OBJECTIVES: We compared the longer-term impact of the two most commonly applied forms of post-discharge management designed to minimize recurrent hospitalization and prolong survival in typically older patients with chronic heart failure (CHF). METHODS: We followed a multi-center randomized controlled trial cohort of Australian patients hospitalized with CHF and initially allocated to home-based or specialized CHF clinic-based intervention for 1368 ± 216 days. Blinded endpoints included event-free survival from all-cause emergency hospitalization or death, all-cause mortality and rate of all-cause hospitalization and stay. RESULTS: 280 patients (73% male, aged 71 ± 14 years and 73% left ventricular systolic dysfunction) were initially randomized to home-based (n=143) or clinic-based (n=137) intervention. During extended follow-up (complete for 274 patients), 1139 all-cause hospitalizations (7477 days of hospital stay) and 121 (43.2%) deaths occurred. There was no difference in the primary endpoint; 20 (14.0%) home-based versus 13 (7.4%) clinic-based patients remained event-free (adjusted HR 0.89, 95% CI 0.70 to 1.15; p=0.378). Significantly fewer home-based (51/143, 35.7%) than clinic-based intervention (71/137, 51.8%) patients died (adjusted HR 0.62, 95% CI 0.42 to 0.90: p=0.012). Home-based versus clinic-based intervention patients accumulated 592 and 547 all-cause hospitalizations (p=0.087) associated with 3067 (median 4.0, IQR 2.0 to 6.8) versus 4410 (6.0, IQR 3.0 to 12.0) days of hospital stay (p<0.01 for rate and duration of hospital stay). CONCLUSIONS: Relative to clinic-based intervention, home-based intervention was not associated with prolonged event-free survival. Home-based intervention was, however, associated with significantly fewer all-cause deaths and significantly fewer days of hospital stay in the longer-term. TRIAL REGISTRATION: Australian New Zealand Clinical Trials Registry number 12607000069459 (http://www.anzctr.org.au/Trial/Registration/TrialReview.aspx?id=81803).

Download Post Print Article