Next Article in Journal
Relationship between Employee Mental Health and Job Performance: Mediation Role of Innovative Behavior and Work Engagement
Previous Article in Journal
Census Tract Demographics Associated with Libraries’ Social, Economic, and Health-Related Programming
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Multilevel Modelling of the Individual and Regional Level Variability in Predictors of Incomplete Antenatal Care Visit among Women of Reproductive Age in Ethiopia: Classical and Bayesian Approaches

by
Teshita Uke Chikako
1,
Reta Habtamu Bacha
2,
John Elvis Hagan, Jr.
3,4,*,
Abdul-Aziz Seidu
5,6,
Kenenisa Abdisa Kuse
7 and
Bright Opoku Ahinkorah
8
1
Wondo Genet College of Forestry and Natural Resource, Hawassa University, Hawassa P.O. Box 05, Ethiopia
2
Department of Statistics, Jimma University, Jimma P.O. Box 378, Ethiopia
3
Department of Health, Physical Education and Recreation, University of Cape Coast, Cape Coast PMB TF0494, Ghana
4
Neurocognition and Action-Biomechanics-Research Group, Faculty of Psychology and Sport Sciences, Bielefeld University, Postfach 10 01 31, 33501 Bielefeld, Germany
5
Centre for Gender and Advocacy, Takoradi Technical University, Takoradi P.O. Box 256, Ghana
6
College of Public Health, Medical and Veterinary Sciences, James Cook University, Townsville, QLD 4811, Australia
7
Department of Statistics, Bule Hora University, Bule Hora P.O. Box 144, Ethiopia
8
School of Public Health, Faculty of Health, University of Technology Sydney, Sydney, NSW 2007, Australia
*
Author to whom correspondence should be addressed.
Int. J. Environ. Res. Public Health 2022, 19(11), 6600; https://doi.org/10.3390/ijerph19116600
Submission received: 31 March 2022 / Revised: 14 May 2022 / Accepted: 25 May 2022 / Published: 28 May 2022

Abstract

:
Background: Antenatal care is an operational public health intervention to minimize maternal and child morbidity and mortality. However, for varied reasons, many women fail to complete the recommended number of visits. The objective of this study was to assess antenatal care utilization and identify the factors associated with the incomplete antenatal care visit among reproductive age women in Ethiopia. Methods: The 2019 Ethiopian Mini Demographic and Health Survey data were used for this study. Multilevel logistic regression analysis and two level binary logistic regression models were utilized. Results: Around 56.8% of women in Ethiopia did not complete the recommended number of antenatal care visits. Women from rural areas were about 1.622 times more likely to have incomplete antenatal care compared to women from urban areas. Women who had no pregnancy complication signs were about 2.967 times more likely to have incomplete antenatal care compared to women who had pregnancy complication signs. Women who had a slight problem and a big problem with the distance from a health center were about 1.776 and 2.973 times more likely, respectively, to have incomplete antenatal care compared to women whose distance from a health center was not a problem. Furthermore, women who had ever terminated pregnancy were about 10.6% less likely to have incomplete antenatal care compared to women who had never terminated pregnancy. Conclusions: The design and strengthening of existing interventions (e.g., small clinics) should consider identified factors aimed at facilitating antenatal care visits to promote maternal and child health related outcomes. Issues related to urban–rural disparities and noted hotspot areas for incomplete antenatal care visits should be given special attention.

1. Introduction

Antenatal care is a pregnancy-related essential health care, which could be given either in a health facility or at home, and is an integral component of maternal and child health [1,2]. It is considered as one of the most important factors for the health of the mother and optimal development of the fetus as well as for preventing or minimizing complications during pregnancy [3]. The complications before the pregnancy period have been the major causes of loss of life and disability among women in the world [3,4]. In 2015, an estimated 303,000 women died as a result of pregnancy- and childbirth-related complications and about 99% of these deaths occurred in low- and middle-income countries, with Sub-Saharan African alone accounting for roughly 66% of these deaths [5].
In addition, maternal mortality is excessively high [6]. In 2017, thousands of women died during and following pregnancy [7]. Ethiopia is among the countries with the highest maternal mortality ratios in sub-Saharan Africa, where the recommended number of Antenatal Care (ANC) visits have still not been achieved in the country [2]. According to the 2016 Ethiopian Demographic and Health survey, pregnancy-related maternal mortality ratios were 412 maternal deaths per 100,000 live births [8]. One of the key strategies in reducing maternal mortality is to access the care given to women during and after pregnancy [9]. Timely and appropriate care given to pregnant women is important for early detection of diseases and provides timely treatment [10,11,12].
The previous World Health Organization’s (WHO) recommended Focused Antenatal Care (FANC) Model. Under normal circumstances, a pregnant woman should have at least four ANC visits [9]. But recently, WHO issued “the 2016 WHO ANC model” with a new series of recommendations to improve the quality of ANC, which in turn help reduce the risk of stillbirths, complications, and ensure a positive pregnancy experience. The new WHO model recommends a minimum of eight contacts [13]. Though the recent 2016 WHO model recommends 8 contacts, the country guideline of Ethiopia still promotes 4 ANCs having slight time differences from the previously WHO recommended FANC model [14]. However, in developing countries, most pregnant women start their first ANC visit later than recommended [12,13]. For instance, globally, only 64% of women receive ANC four or more times throughout their pregnancy. Reports showed slower progress in sub-Saharan Africa than in other regions [15]. Previous research in sub-Saharan Africa showed that different sociodemographic factors, such as age, educational status, wealth index, geographical location (urban-rural), husband’s occupation, marital status, and wealth status, are associated with ANC utilization [16,17,18]. For instance, recent studies in Ethiopia (e.g., [2]), Nigeria (e.g., [19,20]), and other parts of sub-Saharan Africa [21] suggested that the likelihood of not receiving ANC was high among poor women, due to inadequate financing, the non-availability of competent and skilled health care providers in rural areas, distance to ANC facilities, and inadequate or poor-quality health services.
Substantial progress has been made in the past two decades in Ethiopia, where many pregnant women do not receive ANC or receive less than the recommended number of ANC visits and fewer women start ANC at an appropriate time [22]. According to the Ethiopian Demographic and Health Survey (EDHS) 2011 and 2016 reports, 34% and 62%, respectively, of women who had a live birth in the five years before the survey received ANC from a skilled provider at least once for their last birth. In addition, only 19% and 32% of women had four or more ANC visits during their last pregnancy respectively. Furthermore 20% of women made their first ANC visits before the fourth month of pregnancy [8].
The time to first ANC visit, as well as the total number of ANC visit, also affects the quality of the ANC that a pregnant woman receives and not having the recommended ANC service may lead to adverse pregnant outcomes [23]. This is primarily because different services and interventions are available for different gestational ages. ANC visits are vital for the early detection of HIV, malaria, and anemia prophylaxis, and prevention or management of complications [24]. Although scholarly information accounting for incomplete antenatal care services in Ethiopia exists, further research is required to gain a clearer understanding of the issues surrounding incomplete ANC visits in the country. The objective of this study was to assess the prevalence of ANC utilization and identify the factors associated with incomplete ANC among women of reproductive age in Ethiopia.

