Association of an early multidisciplinary HF pathway and inpatient rehabilitation with 1-year survival: a four-group real-world analysis (2020–2025)
R Fujimoto, C Watanabe, T Muneyasu, T Ikegami, M Takeuchi, R Okamoto, T Yoshino, T Yamanaka, T OkaAbstract
Background
Heart failure (HF) care pathways and cardiac rehabilitation (CR) are recommended to standardize multidisciplinary management, yet real-world evidence describing how their implementation relates to long-term survival remains limited, particularly when both are considered jointly.
Purpose
To describe the association of an HF clinical pathway featuring early multidisciplinary involvement and inpatient CR with 1-year all-cause mortality after HF hospitalization.
Methods
We retrospectively analyzed HF hospitalizations between 1 April 2020 and 31 March 2025 in a Japanese hospital-based DPC dataset. The HF pathway targeted early coordinated multidisciplinary involvement (PT/OT, nursing, dietitians, and pharmacists), initiated as early as the day after admission whenever feasible. CR was defined as inpatient rehabilitation. Mortality included in-hospital deaths and follow-up deaths within 1 year. Patients were classified into four groups by pathway implementation (Path+/−) and CR exposure (Rehab+/−): Path−/Rehab−, Path−/Rehab+, Path+/Rehab−, and Path+/Rehab+. Time-to-event was defined from discharge to death (censored at 365 days). Unadjusted survival was assessed using Kaplan–Meier curves and the log-rank test. To account for baseline differences, we additionally performed propensity score–based inverse probability weighting using demographics, comorbidities, HF severity proxies (LVEF, BNP), acute care indicators (emergency admission, ICU/HCU, oxygen therapy, intravenous drug use), and prior HF admissions. Weighted Cox regression with robust variance was used to compare groups, and a global Wald test evaluated overall between-group differences.
Results
In unadjusted analyses (n=1,494; deaths=233), 1-year survival differed significantly across the four groups (log-rank χ²=22.16, p=0.0001; Figure 1). Observed deaths by group were: Path−/Rehab− 92, Path−/Rehab+ 94, Path+/Rehab− 16, and Path+/Rehab+ 31. In the weighted analysis (n=990; deaths=150), between-group differences remained (global Wald χ²=9.72, p=0.0211; Figure 2). Using Path−/Rehab− as reference, hazard ratios were lower for Path−/Rehab+ (HR 0.61, 95%CI 0.42–0.87, p=0.007) and Path+/Rehab− (HR 0.44, 95%CI 0.21–0.94, p=0.034), while Path+/Rehab+ showed a similar direction but did not reach statistical significance (HR 0.62, 95%CI 0.37–1.04, p=0.072).
Conclusions
In this real-world HF cohort, early multidisciplinary pathway implementation and inpatient CR were associated with better 1-year survival compared with neither intervention. The absence of statistical significance in the Path+/Rehab+ group despite a consistent direction underscores the need for further studies to confirm effectiveness and clarify which components, timing, and patient subgroups derive the greatest benefit.Unadjusted KM survival by HF pathway, CRIPTWCox model adjusted survival by HFpath, CR