The methodologic issues involved in evaluating the effectiveness of coronary care units (CCUs) were explored in a state-wide retrospective study of 1156 myocardial infarction discharges from Maryland hospitals during a one-year period. Of all patients treated in CCUs, 30.1% died in hospital compared with 17.8% of those not treated in CCUs (P<.0001). The difference between those treated in CCUs and those not treated in CCUs persisted when analyzed by race, sex, age, anticoagulant use and clinical severity. The combined effects of differences in prognostic factors between CCU and non-CCU patients were assessed using a multiple regression model. Adjusted mortality rates in CCU patients were almost twice as high as in patients not treated in CCUs (P<.0001). These findings are discussed in the context of possible study designs and methodologic issues involved in evaluating CCUs.
|Original language||English (US)|
|Number of pages||9|
|Journal||Johns Hopkins Medical Journal|
|State||Published - Dec 1 1977|
ASJC Scopus subject areas