Risk Factors Associated With Frequency of Antenatal Visits, Number of Items of Antenatal Care Contents Received and Timing of First Antenatal Care Visits in Ethiopia: Multilevel Mixed-Effects Analysis


 Background: An improved quality Antenatal care (ANC) from skilled providers is essential to pregnant women so that ensure the key health circumstances for mother and child during pregnancy. Thus, this study attempted to assess which risk factors are significantly associated with the timing of antenatal care, the number of antenatal care visits, and the number of items of antenatal care contents received from skilled providers in the recent pregnancy among mothers in Ethiopia. Methods: Data was extracted from the 2016 Ethiopian Demographic and Health Survey. A total of 4,685 mothers was included in the analysis. Multilevel mixed-effects logistic regression analysis and multilevel mixed Negative binomial models were fitted to find the factors associated with antenatal care utilization. A 95% Confidence Interval of Odds ratio/Incidence rate ratio excluding one was reported as significant association with timely initiation of the ANC, numbers of ANC visits, numbers of items of antenatal care contents received and predictor variables.Results: About 20% of the mothers initiated ANC within the first trimester, 36.3% visits at least four ANCs, 3.6% have visited at least eight ANCs, and only 53% received at least four items of antenatal care contents. Rural residence (IRR = 0.81; 95%CI: 0.80-0.89), wanted no more children (IRR = 0.83; 95%CI: 0.75-0.91), mother without decision making power (IRR = 0.90; 95%CI: 0.83-0.98) associated with reduced frequency of ANC visits, while being rural residents (IRR = 0.82; 95%CI: 0.75-0.90), wanted no more children (IRR = 0.87; 95%CI: 0.79-0.96), husband alone decision maker of mothers health care (IRR = 0.88; 95%CI: 0.81-0.96) associated with reduced items of ANC content received. Further, birth order six or more (IRR = 0.74; 95%CI: 0.56-0.96), rural residence (IRR = 0.0.41; 95%CI: 0.34-0.51), wanted no more children (IRR = 0.61; 95%CI: 0.48-0.77) associated with delayed antenatal care utilization. Moreover, higher household wealth status, primary or secondary or higher education of mothers and partner, listening to the radio and watching television at least once a week, and a short distance to health facility were positively significantly associated with the frequency of antenatal care visits, numbers of items of ANC contents received and early initiation of ANC visit for the recent pregnancy during the last five years before the survey. Conclusions: Rural residence, poorest household wealth status, no education level of mothers or partners, unexposed to mass media, unwanted pregnancy, mothers without decision making power, and big distance to the nearest health facility, have a significant impact in delaying the timing of ANC visits, reducing the number of ANC visits and items of ANC received in Ethiopia. We ought to timely initiate an ANC visit for a frequent antenatal care visit during pregnancy. Educating girls and encouraging mothers to use modern contraceptives in order to prevent unwanted pregnancies are vital ingredients that should be included in all policies aiming to reduce maternal and neonatal deaths through improved quality of antenatal care utilization.


Introduction
Maternal mortality reduction and enhancements in women's health care are priorities of the third Sustainable Development Goal (SDGs) aimed to reduce the global maternal mortality ratio (MMR) to 70 per 100,000 live births by 2030 [1]. Globally, every day an estimated 810 women died from causes related to pregnancy, where 94% of all these deaths occur in low and lower-middle income countries. Between 2000 and 2017 the global maternal mortality rate (MMR) reduced by 38% between 20000 and 2017 [2]. The maternal mortality rate in Ethiopia was 412 per 100,000 live births in 2016 indicating low utilization of health care services and poor quality of services [3].
An improved quality Antenatal care (ANC) from skilled providers is essential to pregnant women so that ensure the key health circumstances for mother and child during pregnancy [4]. Regardless of health care service utilization for To the best of our knowledge, while numerous studies concentrating on determinants of antenatal care use in Ethiopia have been published [32][33][34][35][36][37] they didn't provide the quality of the services used considering the number of ANC visits received according to the latest WHO recommendations, contents of ANC received, and timing of ANC visits. Thus, this study attempted to assess which risk factors are signi cantly associated with the number of antenatal care visits, the number contents of ANC received and the timing of ANC visits for the recent pregnancy or live birth among women in Ethiopia preceding the survey considering the regional variation (region-speci c random effects). The nds are necessary for policy planners and decision makers to nd gaps in the utilization of maternal health care services and their qualities. Further, it may help to evaluate the SDGs target of reducing maternal mortality to less than 70 deaths per 100,000 live births and newborn deaths to less than 12 deaths per 1,000 live births by 2030 and to take the speci c interventions [1].

