Predicting Adolescent Pregnancy Risk Using Household Survey Data: A Population-Based Approach

Jonathan Patricio, INTITUTO DE INVESTIGACIÓN EN SALUD
Kary Santos, Centro de Desenvolvimento e Planejamento Regional (CEDEPLAR)

Adolescent pregnancy represents a critical public health issue in the Dominican Republic, with significantly high rates, particularly among adolescents from rural and low-income areas. This study aims to develop a predictive model to identify adolescents at high risk of pregnancy, using data from the National Household Survey (ENHOGAR-2018). The sample includes 4,946 adolescents aged 15 to 19, of which 21.4% have been pregnant. The analysis is based on the social determinants of health proposed by the WHO, using algorithms such as logistic regression, neural networks, and decision trees will be employed. These models are expected to reveal significant associations between factors such as early sexual initiation, early unions, and educational level with pregnancy risk. The findings of this research will provide a practical tool to improve the targeting of prevention programs, contributing to the reduction of adolescent pregnancy rates in the country.

Keywords: Children, Adolescents, and Youth, Sexual and Reproductive Health and Rights

See extended abstract.