· The Lancet. Public health 2022 · epidemiological forecasting and demographic decomposition analysis · n=?

Estimation of the global prevalence of dementia in 2019 and forecasted prevalence in 2050: an analysis for the Global Burden of Disease Study 2019.

Cited 5434 times in the scientific literature.

Level 3 - non-randomized controlled study

Non-clinical macro-epidemiological forecasting and demographic modelling study (level assigned by design analogy, not clinical CEBM)

PubMed 34998485 · doi:10.1016/S2468-2667(21)00249-8 · record verified 2026-08-27

What was done

Using data from the Global Burden of Diseases, Injuries, and Risk Factors Study (GBD) 2019, researchers forecasted dementia prevalence from 2019 to 2050 globally, by world region, and by country. The model incorporated relative risks and forecasted prevalence of three GBD risk factors (high body-mass index, high fasting plasma glucose, and smoking), with linear regression incorporating education to estimate non-risk-attributable prevalence. A decomposition analysis assessed the relative contributions of risk factor trends, education, population growth, and population ageing.

What was found

Global dementia cases are forecasted to increase from 57.4 million (95% uncertainty interval [UI] 50.4–65.1) in 2019 to 152.8 million (95% UI 130.8–175.9) in 2050. Global age-standardised both-sex prevalence was projected to remain essentially stable, changing by 0.1% (95% UI -7.5 to 10.8). Female-to-male case ratios were 1.69 (95% UI 1.64–1.73) in 2019 and projected at 1.67 (95% UI 1.52–1.85) in 2050. Projected percentage increases varied geographically, ranging from 53% (95% UI 41–67) in high-income Asia Pacific and 74% (58–90) in western Europe to 357% (323–395) in eastern sub-Saharan Africa and 367% (329–403) in north Africa and the Middle East. Population growth drove most increases in sub-Saharan Africa, whereas population ageing drove most increases in east Asia.

Why it matters

This study provides country-specific projections to guide public health resource allocation and long-term care infrastructure planning over the coming three decades, highlighting disproportionate burden growth in developing regions.

Limits

The abstract reports on mathematical forecasting models rather than direct trial or prospective cohort follow-up, and specific participant sample sizes (n) are not reported. The model incorporates only three lifestyle/metabolic risk factors and education, omitting other established dementia risk factors.

Cited by