OBJECTIVES. Using the public reports of the Pennsylvania Health Care C
ost Containment Council on coronary artery bypass graft surgery for 19
90 to 1992 as a case study, the authors assess the sensitivity of resu
lts to the choice of data and statistical methodology. METHODS. Using
the Council's public-release data, surgical mortality and utilization
were reanalyzed by standard linear models, empirical Bayes methods, Mo
nte Carlo simulations, and hierarchical statistical models. RESULTS. S
tatistical power calculations demonstrate that the annual volume of by
pass surgery for many hospitals and for most surgeons is too small for
meaningful mortality comparisons. The number of hospitals and physici
ans designated as mortality ''outliers'' in the Council's reports resu
lts in part from a failure to adjust critical P values for multiple co
mparisons. Hierarchical statistical models implemented by mixed effect
s logistic regression, by contrast, can detect true differences in per
formance without producing false outliers. Mortality analyses are sens
itive to the choice of comorbidities used for severity adjustment of a
mortality model. Small-area analyses indicate large differences in th
e rates of bypass surgery across Pennsylvania, with lower population-b
ased rates of surgery associated with higher population-based inpatien
t mortality. CONCLUSIONS. Analyses of mortality by operative procedure
, rather than by patient diagnosis, should consider the potential for
selection bias caused by the decision to elect surgery. The clinical a
nd statistical issues of operative mortality are sufficiently complex
to merit review by independent experts before public release of hospit
al and physician performance measures.