PURPOSE: To determine whether the management of heart failure by specialized multidisciplinary heart failure disease-management programs was associated with improved outcomes.
BACKGROUND: The advent of angiotensin-converting enzyme inhibitors, beta-blockers, and spironolactone has revolutionized the management of heart failure. Randomized double-blind studies have demonstrated survival benefits of these drugs in heart failure patients. Nevertheless, in spite of these advances, heart failure continues to be a syndrome of poor outcomes.1–4 There is also evidence that a significant portion of heart failure patients does not receive this evidence-based therapy that reduces morbidity and mortality.5–7 Various disease-management programs have been proposed and tested to improve the quality of heart failure care. Most of these programs are specialized multidisciplinary heart failure clinics lead by cardiologists or heart failure specialists and conducted by nurses or nurse practitioners. Similar to the Department of Veterans Affairs (VA) multidisciplinary geriatric assessment clinics, these clinics also use many other services, including pharmacists, dietitians, physical therapists, and social workers. Some of these programs also have an affiliated home health service. Several observation studies, using mostly pre- and postcomparison designs, have demonstrated the effectiveness of these programs in the process of care, resource use, healthcare costs, and clinical outcomes in patients with heart failure.8 Risk of hospitalization was reduced by 50% to 85% in six of the studies.8 Subsequently, several randomized trials were conducted to determine the effectiveness of these programs. The purpose of this systematic review was to determine the effectiveness of these programs on mortality and hospitalization rates of heart failure patients.
METHODS: Published articles on human randomized trials involving specialized heart failure disease-management programs in all languages were searched using Medline from 1966 to 1999 and other online databases using the following terms and Medical Subject Headings: case management (exp); comprehensive health care (exp); disease management (exp); health services research (exp); home care services (exp); clinical protocols (exp); patient care planning (exp); quality of health care (exp); nurse led clinics; special clinics; and heart failure, congestive (exp). In addition, a manual search of the bibliographies of searched articles was performed to identify articles otherwise missed in the above search. Personal communications were made with three authors to obtain further data on their studies. Using a data abstraction tool, two of the investigators separately abstracted data from the selected articles. Data from the selected studies were combined using the DerSimonian and Laird random effects model and the Mantel-Haenszel-Peto fixed effects model. Meta-Analyst 0.998 software (J. Lau, New England Medical Center, Boston, MA) was used to determine risk ratios (RRs) with 95% confidence intervals (CIs) of mortality and hospitalization for patients receiving care through these specialized programs compared with those receiving usual care. The Cochran Q test was used to test heterogeneity among the studies, and sensitivity analyses were performed to examine the effect of various covariates, such as duration of intervention, and other characteristics of the disease-management programs.
RESULTS: The original search resulted in 416 published articles, of which 35 met preliminary selection criteria. Of these, 11 were randomized trials and were selected for the meta-analysis. Studies that were not randomized trials, did not involve heart failure patients or disease-management programs, or had missing outcomes were excluded. Of the 11 studies selected, nine involved specialized follow-up using multidisciplinary teams and the remaining two involved follow-up by primary care physicians and telephone. These studies involved 1,937 heart failure patients with a mean age of 74. The follow-up period ranged from no follow-up (one study) to 1 year (one study). Patients receiving care from specialized heart failure disease-management programs had a 13% lower risk of hospitalization than those receiving usual care (summary RR = 0.87; 95% CI = 0.79–0.96), but the Cochran Q test demonstrated significant heterogeneity among the studies (P = .003). Subgroup analysis of the nine studies using specialized follow-up by a multidisciplinary team showed similar results (summary RR = 0.77, 95% CI = 0.68–0.86; test of heterogeneity, P> .50). Seven of the nine studies did not show any significant association between intervention and reduced hospitalization, but the two studies that used follow up by primary care physicians and telephone failed to show any significant reduction in hospitalization (summary RR = 0.94, 95% CI = 0.75–1.19). In fact, one of the studies demonstrated a higher risk of hospitalization for patients receiving intervention (RR = 1.26, 95% CI = 1.04–1.52). Of the 11 studies, only six reported mortality as an outcome. None of these studies found any association between intervention and mortality (summary RR = 1.15, 95% CI = 0.96–1.37; test of heterogeneity, P> .15). Five of the studies used quality of life or functional status as outcomes, and, of them, only one demonstrated significant positive association. The results of the sensitivity analyses were negative for any significant association with duration of intervention or follow-up or year of study. Eight studies performed cost analyses and seven demonstrated cost-effectiveness of the intervention.
CONCLUSIONS: The authors concluded that specialized disease-management programs were cost-effective, and heart failure patients cared for by these programs were more likely to undergo fewer hospitalizations, but the study did not provide any conclusive association between these programs and quality of care or mortality. The authors recommend that disease-management programs involve patient education and specialized follow-up by a multidisciplinary team including home health care.