.0%) of the participants were multiparous. In terms of time and frequency of ANC visits, nearly one-third (32.5%) of the women had their first visit during the first trimester, and half (50.6%) had four or more visits. Only 5.0% of the women were enrolled in any health insurance scheme. Nearly three-quarters, 74.0% of the women were autonomous in decision-making. Women who commenced ANC in the third trimester and received only one ANC visit failed to receive full content of care in 97.7% and 97.0% of cases, respectively (Table 3).
The level of missing key contents of ANC
According to the analysis, 88.2% of the women failed to receive full care during their last prenatal visit. The Somali region had the largest proportion of women who missed full contents of care during prenatal visits (94.9%), closely followed by the Oromia region (93.7%). The top three types of care that the majority of women did not receive as part of ANC were HIV tests (59.0%), information about pregnancy complications (55.0%), and receipt of at least two doses of TT vaccination (43.1%) (Table 4).
Results of spatial analysis
Spatial distribution of missing ANC contents.
Western Oromia, northeast of SNNPR, western border of Afar, and southern border of Benishangul Gumuz were identified as regions with a high proportion of women who missed ANC content. In contrast, Tigray, Addis Ababa, and Dire Dawa had lower proportions (Fig 2).
The global spatial autocorrelation analysis found that the geographical distribution of missing contents of care across the country was non-random (i.e., there was significant spatial variation) (global Moran’s I = 0.39, p<0.001, z-score of 12.96). The clustered patterns (on the right side) indicate that missing WHO-recommended items of care occurred frequently throughout the country (Fig 3).
Hot spot (Getis-Ord Gi*) analysis of missing ANC contents.
In the hot spot analysis, significant spatial clusters (Hot Spots) for incomplete contents of care were detected in the eastern and northern areas of the SNNPR, as well as in the western border of the Oromia regions. Eastern Tigray, the central part of Addis Ababa, Dire Dawa, and Harari, on the other hand, had a small proportion of women with incomplete contents of care (Cold Spots) (Fig 4).
Spatial interpolation.
The ordinary Kriging technique was used to estimate the spatial distribution of missing full contents of care for areas where data were not collected. Accordingly, the southern portions of Oromia, the southern and eastern parts of SNNPR, the western parts of Addis Ababa, and the northern part of Afar had the highest expected prevalence of missing the recommended contents of care (red shaded). In contrast, the projected proportion of women with incomplete care items (green shaded) was low in Tigray and Central Addis Ababa (Fig 5).
Spatial scan statistical (SaTScan) analysis.
The SaTScan spatial analysis showed two statistically significant groups of SaTScan clusters, with a high number of women who missed key contents of care. In addition, the analysis revealed 156 significant clusters that accounted for missing contents of care, with 133 and 23 found in the first and second most likely clusters, respectively. The first most likely cluster is located at geographical coordinates of (5.546952 N, 37.666334 E) with a 379.77 Km radius, and LLR of 85.76 at p<0.001. Women living in these areas had a 45% (RR = 1.45) higher risk of missing ANC content than those living elsewhere (Table 5).
Results of multilevel mixed effect logistic regression
Random effect (measures of variation).
For the measures of variation (random effects), the intraclass correlation coefficient (ICC) and Proportional Change in Variance (PCV) were computed. The results of the null model (Model I) revealed that variability between clusters accounted for 17.1% of the total variation in missing ANC content (ICC = 0.171, p<0.001). Furthermore, individual and community variability accounted for 8.1% (ICC = 0.081, p<0.001) and 11.6% (ICC = 0.116, p<0.001) of the variation in the missing care content, respectively. Individual- and community-level factors together accounted for 58.5% of the national variation observed in the null model (PCV = 58.5%). As we moved from Model I (null model) to Model IV (full model), the values of AIC, BIC, and Deviance decreased, indicating that the final model was best fitted (Table 6).
Predictors of missing contents of ANC.
The factors that have a significant association with missing ANC contents were educational status, frequency and timing of ANC, religion, ownership of mobile phone, and residence (Table 7).