Study setting, Data and Population
We used population based, nationally representative data from 2016 Ethiopian Demographic and Health Survey (DHS) [3].  [3].
Our analysis was based on the records of 4,685 (54.3 %) women who have complete information on numbers of ANC visits, the timing of rst ANC visits, and the contents of ANC visits and who gave birth in the ve years preceding the survey. The latest births were referred for all women (Figure 1).

Variables
This study has three response variables: Number of ANC visits during pregnancy (measured as the discrete the number of times that the mother received ANC from the skilled providers), Timing of rst ANC visits (binary outcome categorized into 1 if a mother starts her rst ANC visits within the rst trimester (early initiation, or 12 weeks after the onset of pregnancy) and 0 elsewhere) and the contents of ANC received during pregnancy (discrete outcome measured as numbers of items nationally recommended and recognized contents of ANC in Ethiopia that a mother has received during pregnancy). The standard guidelines for ANC in Ethiopia recognized that every pregnant woman should receive ANC from a skilled provider that comprised of iron supplements, intestinal parasite drugs, at least two doses of Tetanus Toxoid injections, malaria intermittent preventive treatment in pregnancy and health education on danger signs and complications during pregnancy; blood pressure measurement; urine tests; blood tests; health talk on prevention of mother to child transmission of HIV/AIDS and HIV/AIDS counseling, testing and collection of results. The composite index comprises a simple count of the number of items received from skilled providers during the ANC visit was created. The variable had a minimum value of zero indicating that the mother had not taken any item or didn't receive ANC and a maximum value of ten indicating that she has received all nationally recommended and recognized the contents of the ANC. The important explanatory variables explored from previously available literature and theoretical perspectives of maternal health care service are presented in Table 1.

Statistical methods of data analysis
The extracted data were analyzed using "R programming" version 4.0.3. Descriptive statistics such as percentage and frequency tables were used to tabulate the basic characteristics of the respondents. F-test based on analysis of variance (ANOVA) was used for examining the mean difference in numbers of ANC visits and the mean numbers of contents of ANC received. Additionally, the Chi-square test was done to assess the association between numbers of ANC visits, and the timing of the ANC and the predictor variables. Furthermore, the multilevel mixed-effects logistic regression was used to examine the determinants timing of rst ANC visits (Odds ratio (OR) with 95% Con dence Intervals (CI)) and multilevel mixed-effects count models used to examine factor associated with the number of ANC visits and the number of items of contents of ANC received from the skilled provider during pregnancy.
Meanwhile, our response variable number of ANC visits and numbers of items of the components of ANC received are count variables, Poisson regression model with a log link is the benchmark [38]. But, the most serious restriction of Poisson regression is that the assumption of equi-dispersion (it assumes the variance of the distribution is equal to its mean). In practice, most real data experience, overdispersion (variance can be larger mean or under dispersion (variance can smaller than mean) and produce invalid inference. We tested the equi-dispersion assumption using the likelihood test and if it violated, the Negative Binomial (NB) regression model was the alternative for data analysis [39]. Moreover, since, only 62% of mothers' had received ANC from a skilled provider, the data might be experiencing excess zero in this case, both the zero-in ated models (Zero In ated Poisson and Zero In ated Negative Binomial (ZINB)) and Hurdle models (Hurdle Poisson (HP) and Hurdle Negative Binomial (HNB)) were employed [40].
It is also important in addition to inconsistent regression coe cient, as overdispersion and excess zeros in Poisson regression (PR) and negative binomial regression (NBR), the correlation between measurements (intra-cluster correlation (ICC)) needs to be addressed. Thus, to solve the problems linked with ICC we used the multilevel mixedeffects models (cluster/region-speci c random effects) in the standard Poisson, Negative binomial, and zeroin ated models. The use of a multilevel modeling approach accounts for the hierarchical nature of the EDHS data and adjusts the inconsistent regression coe cient and its estimated standard errors for ICC.
