Selfcare management intervention in heart failure (SMART-HF) -a multicentre randomized controlled trial

Abstract

Abstract Background Clinical studies evaluating home-based interventions for heart failure (HF) patients are performed on patients that are too young, too often male and predominately have reduced ejection-fraction (EF). The Selfcare Management Intervention in Heart Failure (SMART-HF) study was designed to evaluate a digital intervention enhancing self-care behaviour in a more generalizable population. The aim was to see if the intervention could reduce in-hospital care due to HF. Methods SMART-HF was a randomized controlled trial, recruiting patients from seven centres in Region Skåne. Patients were randomized 1:1 to a control group (CG) receiving standard care or an intervention group (IG) who were equipped with the digital intervention. The intervention educates the patient about HF and supports the patient with medication and symptom monitoring, also notifying the patient in case of deterioration. The primary outcome was number of in-hospital days due to HF after 240 days follow-up. Results A total of 124 patients were randomized and 118 (CG: 60, IG: 58) were included in the analysis. The mean age was 79 years, 39% were women and 45% had EF <40%. The groups were well balanced at randomization. A total of 367 HF in-hospital days were registered among 33 patients, and patients in the IG lost on average 2.2 days fewer to HF-hospitalizations than patients in the CG, a reduction by 52%, p<0.001. An unadjusted Cox-regression to analyse the event-free survival yielded a hazard ratio = 0.50 (CI:0,24–0,96), p=0.046. The self-care behaviour was 21% better in the IG, p=0.014. The median system adherence was 85% after 240 days. Conclusions The intervention significantly reduced HF-related in-hospital days and improved event-free survival in a representative HF-population. The findings match previous published data from studies evaluating the same intervention. Funding Acknowledgement Type of funding source: None