ABSTRACT
BACKGROUND: Heart failure (HF) clinics are known to improve outcomes of patients with HF. Studies have been limited to single, usually tertiary centres whose experience may not apply to the general HF population. OBJECTIVES: To determine the effectiveness of HF clinics in reducing death or all-cause rehospitalization in a real-world population. METHODS: A retrospective analysis of the Improving Cardiovascular Outcomes in Nova Scotia (ICONS) disease registry was performed. All 8731 patients with a diagnosis of HF (844 managed in HF clinics) who were discharged from the hospital between October 15, 1997, and July 1, 2000, were identified. Patients enrolled in any one of four HF clinics (two community-based and two academic-based) were compared with those who were not. The primary outcome was the one-year combined hospitalization and mortality. RESULTS: Patients followed in HF clinics were younger (68 versus 75 years), more likely to be men (63% versus 48%), and had a lower ejection fraction (35% versus 44%), lower systolic blood pressure (137 mmHg verus 146 mmHg) and lower serum creatinine (121 micromol/L versus 130 micromol/L). There was no difference in the prevalence of hypertension (56%), diabetes (35%) or stroke/transient ischemic attack (16%). The one-year mortality rate was 23%, while 31% of patients were rehospitalized; the combined end point was 51%. Enrollment in an HF clinic was independently associated with reduced risk of total mortality (hazard ratio [HR] 0.69 [95% CI 0.51 to 0.90], P=0.008; number needed to treat for one year to prevent the occurrence of one event [NNT]=16), all-cause hospital readmission (HR 0.27 [95% CI 0.21 to 0.36], P<0.0001; NNT=4), and combined mortality or hospital readmission (HR 0.73 [95% CI 0.60 to 0.89], P<0.0015; NNT=5). DISCUSSION: HF clinics are associated with reductions in rehospitalization and mortality in an unselected HF population, independent of whether they are academic- or community-based. Such clinics should be made widely available to the HF population.