Women without formal education were two times more likely to miss the contents of ANC than women who attended secondary and college education. The odds of missing contents of ANC were 1.94 times higher among women without formal education than among women who attended secondary and college education (aOR = 1.94, 95% CI: 1.24, 3.02). The timing of ANC was identified as significant predictors of missing ANC content. Women who initiated their first ANC visit in the third trimester were 3.05 times more likely to miss ANC content (aOR = 3.05, 95% CI: 1.59, 6.54) than women who began in the first trimester. Similarly, the odds of missing content of care were 4.13 times higher among women who received only one visit than among those who received four or more visits (aOR = 4.13, 95% CI:1.95, 8.74). Women who did not own a mobile phone were 1.44 times more likely to miss the contents of ANC than those who did (aOR = 1.44, 95% CI: 1.07, 1.95). Finally, women who resided in the rural part of the country had a 1.68 times higher chance of missing key contents ANC compared to their urban counterparts (aOR = 1.68, 95% CI:1.47, 2.71) (Table 7).
Discussion
This study assessed the magnitude and determinants of non-compliance with the WHO-recommended contents of ANC in Ethiopia. The vast majority (88.2%) of pregnant women failed to receive full contents of care during prenatal visit. There was significant spatial variation in the missing content of care across regions. Educational status, timing and adequacy of ANC visits, ownership of mobile phones, and residence were identified as significant predictors of missing full contents of ANC among respondents.
In this study, the level of missing key items of care was consistent with a study conducted in East Africa (88.2%) [36]. However, our estimate was higher than previous studies in Nepal (77%) [37], Bangladesh (82%) [38], and Rwanda (75.64%) [39]. This disparity could be attributed to differences in socioeconomic and sociocultural characteristics, availability and accessibility of health service infrastructure, and maternal health service coverage between countries. Significant spatial clusters (Hot Spots) for incomplete contents of care were detected in the eastern and northern borders of SNNPR, as well as in the western border of the Oromia regions. These border areas in Ethiopia are known for their low uptake of maternal health services due to fewer healthcare facilities, limited transportation, inadequate healthcare providers, lower socioeconomic conditions, and a lack of medical supplies and equipment to access maternal health services, making it difficult for pregnant women to access recommended items of care [40, 41].
The odds of missing care content were found to be higher among women with no formal education. This is supported by similar studies conducted in East Africa [36], Nigeria [24], Ghana [42] and Ethiopia [20]. This could be because women with no formal education may have limited access to prenatal care information and may be unaware of the essential elements of care that they should receive. Similarly, these women might encounter difficulties interpreting medical terms, instructions, or advice delivered to them during ANC visits, resulting in missing essential contents of care. Furthermore, these women may encounter financial constraints that limit their access to ANC, which can result in delayed or irregular visits, lowering their chances of receiving the recommended contents of care to a full extent [43].
Late-initiation ANC visits were identified as significant predictors of missing key elements of care. This is supported by studies conducted in Pakistan [44], Rwanda [39], and Ethiopia [45]. This could be because when women start ANC late, it can be difficult to catch up on missing critical elements of care, such as important tests, screenings, and interventions that are usually done earlier in pregnancy. In addition, commencing ANC in the third trimester may result in less time being available to cover all necessary aspects of care during the remaining visits. Moreover, when women begin ANC late, they may not have the usual healthcare provider or facility for their care. This lack of continuity of care, might result in information gaps and make it more likely that they would miss certain important aspects of care [46].
Similarly, having an inadequate number of ANC visits (only one visit) was significantly associated with missing key care contents. This was in tandem with the studies conducted in Bangladesh [47], Nepal [37], Pakistan [44], and Rwanda [39]. This could be because, with only one ANC visit, healthcare providers have limited time to assess the woman’s overall health and perform the relevant screening, tests, education, counselling, and preventive interventions that are usually spread out over numerous visits. It is well recognized that ANC visits allow women to build rapport with their healthcare provider, promoting trust and opening a discussion [9]. This relationship-building aspect is missing in only one visit, which may result in less engagement and compliance with the recommended contents of care.
Women who did not own a mobile phone were more likely to miss the contents of ANC than those who did. Studies conducted in Bangladesh [48], Zanzibar [49], Nigeria [50], and Madagascar [51] support this finding. Women who do not have a mobile phone may have limited access to critical information or updates and may miss appointment reminders or schedule changes. Mobile phones allow people to stay connected and communicate with family, friends, and support networks, which can provide emotional support and aid during the care process [52]. As a result, women without mobile phones may find it more difficult to access their support networks, possibly leading to poor adherence to the contents of care.
