Has COVID-19 Caused a Persistent Transformation in Japan’s Inpatient Respiratory Rehabilitation System? A Hybrid Time-Series Analysis of 10 Years of Claims Data

This study examined the long-term impacts on respiratory rehabilitation services before and after the coronavirus pandemic and its transition to Class 5 status, using insurance data from Kanagawa Prefecture, Japan (2014 to 2024). Using seasonal autoregressive integrated moving average predictions and interrupted time series analysis, we analyzed changes in rehabilitation volume and patient numbers across 3 periods: prepandemic, pandemic, and post-Class 5 transition. The result shows service intensity increased from 12.0 to 14.2 units per patient per month, indicating greater complexity of care needed. Most notably, we identified a consistent lag pattern, with rehabilitation demand surging 1 to 2 months after each infection wave peak, providing a predictable sign for healthcare administrators. The fifth wave triggered the most dramatic response with service volume increasing by approximately 80% and patient numbers by 50% above projected values. Patient composition analysis revealed a consistent presence of both COVID-19 and non-COVID-19 patients throughout the study period. Also, the most important findings were significant increases in monthly rehabilitation claims, persisting irreversibly, even after the pandemic officially ended. The sustained increase in respiratory rehabilitation demand likely exacerbated resource allocation challenges for other rehabilitation services, suggesting potential tradeoffs in care delivery. While significant level and trend changes occurred after the pandemic's emergence, no statistically significant decrease followed the Class 5 transition, with elevated levels continuing. This persistent shift suggests the need to strengthen healthcare system resilience through optimal resource allocation, professional training, and alternative delivery models, including telerehabilitation.