Finally, based on the Vuong statistic [41], likelihood ratio test, the Deviance, AIC, and BIC for model comparison it was found the multilevel mixed-effects negative binomial best t the data and the incidence rate ratios (IRRs) with the corresponding 95% con dence interval were reported for each level of the explanatory variables. Variables with a 95% con dence interval for the risk ratio excluding one were considered as statistically signi cant determinant outcome variables (number of ANC visits and number of items of contents ANC received).

Results
Socioeconomic and demographic characteristics of respondents This study included 4685 mothers who have births, within the past 5 years preceding the survey. The background characteristics of mothers with timing of ANC visits are given in Table 2. Most mothers (70.3%) were from rural areas while only 29.7% of the mothers were from urban areas. Concerning regional administration, a slightly higher percentage of mothers were from Tigray (14.6%), SNNPR (13.3%), Oromia (11.2%), and Amhara (10.7%) while the smallest percent of mothers were from Afar (6.1%), Gambela (6.8%) and Harari (6.8%) regions. The higher percent of mothers were between 20 and 34 years with the median age of 27 years. Around 32 % of mothers were from the richest household, whereas about 23% of mothers were from the poorest household. A majority (49.5%) of mothers had no education at all, 32.8% had primary education and only 6.6% of mothers had a higher education level.
Regarding the media exposure of mothers only 2.8%, 16.3%, and 21.2% of mothers have read a newspaper or magazine, listen to the radio, and watch television at least once a week during their recent pregnancy respectively.
Whereas, the vast majority 88%, 67.6%, and 68.3% of mothers didn't read the newspaper or magazine at all, didn't listen to the radio at all, and didn't watch television at all during their recent pregnancy. On the other hand, concerning decisions on respondent's health care, about two-thirds (65%) of mothers reported both respondent and husband/partner usually decide on respondent's health care.
Further, about 26%, of the mothers reported they have a big problem of getting permission to go in seeking medical care for herself, whereas 50.1% of mothers have a big problem in getting money for treatment in seeking medical care for herself. Similarly, 44.2% of mothers reported they had a big distance to a health facility in seeking medical care for herself and 33.7% of mothers reported a big problem in not wanting to go alone in seeking medical care for herself. The majority (80%) of mothers was reported the last child was wanted at the time of pregnancy, whereas 14.7% of mothers reported the pregnancy was wanted later and 5.3% wanted no more.
Timing of rst ANC visit by some characteristics of mothers Only 20.1% of mothers initiated their rst ANC visits within 12 months of the onset of pregnancy with a median month for rst ANC visits of four months. The percentage of early initiating the rst ANC visits was low in the SNNPR (22.2%), Benishangul-gumuz (23.1%), and Somali regions (32.1%) whereas mothers from Dire Dawa (68.6%) and Addis Ababa (62.5%) cities have a higher percentage of early initiating their rst ANC. More than half (56.2%) of the mothers from rural areas initiated their rst ANC visits within the rst trimester. About 54% of mothers from the richest households initiate their rst ANC visits within the rst trimester compared to 30% of poor or poorer household mothers. Similarly, the percentage of early initiating of rst ANC visits was highest among mothers who have higher levels of education (63.6%) compared to uneducated mothers (31%) and those who have only primary education (40%). Further, mothers whose husbands/partners have higher education level have the highest (57%) of early initiating their rst ANC visit.
The percentage of early initiating rst ANC visit was almost uniform among mothers' age at pregnancy and their occupation. The majority 52% of mothers who reading newspaper or magazine at least once a week, 46% of mothers who listen to the radio at least once a week, and 58% of mothers who watch television at least once a week during their recent pregnancy started their ANC visits within the rst trimester. On the other hand, mothers' autonomy concerning decisions on their own health care increased the percentage of early initiating rst ANC visits.
The majority of mothers whose pregnancy was wanted (40%) early initiated rst ANC visits compared to those mothers wanted no more child (25%). Moreover, mothers reported not a big problem in getting money for treatment in seeking medical care for herself (43%), not a big distance to a health facility in seeking medical care for herself (42.3%), and not a big problem in not wanting to go alone in seeking medical care for herself (41%) have a percentage of early initiating rst ANC visits.