2. Materials and Methods

2.1. Source of Data

The data for this study were taken from the Ethiopian Mini Demographic and Health Survey (EMDHS) of 2019. Ethiopia is located in the horn of East Africa. It has nine regions and two self-governing districts (Addis Ababa and Dire Dawa). The survey drew a representative sample of women of reproductive age (15–49), by administering a questionnaire, which was conducted across the country and available from the EDHS database at https://dhsprogram.com/data/available-datasets.cfm (accessed on 1 February 2022). Data collection took place from 21 March 2019, to 28 June 2019 [14]. The first EMDHS were conducted in 2014 and then subsequently in 2019. It was administered at the household level and funded by the USAID in many middle- and low-income countries.

2.2. Study Variables

2.2.1. Outcome Variable

The number of antenatal care visits was the outcome variable of this study. The recent WHO report recommends women to have at least eight antenatal care visits during their last pregnancy. However, the Ethiopian mini report shows that data were collected using the old recommendation by the WHO (at least 4 visits) [14]. Following the Ethiopia mini report, incomplete antenatal care can be defined as less than four antenatal care visits during their last pregnancy of a woman aged 15-49. Therefore, antenatal care is captured as dichotomous and coded as 1 if a woman aged 15–49 in Ethiopia had less than four antenatal care visits during their last pregnancy and 0 if otherwise.

2.2.2. Independent Variables

Several predictors, such as residence, women’s education level, family wealth index, pregnancy complication signs, media exposure, ever had a terminated pregnancy, distance from a health center, who decided on the respondent’s heath care, religion, mother’s age, and mother’s occupation were considered in this study as factors associated with incomplete antenatal care.

2.3. Statistical Analyses

2.3.1. Multilevel Logistic Regression Model

Since our dependent variable is dichotomous, logistic regression is the popular model to analyze the dataset [13]. This model assumes that all women are independent in the sense that any variables affecting the dependent variable have the same effect in all clusters. Generally, women are nested within the region and lack of independence across levels of nested data as they shared common characteristics. To overcome these problems, multilevel logistic regression model was chosen as the appropriate model. A multilevel logistic regression model can be used to analyze nested sources of variability in hierarchical data, considering of the variability associated with each level of the hierarchy [25]. The study involves a two-level model with different steps for estimation: null and final models (random intercept model). The null model involves no predictors specified at either level. The final model (random intercept model) involves random intercept and explanatory variables from the two levels. In this study, the basic data structure of the two-level logistic regression is a collection of 11 groups (regions), and within group j (1, 2,..., 11), a random sample nj of level-one units (individual women). The outcome variable is given by Yij:
Y i j = { 1   i f   t h e   i th   w o m e n   i n   j th   r e g i o n   h a v e   i n c o m p l e t e   A N C   0   i f   t h e   i th   w o m e n   i n   j th   r e g i o n   h a v e   c o m p l e t e   A N C  
Pij = P(Yij = 1|Xij, Uij) is the probability of having incomplete ANC for the ith women in the jth region and 1 − Pij = P(Yij = 0|Xij), is the probability of having complete ANC for the ith women in the jth region, where Xij stands for set of fixed independent variable and Uij stands for the random part of the model.

2.3.2. The Null Model

This model is given as:
Y i j = β 0 + U 0 j + ε 0 i j   ,   ε 0 i j ~ N ( 0 ,   σ ε 2 )   ,   U 0 j ~ N ( 0 ,   σ u 2 )  
where the index i indicates individual women, j indicates region, U0j is the level two error, ε 0 i j is the level one error, β0 is interpreted as the overall average of antennal care visit status, and Yij is antennal care visit status of ith women in jth region.
This model decomposes the variance at two levels to assess how much of the variation is due to individual themselves and how much is due to region. The interclass correlation (ICC) is given as:
ICC   = ρ = σ u 2 σ ε 2 + σ u 2 = σ u 2 σ Y 2
This measures the correlation between variables on the same level or proportion of group level variance compared to total variance. It can also be interpreted as the expected correlation between two randomly chosen units in the same level and to check within and between variations of regions of women’s antennal care visit status. Next, predicators of the two levels are added to the null model to obtain the final model known as the random intercept model and the fixed explanatory model [26]. The random intercept model expresses the log odds, i.e., the logit (Pij), as a sum of linear functions of the explanatory variables.

2.3.3. The Final Model

The final multilevel model can be written as:
Y i j = β 0 + β 1 X 1 i j + + β m X m i j + ε i j + U j  
where Yij is women’s antenatal care status, ε i j is the level one variance, Uj is the level two variance, and m is the household-level explanatory variables. Since the multilevel logistic regression is nonlinear, the above model can be given by logit transformation of its probability (logit (Pij)) as follows:
l o g i t ( P i j ) = l o g [ P i j 1 P i j ] = β 0 + β 1 X 1 i j + + β m X m i j + U j  
Parameter estimation in multilevel logistic regression is not direct like an ordinary regression model. To estimate unknown parameters, we applied marginal maximum likelihood [27]. This method of parametric estimation is most popular in the multilevel logistic regression model.