Living in a rural part of the country increases the likelihood of missing ANC content. Studies conducted in Bangladesh [47], Vietnam [53], and Ethiopia [45] were in tandem with the current findings. This may be because of a variety of factors. Rural areas in Ethiopia frequently have limited healthcare infrastructure, laboratory setups, and a shortage of healthcare providers. This can lead to long waiting times, delayed appointments, or understaffed clinics, making it challenging for pregnant women to receive comprehensive ANC services. In addition, because poverty is prominent in many rural parts of Ethiopia, pregnant women may face financial difficulties in getting to health facilities on time, and they may prioritize other immediate needs to seek complete ANC contents. Furthermore, owing to a lack of knowledge and information, rural women may be unaware of the potential dangers signs associated with pregnancy; they may have failed to comply with the recommended number of visits and contents of care.
This study had several strengths. First, the results were based on an analysis of nationally representative data, which enhances the generalizability of the findings. Since the findings stem from both spatial and multilevel mixed-effect logistic regression analyses, they can help government and program planners develop geographical, individual, and community-focused public health interventions to minimize the level of missing essential contents of care. However, this finding should be considered in light of some limitations. As the EDHS data are cross-sectional, it is difficult to determine a temporal/causal relationship, and the findings might be exposed to recall- and social desirability biases. In addition, since the study has relied on the secondary data we didn’t consider those variables such as waiting time, healthcare worker ethics, and availability of equipment that might increase the odds of missing contents of care.
Conclusion
The level of missing care content during prenatal visits was high in Ethiopia, with significant spatial variation across regions. Being a rural resident, not having a formal education, commencing ANC late, having inadequate ANC visits, and not having a mobile phone were found to be significant predictors of missing essential contents of care during ANC. Thus, health systems and policymakers should focus on improving access to ANC services, promoting early initiation, and encouraging multiple visits to provide optimal care to pregnant women. In addition, it is essential to focus on enhancing education and healthcare infrastructure in rural countries.
Supporting information
S1 Checklist. STROBE statement—Checklist of items that should be included in reports of observational studies.
https://doi.org/10.1371/journal.pone.0313893.s001
(DOCX)
Acknowledgments
We are grateful to ICF macro (Calverton, USA) for providing the 2016 DHS data of Ethiopia.
References
- 1.
Organization, W.H., Trends in maternal mortality 2000 to 2020: estimates by WHO, UNICEF, UNFPA, World Bank Group and UNDESA/Population Division: executive summary. 2023.
- 2. Zarocostas J., Global maternal mortality rates stagnating. The Lancet, 2023. 401(10377): p. 632. pmid:36842436
- 3.
WHO, UNICEF, UNFPA, World Bank Group, and UNDESA/Population Division. Trends in Maternal Mortality 2000 to 2020. Geneva, World Health Organization, 2023 https://data.worldbank.org/indicator/SH.STA.MMRT.
- 4. Say L., et al., Global causes of maternal death: a WHO systematic analysis. The Lancet global health, 2014. 2(6): p. e323–e333. pmid:25103301
- 5. Kurjak A., Stanojević M., and Dudenhausen J., Why maternal mortality in the world remains tragedy in low-income countries and shame for high-income ones: will sustainable development goals (SDG) help? Journal of Perinatal Medicine, 2023. 51(2): p. 170–181. pmid:35636412
- 6.
Rogers, M., Antenatal Care Services among Pregnant Women in Kampala International University Teaching Hospital Bushenyi-Ishaka Municipality. 2023.
- 7. Koblinsky M., et al., Quality maternity care for every woman, everywhere: a call to action. The Lancet, 2016. 388(10057): p. 2307–2320. pmid:27642018
- 8. Teshale A.B. and Tesema G.A., Prevalence and associated factors of delayed first antenatal care booking among reproductive age women in Ethiopia; a multilevel analysis of EDHS 2016 data. PloS one, 2020. 15(7): p. e0235538. pmid:32628700
- 9.
Organization, W.H., WHO recommendations on antenatal care for a positive pregnancy experience. 2016: World Health Organization.
- 10. Arthur E., Wealth and antenatal care use: implications for maternal health care utilisation in Ghana. Health Econ Rev, 2012. 2:14. pmid:22866869
- 11. Kpordoxah M.R., et al., Assessing the association between an early and recommended number of focused antenatal care visits and the number of prenatal care content received before delivery in Ethiopia. Plos one, 2023. 18(3): p. e0282694. pmid:36867632
- 12. Ghimire P.R., et al., Impact of Antenatal Care on Perinatal Outcomes in New South Wales, Australia: A Decade-Long Regional Perspective. International Journal of Environmental Research and Public Health, 2023. 20(2): p. 977. pmid:36673735
- 13. Caira-Chuquineyra B., et al., Association between inadequate prenatal care and low birth weight of newborns in Peru: Evidence from a peruvian demographic and health survey. Heliyon, 2023. 9(4). pmid:37064470
- 14. Saaka M. and Sulley I., Independent and joint contributions of inadequate antenatal care timing, contacts and content to adverse pregnancy outcomes. Annals of Medicine, 2023. 55(1): p. 2197294.
