Community-based care for the specialized management of heart failure: an evidence-based analysis.Ont Health Technol Assess Ser 2009; 9(17):1-42OH
In August 2008, the Medical Advisory Secretariat (MAS) presented a vignette to the Ontario Health Technology Advisory Committee (OHTAC) on a proposed targeted health care delivery model for chronic care. The proposed model was defined as multidisciplinary, ambulatory, community-based care that bridged the gap between primary and tertiary care, and was intended for individuals with a chronic disease who were at risk of a hospital admission or emergency department visit. The goals of this care model were thought to include: the prevention of emergency department visits, a reduction in hospital admissions and re-admissions, facilitation of earlier hospital discharge, a reduction or delay in long-term care admissions, and an improvement in mortality and other disease-specific patient outcomes.OHTAC approved the development of an evidence-based assessment to determine the effectiveness of specialized community based care for the management of heart failure, Type 2 diabetes and chronic wounds.PLEASE VISIT THE MEDICAL ADVISORY SECRETARIAT WEB SITE AT: www.health.gov.on.ca/ohtas to review the following reports associated with the Specialized Multidisciplinary Community-Based care series.Specialized multidisciplinary community-based care series: a summary of evidence-based analysesCommunity-based care for the specialized management of heart failure: an evidence-based analysisCommunity-based care for chronic wound management: an evidence-based analysisPlease note that the evidence-based analysis of specialized community-based care for the management of diabetes titled: "Community-based care for the management of type 2 diabetes: an evidence-based analysis" has been published as part of the Diabetes Strategy Evidence Platform at this URL: http://www.health.gov.on.ca/english/providers/program/mas/tech/ohtas/tech_diabetes_20091020.htmlPLEASE VISIT THE TORONTO HEALTH ECONOMICS AND TECHNOLOGY ASSESSMENT COLLABORATIVE WEB SITE AT: http://theta.utoronto.ca/papers/MAS_CHF_Clinics_Report.pdf to review the following economic project associated with this series:Community-based Care for the specialized management of heart failure: a cost-effectiveness and budget impact analysis.
The objective of this evidence-based analysis was to determine the effectiveness of specialized multidisciplinary care in the management of heart failure (HF).
TARGET POPULATION AND CONDITION HF is a progressive, chronic condition in which the heart becomes unable to sufficiently pump blood throughout the body. There are several risk factors for developing the condition including hypertension, diabetes, obesity, previous myocardial infarction, and valvular heart disease.(1) Based on data from a 2005 study of the Canadian Community Health Survey (CCHS), the prevalence of congestive heart failure in Canada is approximately 1% of the population over the age of 12.(2) This figure rises sharply after the age of 45, with prevalence reports ranging from 2.2% to 12%.(3) Extrapolating this to the Ontario population, an estimated 98,000 residents in Ontario are believed to have HF. Disease management programs are multidisciplinary approaches to care for chronic disease that coordinate comprehensive care strategies along the disease continuum and across healthcare delivery systems.(4) Evidence for the effectiveness of disease management programs for HF has been provided by seven systematic reviews completed between 2004 and 2007 (Table 1) with consistency of effect demonstrated across four main outcomes measures: all cause mortality and hospitalization, and heart-failure specific mortality and hospitalization. (4-10) However, while disease management programs are multidisciplinary by definition, the published evidence lacks consistency and clarity as to the exact nature of each program and usual care comparators are generally ill defined. Consequently, the effectiveness of multidisciplinary care for the management of persons with HF is still uncertain. Therefore, MAS has completed a systematic review of specialized, multidisciplinary, community-based care disease management programs compared to a well-defined usual care group for persons with HF.
