Importance

Global dementia incidence estimates are limited by underrepresentation of low- and middle-income countries and heterogeneous analytic methods. Due to these limitations, true understanding of the global dementia landscape remains incomplete.

Objective

To identify dementia incidence rates by age and sex in various regions worldwide (high-income countries and low- and middle-income countries) where data on clinically based dementia diagnoses were available.

Data sources

Retrospective community-based cohorts from the Cohort Studies of Memory in an International Consortium (COSMIC); baseline recruitment commenced between January 1990 and December 2011 and follow-up data were collected through the end of 2021. Data were analyzed from June 2024 to June 2025.

Study selection

Eligible studies included longitudinal COSMIC cohort studies with a clinically based diagnosis of dementia. Individual participant-level data from 22 population-based longitudinal cohorts (N = 38 418) spanning 18 countries or territories across 6 continents were included.

Data extraction and synthesis

Data extraction and assessment of study quality and validity were conducted according to Meta-Analyses of Observational Studies in Epidemiology (MOOSE) guidelines. Individual participant-level data were meta-analyzed in a 2-step process using random intercepts for country and study at the global and regional levels and by country income level.

Main outcomes and measures

For each study with a clinically based diagnosis, incidence rates for all-cause dementia were calculated and a 2-step meta-analysis was conducted. Using a person-years at risk framework with inverse-probability censoring weights and Poisson models, age- and sex-specific incidence were estimated and then standardized to the 2023 global age-sex population structure.

Results

Across 287 632 person-years at risk, 3001 participants developed dementia. Age- and sex-standardized incidence was 9.7 (95% CI, 7.89-11.97) per 1000 person-years at risk. Age-standardized incidence was higher in women than men (10.6; 95% CI, 8.61-13.11 vs 8.6; 95% CI, 6.94-10.72). Standardized incidence was higher in low- and middle-income countries (13.6; 95% CI, 9.67-19.24) than in high-income countries (8.3; 95% CI, 6.57-10.54). Applying age- and sex-specific rates to the projected 2050 population yielded a projected 24.7 million new cases, assuming constant incidence over time.

Conclusions and relevance

The harmonized estimates presented in this study provide a comparable benchmark for global dementia incidence and inform planning for prevention, diagnosis, and memory-care services, particularly in underrepresented low- and middle-income country settings.