Reporting on the Safety of Dementia Care for 2010-2021 in South Korea

Abstract This is the second report on the benefit or risk of treatment with anti-Alzheimer’s disease (AD) drugs (AADs). As a tool for monitoring citizen health, we use information from the National Health Service (NHIS) for all citizens via an information retrieval engine to link International Classification of Diseases (ICD)-10 codes and death tables of AADs with big data to create a standard comparison between AADs and deaths over 12 years. According to the International Classification of Disease-10 (ICD-10) codes, the correlation between the trend in the number of deaths and the number of deaths per 100,000 people was analyzed via the National Health Insurance System (NHIS) through the decision of the Open Data Mediation Committee of South Korea (case number: 2020-029) and information disclosure (20240110). The Dementia Management Act (DMA) in South Korea was enacted in 2012 and all dementia subjects are registered at Central Dementia Center. In completely blinded randomized states, we enrolled subjects from 2012 (the year of the Dementia Management Act) to 2018 (the year of the Dementia National Responsibility System was implemented). We observed a correlation between the AD and AAD groups from 2005 to 2019 on Sorok Island and from 2010 to 2021 on the NHIS of South Korea. This study investigated the increase and decrease in mortality in AD patients with AAD. We analyzed the mortality per 100,000 people with the R2. A simple linear equation was applied to trace the increase (+) or decrease (-) in deaths with the Friedman and Kolmogorov‒Smirnov tests for decisions on the listed, delisted, and sustainable drugs based on R2 > 0.75. Total deaths from AD caused by AAD decreased (R2, 0.993) from 2010 to 2021. From 2012 to 2018, the drugs listed to reduce deaths were fluoxetine (R2, 0.826), escitalopram (0.955) and galantamine (0.825) in South Korea. The delisted drugs used were donepezil (0.861), rivastigmine (0.754), risperidone (0.816), quetiapine (0.977), and haloperidol (0.973). The sustainable drugs used were memantine (0.108), olanzapine (0.124), fluvoxamine (0.001), oxcarbazepine (0.52), and trazodone (0.331). With the 12-year trend, galantamine, fluoxetine, could be therapeutics for AD, and sertraline and aripiprazole could be therapeutics; however, additional observation is needed. Deaths and AADs based on NHIS big data could be useful for safe dementia care.