Estimating the Out-of-Hospital Mortality Rate Using Patient Discharge Data

Abstract

This paper is a case study of the use of public use administrative data for the estimation of empirical relations when key dependent variables are not available in the data. It is shown that the out-of-hospital mortality rates can be identified using the patient discharge data without post-discharge death records. Using data on the lengths of hospitalizations and out-of-hospital spells, the mortality rates before and after discharge as well as discharge and re-hospitalization rates are estimated for a sample of heart-attack patients hospitalized in California between 1992 and 1998. The results suggest that ignoring variation of discharge rates among hospital types could be misleading in evaluating hospital performance regarding mortality risks.

Keywords: Mortality; Hospital quality; Duration models; Survival analysis