- 15. Banchani E. and Tenkorang E.Y., Determinants of low birth weight in Ghana: Does quality of antenatal care matter? Maternal and Child Health Journal, 2020. 24: p. 668–677. pmid:32026325
- 16.
Ministry of Health Ethiopia, National Antenatal Care: ensuring positive pregnancy Experiance. Addis Ababa; 2021 http://repository.iifphc.org/bitstream/handle/123456789/1647/ANC-GUIDELINE_Feb-24-2022.pdf?sequence=1&isAllowed=y.
- 17.
EPHI, I., Ethiopian public health Institute (EPHI)[Ethiopia] and ICF. Ethiopia Mini Demographic and Health Survey 2019: Key Indicators, 2019.
- 18. Aragaw F.M., et al., Spatial distribution of delayed initiation of antenatal care visits and associated factors among reproductive age women in Ethiopia: spatial and multilevel analysis of 2019 mini-demographic and health survey. BMJ open, 2023. 13(8): p. e069095. pmid:37620267
- 19. Awoke S.M., et al., Spatiotemporal distribution and bivariate binary analysis of antenatal and delivery care utilizations in Ethiopia: EDHS 2000–2016. BMC Public Health, 2023. 23(1): p. 499. pmid:36922794
- 20. Negash W.D., et al., Magnitude of optimal access to ANC and its predictors in Ethiopia: Multilevel mixed effect analysis of nationally representative cross-sectional survey. Plos one, 2023. 18(4): p. e0284890. pmid:37083707
- 21. Ogu R.N., Agholor K.N., and Okonofua F.E., Engendering the attainment of the SDG-3 in Africa: overcoming the socio cultural factors contributing to maternal mortality. African journal of reproductive health, 2016. 20(3): p. 62–74. pmid:29553196
- 22.
CSA, I., Central Statistical Agency (CSA)[Ethiopia] and ICF. Ethiopia Demographic and Health Survey 2016. Addis Ababa, Ethiopia, and Rockville, Maryland, USA: CSA and ICF; 2016. 2017. 2016.
- 23.
ICF International. 2013. Incorporating Geographic Information into Demographic and Health Surveys: A Field Guide to GPS Data Collection. Calverton, Maryland, USA: ICF Internationa https://dhsprogram.com/publications/publication-dhsm9-dhs-questionnaires-and-manuals.cfm.
- 24. Fagbamigbe A.F. and Idemudia E.S., Assessment of quality of antenatal care services in Nigeria: evidence from a population-based survey. Reproductive Health, 2015. 12(1): p. 88. pmid:26382228
- 25. Legendre P., Spatial autocorrelation: trouble or new paradigm? Ecology, 1993. 74(6): p. 1659–1673.
- 26. Cliff A.D. and Ord K., Spatial autocorrelation: a review of existing and new measures with applications. Economic Geography, 1970. 46(sup1): p. 269–292.
- 27.
Manepalli, U., G.H. Bham, and S. Kandada. Evaluation of hotspots identification using kernel density estimation (K) and Getis-Ord (Gi*) on I-630. in 3rd International Conference on Road Safety and Simulation. 2011. National Academy of Sciences Indianapolis Indiana, United States.
- 28. Songchitruksa P. and Zeng X., Getis—Ord spatial statistics to identify hot spots by using incident management data. Transportation research record, 2010. 2165(1): p. 42–51.
- 29. Mueller T., et al., Map quality for ordinary kriging and inverse distance weighted interpolation. Soil Science Society of America Journal, 2004. 68(6): p. 2042–2047.
- 30. BOSTAN P., Basic kriging methods in geostatistics. Yuzuncu Yıl University Journal of Agricultural Sciences, 2017. 27(1): p. 10–20.
- 31. Oliver M.A. and Webster R., Kriging: a method of interpolation for geographical information systems. International Journal of Geographical Information System, 1990. 4(3): p. 313–332.
- 32. Block R., Software review: scanning for clusters in space and time: a tutorial review of SatScan. Social Science Computer Review, 2007. 25(2): p. 272–278.
