Background-The effectiveness of heart failure disease management programs in patients under cardiologists' care over long-term follow-up is not established. Methods and Results-We investigated the effects of a disease management program with repetitive education and telephone monitoring on primary (combined death or unplanned first hospitalization and quality-of-life changes) and secondary end points (hospitalization, death, and adherence Key Words: heart failure Ⅲ education Ⅲ disease program management Ⅲ case management Ⅲ controlled clinical trials Ⅲ quality of life Ⅲ patient compliance R ecent disease management program (DMP) meta-analyses have reported reductions in mortality and hospitalizations of heart failure (HF) patients. 1-4 However, important issues in DMP for HF remain to be resolved. For example, few investigations include non-high-risk HF for early hospitalization managed by cardiologists or report long-term results. [3][4][5][6] No studies have reported the long-term effects of a repetitive-cyclic reeducation program. 3,4,7,8 Most DMPs have been tested in high-risk HF patients that have been discharged from the hospital, and it has been suggested that DMPs are less effective when patients are already being treated by an HF specialist. 1,3,7,9,10 Improved survival is associated with cardiologist care and with multidisciplinary teams providing specialized follow-up. 4,8 Whether both together could benefit HF is not well defined.
Clinical Perspective see p 124We tested whether a DMP consisting of a long-term repetitive multidisciplinary education program and telephone monitoring could benefit HF outpatients in usual ambulatory care already under the care of a cardiologist with experience in HF. Figure 1). The first patient was randomized on October 5, 1999, and the last on January 18, 2005, in the Heart Institute of the São Paulo University Medical School. At least an 18-month follow-up from inclusion of the last patient was planned to initiate the trial analysis. Referred patients, with no exclusion criteria, were randomized in a 2:1 ratio between the intervention and control parallel groups, respectively. A computer-generated randomization list was drawn up by the statistician. The randomization 2:1 was used based on the previously published benefit of DMP in HF. The 2:1 randomization sequence was developed in blocks of 3, including 2 interventions and 1 control. The order of interventions and control within each block was also randomly assigned. To avoid deduction of the next treatment allocation and for arrangement of education classes, researchers were blinded for block size; each randomization included a number of patients in multiples of 3, with at least 15 eligible participants, except for the last group. The order of subjects in each group was randomized using a computer program. For allocation concealment, sequential, numbered, opaque, and sealed envelopes were used. Investigators ensured that the envelopes were opened sequentially only after the participants' names were written on the app...