What is the effectiveness of specialized, multidisciplinary, community-based care (SMCCC) compared with usual care for persons with HF? LITERATURE SEARCH STRATEGY: A comprehensive literature search was completed of electronic databases including MEDLINE, MEDLINE In-Process and Other Non-Indexed Citations, EMBASE, Cochrane Library and Cumulative Index to Nursing & Allied Health Literature. Bibliographic references of selected studies were also searched. After a review of the title and abstracts, relevant studies were obtained and the full reports evaluated. All studies meeting explicit inclusion and exclusion criteria were retained. Where appropriate, a meta-analysis was undertaken to determine the pooled estimate of effect of specialized multidisciplinary community-based care for explicit outcomes. The quality of the body of evidence, defined as one or more relevant studies was determined using GRADE Working Group criteria. (11) INCLUSION CRITERIA: Randomized controlled trialSystematic review with meta analysisPopulation includes persons with New York Heart Association (NYHA) classification 1-IV HFThe intervention includes a team consisting of a nurse and physician one of which is a specialist in HF management.The control group receives care by a single practitioner (e.g. primary care physician (PCP) or cardiologist)The intervention begins after discharge from the hospitalThe study reports 1-year outcomes
The intervention is delivered predominately through home-visitsStudies with mixed populations where discrete data for HF is not reported
OUTCOMES OF INTEREST
All cause mortalityAll cause hospitalizationHF specific mortalityHF specific hospitalizationAll cause duration of hospital stayHF specific duration of hospital stayEmergency room visitsQuality of Life
SUMMARY OF FINDINGS
One large and seven small randomized controlled trials were obtained from the literature search. A meta-analysis was completed for four of the seven outcomes including: All cause mortalityHF-specific mortalityAll cause hospitalizationHF-specific hospitalization.Where the pooled analysis was associated with significant heterogeneity, subgroup analyses were completed using two primary categories: direct and indirect model of care; andtype of control group (PCP or cardiologist).The direct model of care was a clinic-based multidisciplinary HF program and the indirect model of care was a physician supervised, nurse-led telephonic HF program. All studies, except one, were completed in jurisdictions outside North America. (12-19) Similarly, all but one study had a sample size of less than 250. The mean age in the studies ranged from 65 to 77 years. Six of the studies(12;14-18) included populations with a NYHA classification of II-III. In two studies, the control treatment was a cardiologist (12;15) and two studies reported the inclusion of a dietitian, physiotherapist and psychologist as members of the multidisciplinary team (12;19). ALL CAUSE MORTALITY: Eight studies reported all cause mortality (number of persons) at 1 year follow-up. (12-19) When the results of all eight studies were pooled, there was a statistically significant RRR of 29% with moderate heterogeneity (I(2) of 38%). The results of the subgroup analyses indicated a significant RRR of 40% in all cause mortality when SMCCC is delivered through a direct team model (clinic) and a 35% RRR when SMCCC was compared with a primary care practitioner. HF-SPECIFIC MORTALITY: Three studies reported HF-specific mortality (number of persons) at 1 year follow-up. (15;18;19) When the results of these were pooled, there was an insignificant RRR of 42% with high statistical heterogeneity (I(2) of 60%). The GRADE quality of evidence is moderate for the pooled analysis of all studies. ALL CAUSE HOSPITALIZATION: Seven studies reported all cause hospitalization at 1-year follow-up (13-15;17-19). When pooled, their results showed a statistically insignificant 12% increase in hospitalizations in the SMCCC group with high statistical heterogeneity (I(2) of 81%). A significant RRR of 12% in all cause hospitalization in favour of the SMCCC care group was achieved when SMCCC was delivered using an indirect model (telephonic) with an associated (I(2) of 0%). The Grade quality of evidence was found to be low for the pooled analysis of all studies and moderate for the subgroup analysis of the indirect team care model. HF-SPECIFIC HOSPITALIZATION: Six studies reported HF-specific hospitalization at 1-year follow-up. (13-15;17;19) When pooled, the results of these studies showed an insignificant RRR of 14% with high statistical heterogeneity (I(2) of 60%); however, the quality of evidence for the pooled analysis of was low. DURATION OF HOSPITAL STAY: Seven studies reported duration of hospital stay, four in terms of mean duration of stay in days (14;16;17;19) and three in terms of total hospital bed days (12;13;18). Most studies reported all cause duration of hospital stay while two also reported HF-specific duration of hospital stay. These data were not amenable to meta-analyses as standard deviations were not provided in the reports. However, in general (and in all but one study) it appears that persons receiving SMCCC had shorter hospital stays, whether measured as mean days in hospital or total hospital bed days. EMERGENCY ROOM VISITS: Only one study reported emergency room visits. (14) This was presented as a composite of readmissions and ER visits, where the authors reported that 77% (59/76) of the SMCCC group and 84% (63/75) of the usual care group were either readmitted or had an ER visit within the 1 year of follow-up (P=0.029). (ABSTRACT TRUNCATED)