Number of ANC visits by some characteristics of mothers Table 3 depicts descriptive statistics the mean number of ANC visits, the mean number of ANC contents or items received, and the frequency of ANC visits by the mother's characteristics. The analysis showed that 62% of mothers have received at least one ANC visit from skilled providers and 36.3% received at least four visits from skilled providers. The proportion of mothers who received the recently WHO recommended at least eight ANC visits, was only 3.6%. The result reveals that mothers from Addis Ababa were more likely to have eight or more ANC visits and the higher mean of the number of ANC visits, while mothers from Benishangul-gumuz, Gambela, Somali and Oromia regions were less likely to have eight or more ANC visits and the lower mean number of ANC visits. The proportion of mothers having eight or more ANC visits was higher among urban mothers and has a higher mean number of ANC visits than their rural counterparts. Similarly, household wealth showed a signi cant inverse association with both the mean of ANC visits and the proportion of at least eight ANC visits. The proportion of mothers who have eight or more ANC visits was found to be 12.3% among the richest households, compared to 0.5% among the poorest households.
Educated mothers were more likely to have eight or more ANC visits and a higher mean number of ANC visits than uneducated mothers. The coverage of eight or more ANC visits was found to be 20.3% among mothers with higher education, compared to 1.2% among mothers with no education at all. Likewise, husbands/partner's education level showed a signi cant positive association with the mean of ANC visits and the percentage of mothers who had eight or more ANC visits. On the other hand, the distribution of the mean of ANC visits and the coverage of eight or more ANC visits was uniform among different mother's age groups. Mothers' autonomy of decision making about their own health care and exposure to media such as reading a newspaper/magazine at least once a week, listening to the radio at least once a week, and watching television at least once a week showed a signi cant positive association with the mean of ANC visits and mothers coverage of eight or more ANC visits. Mothers reported not a big problem in getting money for treatment in seeking medical care for herself, not a big distance to the health facility in seeking medical care for herself and not a big problem in not wanting to go alone in seeking medical care for herself showed a signi cant positive association with mean of ANC visits and the proportion of mother who had eight or more ANC visits.

Number of items of ANC service contents received by some characteristics of mothers
Of all mothers who received ANC at least once, 79.9% had their blood pressure measured, 73.8% had a urine sample taken, 79.5% had a blood sample taken, 46.3% had told about pregnancy complications, 45.4% received iron supplementations for at least 180 days, 5.8% received treatment for an intestinal parasite, 69.1% received counseling after tested AIDS, 53.7% informed about birth preparedness, 67% received nutritional counseling and 42.5% received two or more doses of tetanus toxoid vaccine from a skilled provider during their ANC visits ( Figure   2). Figure 3 presents the distribution of the number of ANC visits and the number of items of the ANC services received by mothers during ANC visits. The mean number of ANC visits was 2.63 visits and standard deviation 2.08, while the mean number of items received by the mothers was 3.5 items and standard deviation, indicating that the distribution is overdispersed. About 62% of mothers received ANC from skilled providers and 75% of mothers received at least one of the items of the ANC service contents and whereas, the rest 38% of mothers received no ANC services during their last pregnancy. Nearly 7% of mothers received only one item of ANC services and about 9% received eight items. The percentage of mothers' decreases with the increase in the number of ANC visits and the number of items of ANC services received. For example, 53% of mothers received at least four items of the ANC, while only 14% received at least eight from ten items.
