Regional Differences in Postnatal Care Service Utilisation and Child Morbidity in Nigeria: Evidence from 2018 Nigeria Demographic and Health Survey (NDHS)

Data for this study was extracted from the Nigeria Demographic and Health Survey (NDHS) 2018 birth recoded le dataset, with a sample size of 30713 women (aged 15–49). Data were analysed using descriptive statistics, Chi-Square Test and logistic regression. The main predictor variable was the region, while others included the type of PNC service utilised and various demographic variables of the respondents – age, education, type of residence, religion, ethnicity among others.


Introduction
Due to the changes that are bound to occur during the period of postnatal care (0-6 months after birth) in the lives of mothers and their new-born babies, the World Health Organisation (WHO) has reiterated that the period is a critical phase in the lives of the mothers and the new-born babies [1]. Following the critical nature of this period, the Nigeria Demographic and Health Survey (NDHS 2018) report has also shown that the continuous provisions of health care services during pregnancy and childbirth including after delivery are important for the survival and well-being of both the mothers and their infants. As the report succinctly puts it: "Ensuring access to a continuum of care for women during the antenatal, intrapartum, and postpartum periods is critical for maternal and newborn survival and is a priority of the Federal Ministry of Health" [2].
Despite the importance of providing quality health care services by skilled health care professionals during and immediately after childbirth to mothers and infants, the postnatal periods are usually neglected by most mothers due to sociocultural factors associated with new-borns [3][4][5]. Meanwhile, studies have shown that mothers who attend health facilities during postnatal periods where skilled health professionals can identify, observe and manage any health challenge that may arise in the lives of the mothers and their new-born babies [6][7] stand a chance of preventing the occurrences of body and mental damages [8], as well as child morbidity and mortality than those who did not [9].
Although there has been incredible progress in the global under-ve mortality rate which had reduced from 93 deaths (in 1990) and 76 deaths (in 2000) to 39 deaths per 1,000 live births in 2018 [10], the reduction in this rate is uneven across regions and the Low and Middle-Income Countries (LMICs) of the world. For example, while the death of children under-ve has declined from 182 to 78 deaths per 1000 live births in sub-Saharan Africa between 1990 and 2018, Europe and Northern America recorded a decline from 14 to 6 deaths during the same period [10]. This implies that the current mortality rate in sub-Saharan Africa is 13 times more than those in Europe and Northern America.
In Nigeria, it was reported that between 1990 and 2018, the under-5 mortality rate has declined from 193 deaths to 132 deaths per 1,000 live births [2]. It then means that the under-5 mortality rate as of 2018 is 3 times higher than that of the global one and 2 times higher than that of sub-Saharan Africa. While the trend in the deaths of children under-ve seems to decline at the national level in Nigeria, the rates are not evenly distributed across all regions or geopolitical zones in the country. For example, the rates across the geopolitical zones as at 2018 ranged from 95 deaths per 1000 (North-Central), 134 deaths per 1000 (North-East), 184 deaths per 1000 (North-West), 25 deaths per 1000 (South-East), 73 deaths per 1000 (South-South) to 62 deaths per 1000 (South-West) [2]. Given that the rates of decline in under-ve mortality are not evenly distributed across continents and regions, it is indicative that there are disparities in the rates of socioeconomic development as well as people's attitudes towards prevention of child morbidity and mortality in the country.
While several studies on postnatal care have focused on the use, non-use and barriers to the use of modern health care services during this period [4,5], there is limited or little attention paid to the relationship between regional differences in postnatal care services and child morbidity which is a critical factor in under-ve mortality. Take, for example, Olajubu et al. [5] examined the predictors of postnatal care utilisation among women in a facility-based study in Nigeria, it was found that age and education were strong predictors of postnatal care services utilisation.
Ugboanusi et al. [4] on the other hand, studied factors affecting the utilisation of postnatal care services in Primary Health Care (PHC) facilities in urban and rural settlements in Kaduna (Nigeria), postnatal care service utilisation was lower in the rural areas than the urban centres and there were barriers of transportation, large babies and poor wealth quintiles to the use of PHC facilities. Rao et al. [11] in their study in India pointed out that mothers of new-borns have to bathe water with ashes within seven days after delivery. These practices may be good for the mothers and their new-borns from their cultural viewpoints, but the health implications could be more deleterious than its merits in modern health perspectives.
Meanwhile, Defar et al. [12] have argued that maternal and child health (MCH) care utilisation often varies with geographic location. As such, there is a need for an empirical study examining analysing regional differences in the use of postnatal care services and child morbidity using the recent data from the 2018 NDHS in Nigeria. This is urgently needed to explore the incidences of child morbidity in the country for necessary health programmes and interventions as well as to achieve target 3.2 of the Sustainable Development Goals, which seeks to end preventable deaths of new-borns and children under 5 years … to at least as low as 12 and 25 per 1000 live births. [13] Material And Methods adapted to re ect the population and health issues relevant to Nigeria. In speci c terms, the birth recode dataset which focussed on mothers whose children were between the ages 0-59 months to ascertain the predictive in uence of the postnatal care service utilised by mothers during postnatal periods and child morbidity experience in the last ve years that preceded the survey was used.
In this study, however, the birth recode dataset were extracted where individual women questionnaire was used. The women's questionnaire was used to collect information from all eligible women age 15-49. These women were asked questions on the following topics: background characteristics [including age, education, and media exposure]; birth history and child mortality; knowledge, use, and source of family planning methods; antenatal, delivery, and postnatal care; vaccinations and childhood illnesses [child morbidity]; breastfeeding and infant feeding practices; women's minimum dietary diversity; women's work and husbands' background characteristics, etc. However, for this study, only the socio-demographic characteristics of women, birth history, postnatal care information and child illness [child morbidity] that were extracted for the analysis in this study.

