Medical Cost Trends for Diseases Associated with Bedridden Rural Residents in Japan

OBJECTIVES: The Center of Healthy Aging Program (CHAP) potentially contributed to reduce health expenditures in a Japanese rural area, Hirosaki. This study aimed to assess the effect of CHAP on the incidence and the medical costs for diseases associated with future bedridden residents.

METHODS: We compared the treatment incidence and all-cause medical costs for dementia, stroke, or any fracture from 1st July 2015 to 30th June 2019 (follow-up) using residents’ claims data in Hirosaki between three groups identified through participation in checkups a year before the follow-up period (baseline): CHAP participants (CHAP group), residents who received an annual standard health checkup (checkup group), and others (no checkup group). We included residents aged 40-75 years because the standard checkup is provided only for that age group, and excluded residents with < 1 day of follow-up period. We used age, sex, care level, and Charlson comorbidity index to consider the difference in characteristics between groups.

RESULTS: Of the 44514 eligible residents, the number of the CHAP, checkup, and no checkup groups were 34163 (76.7%), 9946 (22.3%), and 405 (1.0%), respectively. The incidence rate per 1000 person-years of new onsets of dementia, stroke, and fracture during the follow-up period were 0.1, 279.1, and 1155.1, respectively. The adjusted incidence rate did not differ between the three groups for any disease. The annual all-cause costs per patient with newly onset dementia and fracture slightly increased throughout the follow-up period (JPY 2.3 to 3.1 million and JPY 0.4 to 0.7 million, respectively), but remained stable for stroke (approximately JPY 2.3 million), which did not differ between groups.

CONCLUSIONS: We did not observe any reduction in bedridden-associated diseases between checkup participants and non-participants in the 4-year follow-up period, because such diseases involve a long course before increasing in severity. Hence, data linkage throughout long-term is required.