Factors associated with the perceived barriers of health care access among reproductive-age women in Ethiopia: a secondary data analysis of 2016 Ethiopian demographic and health survey

This study was based on secondary data analysis from the 2016 Ethiopia Demographic and Health Survey, which was collected cross-sectionally from January 18, 2016, to June 27, 2016. The study was conducted in Ethiopia, located in the Horn of Africa. The country has nine Regional states (Afar, Amhara, Benishangul-Gumuz, Gambela, Harari, Oromia, Somali, Southern Nations, Nationalities, and People’s Region (SNNP) and Tigray) and two Administrative Cities (Addis Ababa and Dire-Dawa).

For the sampling purposes, the nine regional and two city administrations of Ethiopia were stratified into urban and rural enumeration areas, of which 202 and 443 EAs were initially selected from urban and rural areas, respectively, by probability sampling method. Then, a fixed number of 28 households per cluster were systematically selected. In the interviewed households, 16,583 eligible women identified for interviews of which, 15,683 women had completed the interview and included in the final analysis. The study participants from each EA stratum selected independently by using the probability sampling technique. The detail of the methodology is available in the full report of 2016 EDHS [3].

Five data collection questionnaires were used for the 2016 EDHS, women’s questionnaire was one of the tools used to collect data about women and child health characteristics. For this study, women’s Individual Record (IR) file was used to extract, socio-demographic and reproductive traits were extracted from the most substantial dataset.

The Perceived barriers to health care access were the response variable. On the other hand, age, residence, wealth (economic) status, level of education, marital status, working status, health insurance coverage, and reproductive characteristics such as contraceptive use and intention, place of delivery, ANC follow up, and pregnancy during data collection were independent variables.

Each woman was interviewed to rate the difficulties of accessing health care based on obtaining money, health facility’s distance, permission to consult the doctor, and not wanting to go alone. Women reported at least one challenge of healthcare access (money, distance, companionship, and permission) considered as having perceived barriers of health care access, which coded as “1”. On the other hand, if a woman didn’t report challenges of the obstacles mentioned above, like obtaining money, distance, companionship, and permission, it was considered no perceived barrier to health care access, coded as “0” [3, 12].

The data were weighted using sampling weight, primary sampling unit, and strata before any statistical analysis to restore the representativeness of the survey and take into account the sampling design to get reliable statistical estimates. Descriptive and summary statistics were conducted using STATA version 14 software. To identify factors associated with the perceived barriers of health care access among reproductive-age women, generalized estimating equation (GEE) model was fitted [25]. The data nature of EDHS showed that women were nested within a cluster, and those who reside within the same clusters were correlated to each other compared to the other clusters. This data showed that the Intra Class Correlation (ICC) was calculated and found to be 40% indicated a correlation among observations in the clusters. These violated the conditional logistic regression model’s assumptions that consider the independence of observations and equal variance across clusters. The Generalized Estimating Equation (GEE) model was fitted with a logit link function and binomial family and working correlation structures of independent, exchangeable, unstructured, and autoregressive compared simultaneously. The model with the smallest standard error differences between robust and model-based standard errors that were the model with the exchangeable correlation structure was selected. To assess the strength of association between outcome and independent variables crude and adjusted odds ratio with a 95% confidence interval (CI) computed and presented on the table. Variables had less than 5% p-value in the multivariable GEE model considered independent factors associated with the perceived barriers of health care access.

In this study, we fitted two models GEE which is a marginal model that considers correlation among clusters and a conventional logistic regression model with a robust standard error that also controlled for within-cluster correlations. To select the best-fitted model, we used quasi information criteria (QIC) for the model comparison and the result presented on the Table 1 as follow.

A total of 15,683 reproductive-age women were included in the final analysis of the study. The median age of women was 27 with (IQR: 20 to 35) years; nearly two-thirds (67.5%) of women gave birth previously, and 18 years is the median age to the first birth. More than three-four (77.8%) women were rural dwellers, 47.8% had no formal education, 63.8% were married, 46.3% rich wealth index, and 43.2% were orthodox Christians. The majority (94.7%) of women had no health insurance coverage; among interviewed women, 7.2% of them were pregnant during data collection, and 41.6% of them visited health facilities in the past 12 months (Table 2).

The chi-square analysis showed a correlation between reproductive health service (contraceptive utilization history, place of delivery, and previous ANC follow-up). The reproductive variables cannot be included in the GEE marginal analysis model because no use of these services may not be an indicator of health care perceived barriers (Table 3).

In this study, more than two-thirds (69.9, 95%CI: 69.3 to 70.7) of reproductive-age women had at least one perceived barrier to access health care, of which money (54.8%), the distance of health facilities (50.3%) were the most frequently mentioned challenges. Furthermore, out of the currently pregnant women during the data collection time, about 73% of them had perceived barriers to healthcare access. Of the parameters used to assess perceived barriers of healthcare access, about 21.5% of women had multiple challenges (money, distance, companionship, and permission) (Fig. 1).

The result of the bi-variable analysis showed that all explanatory variables were associated with healthcare access’s perceived barriers at a 20% level of significance. The multivariable generalized estimated equation (GEE) model showed that variables like residence, marital status, age group, educational level, wealth index, and health insurance were significant determinants of perceived barriers of health care access at a 5% level of significance (Table 4).

Those women who reside in rural areas were 2.13 times more likely to had perceived barriers of health care access than urban residents (AOR = 2.13, 95%CI: 1.79 to 2.53). Similarly, women aged 35–49 years, the odds of perceived barriers to access health care were 1.24 times higher than those aged 15–19 years (AOR = 1.24, 95%CI: 1.09 to 1.40). The likelihood of having perceived barriers to access health care among divorced/separated women was increased by 34% compared to married/live together (AOR = 1.34, 95%CI: 1.17 to 1.54). Similarly, the odds of perceived challenges of health care access to those who had no formal education (AOR = 2.30, 95%CI: 1.95 to 2.72), primary (AOR = 1.84, 95%CI: 1.58 to 2.15) and secondary (AOR = 1.31, 95%CI: 1.13 to 1.51) higher compared to those who attended college and above. For women who had poor and middle wealth status, the odds of perceived barriers of health care access were 2.09 and 1.57 times higher, compared to wealthier women, respectively (AOR = 2.09, 95% CI: 1.86, 2.35) and (AOR = 1.57, 95% CI: 1.38, 1.79). Those women who had no health insurance coverage, the odds of perceived barriers of health care access increased by 19% compared to those insured (AOR = 1.19, 95% CI: 1.01, 1.45) (Table 3).