Sample and sampling techniques
The total number of women captured in the dataset was 127,545 of the NDHS. Of this number, a total number of 30,713 mothers whose children were under the age of ve were selected as the sample size for this study. The sample for the 2018 NDHS was a strati ed sample selected in two stages. In the rst stage, 1,400 EAs were selected with probability proportional to EA size. In the second stage's selection, a xed number of 30 households were selected in every cluster through equal probability systematic sampling, resulting in a total sample size of approximately 42,000 households. The household listing was carried out using tablets, and random selection of households was carried out through computer programming. Details of the sampling techniques can be found in the nal report of 2018 NDHS.

Variable Used in the Analysis
The Outcome Variables The outcome variables of interest in this study were: (i) postnatal care services utilisation, and (ii) child morbidity experienced by mothers. The postnatal care services utilised by the respondents was measured by the place of delivery of their babies. This was captured by the facility where a respondent was delivered of a baby in the last ve years that preceded the survey. These facilities were ranged from home, respondent's home, other homes, other; public sector, government hospitals, government health centre, government post, other public sectors, private sector, private hospitals/clinics to other private sectors. These facilities were further re-categorised into traditional {0} or modern {1}. It is traditional if a respondent indicated that she was delivered of a baby either at home, respondent's home, other homes or any other place rather than a modern facility. And modern, if a respondent signi ed that she was delivered of a baby either in the public sector, government health centre, government health post, other public sectors, private sector, private hospitals/clinics, or any other private sector.
Child morbidity, on the other hand, was measured by the experience of any under-ve disease diagnosed. These were captured by series of questions which ranged from whether children under the age of ve had been diagnosed of diarrhoea recently; have had a fever in the last two weeks; have had a cough in the last two weeks; had short rapid breaths; had a problem in the chest or blocked or running nose and severe; or had mild or moderate anaemia level. The responses were all in either yes (1) or no (0). It is "yes" a respondent had experienced the disease diagnosed during the period indicated and "no" if the respondent did not have the experience of such disease of under-ve in the period. Respondents' responses were further re-grouped as no child morbidity {0} if there was no disease diagnosed with under-ve children, and child morbidity {1} if a respondent had ever had the experience of any disease.

The Independent/Explanatory Variables
The selection of the explanatory variables in this study was guided by Anderson-Newman's [14] Model of healthcare utilisation which proposed that the utilisation of any healthcare facility by an individual is determined by three factors, namely: the predisposing factors, enabling factors, and the need factor. However, only the variables related to the predisposing factors and the enabling factors were included in the analysis of this study. The predisposing factors are the socio-economic cum cultural traits of individuals that precede their health conditions and statuses such as region, age, educational level, and location. This study factored into the analysis the region of the respondent [South-West, South-East, South-South, North-West, North-East, North-Central], age, educational level [nor formal education, primary, secondary, tertiary], residence [rural/urban]. This study also factored in the ethnicity [Yoruba, Igbo, Hausa, The Minority] of the respondent, the number of children-born [0-3, 4-6, 7+] as the enabling factors for PNC services utilization and child morbidity; while the need factor is the most immediate cause of health care service utilisation. Table 1 presents the summary of the measurements and de nition of both the dependent and independent variables as used in the study.

Results
Socio-demographic characteristics of the respondents Table 2 presents the socio-demographic characteristics of the respondents, which ranged from the region (geo-political zones), age, residence, educational level, religion, ethnicity and the number of children everborn by the respondents. However, it was revealed that North West region has the highest proportion of the respondents (29.1%), followed by those in North East (21.1%), and North Central (17.6%), while among those in the South East region has the highest proportion of respondents (11.5%) in the southern part of the country when compared to other categories of respondents. Regional differences in PNC services utilisation in Nigeria Table 3 simultaneously examines the distribution of respondents by PNC services utilised and their regions while also exploring the association between regions, socio-demographic characteristics and PNC services utilised (traditional or modern).

Association between regions, PNC service utilisation and child morbidity in Nigeria
To determine the predictive in uence of regions and PNC service utilized by women of reproductive age on child morbidity, Table 4   This shows that there are variations in the prevalence of child morbidity across the regions in Nigeria with the North East having the highest odds. In the same vein, those who resided in the rural areas at the time of the survey are observed to be 1.2 times more likely to experience child morbidity than those in the urban centre. This suggests that the place of residence of the respondents can have a signi cant in uence on child morbidity.
In the second model when the socio-demographic variables of the respondents were factored into the analysis, the statistical signi cance in uence of the region on child morbidity persisted though with a drop in the propensity especially with those in North Central and North West which had no statistically signi cant in uence on child morbidity. On the rural-urban dichotomy, although statistically signi cantly associated with child morbidity, there is a drop compared to the rst model when socio-demographic characteristics of the respondents were not included in the model.
Further analysis revealed that the age of the respondents is statistically associated with child morbidity.
Those who are in the age categories of 25 and above are found to be less likely to experience child morbidity than those between 15-19 years. This may be largely due to the fact that as the mother advance in age, the management of under-ve children relative to morbidity increases with age which those in the lower categories of age may not have the capacity.
Relative to the in uence of residence on child morbidity, there is a statistically signi cant association between residence and child morbidity. Those who resided in the rural areas of Nigeria are 1.14 times more likely to experience child morbidity than those in the urban centre. The education of the respondents is indeed fundamental to child morbidity. While those who had attained primary educational level are 1.14 times more likely to experience child morbidity than those who did not attend any formal education, those with tertiary educational level are 19.3% less likely to experience child morbidity than those without any formal education.
Religious adherent is also found to be critical among the study population on child morbidity. Those who adherents of Islam are 1.11 times more likely to experience child morbidity than those who were followers of Catholic, while the traditionalists were 36.5% less likely to experience child morbidity when compared to those in the reference category. It then means that the risk of child morbidity is higher among the followers of Islamic religion than those in other categories of religion.
Ethnicity and child morbidity are also statistically signi cantly related. Those who are Igbo and Hausa, in particular, are 20% and 24.3% more likely to experience child morbidity than those in Yoruba. Although there is no statistically signi cant association between those in the minorities and child morbidity in Nigeria, yet variations existed in the experience of child morbidity in the country.
On the account of the number of children a respondent has, there is a statistically signi cant association between those who had between four and six children, seven children and above and child morbidity in the country. For example, those who had between 4-6 children and 7 + children are 6.6% and 26% more likely to experience child morbidity than those who had between zero and three children. This suggests that as the number of children increases, the more likely a woman exposes to the risk of child morbidity in the country.

