Prediction rules for patients with suspected myocardial infarction. Applying guidelines in community hospitals.

Several clinical prediction rules have been developed to assist physicians in managing intensive care resources for patients with suspected myocardial infarction. These guidelines, developed in university settings, attempt to identify patients at high or low risk for developing life-threatening complications or death. Since some prediction rules have not performed well when applied to different patient populations, we applied these rules to 397 patients with suspected myocardial infarction who were admitted to community hospital coronary care units. The relative risk of dying associated with an abnormal initial electrocardiogram declined from 17 in the academic center to 2.9 in the community hospital. In contrast, a guideline that uses data available after 24 hours of observation did segregate patients at higher and lower risk in both the community and academic hospitals. This study shows that clinical prediction rules that were developed in academic medical centers should be validated before applying them in community hospital settings.