An Attempt to Refine DRGs (Diagnosis Related Groups) for Navy Medical Department Use by Including Military Unique Variables and an Estimate of Disease Severity,
Abstract
This report presents the results of a study to determine if differences in length of stay (LOS) within DRGs could be explained by additional variables on the Navy Inpatient Data System and by a measure of disease severity. Using the same methodology as the Yale researchers, findings indicate that the additional inpatient variables (patient related, hospital characteristics, and military related) increased the explained variance in LOS from 25 to 30 percent and created 345 subgroups. Two-thirds of these subgroups were accounted for by five factors: number of diagnoses, number of procedures, admitted by transfer, active duty enlisted status, and large teaching hospital. Appendix B details the average length of stay, standard deviation, and number of cases by DRG subgroups. Severity of illness within seven DRGs was measured by a valid and reliable nursing acuity tool. Findings indicated that maximum patient classification category and points were positively correlated with LOS and explained significantly more variance than accounted for by the additional inpatient variables for most study DRGs. The report concludes with specific recommendations for monitoring and comparing hospital performance using these findings.
Document Details
- Document Type
- Technical Report
- Publication Date
- Feb 01, 1985
- Accession Number
- ADA157676
Entities
People
- K. A. Rieder
- Richard J. Hall
- T. L. Kay