National Surveillance-Based Retrospective Longitudinal Analysis of Dementia Prevalence Trends in the Republic of Korea, 2016–2025, with a One-Year 2026 Forecast Using National Health Insurance Service Administrative Data

Background/Objectives: Dementia poses a major public health challenge in South Korea, where population ageing has accelerated rapidly. We aimed to describe national 2016–2025 dementia prevalence trends and to generate a one-year 2026 forecast using NHIS-linked administrative data. Methods: We performed a retrospective longitudinal analysis of de-identified aggregate administrative data from the Ministry of Health and Welfare Municipal and District Dementia Status Report (2016–2025) and NHIS age–sex pivot table records. This was a purely descriptive and forecasting analysis; no inferential statistical tests were applied. Descriptive statistics were computed for national, sex-stratified, regional, and age-stratified outcomes. For the 2026 forecast, we used a prespecified demographic-offset time-series approach: the population aged 65 years or older and the estimated prevalence rate were modelled separately using damped-trend Holt exponential smoothing, then recombined to estimate patient count. Residual bootstrap resampling (20,000 iterations) was used to derive a 95% prediction interval for the patient count. A sensitivity analysis using ARIMA(1,1,0). Results: 602,633 (8.89%) in 2016 to 955,585 (9.09%) in 2025, a 58.6% increase. Females consistently showed higher prevalence than males (2025: 9.55% vs. 8.52%), and regional analysis identified South Jeolla Province (10.17%) as the highest-prevalence region and Ulsan Metropolitan City (8.10%) as the lowest. The female-to-male cumulative rate ratio reached 2.49 in the ≥85-year group. The primary 2026 forecast estimated 1,003,407 dementia patients among adults aged 65 years or older, with an estimated prevalence of 9.19% and a 95% prediction interval of 983,838–1,022,158. Conclusions: South Korea’s rising dementia burden is primarily driven by population ageing rather than a sharp increase in age-specific prevalence. The 2026 forecast supports urgent planning for a national dementia care population of approximately one million people.