A multivariate hierarchical Bayesian framework for healthcare predictions with application to medical home study in the Department of Veteran Affairs

TL;DR

This paper introduces a multivariate hierarchical Bayesian model to predict healthcare demand and supply balance, improving clinical workload predictions in the Veteran Health Administration's medical home study.

Contribution

It extends hierarchical generalized linear models to include multivariate responses for better healthcare demand forecasting.

Findings

Model achieves high predictive performance on veteran healthcare data.

Balances supply and demand to reduce under and over-utilization.

Supports improved planning in patient-centered medical homes.

Abstract

Recently the patient centered medical home (PCMH) model has become a popular approach to deliver better care to patients. Current research shows that the most important key for succession of this method is to make balance between healthcare supply and demand. Without such balance in clinical supply and demand, issues such as excessive under and over utilization of physicians, long waiting time for receiving the appropriate treatment, and non continuity of care will eliminate many advantages of the medical home strategy. To reach this end we need to have information about both supply and demand in healthcare system. Healthcare supply can be calculated easily based on head counts and available hours which is offered by professionals for a specific time period while healthcare demand is not easy to calculate, and it is affected by some healthcare, diagnostic and demographic attributes. In this paper, by extending the hierarchical generalized linear model to include multivariate responses, we develop a clinical workload prediction model for care portfolio demands in a Bayesian framework. Our analyses of a recent data from Veteran Health Administration indicate that our prediction model works for clinical data with high performance.