Predicting the Frequency and Amount of Health Care Expenditures

Authors:
  • EFEdward W. Frees
    Edward W. Frees
  • MRMarjorie A. Rosenberg
    Marjorie A. Rosenberg
Publisher website
Google Scholar
Abstract
This article extends the standard two-part model for predicting health care expenditures to the case where multiple events may occur within a one-year period. The first part of the extended model represents the frequency of events, such as the number of inpatient hospital stays or outpatient visits, and the second part models expenditure per event. Both component models also use independent variables that consist of an individual’s demographic and access characteristics, socioeconomic status, health status, health insurance coverage, employment status, and industry classification. The second part of the model also includes a variable representing the number of events to predict the expenditure per event, thus capturing dependencies between the first and second parts. This article introduces closed-form predictors of annual total expenditures and demonstrates how to create simulated predictive distributions for individuals and groups. The data for this study are from the Medical Expenditure Panel Survey (MEPS). MEPS panels 7 and 8 from 2003 were used for estimation; panels 8 and 9 from 2004 were used to validate predictions. This annual expenditures model provided a better fit to the data than standard two-part models. The count variable was significant in predicting outpatient expenditures. The aggregate expenditures model provided better point predictions of held-out total expenditures than competing models, including the standard two-part model. The predictive distribution for aggregate expenditures for small groups is long tailed, with both the variability and skewness decreasing as the group size increases, an important point for programs designed to manage expenditures.

This publication has 16 references indexed in Scilit: