REFERRAL SELECTION BIAS IN THE MEDICARE HOSPITAL MORTALITY PREDICTIONMODEL - ARE CENTERS OF REFERRAL FOR MEDICARE BENEFICIARIES NECESSARILY CENTERS OF EXCELLENCE

Objective. Although the Health Care Financing Administration (HCFA) us es Medicare hospital mortality data as a measure of hospital quality o f care, concerns have been raised regarding the validity of this conce pt. A problem that has not been fully evaluated in these data is the p otential confounding effect of illness severity factors associated wit h referral selection and hospital mortality on comparisons of risk-adj usted hospital mortality. We address this issue.Data Sources and Study Setting. We analyzed the 1988 Medicare hospitalization data file (MED PAR). We selected data on patients treated at the two Mayo Clinic-asso ciated hospitals in Rochester, Minnesota, and a group of seven other h ospitals that treat many patients from large geographic areas. These h ospitals have had observed mortality rates substantially lower than th ose predicted by the HCFA model for the period 1987-1990. Study Design . Using the multiple logistic regression model applied by HCFA to the 1988 data, we evaluated the relationship between distance from patient residence to the admitting hospital and risk-adjusted hospital mortal ity. Principal Findings. Among patients admitted to Mayo Rochester-aff iliated hospitals, residence outside Olmsted County, Minnesota was ind ependently associated with a 33 percent lower 30-day mortality rate (p < .001) than that associated with residence in Olmsted County. When p atients at Mayo hospitals were stratified by residence (Olmsted County versus non-Olmsted County), the observed mortality was similar to tha t predicted for community patients (9.6 percent versus 10.2 percent, p = .26), whereas hospital mortality for referral patients was substant ially lower than predicted (5.0 percent versus 7.5 percent, p = < .001 ). After incorporation of the HCFA risk adjustment methods, distance f rom patient residence to the hospitals was also independently associat ed with mortality among the Mayo Rochester-affiliated hospitals and se ven other referral center hospitals. Conclusions. The HCFA Medicare ho spital mortality model should be used with extreme caution to evaluate hospital quality of care for national referral centers because of res idual confounding due to severity of illness factors associated with g eographic referral that are inadequately captured in the extant predic tion model.