A multilevel Bayesian Markov Chain Monte Carlo Poisson modelling of factors associated with components of antenatal care offered to pregnant women in Nigeria

Although improvements have been recorded in maternal and child health globally, most developing countries in sub-Saharan Africa (SSA) have continued to lag in the pace of this improvement [1]. Most maternal and child health indicators put the Low and Middle-Income Countries (LMIC) including Nigeria at the lower end of the scale [1‐3]. The Maternal and Child Health (MCH) situation in the SSA remained worrisome when compared to the advanced countries and some other developing countries. A childbearing woman in developed countries has only 1 in 5,400 chance of dying from pregnancy and childbirth-related complications compared with only 1 in 22 chances in SSA [1].

To facilitate the attainment of Sustainable Development Goals (SDGs) on improved health and wellbeing [4], in its Strategy for Health 2016 to 2030, the UNICEF envisioned “a world where no child dies from a preventable cause and all children reach their full potential in health and wellbeing” [5]. Efforts focused on achieving this vision birthed the goal to “end preventable maternal, newborn and child deaths” [5]. This underscores the need to ensure guaranteed maternal and child health without inequality at any level. Antenatal care (ANC) is recognised worldwide as the main approach to ensure safe maternal and infant health outcomes. Compliance with ANC protocol is important in the management of pregnancy conditions and is essential for safe delivery [6‐9]. The standards in the existing ANC protocol were reviewed in 2016 to reinvigorate the ANC program to guarantee the realization of the relevant goal of the SDGs by 2030 [9]. The reviewed WHO guideline on ANC utilization established three main pillars that can help maximize the gains and importance of ANC utilization. They are (i) initiation of first ANC contact in the first trimester of gestation (ii) having a minimum of 8 ANC contacts and (iii) receiving all components of ANC during the contacts with ANC providers. Therefore, it is not just sufficient to initiate ANC early and have at least 8 ANC contacts, a pregnant woman must receive all the components of ANC services. WHO therefore prescribed the components of ANC to achieve effective results all over the world [9]. Thus, among other reviewed standards, several services were packaged to make up for the minimum quality of ANC available to a pregnant woman. Emphasis was laid on the compulsory administration of specific components of ANC [9]. WHO identified and grouped the components to ensure standard quality of ANC into three (i) history taking, assessment, physical examination and laboratory tests, (ii) health promotion and education on nutrition, delivery, danger signs, and child health and (iii) care provisions such as tetanus toxoid immunization, and others [9‐12]. The specifics of these components vary slightly across countries.

Although evidence shows that ANC coverage has increased in Nigeria [13‐16], improvement in coverage alone cannot guarantee safe maternal and infant health outcomes [17, 18]. A Nigerian study showed evidence of an unacceptable quality of ANC with only about 1 in 8 pregnant women who had ANC contacts being provided with the minimum quality standard [17]. The study further emphasized that the poor quality of ANC among childbearing women might have contributed to the non-realisation of the defunct Millennium Development Goals (MDGs) on child and maternal health in Nigeria.

Early initiation of ANC, a sufficient number of ANC contacts, attention by skilled health workers, living in urban areas, better education, short distance to health facilities, media exposure, and household wealth quintile have been associated with the quality of ANC received by pregnant women in Nigeria [2, 17, 19‐21]. Older pregnant women, higher parity, higher level of education, and higher household wealth status were identified as factors associated with receiving a good quality of ANC by a similar study in Nepal [20]. Also, a Kenyan study pointed out that the level of women’s literacy, employment status, and receiving all ANC services from a health facility had a relationship with having a good quality of ANC by pregnant women [22, 23]. Studies have identified that the commonest components of ANC services received by pregnant women were blood pressure measurement, iron supplement, and urine test [17, 22]; while education on PMTCT was low in Nigeria as only 2 in 5 tested for HIV during ANC visit [17] and negligible proportion had an ultrasound performed.

While most studies focused on the time of initiation of ANC contacts and the minimum number of ANC contacts made, such emphasis was rarely placed on the components of ANC offered to women during the ANC contacts. Whereas, up-to-date information on the components of ANC services could provide a good understanding of the level of compliance with the recommended components and standards of ANC in Nigeria and could provide relevant evidence-based information necessary to inform policy on the implementation of the WHO ANC guidelines. This study is also a response to the United Nations' call for periodical monitoring and evaluation of how close countries of the world are to attaining the specific targets of the Sustainable Development Goals (SDG) [4]. The call also encouraged sub-country analysis. Our study shows how close Nigeria is to SDG-3 target 3.1: “By 2030, end preventable deaths of newborns and children under 5 years of age, with all countries aiming to reduce neonatal mortality to at least as low as 12 per 1,000 live births and under-5 mortality to at least as low as 25 per 1,000 live births” [4]. The services received during ANC contacts are of paramount importance to the attainment of SDG-3. Therefore, this study assessed the number of components of ANC received by pregnant women in Nigeria and its associated factors. We also assessed state-level differentials and inequalities in the quality of ANC care across the states in Nigeria and examined the clustering of the ANC components (or its lack) to inform policy intervention and research translation. The study provided recommendations on the pathways for the improvement of maternal and child health outcomes in Nigeria.

The data was collected from a cross-section of women selected using two-stage probability sampling across the 37 states in Nigeria. In all, 11,867 women provided information about the ANC components received during ANC visits for their most recent pregnancy during the 5 years preceding the survey. According to the owners of the data, in the households interviewed, 42,121 women aged 15–49 years were identified for individual interviews; interviews were completed with 41,821 women, yielding a response rate of 99.3% with 99.2% in urban areas and 99.4% in the rural areas [15]. However, we do not have any information on the population composition of pregnant women in Nigeria. The availability of this would have enabled the assessment of the level of representativeness of the respondents. During the last pregnancy, 94% had their blood pressure measured, nine of every ten women (91%) received a tetanus injection, 89% were given an iron supplement, 87% had a blood test, 86% had a urine test, 82% received counselling on HIV/PMTC, 80% were told about danger signs while only 24% got IPTp and only 22% received IPD (Fig. 2).

Figure 3 shows that only 5% of the respondents received all 9 components (Fig. 3). The commonest (35%) number of components received was 7. About 61% and 90% received at least 7 and at least 5 components respectively.

Table 1 shows the distribution of the respondents by their background characteristics and the prevalence of receiving all the 9 studied components. The prevalence of having the 9 components was highest (6%) among women aged 30 to 39 years compared with 2% among those aged 15 to 19 years. Nearly a quarter (10%) of the pregnant women with higher education who had at least one ANC contact received all the 9 ANC components while less than 2% of those with no education did. The proportion of all the 9 ANC components was lowest among women whose spouse alone decide health care access (3%), no media exposure (2%), Hausa/Fulani (2%), from households in the lowest wealth tertile (2%), North East (3%) North West (2%) and rural areas (4%) as shown in Table 1. All the explanatory variables considered were significantly (p < 0.001) associated with receiving all the ANC components received except whether the pregnancy was wanted or not and the type of ANC facility attended. Also, all characteristics in Table 1 were associated with having all 9 components, except marital status, sex of household head, whether the pregnancy was wanted or not, health insurance, and community media access. Having all the components was significantly higher among women from communities with low rates of illiteracy, poverty, and community socioeconomic disadvantage.

Figure 4 shows the map of Nigeria with the geographical distribution of the percentage of ANC attendees that received all nine components across the states. The highest prevalence of receiving all the 9 studied components was highest in Abia (32%), none (0%) in Zamfara, Borno (0.1%) and Kebbi (0.3%). Benue (8.3%), Kwara (8.6%) and Adamawa (9.3%) are the only states in the North with at least 5% women receiving all 9 components. The only five states with 10% or higher were from the southern part of Nigeria. The distribution of having all ANC components received during the most recent pregnancy by States and regions in Nigeria is shown in Supplementary Table A.

The overall prevalence of receiving all the 9 components of ANC received during ANC contacts in Nigeria is poor. There is a need to ensure that all pregnant women receive an adequate and optimal number of ANC components. Different factors influenced the number of ANC components received. Notable among them are the institution and skill of the ANC provider, education, and social-economic status of the women. The study shows a strong relationship between the timeliness of ANC initiation and having a minimum of 8 ANC contacts as they both increased the adequacy of ANC components received during the contacts. Although there are some states with high ANC utilization but inadequate ANC components. Our study suggested that early initiation of ANC during the first trimester and having the recommended number of ANC contacts are critical to having all ANC components, which in turn ensures that a pregnant woman and her unborn child take optimal advantage of ANC services.