Abstract
Despite declines in recent years, inpatient care remains the largest component of national medical expenditures. The U.S. Department of Health and Human Services currently sponsors five data collection efforts that can be used to estimate the utilization of inpatient hospital care in the United States. Through a comparison of 1996 inpatient utilization estimates, this paper describes important methodological considerations when using the different data sources for measuring inpatient use. Our results show that surveys with similar target populations and methodologies produced similar estimates. In particular, estimates for the household based surveys (NHIS and MEPS-HC) were comparable while those for the hospital discharge surveys (NHDS and HCUP-NIS) were comparable. Hospital discharge surveys produced substantially higher estimates of total discharges than household surveys, which is partly attributable to differences in target populations. Also, data from the MCBS suggest that underreporting may be another explanation for lower estimates from the household surveys.