FACTORS ASSOCIATED WITH USE OF MATERNAL HEALTHCARE SERVICES AMONG WOMEN WHO HAD HOME DELIVERIES AMID BOKO HARAM INSURGENCY IN NORTH-EAST NIGERIA

 

Bola Lukman Solanke

Department of Demography and Social Statistics, Obafemi Awolowo University, Ile-Ife, Nigeria, modebolasolanke@gmail.com, bsolanke@oauife.edu.ng

 

 

Abstract

Background:

Since 2009, North-East Nigeria has been undergoing tragic Boko Haram insurgency that has not only impacted negatively on every aspects of social life in the geo-political zone, but has also manifested serious implications for public health. The persistence of the insurgency not only undermines access, quality and use of maternal healthcare services in the region, it also elevates the risks of maternal morbidity and mortality. How the Boko Haram insurgency interacts with social factors in the North-East context, to affect use of maternal healthcare services needs further investigation. Previous studies have examined a range of factors associated with use of maternal healthcare services. Most of these studies identified individual and community factors. However, none of the Nigerian studies has specifically examined whether amid the Boko Haram insurgency, the individual and community factors are still important for explaining the use of maternal healthcare in North-East Nigeria. This study attempts to fill this knowledge gap by raising the research question: to what extent are individual and community characteristics still important for explaining use of maternal healthcare services amid Boko Haram insurgency in North-East Nigeria among women who had home deliveries in North-East Nigeria? This group of women was the target of the study because home delivery is the dominant form of child delivery in Northern Nigeria.  

Methods:

The data analysed in the study were extracted from the 2013 Nigeria Demographic and Health Survey (NDHS). The weighted sample size was 2,729 women of reproductive age. The outcome variable in the study was use of maternal healthcare services, measured by antenatal care attendance and postnatal check-up. Antenatal care attendance was grouped into two categories of women who reported four or more antenatal care attendance before delivery, and group of women who did not report any antenatal attendance or reported less than four antenatal attendances. Postnatal check-up was divided into two categories of women who ‘received postnatal check-up’ and women who ‘did not receive a postnatal check-up’. Two sets of explanatory variables, namely individual and community characteristics are analysed in the study. The individual characteristics are maternal age, parity, work status, religious affiliation, education, age at first birth, and current marital status. The community characteristics analysed in the study are community education (proportion of illiterate women in the community), community poverty concentration (proportion of women in poorest households), community perception of distance to health facility (proportion of women who perceived distance to facility as not a barrier to healthcare), community media exposure (proportion of women who reads the newspaper, listen to radio or watch television at least once per week), residence type (urban or rural), and state of residence. Data were analysed using Stata 12. The multilevel mixed effects logistic regression was used to investigate the association between the research variables. The multilevel model with three replications in addition to an empty model was fitted in the study. The multilevel analysis was performed differently for the two measures of maternal healthcare use. Model 1 was based solely on individual characteristics, while Model 2 combined individual and community characteristics. Model 3 was the full model which included the explanatory and control variables. The fixed effects of the multilevel model were measured by odds ratio of logistic regression, while the random effects were measured by Intra-Class Correlation (ICC).

Results:

More than two-thirds of respondents did not attained four antenatal care attendance before delivery; however, slightly less than one-third of respondents had four or more antennal care attendances. The majority of respondents did not receive postnatal check-up after delivery. Based solely on individual level variables (Model 1), women’s work status, education, and marital status are the individual characteristics with significant association with antenatal attendance. With inclusion of the community level variables in Model 2, women’s work status, education and marital status remained significantly associated with antenatal attendance. In Model 3 (full model), with the exclusion of women’s work status, the individual level variables did not reveal significant association with antenatal attendance. Two community level variables, namely, community media exposure and community residence type did not reveal significant association with antenatal care visits. With respect to postnatal check-up, in Model 1, women’s work status and education are the individual characteristics that significantly associate with postnatal check-up. In Model 2, women’s education was the only individual characteristic that remained significantly associated with postnatal check-up. Community media exposure and State of residence are the two community level variables that significantly associate with postnatal check-up in the model. In the full model, primary and secondary education remained significantly associated with postnatal check-up at the individual level, while community media exposure and State of residence remained significantly associated with postnatal check-up at the community level.                   

The random effects of the multilevel models for both antenatal attendance and postnatal check-up revealed that with the inclusion of individual characteristics in Model 1 for antenatal attendance, the community level variables accounted for 43.3% of total variation in antenatal attendance, while in Model 2, the variation attributable to the community characteristics reduced to 17.3%, and further to 16.2% in Model 3. Likewise, with the inclusion of individual characteristics in Model 1 for postnatal check-up, the community level variables accounted for 38.1% of total variation in postnatal check-up, while in Model 2, the variation attributable to the community characteristics reduced to 12.1%, and further reduce to 9.3% in the full model.

Conclusion:

Findings from the study revealed that many individual socio-demographic characteristics do not significantly affect use of maternal healthcare services in Boko Haram affected areas. However, community characteristics appeared to have more significance for use of maternal healthcare services particularly, antenatal care attendance. Individual and community characteristics are still important for explaining use of maternal healthcare services amid Boko Haram insurgency in North-East Nigeria.

Keywords: Boko Haram, maternal healthcare, North-East Nigeria