Discussion Of Findings
The study found that North West region has the highest proportion of respondents, followed by those in North East, and North Central with those in the South-South having the least proportion in the country. This simply suggests that women who were within the reproductive age in the northern part of Nigeria are far higher than those in the southern regions. It was also revealed that the respondents' age was 30 years. This is by implication suggesting that majority of those who were captured in the survey were young women who were also reproductively active enough to have children and may require postnatal care services.
Finding also indicated that majority of the respondents resided in rural areas while only a few of the respondents were urban residents. This simply means that there are more rural areas in the country than the urban centres which may inform the nature of postnatal care facilities that will be available for the teeming population of mothers in the country. Findings on the educational attainment of the respondents showed that almost half of them did not have formal education, while the highest proportion of the respondents that had attended formal education was those who had secondary school certi cate compared to just one in ten who had tertiary educational attainment. This further explains the fact that the majority of the women in the survey were not residing in the urban centres where there are diverse opportunities to attain higher educational level than the rural areas.
It further revealed that majority of the respondents were adherents of Islam. This is true of the fact that majority of the respondents were from northern Nigeria depicting the fact that majority of the adherents of Islam in the country are Hausa who are also domiciled in the northern part of the country when compared to the southern part of Nigeria where there are more adherents of Christianity.
Earlier studies have shown that there are variations in the use of PNC services by regions [12] based on sociocultural differences and some factors affecting PNC service utilization in some facilities [4,11,14].
This study also found that PNC services utilization varied signi cantly by regions and socio-demographic characteristics with those in the North East and North West using more of the traditional facilities more than the modern health care facilities for PNC services. The variations in the use of modern and traditional PNC services with the North East and North West having higher percentages of traditional PNC service utilization than the modern PNC facilities could be attributed to some forms of barriers identi ed by Ugboanusi and colleagues [4] that there are barriers of transportation and poor wealth quintiles to the use of PHC facilities.
Takai et al. [15] found that there are factors affecting the under-utilisation of modern PNC services which Somefun and Ibisomi [14] succinctly mentioned that distance could be a barrier. Hence, mothers may not have an option other than the use of traditional facilities at their disposal.
Findings revealed that there is signi cant variation in the likelihood of child morbidity by regions in Nigeria.
While the ndings indicated that those in the South-South, South East, North Central and North West are 1.5, 1.5, 1.3 and 1.3 times more likely to experience child morbidity than those in the South West region respectively, mothers in the North East are two times more likely to experience child morbidity than those in the South West. This suggests that the type of PNC facilities a mother uses for both ANC and PNC services are strong predictors of child morbidity, which consequently accounted for the increase in child deaths per 1000 live births as demonstrated in the report of NPC and ICF International [2] in Nigeria.
It was also revealed that the socio-demographic characteristics of the respondents predicted child morbidity in Nigeria. This nding corroborates the work of Sotunsa et al. [7] who found that there is a signi cant relationship between postnatal care utilization and maternal and child mortality in a population. This means that this can affect child morbidity. It further explains the Anderson-Newman Model [13] that sociodemographic factors of the respondents are the predisposing factors of health condition that in uence the use of PNC service utilization.

Conclusion And Recommendations
Following the logic of this research, it can be deduced that the occurrence of any disease in Nigeria population among the children under the age of ve varied signi cantly by regions as well as the nature of PNC service utilized. Although other variables such as the socio-demographic characteristics of the respondents in uenced the variation to some extent, it is quite observable that the variations that existed in PNC service utilization by regions have greatly in uenced the disparities in the prevalence of child morbidity across the regions in the country. As such, there should be region-speci c programmes of action that will facilitate the use of modern PNC service utilization across all the regions.
Given the ndings of the study, the following are recommended for the health policymakers to take action: There should region-speci c sensitization of women on the need for the use of modern health care for both antenatal care and postnatal care services to reduce the high prevalence of child morbidity across the regions.
The government and other stakeholders in the health sector should assist the provision of adequate modern health care facilities across the regions while laying more emphasis on maternal and child health care facilities.
The provision of Primary Health Care (PHC) facilities should be encouraged in all communities as this remains the closest health care facilities for easy accessibility for mother and children to combat child morbidity that informed child deaths. Percentage distribution of child morbidity by regions in Nigeria