Health Disparities
Cost-Constrained Dynamic Interventions for Health Equity Cuong* Cuong Pham Pham Pham Pham Pham Pham Departments of Epidemiology, Johns Hopkins Bloomberg School of Public Health
Hypertension is a leading risk factor for cardiovascular disease. In the United States, it disproportionately affects African Americans, Hispanics, and individuals living in underserved rural communities. These disparities are driven by social, cultural, and structural factors that impact disease risk and access to effective care.
Although multilevel interventions have been developed to reduce hypertension disparities, their implementation is constrained by limited resources. Hence, healthcare organizations face difficult decisions about how and whom to treat. These decisions require balancing the goal of optimizing health outcomes with equity and cost constraints. However, identifying optimal solutions to these trade-offs is challenging.
We address this challenge by incorporating the concept of allowability into the dynamic treatment regime (DTR) framework. A DTR is a set of decision rules that adapt treatment assignment over time based on patients’ characteristics and evolving health status, while allowability permits adjustment for fair sources of variation when measuring health disparities. Although recent methods enable the estimation of outcome-optimal DTRs under cost constraints, such regimes do not necessarily reduce disparities. We extend the DTR framework to explicitly target disparity reduction while accounting for allowable covariates. This extension enables evaluation of existing care policies and construction of optimal, cost-constrained DTRs that improve health outcomes and reduce disparities.
We apply this framework to a collaborative care model for hypertension management that integrates community health workers (CHWs) within a care team. Our aims are to: (1) assess the extent to which existing care regimes improve hypertension outcomes and reduce disparities across racial, ethnic, and geographic groups; and (2) determine when and for whom adding CHWs yields the greatest population-level benefit while minimizing disparities under cost constraints.