The proposed research will investigate: (1) whether will-documented treatment guidelines, which could result in improved survival and reduced re-hospitalization in post-myocardial infarction patients, are being adopted by physicians in primary care settings; and (2) whether a retrospective cohort study using large, linked Medicare and two state drug-claims data bases (N=9,600 AMI patients) can replicate the results of randomized controlled trials (RCTs) of beta blocker therapy on patient outcomes. Several large RCTs have already confirmed the value of chronic beta blockade following AMI in increasing survival and reducing the risks and costs of post-AMI cardiac events. Two-year mortality rates and rates of non-fatal reinfarctions appear to be 20-40% lower in patients receiving long-term beta blockade. Yet, few data exist from large, "real-world" populations on the predictors and outcomes of use of these effective agents, particularly among women, minorities, and elderly people, who are well-represented in this study. The investigators will link and analyze three New Jersey claims data- bases in their possession, including eight years of Medicare claims data, computer drug claims data for poor elderly Medicaid patients (n=about 60,000/yr), and for moderate-income elderly enrollees in a state drug benefit program for the elderly (n= about 250,000/yr). Specific research questions include (but are not limited to): 1. What proportion of eligible elderly patients with first AMIs (that is, those without known contraindications) receive beta blockers after AMI? 2. What proportion of such patients receive these medications on a regular basis? 3. Has the publication of numerous clinical reports in the mid-1980s recommending beta blocker therapy routinely after AMI increased utilization over time (1986-92)? 4. Do patient characteristics (e.g., age, sex, race, income, Medicaid vs. non-Medicaid, prior AMI) predict use of beta blockers after AMI? 5. Do specific physician characteristics (e.g., cardiovascular specialty, group practice) predict higher rates of appropriate use of beta blockers for eligible patients? 6. Using survival analysis methodologies which control for all patient and physician characteristics predicting use of beta blockers, what are the estimated effects of post-AMI beta blocker therapy on the rate of reinfarctions and survival? Are claims-based epidemiological estimates similar to results reported in large RCTs?