Selective Intervention Planning using Restless Multi-Armed Bandits to Improve Maternal and Child Health Outcomes

TL;DR

This paper develops machine learning models and a restless multi-armed bandit approach to identify and intervene with women less likely to engage in health call programs, significantly improving maternal and child health outcomes in India.

Contribution

It introduces a novel application of RMABs for intervention planning in health programs and demonstrates improved engagement through predictive modeling and targeted interventions.

Findings

Predictive models achieve high accuracy in identifying low-engagement women.

Interventions based on model predictions increase engagement by over 61%.

Preliminary RMAB results suggest effective resource allocation for health interventions.

Abstract

India has a maternal mortality ratio of 113 and child mortality ratio of 2830 per 100,000 live births. Lack of access to preventive care information is a major contributing factor for these deaths, especially in low resource households. We partner with ARMMAN, a non-profit based in India employing a call-based information program to disseminate health-related information to pregnant women and women with recent child deliveries. We analyze call records of over 300,000 women registered in the program created by ARMMAN and try to identify women who might not engage with these call programs that are proven to result in positive health outcomes. We built machine learning based models to predict the long term engagement pattern from call logs and beneficiaries' demographic information, and discuss the applicability of this method in the real world through a pilot validation. Through a pilot service quality improvement study, we show that using our model's predictions to make interventions boosts engagement metrics by 61.37%. We then formulate the intervention planning problem as restless multi-armed bandits (RMABs), and present preliminary results using this approach.