Exploring trend and barriers of antenatal care utilization using data mining:evidence from EDHS of 2000 to 2016

Background The healthcare industry is paying attention to pregnancy and Antenatal care (ANC) for mothers. Thus, the presented study aimed at exploring the trend and identifying the barrier for ANC utilization of mothers in Ethiopia. Data mining is a field of big data science used to discover patterns and knowledge from big data. Methods All EDHS datasets from 2000 to 2016 were used for this study. The pooled cross-sectional study was conducted using the knowledge discovery process having steps; selection, cleaning, integration, transformation, and data mining algorithms. These algorithms are; Classification, clustering, association rules, and attribute ranking with pattern prediction. Results The proportion of ANC utilization was 27.6%, 28.2%, 34.5%, and 62.9% in 2000, 2005, 2011, and 2016 respectively. The pooled data; contained 28,631 mothers which were included in the study. Of these, more than half (56.09%) of them were not utilizing ANC during a pregnancy. Pregnancy complication, educational status of mothers and husbands, mothers’ residence, economic status, and media exposure had an association with ANC utilization having a confidence level of 95% and above. Conclusion ANC utilization in Ethiopia was increased significantly from 27.6%in 2000 to 62.9% in 2016. Despite this increment, the pooled proportion of ANC utilization is still low. The barriers to this low utilization were; Pregnancy complication, poor education of mothers and their husbands, living in rural, poor economic status, and media exposure. This study will recommend; firstly, pregnant mothers have to attend ANC service even though she had no pregnancy complication. Secondly, Education and poverty reduction are key strategic area to be addressed in improving women’s awareness towards ANC during a pregnancy. Thirdly, Expansion of infrastructure among the rural communities having good media coverage needs to be prioritized to improve ANC service utilization.