Menu
Menu
Menu
Menu
Despite the significant benefits of giving birth at a health facility to improve maternal and child health, the practice remains lower than expected in pastoralist communities of Ethiopia. Understanding the intentions of pregnant women to use health facilities for delivery predicts the adoption of the behavior, yet documented evidence of intention in the context of pastoralist populations remains scarce. The current study aimed to assess pregnant women’s intentions to use a health facility for delivery in the Afar region of Ethiopia using the framework of the health belief model (HBM). A community-based, cross sectional survey was conducted from April 1 to April 30 2016 among 357 randomly sampled pregnant women using an interviewer-administered, semi-structured questionnaire. Data were entered into EpiData and exported to SPSS version 20.0 for analysis. Principal component factor analysis was done to extract relevant constructs of the model, and the reliability of items in each construct was assessed for acceptability. Multivariate logistic regressions were applied to identify predictors of pregnant women’s intentions to give birth at a health facility. The odds ratio was reported, and statistical significance was declared at 95% CI and 0.05 p value. Three hundred fifty seven pregnant women participated in the study (104.6% response rate indicating above the minimum sample size required). Among the respondents, only 108 (30.3%) participants intended to use a health facility for the delivery for their current pregnancy. Higher household average monthly income [AOR = 1.23, 95% CI = (1.10 − 2.90), antenatal clinic (ANC) attendance for their current pregnancy [AOR = 1.41, 95% CI = (1.31 − 2.10), perceived susceptibility to delivery-related complications [AOR = 1.52, 95% CI = (1.30 − 2.70), and perceived severity of the delivery complications [AOR = 1.66, 95% CI = (1.12 − 2.31) were positively associated with pregnant women’s intentions to deliver at a health facility. Intention was negatively associated with participants’ perceived barriers to accessing a health facility [AOR = 0.62, 95% CI = (0.36 − 0.85). Conclusions: A low proportion of pregnant women in the sampled community intended to deliver at a health facility. Pastoralist communities may have special needs in this regard, with household income, antenatal care attendance, perceived risk of complications, and perceived barriers to accessing a health facility largely explaining the variance in intention. Community-based interventions providing counseling and messaging on danger signs in the perinatal period and emphasizing benefits of delivering at a facility are recommended, alongside improving access.
A community-based cross sectional study was conducted in Zone 3 (Gabi Rasu zone) of Afar region from April 1 to April 30, 2016. Zone 3 of Afar region is located 365 km from the south of Samara, the administrative city of Afar national regional state. The zone is bordered in the south by Oromia region, in the southwest by Amhara region, and in the east by Somali region. The total population of the zone was estimated to be 257,068 in 2016, 123,393 of whom were women residing in seven districts [22]. In terms of access to health facilities, the zone has only 1 primary hospital, 13 health centers, and 74 health posts (all of which are potential sites for health facility delivery). Sample size was determined using single population proportion statistical formula with the following parameters: p = proportion of women who give birth at health facility in Afar region = 16% [9]; d = margin of error = 5%; confidence interval = 95%, and design effect = 1.5. Then, considering a design effect of 1.5 i.e., 1.5 ∗ 207 = 310 and 10% contingency, i.e., 310 + (10% ∗ 310), the minimum sample size required for the study was 341. Twelve kebeles (the smallest administrative unit) from three randomly selected districts (Amibara, Gewane, and Argoba) were selected by lottery method. The sample size was proportionally allocated to each kebele based on the projected number of pregnant women. In each selected kebele, a list of pregnant women was developed to construct the sampling frame. Women who were at least 3 months gestational age (ascertained by self-report of last menstrual period) and who had lived in the study area at least for six months were included in the study. Finally, systematic random sampling was employed to select pregnant women from the list. Data collectors approached each selected pregnant woman at their home for an interview. A semi-structured questionnaire was adapted from validated examples in the literature, translated to the local language, and pretested on 10% of the sample size in a similar setting. The first section of the questionnaire contained sociodemographic and previous history of birth. The second consisted of items designed to assess respondents’ response to the constructs of the health belief model, namely (1) perceived susceptibility to birth-related complications, (2) perceived severity of the complications, (3) perceived barriers to delivery at a health facility, (4) perceived benefits of delivery at health facility, and (5) self-efficacy to use a health facility for delivery. For each item, respondents were asked their level of agreement or disagreement to items using a five-point Likert scale ranging from strongly agree (5) to strongly disagree (1). Consequently, 26 items were used to measure the constructs of the health belief model. The items were subjected to exploratory factor analysis with principal component analysis method, with a fixed number of constructs (i.e., five factors). Accordingly, the analysis identified perceived susceptibility (5 items), perceived severity (4 items), perceived benefits (5 items), perceived barriers (5 items), and perceived self-efficacy (2 items) as those to be extracted as constructs. All the extracted constructs explained jointly 52.1% of the variance of intention while perceived susceptibility alone explained 19.3% of the variation in the intention to use health facility for delivery. Then, reliability testing of items in each construct was assessed before using the constructs for further analysis. The result of the test showed that Chronbanch’s α was above 0.70% for all constructs. For each construct, the items were summed up to produce a composite score, and the mean score was used for further analysis. Outcome variable: The outcome variable was intention to use a health facility for birth. It was measured by asking pregnant women about their plan for where they would deliver their baby for their current pregnancy. The women were asked to choose either home or a health facility. Data Collection: Trained diploma holder nurses who were fluent in local languages collected the data. The principal investigators trained data collectors and closely supervised the data collection process. Data management and analysis: The data were entered into EpiData version 3.1 and then exported to SPSS version 20.0 for analysis. Descriptive statistics were used to summarize the results. The association between each independent variable and outcome variable was first assessed using binary logistic regression analysis. Variables with a p value of less than 0.05 were entered into multivariate logistic regression models. Adjusted odds ratios were reported at 95% confidence interval and a level of significance less than 0.05 was used to declare an association. Ethical considerations: Ethical clearance was obtained from the ethical review committee of Samara University, Ethiopia. The purpose of the study was explained to all respondents, and written informed consent was obtained from each respondent after they were assured of its confidentiality.
N/A