Further, the results of the mean of the number of contents/items of ANC services received in Table 3 across characteristics of mothers indicate the timing of the rst ANC visit, and the numbers of ANC visits have a signi cant association with the number of items of ANC services received. The mean of items of ANC services received by the mothers during their ANC visits differ signi cantly across administrative regions, place of residence, household wealth, birth order, mother and partner's education, media exposure, wanted pregnancy, mother's decision on own health care, the problem in getting money for treatment in seeking medical care for herself, distance to the health facility in seeking medical care for herself and problem in not wanting to go alone in seeking medical care for own health. The frequency of ANC visits and early initiation of the rst ANC visit showed a signi cant positive association with the mean number of items of ANC services received. For instance, mothers initiated rst ANC within the rst trimester had received on average 6.2 items of ANC service, while mothers with only one ANC visits had received on an average 3.8 items of ANC services, compared to virtually 6 items on average among the mothers with four or more ANC visits. Factors associated with early initiation of the rst ANC visits: multilevel mixed-effects logistic regression analysis In the multivariable multilevel logistic regression, factors such as birth order of the child, place of residence, education level of the mother as well as husband/partner, household wealth, age of mother, frequency of listening to the radio and watching television, wanted last pregnancy and distance to health facility were found to be independently associated with the timing of rst ANC visit ( Table 4). The likelihood of early initiating rst ANC visit was lower among six or more birth order of children (AOR = 0.74; 95%CI: 0.56-0.96) as compared to the rst birth childbirth order. Moreover, rural mothers were less likely to be starting their rst ANC visit within the rst trimester (AOR = 0.41; 95%CI: 0.31-0.54) as compared to urban counterparts.
The odds of early initiation of rst ANC were also higher among mothers listen to the radio less than once a week (AOR = 1.56; 95%CI: 1.25-1.93), listen to the radio at least once a week (AOR = 1.49; 95%CI: 1.20-1.85) and watching television at least once a week (AOR = 1.58; 95%CI: 1.11-2.23) compared to those who didn't listen to the radio and watching television at all respectively. Pregnant mothers who want no more children were 39% (AOR = 0.61; 95%CI: 0.48-0.77), less likely to start the rst ANC visit within the rst trimester than an wanted child at the time of pregnancy. Furthermore, mothers reported no problem or not, a big problem in distance to a health facility in seeking medical care for herself was 55% (AOR = 1.55; 95%CI: 1.35-1.78) more likely to start rst ANC visit within the rst trimester. In contrast, variables such as sex of household head, mother's occupation, frequency of reading newspaper/magazine, decision making power on health care, the problem of getting permission to go in seeking medical care for herself, the problem in getting money for treatment in seeking medical care, and problem in not wanting to go alone in seeking medical care for herself had no signi cant effect on the timing of rst ANC (P > 0.05) after adjusting for other variables within the model (Table 4).

Factors associated with reduced numbers of ANC visits: multilevel mixed-effects Negative binomial analysis
The multivariable multilevel mixed effect negative binomial model with random effects presented in Table 4, indicated that birth order, place of residence, mother's as well as husband's education level, household wealth, women's media exposure such as listening to the radio and watching television, wanted last pregnancy, mother's decision making power on their own healthcare issues, mother's problem of getting permission to go in seeking their own medical care, and distance to a health facility in seeking own medical care have highly signi cant effects on the frequency of antenatal care use. The result indicated that the number of ANC visits a mother had received, reduced by 12% (IRR = 0.88; 95%CI: 0.80-0.97) for six or more birth order as compared to the rst birth. Likewise, mothers from rural were 19% (IRR = 0.81; 95%CI: 0.74-0.89) times less likely to have a higher number of ANC visits during their pregnancy as compared to urban mothers. As household wealth was concerned, compared to being from the poorest households wealth status mothers from the richest households were 1 1.04-1.37), mothers with no problem of getting permission to go in seeking their own medical care (IRR = 1.07; 95%CI: 1.01-1.14) no big problem of distance to health facility in seeking own medical care (IRR = 1.12; 95%CI: 1.06-1.19) had a signi cantly higher frequency of ANC visits, while mothers whose husband/partner alone make decision on healthcare issues (IRR = 0.90; 95%CI: 0.83-0.98), and mothers wanted no more children (IRR = 0.83; 95%CI: 0.75-0.91) and whose pregnancy wanted later (IRR = 0.94; 95%CI: 0.88-0.99) had a signi cantly lower frequency of ANC visits.
Factors associated with contents ANC service visits: multilevel mixed-effects Negative binomial analysis The multivariable multilevel negative binomial analysis of predicting the number of items of the content of ANC services received by a mother during her pregnancy showed factors such as place of residence, household wealth status, mothers as well as partners education status, mass media exposure such as frequency of listening to the radio and watching television, decision making power of mothers on their own health care, distance to the health facility, permission to go in seeking medical care for herself, wanted last pregnancy and frequency of ANC visits were independently associated with the incidence rate ratio (IRR) of numbers of items the contents of ANC services received from skilled providers.
