Abstract:

West Papua, Indonesia has an excessively high maternal and infant mortality rate. In addition, health service utilization is hindered by geographic, economic, political, and cultural constraints. This study, a secondary data analysis, evaluates whether training and placement of a village midwife in every village of Paniai District of West Papua has an impact on reproductive health service utilization. In addition, it describes demographic differences between women who utilize selected reproductive health services and those who do not before and after the intervention.

For the original study, two cross-sectional surveys were conducted with two different random samples in Paniai District almost four years apart. For this analysis, two hundred forty records for both the baseline and final data sets were analyzed in univariate and multivariable logistic regression models. Demographic variables were chosen based on a review of literature and the study hypotheses. Statistically significant interaction terms were added to the multivariable models.

The two samples differed significantly in ethnic and employment status composition. The percentages of the sample having at least one prenatal exam during pregnancy, receiving prenatal care from a health worker, and delivering attended by a health worker all increased significantly between the baseline and final surveys.

For the baseline (pre-intervention) sample, women born in rural areas of the district were more likely than women born in coastal areas or outside of the province to receive antenatal care, and Protestant women were less likely than Catholic or Muslim women to receive antenatal care, adjusting for all other variables in the model. A statistically significant interaction term indicates that the odds ratios calculated vary with education and religion status, adjusting for all other variables. In the final sample (post-intervention), Protestant women were more likely than Catholic or Muslim women to receive antenatal care, and literate women and women who work away from home were less likely than their counterparts to receive antenatal care, adjusting for all other variables. There were no statistically significant interaction terms.

In the baseline sample, women who were 27 and older and women who were born in a rural area of the province were less likely than their counterparts to have a health worker attend birth, adjusting for all other variables. In addition, these odds ratios vary with age and religion status, adjusting for all other variables. In the final sample, women who were Protestant and women who were born in a rural area of the province were less likely than their counterparts to have a health worker attend birth, adjusted for all other variables. Women who work outside the home are more likely than women who do not to have a health worker attend birth, adjusting for all other variables. Finally, these odds ratios vary with religion and work status, adjusting for all other variables.

The results show that demographic characteristics cannot fully predict either outcome, but support prediction of health service utilization in certain cases. Overall, religion is a statistically significant main predictor, and should be considered in the development of health programs and policy in West Papua.