Reply

Drs. Thornton and Ahmed have identified one important message from our population cohort study (1). We agree that beta-blockers and angiotensin-converting enzyme (ACE) inhibitors/receptor blockers were underutilized in heart failure (HF) patients in Alberta, Canada (1), and elsewhere (2–5). We found that adjusted one-year mortality was lower in seniors hospitalized for HF who were prescribed beta-blockers (18.2%; 95% confidence interval [CI] 14.2 to 22.2) or ACE inhibitor/receptor blockers (22.3%; 95% CI 20.9 to 23.7) within three months of hospital discharge than in those with no prescriptions (29.9%; 95% CI 28.8 to 31.0). The use of both beta-blockers and ACE inhibitors/receptor blockers was associated with an even lower mortality (16.6%; 95% CI 13.3 to 20.0) than ACE inhibitor/receptor blockers alone (22.3%; 95% CI 20.9 to 23.7).

The underutilization of effective therapy is not unique (6)even for therapies such as these that have been repeatedly documented to be beneficial in clinical trials that span over a decade (7). To improve the utilization of beta-blockers and ACE inhibitors/receptor blockers, their utilization has been flagged as a marker of quality of care. Our study may provide some clues as to why underutilization persists. Heart failure is not defined in terms of preservation or loss of systolic function; rather, it is a clinical constellation of symptoms (8). Both systolic and diastolic impairment in HF are independently associated with increased mortality risk (9). Clinical trials are not representative of the general population with HF, as younger, male, white, and decreased ejection-fraction patients are over-represented (10). The use of the exclusion criteria of decreased ejection fraction has the effect of under-recruitment of older female patients. In Alberta, seniors (age 65 years and older) constituted about 85% of all new diagnoses for HF during the study period (April 1, 1994, to March 31, 1999). Hospitalized patients age 75 years and older constituted about 60% of all incident, prevalent, and total hospitalization cases during this study period. In those age 75 years and older, women constituted over 56% of hospitalized HF patients. The population in Alberta was similar to that noted in the United States (10).

Thus, the reflective clinician may quite rightly question whether clinical trial results are generalizable to the demographically different population of hospitalized HF patients. Our study can only associate decreased mortality to the use of beta-blockers and ACE inhibitors/receptor blockers while statistically controlling for age, gender, and comorbidity. We cannot exclude residual confounding or confounding by intent, in that clinicians are better able to differentiate patients more or less likely to benefit from therapy. We hope that the large survival benefit noted in seniors with the use of beta-blockers and ACE inhibitors/receptor blockers will promote clinical trials that enroll patients who are more representative of the general population. We are hopeful that this will occur, because a more representative trial has now been published and has demonstrated some benefit from therapy on composite outcomes (11).

• American College of Cardiology Foundation