2.4. Bayesian Multilevel Logistic Regression Model

The three pillars Bayesian analysis were the likelihood function, which reflect information about the parameters contained in the data, the prior distribution, which quantifies what is known about the parameters before observing data, and the prior distribution and likelihood can be easily combined to form the posterior distribution, which represents the total knowledge about the parameters after the data have been observed [28,29,30]. The study considers normal for fixed effect and inverse gamma for random effect as prior distribution. The likelihood contribution from the ith subject in the jth group is Bernoulli and given by:
B e r n o u l l i   ( P i j ) = P i j y i j ( 1 P i j ) 1 y i j  
where Pij represents the probability of the event for subject i in the j group, which has covariate vector xij, and yij indicates the presence, yij = 1, or absence, yij = 0, of the event for that subject. Thus, the likelihood contribution for the ith subject in the jth region is:
L ( y | β i , σ u 2 ) = ( e β 0 + β 1 X 1 i j + + β m X m i j + U j 1 + e β 0 + β 1 X 1 i j + + β m X m i j + U j ) y i j ( 1 e β 0 + β 1 X 1 i j + + β m X m i j + U j 1 + e β 0 + β 1 X 1 i j + + β m X m i j + U j ) 1 y i j
Since individual subjects in the group are assumed independent from each other, the likelihood function over a dataset of n subjects in the 11 region is then:
L ( y | β i , σ u 2 ) = i = 1 n j = 1 11 [ ( e β 0 + β 1 X 1 i j + + β m X m i j + U j 1 + e β 0 + β 1 X 1 i j + + β m X m i j + U j ) y i j ( 1 e β 0 + β 1 X 1 i j + + β m X m i j + U j 1 + e β 0 + β 1 X 1 i j + + β m X m i j + U j ) 1 y i j ]
A non-informative prior distribution that is used to express complete ignorance of the value before the data are collected was considered. We prefer non-informative because no value is favored over any other, and they are also described as diffuse or at prior due to their reason and their shape. The log odd of fixed effect is approximately normal and the random effect is random. This leads the authors to use normal distribution prior for the fixed effect ( β i ) and Inverse gamma distribution prior for random effect ( σ i 2 ).
Combining the likelihood function with the prior distribution on ( β , σ u 2 ) and the full conditional distribution for unknown parameters, the posterior distribution can be written as:
f ( β , σ u 2 | y ) = L ( y | β , σ u 2 ) × π ( β ) × π ( σ u 2 )
After posterior distribution is determined, the assessment of algorithm convergence was checked to identify whether chain has reached its stationary or not. The term convergence of an algorithm refers to whether the algorithm has reached its equilibrium (target) distribution. If the algorithm has reached its equilibrium, then the generated sample comes from the right target distribution. Thus, monitoring the convergence of the algorithm is crucial for generating effective results from the posterior distribution. Therefore, to monitor the convergence of the algorithm, we used the most popular and straight forward convergence assessment methods such as Rhat, Bulk effective sample size (Bulk_ESS), and Tail effective sample size (Tail_ESS). If Rhat = 1 and Bulk_ESS and Tail_ESS were greater than 1000, then we have evidence that the chain has converged.

2.5. Software

The data analysis was done using R statistical software ((Version 4.0.4, Foundation for Statistical Computing, Vienna, Austria). The authors used R software with lme4 package for the classical multilevel logistics and R software with brms package for the Bayesian multilevel logistics analysis. Based on previous literature [30], we used non informative normal prior distribution with mean = 0 and precision = 0.001 for the fixed effect and inverse gamma distribution (0.1, 0.1) for the random effect, with iteration = 20,000, warm-up = 2000 (number of iterations that was discarded), chains = 2, initials (the starting values of the iterations) = 0, cores (specifies the number of cores used for the algorithm) = 2, and adapt delta (controls divergent transition) = 0.95. Summary statistics were carried out from the posterior distribution after the model was converged and the 95% credible interval was used for the test of significance.

2.6. Ethics Approval

Ethical clearance was obtained by the Institutional Review Board of ICF International. Permission was also sought from each woman during the fieldwork. The authors requested and officially received approval from the DHS Program to use the dataset for this study. Detailed information about the DHS data usage and ethical standards are available at http://goo.gl/ny8T6X (accessed on 1 February 2022).

3. Results

3.1. Descriptive Results

The total number of women aged 15 to 49 years covered in this study was 3190. From these, 1812 women aged 15–49 years (56.8%) were not following the recommended number of antenatal care visits (Table 1).

3.2. Result of the Variance Component Model

In this model, we first fitted a simple model with no predictors, (i.e., an intercept-only model) that predicted the probability of incomplete antenatal care status. This model contains no explanatory variables, and it can be considered as a parametric version of assessing heterogeneity among regions with respect to incomplete ANC visits (Table 2).
From Table 2, we consider the variance of the random factor (between region variance) as significantly different from zero, which indicates that there are regional differences in incomplete ANC.
To get an idea of how much of variation in incomplete antenatal care was attributable to the individual level and regional level, it is useful to see the intra-region correlation coefficient (ICC). ICC = 0.081 indicates that around 8.1% of variation in incomplete ANC among women is due to the variation across (between) regions, whereas the remaining 91.9% was attributable to individual level (i.e., within-region) differences.

3.3. Result of the Classical Multilevel Model

Table 3 results show that residence, women’s education level, family wealth index, pregnancy complication signs, media exposure, ever had a terminated pregnancy, distance from a health center, and who decided on the respondent’s heath care are significant predictors of incomplete ANC among women aged 15–49 years in Ethiopia, while religion, mother’s age, and mother’s occupation were non-significant predictors of incomplete ANC among women aged 15–49 years in Ethiopia.
Women from rural areas were about 1.622 times more likely to have incomplete ANC visits compared to women from urban areas. Additionally, women who ever had a terminated pregnancy were about 10.6% less likely to have incomplete ANC compared to those women who never had a terminated pregnancy. The results also indicate that women who had secondary and higher education were about 20.5% and 31.1% less likely to have incomplete ANC compared to those women who do not have formal education. Table 3 also shows that women who had slight problem and big problem with distance from a health center were about 1.776 and 2.973 times more likely, respectively, to have incomplete ANC compared to women who saw distance to a health center as not a problem.
Table 3 also revealed that rich women were 15.4% less likely to have incomplete ANC compared to poor women. Additionally, women who had no pregnancy complication signs were about 2.967 times more likely to be incomplete ANC compared to the women who did have pregnancy complication signs. Likewise, women who were exposed to all media sources at least once a week were about 28.2% and 8.7% less likely to have incomplete ANC compared to women who have no media exposure at all. Concerning a person who usually decides on the respondent’s health care, respondents whose health care was decided by the respondent and husband and respondent and another person were 29.7% and 42.4% less likely to have incomplete ANC compared to respondents whose health care was decided by the respondent alone.
From the result of the variance component model and random intercept model, AIC = 13,149.9 and BIC = 13,411.7 for the random intercept model were the smallest compared to the variance component model. This estimation shows that the random intercept model better fits the data to predict the factors for incomplete ANC among women aged 15–49 in Ethiopia.
The contrast between the two prominent statistical views, classical and Bayesian approaches, should be based on the better fit model (random intercept model), and then, a rigorous decision should be made for having accurate implementation.