The estimated IRR shown in Table 4

Discussion
This study used the 2016 Ethiopian Demographic and Health Survey data to examine the risk factors associated with reduced frequency of ANC, numbers of items of ANC contents received from skilled providers, and timing of rst ANC visits among mothers during their recent pregnancy in Ethiopia. The multilevel mixed-effects analyses were used to account for clustering. This study differed from previous conducted in Ethiopia [32][33][34][35][36][37] that it considered intra-regional/cluster correlation and used multilevel mixed effects analysis adjusting the inconsistent regression coe cient and its estimated standard errors for intra-cluster correlation. The study found an overall 62% of the mothers had received at least one ANC visit, 36.3% received at least four visits, and 3.6% received at least eight ANC from skilled providers respectively. On the other hand, 53% of mothers received a minimum of recommended at least four items of ANC services, while 20% of mothers started their rst ANC visit within the rst trimester.
For frequency of ANC visits as a response, the multilevel negative binomial regression analysis revealed that the covariates: higher birth order six or more, being rural resident, an unwanted child at the time of pregnancy and wanted no more children, lower household wealth status, lower education of mothers as well as her husband/partner, mass media unexposed, the big problem of getting permission to go in seeking medical care, big distance to the health facility in seeking medical care, and husband/partner alone as decision maker of mothers health care were signi cantly associated with the lower incidence rate ratio of numbers of ANC visits. For items of ANC contents received, being rural residents, lower household wealth status, mothers and partners with lower education, female household head, mass media unexposed, wanted no more children, the big problem of getting permission to go in seeking medical care, big distance to the health facility in seeking medical care, husband/partner alone as decision maker of mothers health care, and lower frequency of ANC visits were signi cantly associated with a lower incidence rate ratio of numbers of items of ANC services received. Further, for the timing of rst ANC visit as an outcome variable, birth order six or more, rural residence, lower household wealth status, mass media unexposed, mothers and partners with lower education, mothers age 40-44, wanted no more children, and a big distance to the health facility in seeking medical care was signi cantly associated with lower odds early initiating rst ANC for their recent pregnancy during the last ve years before the survey.
The result shows that the birth order of the child inversely associated with the timing of the rst ANC visit and frequency of ANC visits. Mothers are less probable to start rst ANC visit early and frequently receive ANC services to their sixth or higher birth order child. This is in line with a previous study in Uganda [42] that has found mothers with third birth order, compared to those with the rst, are about 6-7% less likely to attain the four antenatal visits, and mothers with at least the third birth order, are 4-5% times less likely to initiate the rst visit in the rst trimester.
Muchie [32] using Ethiopian Mini DHS also found that 38 and 36% lower odds of completing four or more visits of ANC utilization for birth order of child four or ve and six or more respectively.
Rural mothers are less probable to have a higher number of ANC visits and items of ANC contents from skilled providers and less probably start rst ANC visits within the rst trimester than the urban mothers. This nding is congruent with that of [36] reported higher odds of delaying rst ANC visits and ANC visits less than four among rural mothers, while [43] found rural mothers are 17% times less likely to attend a higher number of ANC visits compared with urban mothers. Similar ndings from Bangladesh [30] reported urban mothers were 1.16 times more frequent visits ANC and 1.35 times more likely received more items of contents of ANC services from a skilled provider than rural counterparts. The reason might be that in the rural areas of Ethiopia there is a lack of skilled health care providers, lack of information on antenatal care services, lack of infrastructure, and long distances from health facilities. Further, most mothers in rural Ethiopia were uneducated. Contrary to our ndings Gebremeskel et al [44] and Weldearegawi et al [28] reported place of residence was not associated with the timing of the rst ANC visit. This inconsistency might be due to the statistical methodology used and the smaller sample size used (n=409), whereas the EDHS 2016 used (n= 4685).
In addition, the result shows that the incidence rate ratio of the frequency of ANC visits and the number of items of ANC services received during pregnancy is higher among mothers from richest, richer, and middle household wealth status compared to the poorest household wealth. Similarly, mothers from the richest, richer, and middle household wealth status are more likely to initiate the rst ANC visit within the rst trimester compared to the reference group.
