Factors determining angiotensin-converting enzyme inhibitor underutilization in heart failure in a community setting

Background: Angiotensin-converting enzyme (ACE) inhibitors were underprescribed for patients with congestive heart failure (CHF) treated in the community setting in the early 1990s despite convincing evidence of benefit. Hypothesis: We postulated that (1) the prevalence of ACE inhibitor use has increased, and (2) prescribing biases have narrowed, as community physicians have gained additional clinical experience with these drugs for treatment of CHF. Methods: We examined rates of ACE inhibitor use among 1,150 patients with CHF hospitalized at 10 community hospitals in 1995, evaluated determinants of ACE inhibitor pre prescription, and compared the results with survey data gathered among similar patients during 1992. Results: Compared with 1992, ACE inhibitor use prior to hospital admission was increased among all patients (42 vs. 33%, p < 0.001) and the subset with a history of CHF (53 vs. 39%, p < 0.0005). Angiotensin-converting enzyme inhibitor prescription at hospital discharge also increased among all survivors (64 vs. 51%, p < 0.00005) and the subset eligible for ACE inhibitor treatment based on clinical trial criteria (77 vs. 66%, p = 0.04). Multivariate analysis suggested no change in the prescribing biases previously observed; ACE inhibitor use was related to lower ejection fraction, lower serum creatinine, documentation of left ventricular systolic function, younger patient age, prescription of any diuretic drug, and nonprescription of alternate vasodilators and calcium blockers. In multivariate analyses, physician specialty did not predict ACE inhibitor use. Conclusions: Angiotensin-converting enzyme inhibitor use among patients with CHF is increasing but remains below the 80-90% rates of drug tolerance documented in randomized clinical trials. This discrepancy is partially explained by the prevalence of renal impairment and "diastolic" heart failure in the community setting. However, age bias, use of alternative vasodilators, and substandard quality of care may also play a role.