3.4. Results of the Bayesian Multilevel Model

From Table 4, the Rhat value is 1, and all effective sample sizes (both Bulk_ESS and Tail_ESS) are greater than 1000, which suggests that the model was converged and reached its equilibrium. Table 4 revealed that residence, women’s education level, family wealth index, pregnancy complication signs, media exposure, ever had a terminated pregnancy, distance from a health center, and who decided on the respondent’s heath care were significant predictors of incomplete ANC of women aged 15–49 years in Ethiopia (since their 95% confidence intervals does not contain zero). Religion, mother’s age, and mother’s occupation were not significant predictors of incomplete ANC of women aged 15–49 years in Ethiopia (since their 95% confidence intervals contains zero).
The results of the classical multilevel and Bayesian multilevel show that the data were well fitted and the results of both approaches were almost consistent, but almost all parameters in the Bayesian result had small standard errors compared to the corresponding classical multilevel logistic model. This is because the Bayesian analysis gives additional solutions as the posterior distribution of the parameters.

4. Discussion

This study investigates the predictors of incomplete ANC among women aged 15–49 years in Ethiopia. The results revealed that residence, women’s education level, family wealth index, pregnancy complication signs, media exposure, ever had a terminated pregnancy, distance from a health center, and who decided on the respondent’s heath care were found to be significant predictors of incomplete ANC of women aged 15–49 years in Ethiopia, while religion, mother’s age, and mother’s occupation were non-significant predictors of incomplete antenatal care of women aged 15–49 years in the country.
The study shows residence has significant relation with incomplete ANC and the utilization of ANC varies with place of residence, such that women from rural areas have less ANC visits than women from urban areas. This finding is consistent with other studies done in Ethiopia and Nigeria [2,17,31]. The variations noted suggest that urban women could have more access to health services and have considerable information and education about ANC than their rural counterparts. This finding suggests that health service inequity is preventable and could be biased by the systematic dissimilarities in the health of residents rooted in social determinants of health [32]. Therefore, health equity and social determinants become crucial components of the postmillennium goal and sustainable development global agendas as a part of progressive accomplishment for collective health coverage [33,34]. The social determinants of health, such as the conditions in which people are born, grow, work, and live, including the wider set of social systems, such as economic policies and systems, development agendas, social norms, social policies, and political systems, shape the conditions of peoples’ daily life and bring health inequity [35]. Worldwide, the inequalities in the distribution of health facilities are reflected by the higher proportion of ANC use in urban centers as compared to rural areas [33,34,36]. These discrepancies might be because of the local inaccessibility of obstetric care services in rural areas as well as disparities in some district administrations. Therefore, stakeholders and other service providers must work together for improving societies’ easy access to healthcare services.
Similarly, women who had ever terminated pregnancy had more ANC visits than those who had never terminated pregnancy. This finding was supported by a study carried out in Ethiopia, Malawi, and other low-income and middle-income countries [37,38,39]. Usually, illness experienced, complications associated with past pregnancy termination, and perceived susceptibility to illnesses connected to pregnancy termination could trigger more antenatal care utilization of women in reproductive age [40,41,42].
Women with a secondary/higher level of educational and high wealth index were less likely to have incomplete antenatal care compared to women with no formal education. Women with a higher educational level are more likely to be knowledgeable about the benefits of ANC services. Logically, higher-educated women could be more flexible, open to new ideas, and make appropriate decisions about their families’ health, including ANC [43]. Furthermore, when women are of low socio-economic standing (i.e., education and wealth) and do not have adequate resources or limited resources could create inadequate resource allocation to women and to ANC; hence, their adherence to ANC could be minimized [44,45,46]. Therefore, high educational level and wealth could serve as a protective mechanism against incomplete antenatal care [47,48,49]. The study discovers that having wanted pregnancy was significantly associated with the odds of having fewer antenatal care visits as compared to unwanted pregnancies. This result was similar with other studies done in Tanzania, Nigeria, and Ghana, as women who had a wanted pregnancy were more likely to have antenatal care [50,51,52]. Women with wanted pregnancies might feel a higher desire for a healthy pregnancy and childbirth, and thus, may pay greater attention to antenatal care services.
Similar to previous studies [53,54,55], the distance to the nearest health center with ANC-providing facilities in a typical region was significantly associated with the odds of having incomplete antenatal care. For example, traveling long distances and having good access routes to ANC centers could restrict regular attendance for ANC services. Conversely, those women who live close to obstetric health facility were more likely to use antenatal care services [56]. This finding indicates that geographic accessibility, measured in either distance or travel time, has a greater influence on maternal health service utilization. Furthermore, media exposure might also play a vital role in increasing the use of more antenatal care and other maternal health services. In this study, women who had almost all and at least once a week access to mass media were more likely to have complete antenatal care, which is recommended by WHO, than those who had no media exposure at all. This result was in line with previous studies done in Nepal, Ghana, and Ethiopia [57,58,59]. Utilizing mass media to spread an antenatal care message could be an effective approach to improve the uptake of early initiation of the recommended antenatal care. The frequency of mass media (e.g., regular listening of radio messages) has been shown to positively impact on the utilization of maternal healthcare services, including the regularity of ANC visits by women [57].

Strength and Limitations of the Study

The strength of this study was the incorporation of a large dataset that was nationally-representative. This could increase the generalization of the study finding to the whole region of Ethiopia. However, the cross-sectional design used restricted causal interpretation of the findings, but only associations. The study was based on a secondary dataset, which limits control over the types of variables available. Hence, the data analyses used were restricted to only the variables captured in the DHS. It is, therefore, possible that other interconnected variables were overlooked during the survey. For example, future research on ANC should use the recent WHO recommendations of 8 visits or contacts as a standard reference for comparison. Due to the self-reported nature of the survey, issues related to recall and social desirability biases cannot be ignored in this study.

