Survival after Treatment during Hospitalization for Acute Myocardial Infarction:
A Community-Based Perspective
Background: The wealth of data from clinical trials has done as much to diversify standard therapy for acute myocardial infarction (MI) as it has to define it. To better characterize the use of treatments during hospitalization for MI in clinical practice, we analyzed one-year post MI survival in relation to common treatments in the Atherosclerosis Risk in Communities (ARIC) Study. Methods: Between 1987 and 1996 potential cases of MI among men and women, ages 35-74 were identified through hospital surveillance in Forsyth Co. NC, Jackson, MS, Minneapolis, MN, and Washington Co. MD. Diagnosis was validated using a standardized algorithm. Information on treatments was obtained from the medical record. Medication use included those given or continued during the hospitalization or prescribed at discharge. Cox proportional hazards models were used to estimate survival associated with treatments. Results: Of the 15,032 MI events, 34% were female and 18% black. One-year case-fatality of hospitalized MI events ranged from a high of 12% in 1987 to a low of 6.5% in 1994. Between 1987 and 1996, use of thrombolytic therapy (TT) increased from 11% to 16% of patients, beta-blockers (BB) 45% to 66%, and aspirin (AS) 49% to 87%. Coronary artery bypass graft (CABG) surgery increased from 13% to 19% and coronary angioplasty (PTCA) from 11% to 26%. After adjusting for age, race, sex, center, history of MI, severity indicators, complications, and the other therapies, AS (hazard ratio (HR) =0.47, 0.40-0.54) and PTCA (HR=0.47, 0.39-0.57) held the strongest association with survival, followed by CABG (HR=0.61, 0.50-0.75), BB (HR=0.62, 0.53-0.72), and TT (HR=0.71, 0.59-0.87). In 1996, the use of PTCA varied by center from 12% to 36% and AS use varied from 79% to 92%. Conclusions: Although not a randomized trial and therefore subject to selection bias, the ARIC data suggest that established treatments may be associated with better MI survival in community practice. Although no other therapy studied was more strongly associated with survival or has fewer contraindications than aspirin, it is underused in some communities.