We present alternative methods for estimating hospital-level mortality
rates to those used by the Health Care Finance Administration for Med
icare patients. We use an empirical Bayes model to represent the diffe
rent sources of variation in observed hospital-specific mortality rate
s and we use a logistic regression model to adjust for severity differ
ences (in patient mix) across hospitals. In addition to providing a pr
incipled derivation of a standard error for the commonly used estimato
r, our fully model-based formulation produces much more accurate estim
ates and resolves the severe problem of multiple comparisons that aris
es when extreme estimates are used to identify exceptional hospitals.
We estimate models for each of four disease conditions using the natio
nal Medicare mortality data base which does not contain patient severi
ty descriptors, and mortality data from national samples which do incl
ude patient severity descriptors. We find substantial between-hospital
variation in the unadjusted death rates from the national data base.
Mortality rates differ substantially with patient severity in our mode
ls, but the sample sizes are too small to yield reliable estimates of
the between-hospital variation in adjusted mortality rates.