5. Conclusions

The results show that the prevalence of incomplete ANC among women aged 15–49 years was 56.8% in Ethiopia. The major determinants associated with incomplete ANC in Ethiopia were residence, ever had a terminated pregnancy, women’s education level, distance from a health center, wealth index, pregnancy complication signs, media exposure, and who decided on the respondent’s health care. To enhance the number of ANC visits, the design and strengthening of existing interventions (e.g., small clinics) should consider the identified factors when facilitating antenatal care visits to promote maternal and child health-related outcomes. Issues related to urban–rural disparities and noted hotspot areas for incomplete ANC visits should be given special attention.

Author Contributions

T.U.C. and R.H.B., developed the study concept and performed the analysis. T.U.C., R.H.B., J.E.H.J., A.-A.S., B.O.A. and K.A.K. drafted and revised the manuscript for its intellectual content. All authors have read and agreed to the published version of the manuscript.

Funding

We sincerely thank Bielefeld University, Germany for providing financial support through the Open Access Publication Fund for the article processing charge.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Ethical clearance was obtained by the Institutional Review Board of ICF International. Permission was also sought from each woman during the fieldwork. The authors requested and officially received approval from the DHS Program to use the dataset for this study. Supplementary information about the DHS data usage and ethical standards are available at http://goo.gl/ny8T6X (accessed on 1 February 2022).

Data Availability Statement

The dataset used for the study is available at: https://dhsprogram.com/data/available-datasets.cfm (accessed on 1 February 2022).

Conflicts of Interest

The authors declare that there are no competing interests.

