Leveraging the DeSC Healthcare Database for Alzheimer's Disease (AD) Research in an Aging Population in a Real-World Setting in Japan

OBJECTIVES: Dementia research using secondary databases was challenging mainly due to fragmentation of healthcare service facilities covered and the overrepresentation of the working population in existing databases in Japan. DeSC Healthcare database covers 15.3 million individuals from multiple insurance systems including the Late-Stage Elderly Health Scheme (LSEHS) file having the largest elderly population coverage (20.5%), among commercially available databases in Japan. DeSC enables linkage of claims data to patient-reported outcomes (PROs) via the proprietary app, ‘kencom’. We aim to describe features of DeSC in researching AD in Japan.

METHODS: LSEHS data was used to identify the first record of AD diagnosis using the International Classification of Diseases, 10^th Revision (G30/F00) between April 2014 and March 2023. Overall follow-up period, patients with AD diagnosis and anti-dementia treatment received were summarised descriptively.

RESULTS: For the 5,396,490 individuals included in LSEHS between 2014 and 2023, the mean (standard deviation, SD) follow-up period was 45.9 (24.0) months, and 847,345 (15.7%) had an AD diagnosis. Mean (SD) age of first record of AD was 84.4 (6.0) years, 66.7% were female. Throughout the study period, the AD prevalence in LSEHS was approximately 10% per year. Prevalence was higher among patients aged ≥85 years (20%) than those aged <85 years (<10%). The proportion of AD patients treated with anti-dementia drugs decreased from 76.0% in 2014 to 68.3% in 2022. The most prescribed anti-dementia drugs were donepezil, followed by memantine and galantamine throughout the study period. DeSC offers the potential to assess qualitative outcomes in AD research, via customized surveys including caregiver burdens (e.g., Dementia Caregiver Positive Feeling Scale) using ‘kencom’.

CONCLUSIONS: DeSC is a robust healthcare data source for conducting research in various fields, particularly the elderly, offering longitudinal data. Additionally, PROs and caregiver voices can be assessed, a feature lacking in most healthcare data sources.