This nding is in line with that of [30,36]. Inconsistent to our ndings, poorer and middle wealth status did not signi cantly associate with the number of ANC utilization in Bangladesh [43], whereas the richer and richest wealth status did signi cantly increase the incidence of ANC utilization by 28% and 46% respectively. Furthermore, we found that the mothers and partner's education level is an important risk factor that signi cantly affects the antenatal care service utilization in Ethiopia. Mothers with at least primary education levels are more likely to start the rst ANC visit, frequently receive ANC visits, and receive the highest number of items of ANC contents from skilled providers. Similarly, mothers whose partners attained at least primary education are more likely to attain a higher number of ANC visits and items of contents of ANC services as compared to the uneducated category. Besides, a higher level of education of husbands associated with higher odds of early initiating rst ANC visit than a husband who attended no education. Further analysis of the 2011 Ethiopian DHS showed that mothers who had primary education were 79%, secondary education was 62%, and higher education was 45% times less likely to delay initiation of ANC visits, while mothers who had primary education were 1.77 times, secondary education was 3.41 times, and higher education was 2.96 times more likely to have at least four ANC visits for their recent birth [36]. Similar studies reported a nding of 1.22, 1.49, and 1.59 incidence rate ratio of the frequency of ANC visits among mothers with primary, secondary and higher education in Bangladesh [30]. Consistent to our nding [30] also found that mothers with primary 1.12 times, secondary 1.26 times, and higher education 1.39 times more likely to receive higher numbers of items of ANC contents in Bangladesh. Contrary [43], found having a secondary or higher education did signi cantly increase the incidence frequency of ANC utilization by 35% and 63%, respectively, while a primary education did not signi cantly increase the incidence of utilizing the ANC service in Bangladesh.
Primary level education of partners' has not signi cantly increased the incidence of receiving the contents of ANC services in Bangladesh, whereas partners having a secondary or higher education did signi cantly increase the incidence of the contents of ANC services mothers received. But the number of ANC visits increases with primary, secondary, and higher education level partners compared to uneducated [30]. In contrast, Ghana [45] found no signi cant effect of husband/partner level education on the timing of ANC visits. Additionally, a systematic review analysis in sub-Saharan Arica found that husband education signi cantly associated with uptake, frequency, and timing of ANC visits [46]. This result could be due to the fact that educated mother has better access to information, make decisions on own health care and could empower them to exercise, and able to change traditional attitudes of utilizing the ANC service as compared to uneducated. This study suggests that there is an urgent need to focus on mothers' education. The study also showed the need of advocating primary education for girls not to drop out from school and encourage them to pursue their secondary or higher education is essential in order to attain a tangible change to achieve the maternal and infant mortality reduction of the sustainable development goals through effective implementation of maternal health care services [32,47].
The result suggests that mothers frequently listening to the radio and watching on television at least once a week have a higher incidence rate ratio of frequently visiting ANC and receive more items of the content of ANC services.
Additionally, mothers frequently listening to the radio and watching television at least once a week are more likely to early initiate the rst ANC visit than mothers didn't listen to the radio and watch television at all. This result is in agreement with ndings of Yaya et al [36] where mothers listening to the radio at least once a week 1.29 times more likely to visit at least four ANC services as compared to not listening to the radio at all, and those mothers watching television were 2.29 times more likely to receive a minimum of four ANC visit and 40% less likely to delay their rst ANC visit compared to those didn't watch television at all. But they didn't nd an association between listening to the radio and the timing of the rst antenatal care visit. This variation might be that the difference in the methodology used. In Bangladesh [30,43] mass media exposure at least once a week was associated with increased numbers of utilizing ANC and items of ANC contents used.
