Menu
Menu
Menu
Menu
Background: Poor maternal health delivery in developing countries results in more than half a million maternal deaths during pregnancy, childbirth or within a few weeks of delivery. This is partly due to unavailability and low utilization of maternal healthcare services in limited-resource settings. The aim of this study was to investigate the access and utilization of maternal healthcare in Amansie-West district in the Ashanti Region of Ghana. Methods: An analytical cross-sectional study, involving 720 pregnant women systematically sampled from antenatal clinics in five sub-districts was conducted from February to May 2015 in the Amansie-West district. Data on participants’ socio-economic characteristics, knowledge level and access and utilization of maternal health care services were collected with a structured questionnaire. Odds ratios were estimated to describe the association between explanatory variables and maternal healthcare using generalized estimating equations (GEE). Results: 68.5, 83.6 and 33.6% of the women had > 3 antenatal care visits, utilized skilled delivery and postnatal care services respectively. The mothers’ knowledge level of pregnancy emergencies and newborn danger signs was low. Socio-economic characteristics and healthcare access influenced the utilization of maternal healthcare. Compared to the lowest wealth quintile, being in the highest wealth quintile was associated with higher odds of receiving postnatal care (adjusted odds ratio [aOR]; 95%CI: 2.84; 1.63, 4.94). Use of health facility as a main source of healthcare was also associated with higher odds of antenatal care and skilled delivery. Conclusion: This study demonstrates suboptimal access and utilization of maternal healthcare in rural districts of Ghana, which are influenced by socio-economic characteristics of pregnant mothers. This suggests the need for tailored intervention to improve maternal healthcare utilization for mothers in this and other similar settings.
The details of the methods of the study have been described elsewhere [21]. An analytical cross-sectional study was conducted from February to May 2015 in the Amansie-West district of Ghana. The district is one of the most deprived districts in the Ashanti region and uniformly rural. It had a population of 149,437 and an annual growth rate of 2.7% as of 2014. The health system in the district is very weak, low health staff-to-patients (1: 74); doctor to population (719); nurse to population (1:2, 767) and midwife to women in reproductive age (WIRA) (1:4528) [22] ratios. The study population was defined as confirmed pregnant women from 4 to 9 months. Seven hundred and twenty (720) pregnant women were systematically sampled from the various ANCs. The sample size was calculated with recourse to Cochran [23]; n=Z2p(1-p)d2. Where; n = the sample size Z = the number relating to the degree of confidence anticipated in the result; in this case 95% confidence interval (Z = 1.96 which is the abscissa of the normal curve). p = an estimate of the proportion of people falling into the group in which we are interested clients of health care, where q = 1-p d = proportion of error we are prepared to accept (sampling error; 5% anticipated error). n = 1.96 2 × 0.50 (1–0.50) ÷ 0.04 2 Due to attrition and incomplete data, an extra 20% (120 women) was added leading to a total of 720 respondents. The participating ANCs were selected from five of the 10 sub-districts in the district. The required respondents from selected health facilities were proportional to the size of total eligible population per community. The distribution of respondents according to the sub-district was Manso Nkwanta 120, Edubia 141, Agroyesum 114, Antoakrom 140 and Esuowin 205 (Additional file 1: Table S1). At the five selected sub-districts, systematic random sampling technique was employed to select respondents from ANCs of private and public hospitals and health centers. This was guided by the sampling interval, K, estimated as the required sample size divided by the total attendants per facility. During the visit hours, a first participant was identified and interviewed as the starting point and the Kth respondent is approached, starting the count at the selected starting participant. This was repeated until the required sample size was attained All participants involved in the study signed an informed consent form after explaining the objectives of the study. Participants had the right to withdraw from the study at any point in time during the data collection process. Data on respondents’ socioeconomic characteristics, access, and utilization to maternal health care services were collected using structured questionnaire after checking for clarity, consistency, and acceptability by pretesting. Data entry and analysis were done with SPSS for Windows (version 22) [24]. The outcome variable was maternal healthcare during the previous pregnancy, defined as ANC visits, skilled delivery, and PNC during previous pregnancy. The explanatory variables were socio-economic characteristics (age, education, religion, marital status, employment status, number of children, household wealth), access to healthcare (valid health insurance, proximity to health facility), healthcare seeking behavior (breastfeeding, use of family planning, preference of healthcare) and knowledge about pregnancy and danger signs. We calculated household wealth index with recourse to the procedure adopted in the demographic and health study (DHS) [25], using a simplified data on a household’s ownership of selected assets, such as televisions, bicycles, and farmlands. Scores were assigned to assets by using a Principal Component Analysis (PCA) and then standardized before grouping into quartiles. Details of the study variables are shown in Table 1. Univariable associations were tested using Chi-squared test and student t-test for categorical and continuous or discrete variables respectively. The influence of the explanatory variables on the odds of antenatal care, skilled delivery and postnatal care was estimated using generalized estimating equations (GEE) [26]. This helped to address the possible correlations of data within clinic groups. All statistical tests were performed at a significance level of p < 0.05. Study variables NHIS National Health Insurance Scheme, JHS Junior High School
