Factors associated with non-fatal discontinuation of a heart failure self-monitoring program in a rural super-aging japanese city
K Yonezu, S Saito, H Yamasaki, T Zaizen, M Hara, K Takahashi, S Kodama, T Ono, K Masutomo, H Zaizen, S Nobe, O Yoshiga, M Inoue, Y Tatsukawa, N TakahashiAbstract
Background
Japan is a rapidly aging country, particularly in rural regional cities. The Prefecture, where this study was conducted, has a population of about 1.09 million and an aging rate of 34.4%. In such settings, heart failure (HF) self-monitoring programs are used to reduce hospitalization, but real-world adherence remains challenging.
Purpose
To identify clinical, functional, and social factors associated with non-fatal discontinuation of an HF self-monitoring program in a rural super-aging Japanese city.
Methods
We conducted a retrospective observational study of patients who used an HF self-management form in the Prefecture between December 2020 and May 2024. Data were collected via standardized questionnaires from multiple institutions. To ensure clinical homogeneity, only patients managed at cardiac catheterization-capable hospitals were included. Patients were classified into continuation (≥1 year) and non-fatal discontinuation groups. Clinical characteristics at the index HF hospitalization and during follow-up were compared. Multivariable logistic regression analysis was performed to identify factors independently associated with non-fatal discontinuation.
Results
Among 1,508 registered patients, 663 met the inclusion criteria. Median age was 79 years (IQR 70–85); 59% were men. Overall, 531 patients (80%) continued the program, whereas 132 (20%) discontinued it. The main reason for discontinuation was relocation or hospital transfer (60%), followed by non-cardiac comorbidities or caregiver loss (26%). Compared with the continuation group, the discontinuation group was older (82 vs. 78 years, p=0.0002), had higher frailty scores (4 vs. 3, p=0.0028), and was less likely to independently record HF symptoms (80% vs. 88%, p=0.035) or engage in hobbies (19% vs. 47%, p<0.0001). At the index HF admission, they more frequently presented with NYHA class III–IV symptoms (62% vs. 41%, p<0.0001) and higher NT-proBNP levels (4590 vs. 2275 pg/mL, p=0.0002). During follow-up, the discontinuation group was more likely to remain in NYHA class III–IV (7% vs. 3%, p=0.0299) and had persistently elevated NT-proBNP levels (1241 vs. 859 pg/mL, p=0.008). Functional dependency was more common, with a Barthel Index <85 observed more frequently at discharge or initial outpatient introduction (55% vs. 39%, p=0.0022). In multivariable logistic regression, advanced age (OR 1.08 per year, 95% CI 1.01–1.14; p=0.008), NT-proBNP >3555 pg/mL (OR 2.92, 95% CI 1.26–6.78; p=0.011), and Barthel Index <85 at discharge (OR 2.78, 95% CI 1.05–7.32; p=0.038) were independently associated with non-fatal discontinuation.
Conclusion
In a real-world HF self-monitoring program in a rural super-aging Japanese city, one in five patients experienced non-fatal discontinuation. Advanced age, greater disease severity, and reduced activities of daily living were key factors, highlighting the need for tailored implementation strategies for vulnerable older adults.