• risk-adjustment;
  • severity of illness;
  • outcomes;
  • AIDS;
  • mortality;
  • length of stay

A comparison of three strategies for risk-adjustment of outcomes for AIDS patients hospitalized for Pneumocystis carinii pneumonia

Background: The need for risk-adjustment of patient outcomes has been driven by the competitive health care market and the subsequent increase in comparative outcome reporting among health care institutions, among managed care plans, and among individual providers for some procedures (e.g. coronary artery bypass graft surgery). However, if the outcomes reported do not take into account patient characteristics that can be considered dimensions of risk for poor clinical outcomes or increased utilization of services, there is the possibility that inaccurate conclusions will be drawn about the quality of care provided. Objective: The specific purpose of this study was to examine the ability of four measures, APACHE III — acute physiology scale, Quality Audit Marker — ambulation score, Quality Audit Marker — self-care ability score, and Nursing Severity Index, to predict mortality and hospital length of stay in a convenience sample of 140 males with Pneumocystis carinii pneumonia. Methods: The study utilized a descriptive, longitudinal design. Results: APACHE III — acute physiology scale (P = 0·006, odds ratio = 1·40), and Quality Audit Marker — ambulation (P = 0·037, odds ratio = 0·50), were significant predictors of hospital mortality and the APACHE III — acute physiology scale was also a predictor of mortality within 3 (P = 0·004, odds ratio = 1·13) and 6 months (P = 0·009, odds ratio = 1·10) following hospitalization. Only Quality Audit Marker — ambulation (P = 0·0001) was a significant predictor of length of stay. Conclusions: The findings of this study confirm the findings of other investigators that measures of acute clinical stability and functional status have utility as risk-adjustment approaches for the outcomes of mortality and length of stay. Further research is needed that compares the utility of generic vs. disease-specific measures for prediction of outcomes in HIV/AIDS.