Are persons with unknown health status identified by the National Health Insurance Database (KDB) system at high-risk of requiring long-term care and death?

As Japan is experiencing an aging population, the “Integrated implementation of health services and long-term care (LTC) prevention for older people” (Integrated Implementation) program has been introduced to extend the healthy life expectancy.1 Based on the National Health Insurance Database (KDB) system, municipalities can identify older people that have not received “medical care” and have not participated in “health checkups”. The Integrated Implementation requires municipalities to check those older people with unknown health status.2 In fact, people who did not participate in health checkups and the postal Kihon Checklist survey (a tool for screening frailty) had a higher risk of disability (LTC Insurance certification) and mortality.3-5 This suggests that persons with unknown health status might be at high risk of needing LTC, but we have not found any studies that have verified this speculation. Therefore, the present study aimed to clarify whether persons with unknown health status identified by the KDB system are at high risk of needing LTC or mortality. The study participants consisted of 27 300 people, all older people aged ≥75 years, living in Yamato City, Kanagawa, Japan. Data on whether the participant received medical care or participated in health checkups (annual medical expenses and dates of health checkups) in 2017 were extracted using the KDB system and divided into four groups, which followed for 2 years. Regarding medical care, individuals with annual medical expenses of 1 yen or more were defined as “persons who received medical care”. Those with annual medical expenses of 0 yen were defined as “persons who did not receive medical care”. Then, we classified the participants into four groups. Group 1 consisted of persons who received medical care and participated in health checkups; group 2 consisted of persons who received medical care, but did not participate in health checkups; group 3 consisted of persons who did not receive medical care, but participated in health checkups; and group 4 consisted of persons who did not receive medical care and did not participate in health checkups. Group 4 represented persons with unknown health status. The primary outcome was a composite end-point of “needing LTC of care level ≥2 (i.e. moderate-to-severe functional disability) or all-cause death”. The study flow diagram is shown in Figure S1. Hazard ratios (HRs) and 95% confidence intervals were calculated for each outcome by using the Cox proportional hazards model. Of the 27300 participants in the present study, 18 156 were followed up. The mean age was 79.7 years, and 46.5% were men. The comparison of primary outcomes between the four groups is shown in Table 1. The sex- and age-adjusted HRs of the primary outcome for group 2 and group 4 were significantly higher than group 1. After stratifying by sex and age groups, the sex- and age-adjusted HRs of the primary outcome for group 2 and group 4 were significantly higher than group 1 (Table 1). The present study had several limitations. First, as only data from Yamato City were used in this study, it is unclear whether the HRs would be the same for all of Japan. Second, the proportion of those who received medical care or participated in health checkups among older people (≥75 years) in Yamato City might differ from that in other areas. For example, if frail older people in Yamato City have easier access to medical care or health checkup services, the data from Yamato City would include high-risk persons into the known groups (group 1 to group 3). If so, the HR would be lower than in other municipalities. Therefore, it is unclear whether the results (HRs) in this study from Yamato City would be consistent with those estimated for all of Japan. Third, the causes of death and disability were not available from the KDB system. For example, persons with dementia of frailty might have difficulty accessing medical care or health checkups. Therefore, the causes should be investigated to seek preventive measures. In conclusion, the use of the KDB system to identify persons with unknown health status might be useful to predict which individuals are at high risk of needing LTC or dying. As persons with unknown health status have not been covered by conventional health services, these persons should be checked as part of the Integrated Implementation program. We thank the Health Promotion Division of Yamato City for their cooperation in our research. This research was funded by the Health and Labor Sciences Research Grants (No. 20AA2006). The authors declare no conflict of interest. This study was approved by the research ethics review committee of Kanagawa University of Human Services (approval no. Hodai 5–22-3, date of approval: 10 May 2022). Data sharing is not applicable to this article as no new data were created or analyzed in this study. FIGURE S1. The study flow diagram. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.