- Design
- Individual participant data meta-analysis of 22 population-based cohorts (COSMIC)
- Population
- 38,418 older adults in 18 countries or territories; 3,001 incident dementia cases
- Primary outcome
- Age- and sex-standardised incidence of all-cause dementia
- Effect
- 9.7 per 1,000 person-years (95% CI 7.89 to 11.97); women 10.6 vs men 8.6; low- and middle-income 13.6 vs high-income 8.3
This individual participant data meta-analysis combined 22 population-based cohorts from the COSMIC consortium: 38,418 people across 18 countries or territories on six continents. Across 287,632 person-years, 3,001 people developed dementia, based on clinical diagnosis.
Standardised incidence was 9.7 per 1,000 person-years (95% CI 7.89 to 11.97). Age-standardised incidence was higher in women (10.6) than men (8.6), and higher in low- and middle-income countries (13.6, 95% CI 9.67 to 19.24) than in high-income countries (8.3). Applying the rates to the projected 2050 population, the authors estimated 24.7 million new cases, assuming incidence does not change.
The figures give a benchmark for planning memory services. The intervals are wide, the cohorts differ in their diagnostic methods, and the projection assumes constant incidence, which may not hold.
- Standardised dementia incidence was 9.7 per 1,000 person-years across 22 cohorts.
- Incidence was higher in women (10.6) than in men (8.6).
- Incidence was higher in low- and middle-income countries (13.6) than in high-income countries (8.3).
- The 24.7 million projected new cases in 2050 assumes constant incidence.
- The study does not report data specific to India.
Why it matters
Most prior incidence data came from high-income countries, so this widens the evidence base for planning.
Don't overread it
The country and income-group comparison has overlapping intervals and the projected case numbers are model-based.
The statistics, in plain English
The confidence intervals for the income groups overlap (9.67 to 19.24 vs 6.57 to 10.54), so the difference between them is suggestive and not firmly established. A projection that assumes constant rates ignores changes in education, vascular risk and ageing patterns.
Read the rest in the app
You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.
Scan to keep reading on your phone. No account needed to start.
Tomorrow morning, before your first patient
One edition a day for neurology, written by the desk, every claim tied to its paper. Six minutes.
Get the app — free