References

  1. March of Dimes. Your First Prenatal Care Checkup. 2011. Available online: http://www.marchofdimes.org/pregnancy/your-first-prenatal-care-checkup.aspx (accessed on 20 October 2021).
  2. Tegegne, T.K.; Chojenta, C.; Getachew, T.; Smith, R.; Loxton, D. Antenatal care use in Ethiopia: A spatial and multilevel analysis. BMC Pregnancy Childbirth 2019, 19, 399. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  3. Black, R.; Laxminarayan, R.; Temmerman, M.; Walker, N. Disease Control Priorities, 3rd ed.; Reproductive, Maternal, Newborn, and Child Health; World Bank: Washington, DC, USA, 2016; Volume 2, Available online: https://openknowledge.worldbank.org/handle/10986/23833 (accessed on 13 January 2022).
  4. Ashraf, F.; Thaver, I.; Imtiaz, F.; Ayub, A. Quality Assessment of Focused Antenatal Care Service Delivery in Tertiary Care Health Facility. J. Ayub Med Coll. Abbottabad 2017, 29, 219–224. [Google Scholar] [PubMed]
  5. WHO (World Health Organization). Trends in Maternal Mortality: 1990 to 2015. Estimates by WHO, UNICEF, UN-FPA, the World Bank and the United Nations Population Division; WHO: Geneva, Switzerland, 2015. [Google Scholar]
  6. WHO. Global Maternal Mortality; WHO: Geneva, Switzerland, 2016. [Google Scholar]
  7. Wenjuan, W.; Alva, S.; Wang, S.; Fort, A. Levels and Trends in the Use of Maternal Health Services in Developing Countries; DHS Comparative Reports No. 26; ICF Macro: Calverton, MD, USA, 2011. [Google Scholar]
  8. Central Statistical Agency (CSA); ICF. Ethiopia Demographic and Health Survey 2016; CSA: Addis Ababa, Ethiopia; ICF: Calverton, MA, USA, 2016. [Google Scholar]
  9. World Health Organization. Integrated Management of Pregnancy and Childbirth WHO Recommended Interventions for Improving Maternal and Newborn Health; WHO: Geneva, Switzerland, 2010. [Google Scholar]
  10. Halim, N.; Bohara, A.K.; Ruan, X. Healthy mothers, healthy children: Does maternal demand for antenatal care matter for child health in Nepal? Health Policy Plan. 2011, 26, 242–256. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  11. Bloom, S.S.; Wypij, D.; Das Gupta, M. Dimensions of women’s autonomy and the influence on maternal health care utilization in a North Indian city. Demography 2001, 38, 67–78. [Google Scholar] [CrossRef] [PubMed]
  12. Wado, Y.D.; Afework, M.F.; Hindin, M.J. Unintended pregnancies and the use of maternal health services in southwestern Ethiopia. BMC Int. Health Hum. Rights 2013, 13, 36. [Google Scholar] [CrossRef] [Green Version]
  13. World Health Organization. WHO Recommendations on Antenatal Care for a Positive Pregnancy Experience; World Health Organization: Geneva, Switzerland, 2016. [Google Scholar]
  14. Ethiopian Public Health Institute (EPHI) [Ethiopia] and ICF. Ethiopia Mini Demographic and Health Survey 2019: Key Indicators; EPHI and ICF: Rockville, MD, USA, 2019. [Google Scholar]
  15. Lincetto, O.; Mothebesoane-anoh, S.; Gomez, P.; Munjanja, S. Antenatal Care: Opportunities for Africa’s Newborns. 2013. Available online: https://www.scirp.org/(S(351jmbntvnsjt1aadkposzje))/reference/ReferencesPapers.aspx?ReferenceID=1809124 (accessed on 18 January 2022).
  16. Fagbamigbe, A.F.; Idemudia, E.S. Wealth and antenatal care utilization in Nigeria: Policy implications. Health Care Women Int. 2017, 38, 17–37. [Google Scholar] [CrossRef] [Green Version]
  17. Fagbamigbe, A.F.; Idemudia, E.S. Barriers to antenatal care use in Nigeria: Evidences from non-users and implications for maternal health programming. BMC Pregnancy Childbirth 2015, 15, 95. [Google Scholar] [CrossRef] [Green Version]
  18. Tessema, Z.T.; Teshale, A.B.; Tesema, G.A.; Tamirat, K.S. Determinants of completing recommended antenatal care utiliza-tion in sub-Saharan from 2006 to 2018: Evidence from 36 countries using Demographic and Health Surveys. BMC Pregnancy Childbirth 2021, 21, 192. [Google Scholar] [CrossRef]
  19. Bolarinwa, O.A.; Tessema, Z.T.; Frimpong, J.B.; Seidu, A.-A.; Ahinkorah, B.O. Spatial distribution and factors associated with modern contraceptive use among women of reproductive age in Nigeria: A multilevel analysis. PLoS ONE 2021, 16, e0258844. [Google Scholar] [CrossRef]
  20. Nwosu, C.O.; Ataguba, J.E. Socioeconomic inequalities in maternal health service utilisation: A case of antenatal care in Ni-geria using a decomposition approach. BMC Public Health 2019, 19, 1493. [Google Scholar] [CrossRef] [Green Version]
  21. Tessema, Z.T.; Akalu, T.Y. Spatial Pattern and Associated Factors of ANC Visits in Ethiopia: Spatial and Multilevel Modeling of Ethiopian Demographic Health Survey Data. Adv. Prev. Med. 2020, 2020, 4676591. [Google Scholar] [CrossRef] [PubMed]
  22. Tesfaye, B.; Mathewos, T.; Kebede, M. Skilled delivery inequality in Ethiopia: To what extent are the poorest and uneducated mothers benefiting? Int. J. Equity Health 2017, 16, 82. [Google Scholar] [CrossRef] [PubMed]
  23. Mpembeni, R.N.; Killewo, J.Z.; Leshabari, M.T.; Massawe, S.N.; Jahn, A.; Mushi, D.; Mwakipa, H. Use pattern of maternal health services and determinants of skilled care during delivery in Southern Tanzania: Implications for achievement of MDG-5 targets. BMC Pregnancy Childbirth 2007, 7, 29. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  24. Belayneh, T.; Adefris, M.; Andargie, G. Previous Early Antenatal Service Utilization Improves Timely Booking: Cross-Sectional Study at University of Gondar Hospital, Northwest Ethiopia. J. Pregnancy 2014, 2014, 132494. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  25. Breslow, N.E.; Clayton, D.G. Approximate Inference in Generalized Linear Mixed Models. J. Am. Statist. Assoc. 1993, 88, 9–25. [Google Scholar]
  26. Goldstein, H. Multilevel Statistical Models; John Wiley & Sons: Hoboken, NJ, USA, 2011. [Google Scholar]
  27. Bock, R.D.; Aitkin, M. Marginal maximum likelihood estimation of item parameters: Application of an EM algorithm. Psychometrika 1981, 46, 443–459. [Google Scholar] [CrossRef]
  28. Dienes, Z. Bayesian Versus Orthodox Statistics: Which Side Are You on? Perspect. Psychol. Sci. 2011, 6, 274–290. [Google Scholar] [CrossRef] [Green Version]
  29. Kruschke, J.K. Bayesian data analysis. Wiley Interdiscip. Rev. Cogn. Sci. 2010, 1, 658–676. [Google Scholar] [CrossRef]
  30. DePaoli, S.; Winter, S.D.; Visser, M. The Importance of Prior Sensitivity Analysis in Bayesian Statistics: Demonstrations Using an Interactive Shiny App. Front. Psychol. 2020, 11, 608045. [Google Scholar] [CrossRef]
  31. Haftu, A.; Hagos, H.; Mehari, M.-A.; Gher, B. Pregnant women adherence level to antenatal care visit and its effect on perinatal outcome among mothers in Tigray Public Health institutions, 2017: Cohort study. BMC Res. Notes 2018, 11, 872. [Google Scholar] [CrossRef]
  32. Seid, A.; Ahmed, M. Survival time to first antenatal care visit and its predictors among women in Ethiopia. PLoS ONE 2021, 16, e0251322. [Google Scholar] [CrossRef] [PubMed]
  33. Tesema, G.A.; Mekonnen, T.H.; Teshale, A.B. Individual and community-level determinants, and spatial distribution of institutional delivery in Ethiopia, 2016: Spatial and multilevel analysis. PLoS ONE 2020, 15, e0242242. [Google Scholar] [CrossRef]
  34. Okedo-Alex, I.N.; Akamike, I.C.; Ezeanosike, O.; Uneke, C.J. Determinants of antenatal care utilisation in sub-Saharan Africa: A systematic review. BMJ Open 2019, 9, e031890. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  35. Fagbamigbe, A.F.; Mashabe, B.; Lepetu, L.; Abel, C. Are the timings and risk factors changing? Survival analysis of timing of first antenatal care visit among pregnant women in Nigeria (2003–2013). Int. J. Women’s Health 2017, 9, 807–819. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  36. Konje, E.T.; Magoma, M.T.N.; Hatfield, J.; Kuhn, S.; Sauve, R.S.; Dewey, D.M. Missed opportunities in antenatal care for improving the health of pregnant women and newborns in Geita district, Northwest Tanzania. BMC Pregnancy Childbirth 2018, 18, 394. [Google Scholar] [CrossRef] [PubMed]
  37. Terfasa, T.G.; Afework, M.F.; Berhe, F.T. Antenatal Care Utilization and It’s Associated Factors in East Wollega Zone, Ethiopia. J. Pregnancy Child Health 2017, 4, 4172. [Google Scholar] [CrossRef]
  38. Wang, W.; Winter, R.; Mallick, L.; Florey, L.; Burgert-Brucker, C.; Carter, E. The Relationship between the Health Service Environ-ment and Service Utilization: Linking Population Data to Health Facilities Data in Haiti and Malawi; DHS Analytical Studies No. 51; ICF International: Rockville, MD, USA, 2015. [Google Scholar]
  39. Kuhnt, J.; Vollmer, S. Antenatal care services and its implications for vital and health outcomes of children: Evidence from 193 surveys in 69 low-income and middle-income countries. BMJ Open 2017, 7, e017122. [Google Scholar] [CrossRef] [Green Version]
  40. Fekede, B.; Gebremariam, A. Antenatal care service utilization and factors associated in Jimma town (Southwest Ethiopia). Ethiop. Med. J. 2007, 45, 123–133. [Google Scholar]