Mothers whose pregnancies are unwanted or wanted later were less likely to have frequently visited and received the highest number of items of contents of ANC services. Similarly, mothers whose pregnancies, unwanted were less likely to start the rst ANC early compared to wanted pregnancies. A similar study of the Bangladeshi DHS found that wanted pregnancy was associated with a higher incidence of receiving higher items of ANC contents [30]. Another study from Bangladesh [43], Southern Ethiopia [44], Bahir Dar [29], and Eastern Tigray [28] found unwanted pregnancy signi cantly associated with delayed initiation of ANC service utilization. Further, [48] found that mothers whose last pregnancy wanted later were less likely to have a higher frequency of ANC visits. This might be mothers with unwanted pregnancies have anxiety and poor psychological wellbeing [49] and less attention to pregnancy linked complications and do not use supplements such as folic acid, vaccinations, health information, and nutritional counseling [50]. Thus, mothers ought to be encouraged to use modern methods of contraceptives in order to prevent unwanted pregnancies.
Furthermore, mothers' decision making power of her own health signi cantly associated with the frequency of ANC visits and items of contents of ANC services received. Mothers without decision making power or whose husband/partner alone makes the decision on her own health care strongly associated with a lower frequency of ANC visits and a lower number of items of contents of ANC services received. This result was congruent with those of studies in North West Ethiopia [51], Bangladesh [43], systematic review sub-Saharan Africa [46], and Tanzania [52]. Unlike our ndings, [53] found that decision making on own health care seeking was not signi cantly associated with the timing of the rst ANC visit.
Mothers with no big problem of distance from health facility had better odds of early initiating rst ANC visit; receive high numbers of the ANC and items of contents of ANC services. Similar ndings were found in a study in Bahir Dar, Ethiopia [29]. In the Eastern Tigray zone Northern Ethiopia, distance to the nearest health facility was not signi cant predictors of late antenatal care follow up [28], and in Rwanda, distance to the health facility was not signi cant predictors of poor quality of antenatal utilization [54]. Likewise, mothers who had permission to go in seeking medical care for her own are more likely to frequently visited ANC and received more items of ANC contents.
Lastly, the results indicate that the number of ANC visits and timing of the rst ANC visit during pregnancy positively associated with the number of items of ANC contents mothers received from skilled providers. Mothers early initiated rst antenatal care were more likely to receive a higher number of items of contents of ANC services compared to mothers late initiated rst ANC visit. Likewise, the numbers of items of contents of ANC services monotonically increase with frequent ANC visits. The nding is in line with that of [55,56].

Conclusions
Findings of this study suggest that rural residence, poorest household wealth status, no education level of mothers or partners, unexposed to mass media, unwanted pregnancy, mothers without decision making power, and big distance to the nearest health facility, have a signi cant impact in delaying the timing of ANC visits, reducing the number of ANC visits and items of ANC received in Ethiopia. Further, timely initiation of the ANC and the number of ANC visits was signi cantly associated with the increase in the number of items ANC a mother has received during pregnancy. This study recommends that for the quality of ANC received from skilled providers, we ought to timely initiate ANC visits and frequent antenatal care visits during pregnancy. Another implication of this study is that education of girls empowerment particularly in the rural areas, encouraging mothers to use modern contraceptive in order to prevent unwanted pregnancies, expansion of health education on media, and expansion of health facilities are vital ingredients that should be included in all policies aiming to reduce maternal and neonatal deaths through improved quality of antennal care utilization. Moreover, mothers in the low economic level and rural residences should be given special emphasis. We are grateful to ICF macro (Calverton, USA) for providing the 2016 DHS data of Ethiopia.

Declaration
Author, declare that the research is our original work, and all sources of materials used have been duly acknowledged.
Authors' contributions TAB involved from the inception to design, acquisition of data, analysis and interpretation, drafting the manuscript, BT involved from the inception to design, acquisition of data, analysis and interpretation and drafting the manuscript and edit the manuscript for the nal submission. Both authors read and approved the nal manuscript.

Funding
The author has no support or funding to report.

Availability of data and materials
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Ethics approval and consent to participate
Ethics approval for this study was not required since the data is secondary and is available in the public domain. Anyone can download datasets after registered as a DHS data user at https://dhsprogram.com › data › new-userregistration for a reasonable request.

Consent for publication
Not applicable. Problem in distance to health facility in seeking medical care for herself: 1 = big problem, 2 = no problem/Not a big problem Not wanting to go alone Problem in not wanting to go alone in seeking medical care for herself: 1 = big problem, 2 = no problem/Not a big problem * Professional/technical/managerial, clerical, unskilled manual, etc