- 33. Tessema Z.T., et al., Geographical variation in determinants of high-risk fertility behavior among reproductive age women in Ethiopia using the 2016 demographic and health survey: a geographically weighted regression analysis. Archives of Public Health, 2020. 78(1): p. 1–12. pmid:32832078
- 34. Moon J. and Jung I., A simulation study for geographic cluster detection analysis on population-based health survey data using spatial scan statistics. International Journal of Health Geographics, 2022. 21(1): p. 1–8.
- 35.
Anselin, L., Review of cluster analysis software. Report in fulfillment of consultant agreement# 2003-04-01 with the North American Association of Central Cancer Registries, Inc, 2004.
- 36. Raru T.B., et al., Quality of antenatal care and associated factors among pregnant women in East Africa using Demographic and Health Surveys: A multilevel analysis. Women’s Health, 2022. 18: p. 17455065221076731. pmid:35114855
- 37. Joshi C., et al., Factors associated with the use and quality of antenatal care in Nepal: a population-based study using the demographic and health survey data. BMC Pregnancy Childbirth, 2014. 14.
- 38. Akter E., et al., Levels and determinants of quality antenatal care in Bangladesh: Evidence from the Bangladesh Demographic and Health Survey. Plos one, 2023. 18(5): p. e0269767. pmid:37134074
- 39. Uwimana G., et al., Determinants of timing, adequacy and quality of antenatal care in Rwanda: a cross-sectional study using demographic and health surveys data. BMC Health Services Research, 2023. 23(1): p. 1–12.
- 40. Muchie K.F., Quality of antenatal care services and completion of four or more antenatal care visits in Ethiopia: a finding based on a demographic and health survey. BMC pregnancy and childbirth, 2017. 17(1): p. 1–7.
- 41. Tarekegn S.M., Lieberman L.S., and Giedraitis V., Determinants of maternal health service utilization in Ethiopia: analysis of the 2011 Ethiopian Demographic and Health Survey. BMC pregnancy and childbirth, 2014. 14(1): p. 1–13. pmid:24886529
- 42. Atinga R.A. and Baku A.A., Determinants of antenatal care quality in Ghana. International Journal of Social Economics, 2013. 40(10): p. 852–865.
- 43. Fagbamigbe A.F. and Idemudia E.S., Barriers to antenatal care use in Nigeria: evidences from non-users and implications for maternal health programming. BMC pregnancy and childbirth, 2015. 15: p. 1–10.
- 44. Agha S. and Tappis H., The timing of antenatal care initiation and the content of care in Sindh, Pakistan. BMC pregnancy and childbirth, 2016. 16: p. 1–9.
- 45. Woldeamanuel B.T. and Belachew T.A., Timing of first antenatal care visits and number of items of antenatal care contents received and associated factors in Ethiopia: multilevel mixed effects analysis. Reproductive health, 2021. 18(1): p. 1–16.
- 46. Yeji F., et al., Continuum of care in a maternal, newborn and child health program in Ghana: low completion rate and multiple obstacle factors. PloS one, 2015. 10(12): p. e0142849. pmid:26650388
- 47. Islam M.M. and Masud M.S., Determinants of frequency and contents of antenatal care visits in Bangladesh: Assessing the extent of compliance with the WHO recommendations. PloS one, 2018. 13(9): p. e0204752. pmid:30261046
- 48. Tang S., et al., Women using mobile phones for health communication are more likely to use prenatal and postnatal services in Bangladesh: cross-sectional study. JMIR mHealth and uHealth, 2019. 7(2): p. e10645. pmid:30816850
- 49. Lund S., et al., Mobile phones improve antenatal care attendance in Zanzibar: a cluster randomized controlled trial. BMC pregnancy and childbirth, 2014. 14(1): p. 1–10.
- 50. McNabb M., et al., Assessment of the quality of antenatal care services provided by health workers using a mobile phone decision support application in northern Nigeria: a pre/post-intervention study. PLOS one, 2015. 10(5): p. e0123940. pmid:25942018
- 51. Benski A.C., et al., Improving the quality of antenatal care using mobile health in Madagascar: five-year cross-sectional study. JMIR mHealth and uHealth, 2020. 8(7): p. e18543. pmid:32673263
- 52. Shiferaw S., et al., Designing mHealth for maternity services in primary health facilities in a low-income setting—lessons from a partially successful implementation. BMC Medical Informatics and Decision Making, 2018. 18(1): p. 1–14.
- 53. Tran T.K., et al., Factors associated with antenatal care adequacy in rural and urban contexts-results from two health and demographic surveillance sites in Vietnam. BMC health services research, 2012. 12(1): p. 1–10. pmid:22335834