  41. Jira, C.; Belachew, T. Determinants of Antenatal care utilization in Jimma Town. Ethiop. J. Health Sci. 2005, 15, 49–61. [Google Scholar]
  42. Tewodros, B.; Dibaba, Y. Factors affecting antenatal care utilization in Yem special woreda, southwestern Ethiopia. Ethiop. J. Health Sci. 2009, 19, 45–51. [Google Scholar]
  43. Desalew, A.; Semahegn, A.; Birhanu, S.; Tesfaye, G. Incomplete Vaccination and Its Predictors among Children in Ethiopia: A Systematic Review and Meta-Analysis. Glob. Pediatr. Health 2020, 7, 2333794X20968681. [Google Scholar] [CrossRef] [PubMed]
  44. Peltzer, K.; Pengpid, S. Socioeconomic Factors in Adherence to HIV Therapy in Low- and Middle-income Countries. J. Health Popul. Nutr. 2013, 31, 150–170. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  45. Ahmed, S.; Creanga, A.A.; Gillespie, D.G.; Tsui, A.O. Economic Status, Education and Empowerment: Implications for Maternal Health Service Utilization in Developing Countries. PLoS ONE 2010, 5, e11190. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  46. Mishra, P.; Hansen, E.H.; Sabroe, S.; Kafle, K.K. Socio-economic status and adherence to tuberculosis treatment: A case-control study in a district of Nepal. Int. J. Tuberc. Lung Dis. 2005, 9, 1134–1139. [Google Scholar]
  47. Yaya, S.; Bishwajit, G.; Ekholuenetale, M. Factors associated with the utilization of institutional delivery services in Bangladesh. PLoS ONE 2017, 12, e0171573. [Google Scholar] [CrossRef] [Green Version]
  48. Huda, T.M.; Chowdhury, M.; El Arifeen, S.; Dibley, M.J. Individual and community level factors associated with health facility delivery: A cross sectional multilevel analysis in Bangladesh. PLoS ONE 2019, 14, e0211113. [Google Scholar] [CrossRef]
  49. Assefa, L.; Alemayehu, M.; Debie, A. Magnitude of institutional delivery service utilization and associated factors among women in pastoral community of Awash Fentale district Afar Regional State, Ethiopia. BMC Res. Notes 2018, 11, 162. [Google Scholar] [CrossRef] [Green Version]
  50. Bishanga, D.R.; Drake, M.; Kim, Y.M.; Mwanamsangu, A.H.; Makuwani, A.M.; Zoungrana, J.; Lemwayi, R.; Rijken, M.J.; Stekelenburg, J. Factors associated with institutional delivery: Findings from a cross-sectional study in Mara and Kagera regions in Tanzania. PLoS ONE 2018, 13, e0209672. [Google Scholar] [CrossRef]
  51. Fagbamigbe, A.F.; Abel, C.; Mashabe, B.; Adebowale, A.S. Survival analysis and prognostic factors of the timing of first antenatal care visit in Nigeria. Adv. Integr. Med. 2019, 6, 110–119. [Google Scholar] [CrossRef]
  52. Manyeh, A.K.; Amu, A.; Williams, J.; Gyapong, M. Factors associated with the timing of antenatal clinic attendance among first-time mothers in rural southern Ghana. BMC Pregnancy Childbirth 2020, 20, 47. [Google Scholar] [CrossRef] [Green Version]
  53. Aliyu, A.A.; Dahiru, T. Predictors of delayed Antenatal Care (ANC) visits in Nigeria: Secondary analysis of 2013 Nigeria De-mographic and Health Survey (NDHS). Pan Afr. Med. J. 2017, 26, 124. [Google Scholar] [CrossRef] [PubMed]
  54. Tolefac, P.N.; Halle-Ekane, G.E.; Agbor, V.N.; Sama, C.B.; Ngwasiri, C.; Tebeu, P.M. Why do pregnant women present late for their first antenatal care consultation in Cameroon? Matern. Health Neonatol. Perinatol. 2017, 3, 29. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  55. Sanogo, N.A.; Yaya, S. Wealth Status, Health Insurance, and Maternal Health Care Utilization in Africa: Evidence from Gabon. BioMed Res. Int. 2020, 2020, 4036830. [Google Scholar] [CrossRef] [PubMed]
  56. Kibusi, S.M.; Sunguya, B.F.; Kimunai, E.; Hines, C.S. Health insurance is important in improving maternal health service utilization in Tanzania—analysis of the 2011/2012 Tanzania HIV/AIDS and malaria indicator survey. BMC Health Serv. Res. 2018, 18, 112. [Google Scholar] [CrossRef]
  57. Acharya, D.; Khanal, V.; Singh, J.K.; Adhikari, M.; Gautam, S. Impact of mass media on the utilization of antenatal care services among women of rural community in Nepal. BMC Res. Notes 2015, 8, 345. [Google Scholar] [CrossRef] [Green Version]
  58. Twum, P.; Qi, J.; Aurelie, K.K.; Xu, L. Effectiveness of a free maternal healthcare programme under the National Health Insurance Scheme on skilled care: Evidence from a cross-sectional study in two districts in Ghana. BMJ Open 2018, 8, e022614. [Google Scholar] [CrossRef]
  59. Nageso, D.; Tefera, K.; Gutema, K. Enrollment in community based health insurance program and the associated factors among households in Boricha district, Sidama Zone, Southern Ethiopia; a cross-sectional study. PLoS ONE 2020, 15, e0234028. [Google Scholar] [CrossRef]
Table 1. Descriptive result of outcome variables.
Table 1. Descriptive result of outcome variables.
VariableLess than 4 VisitGreater or Equal to 4 Visits
CountPercentCountPercent
Number of antenatal care visits181256.8%137843.2%
Table 2. Result of the variance component model.
Table 2. Result of the variance component model.
Fixed EffectEstimateStd. ErrorZ-Valuep-Value
Intercept−0.45620.1640−2.7820.028
Random Effect
σ ^ u 2 0.290.0654.4620.003
ICC (ρ)0.081
AIC = 13,419.7, BIC = 13,434.3, logLik = −6707.9, deviance= 13,415.7.
Table 3. Result of the classical multilevel model.
Table 3. Result of the classical multilevel model.
VariableCategoryEstimateStd. ErrorZ-Valuep-ValueOdds Ratio
Intercept 0.5910.1204.920.0026
ResidenceUrban (ref.)
Rural0.4970.0895.5450.00021.622
Ever had a terminated pregnancyNo (ref.)
Yes−0.1660.076−2.1840.04870.894
Women’s education levelNo Formal Education (ref.)
Primary−0.0560.058−0.9640.3340.658
Secondary−0.2090.101−2.0760.0380.795
Higher−0.4240.144−2.9320.0030.689
ReligionOrthodox (ref.)
Muslim0.1040.2640.3950.6921.724
Protestant0.0090.0880.0960.9231.810
Other0.0820.0731.1190.2631.737
Distance from a health centerNo problem (ref.)
Slight problem0.5660.2232.5380.0251.776
Big problem0.1080.0482.2510.0382.973
Wealth indexPoor (ref.)
Middle0.0880.0561.5830.1130.771
Rich−0.2670.094−2.8310.0140.846
Pregnancy complication signsYes (ref.)
No0.1320.052.6410.0022.967
Media exposureNot at all (ref.)
Less than Once a week−0.0420.057−0.7250.4680.667
At least Once a week−0.2440.056−4.3480.00010.718
Almost all−0.1520.071−2.1450.0320.913
Mother’s age15–19 (ref.)
20–24−0.1030.115−0.8980.3690.855
25–29−0.0520.118−0.4370.6620.862
30–34−0.1010.122−0.8980.3690.772
35–39−0.1550.142−1.0940.2740.812
40–440.1530.1880.8160.4150.767
45–49−0.3110.1753.1730.0750.732
Who decided on the respondent’s health careRespondent alone (ref.)
Respondent and husband−0.1310.057−2.290.0370.703
Respondent and another person−0.1860.070−2.6350.0080.576
Husband alone0.0730.0561.6870.0571.194
OccupationNot working (ref.)
Managerial0.0550.1510.3660.7141.340
Clerical0.060.0670.8830.3771.167
Sales−0.1740.317−0.5500.5821.419
Agricultural−0.2920.171−1.7080.0881.424
Other0.1080.3530.0940.7601.114
σ ^ u 2 0.2710.068
ICC (ρ) 0.076
AIC = 13,149.9, BIC = 13,411.7, logLik = −6538.9, deviance = 13,077.9.
Table 4. Result of the Bayesian multilevel model.
Table 4. Result of the Bayesian multilevel model.
VariableCategoryEstimateStd. Error95% CIRhatBulk_ESSTail_ESS
LowerUpper
Intercept 0.5940.118 0.3410.8471.0087958954
ResidenceUrban (ref.)
Rural0.509 0.086 0.3680.6511.0011,54210,598
Ever had a terminated pregnancyNo (ref.)
Yes−0.154 0.072 −0.306−0.0511.0015,87412,698
Women’s education levelNo Formal Education (ref.)
Primary −0.047 0.055 −0.1680.1741.0018,59617,849
Secondary−0.203 0.091 −0.328−0.0781.0019,87219,877
Higher−0.422 0.141 −0.563−0.1811.0020,18719,874
ReligionOrthodox
Muslim0.103 0.265 −0.1050.2181.0058474526
Protestant0.010 0.09 −0.1540.1691.0034854596
Other0.085 0.071 −0.2510.3361.0067896875
Distance from a health centerNo problem (ref.)
Slight problem0.5680.2210.3160.8181.0010,5649865
Big problem0.104 0.0450.0120.1981.0011,8959987
Wealth indexPoor (ref.)
Middle0.089 0.051−0.1250.0531.0015,78412,589
Rich−0.263 0.092 −0.378−0.1481.0013,58912,485
Pregnancy complication signsYes (ref.)
No0.1360.0480.0510.2211.0021,44819,874
Media exposureNot at all (ref.)
Less than Once a week−0.045 0.053 −0.1790.0891.0012,54812,478
At least Once a week−0.241 0.054−0.346−0.1361.0012,45712,478
Almost all−0.150 0.069 −0.195−0.1031.0013,48713,154
Mother’s age15–19 (ref.)
20–24−0.101 0.112−0.1450.1391.0095349845
25–29−0.054 0.116−0.1680.1221.0078467256
30–34−0.103 0.121−0.1960.1281.0054876547
35–39−0.152 0.141−0.2350.1121.0045143984
40–440.1500.185−0.2240.1451.0065896847
45–49−0.3130.176−0.4870.1891.0054785894
Who decided on the respondent’s health careRespondent alone (ref.)
Respondent and husband−0.1290.055 −0.194−0.0641.0012,45712,254
Respondent and another person−0.182 0.069 −0.281−0.0841.0011,48611,345
Husband alone0.071.0540−0.0450.1481.0010,64810,245
OccupationNot working (ref.)
Managerial0.052 0.149−0.1150.1591.0098459632
Clerical0.061 0.066 −0.0680.1861.0054785148
Sales−0.173 0.316 −0.2460.0481.0064876245
Agricultural−0.293 0.168 −0.3750.0821.0034873987
Other0.1050.351−0.1540.3451.0045785986
σ ^ u 2 0.2830.064 1.0025843025
ICC (ρ) 0.079
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Share and Cite

