Smartphone-based remote monitoring (RM) in chronic heart failure reduces emergency hospital attendances, unplanned admissions and secondary care costs: a retrospective cohort study

BACKGROUND: Despite effective therapies, the economic burden of heart failure with reduced ejection fraction (HFrEF) is driven by frequent hospital attendances [1]. Treatment optimisation and admission avoidance relies on frequent symptom review and monitoring of vital signs [2]. RM programmes aim to prevent admissions and improve system efficiency by enabling self-management [3]. Few studies evaluate the economic impact of RM in HFrEF, compared to real-world matched controls [4]. We compare hospital attendances and costs between patients using Luscii, a novel smartphone-based RM platform, and matched controls receiving usual care for 3 months. PURPOSE: To assess the impact of RM on emergency department (ED) attendances, unplanned admissions and associated healthcare costs over 3 months. METHODS: A retrospective cohort study of new HFrEF referrals to our service was undertaken using the Discover dataset [5] for two cohorts (i) “RM group”: patients who used the RM platform for at least 3 months and (ii) “control group”: consecutive patients referred before the RM platform was available. The groups were matched 1:1 for age, sex, ethnicity, New York Heart Association grade and left ventricular ejection fraction. Medical co-morbidities, ED attendances, unplanned admissions and costs were extracted over 3 months from platform onboarding (RM group) or accepted referral (control group). Platform costs were added for the RM group. Differences between outcomes were analysed using t-tests, Kaplan-Meier event analysis and Cox's proportional hazard modelling. RESULTS: 146 patients (mean age 63 years; 23% female) were included in the analyses (73 “RM group”; 73 “Control group”). The groups were well-matched for all baseline characteristics except hypertension (p=0.03). Compared to the control group, after 3 months follow-up the RM group had significantly fewer ED attendances (p<0.01) and unplanned admissions (p<0.01). Accounting for RM platform costs, there was no difference between ED costs (p=0.42), but significantly lower unplanned admissions costs in the RM group (p=0.02) (Table 1). RM was protective against ED attendances (HR=0.43, p=0.02) and unplanned admissions (HR=0.26, p=0.02), which was sustained after controlling for hypertension (Table 1). Kaplan-Meier analyses found significantly lower probability of ED attendances (p=0.02) and unplanned admissions (p=0.01) in the RM group (Figure 1). CONCLUSIONS: HFrEF patients with RM were half as likely to attend ED and approximately four times less likely to need short-term unplanned admissions. The economic benefit of RM is driven by lower unplanned admission costs; the cost benefit is equivocal at the ED stage. Participants were younger than the typical HFrEF cohort. RM use could free up valuable resources to enhance standard care for older patients who decline or are unable to use RM. Further evaluation is required of the long-term impact of RM and its effect on outpatient encounters and costs. FUNDING ACKNOWLEDGEMENT: Type of funding sources: Private grant(s) and/or Sponsorship. Main funding source(s): Discover data extraction and analyst time were funded by Astra Zeneca. Astra Zeneca did not have any input to study design, analyses or reporting.