MDPI and ACS Style

Chikako, T.U.; Bacha, R.H.; Hagan, J.E., Jr.; Seidu, A.-A.; Kuse, K.A.; Ahinkorah, B.O. Multilevel Modelling of the Individual and Regional Level Variability in Predictors of Incomplete Antenatal Care Visit among Women of Reproductive Age in Ethiopia: Classical and Bayesian Approaches. Int. J. Environ. Res. Public Health 2022, 19, 6600. https://doi.org/10.3390/ijerph19116600

AMA Style

Chikako TU, Bacha RH, Hagan JE Jr., Seidu A-A, Kuse KA, Ahinkorah BO. Multilevel Modelling of the Individual and Regional Level Variability in Predictors of Incomplete Antenatal Care Visit among Women of Reproductive Age in Ethiopia: Classical and Bayesian Approaches. International Journal of Environmental Research and Public Health. 2022; 19(11):6600. https://doi.org/10.3390/ijerph19116600

Chicago/Turabian Style

Chikako, Teshita Uke, Reta Habtamu Bacha, John Elvis Hagan, Jr., Abdul-Aziz Seidu, Kenenisa Abdisa Kuse, and Bright Opoku Ahinkorah. 2022. "Multilevel Modelling of the Individual and Regional Level Variability in Predictors of Incomplete Antenatal Care Visit among Women of Reproductive Age in Ethiopia: Classical and Bayesian Approaches" International Journal of Environmental Research and Public Health 19, no. 11: 6600. https://doi.org/10.3390/ijerph19116600

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop