Twenty-one brain-health statistics for 2026, every one cited to its primary source. Most numbers in popular brain-health writing are repeated from secondary sources without verification, and accumulate distortions on the way. This list pulls each figure from a peer-reviewed paper, government health agency, or named cohort study. Use it as a reference; cite it only when you've checked the source for yourself.

Scope: Statistics current as of 2026, focused on adults in the United States and Europe unless otherwise noted. Population-level figures unless stated. Effect sizes from randomized trials are noted as such; observational findings are flagged as observational.

Prevalence and incidence

1. 57 million people lived with dementia globally in 2024, per the World Health Organization's 2025 dementia fact sheet. The figure is projected to reach 78 million by 2030 and 139 million by 2050. (WHO, 2025)

2. Approximately 6.9 million Americans aged 65+ are living with Alzheimer's disease in 2024, per the Alzheimer's Association 2024 Alzheimer's Disease Facts and Figures report. The figure is projected to reach 13.8 million by 2060 absent breakthrough treatment. (Alzheimer's Association, 2024)

3. Age-adjusted dementia incidence has been falling in multiple Western cohorts since around 2000. The Framingham Heart Study reported a roughly 20% per-decade decline in age-specific dementia incidence between 1977 and 2008 (Wolters et al., 2020). The decline is attributed to improved cardiovascular care, education, and lifestyle factors. (Wolters et al., 2020, Neurology)

Modifiable risk factors

4. 45% of dementia cases globally are linked to 14 modifiable risk factors, per the Lancet Commission's 2024 update. This is a population-level estimate of the maximum theoretical risk reduction if all 14 factors were addressed at the population level. (Livingston et al., 2024, The Lancet)

5. Hearing loss is the single largest modifiable risk factor in the Lancet 2024 framework. The 2023 ACHIEVE randomized trial found hearing aids slowed cognitive decline by approximately 48% over three years in adults at elevated dementia risk, the strongest interventional evidence for any single Lancet factor. (Lin et al., 2023, Lancet)

6. Two new factors were added in 2024: high LDL cholesterol and untreated vision loss, joining the original twelve factors from the 2020 Lancet update. The full list, with what to do about each, is in our breakdown of the 14 factors.

7. Loneliness in older adults is associated with approximately 40% higher dementia risk across observational studies. Social isolation is a Lancet-listed factor with consistent observational evidence; the Verghese et al. 2003 NEJM study on leisure activities found participation in cognitively-engaging social activities was associated with reduced dementia risk. (Verghese et al., 2003, NEJM)

Cognitive aging timeline

8. Working memory and processing speed begin measurable decline in the late 20s to early 30s, per Salthouse's 2009 longitudinal data in Neurobiology of Aging. By age 60, the average decline from peak is roughly one standard deviation in laboratory tasks, though individual variance is substantial. (Salthouse, 2009)

9. Vocabulary and crystallized knowledge are stable into the 70s, with some studies showing continued growth into the 60s before plateau. The dissociation between fluid abilities (which decline) and crystallized knowledge (which holds) is one of the most replicated findings in cognitive-aging research.

Cognitive training and exercise

10. A year of moderate aerobic exercise grew the hippocampus by approximately 2% in older adults in Erickson et al.'s 2011 randomized trial in PNAS, with corresponding improvements in spatial memory. The control group continued to show age-related hippocampal shrinkage. (Erickson et al., 2011)

11. Speed-of-processing training reduced the 10-year hazard of dementia by 29% in adults 65+ in the ACTIVE trial's long-term follow-up. The intervention was 10 one-hour sessions over five to six weeks, with optional booster sessions. This is the strongest single result for cognitive training in older adults. (Edwards et al., 2017, Alzheimer's & Dementia)

12. The FINGER multimodal intervention improved overall cognition by approximately 25% versus controls over two years in 1,260 older Finnish adults at elevated risk. The intervention combined diet, exercise, cognitive training, and vascular monitoring. Single-component effects could not be isolated. (Ngandu et al., 2015, Lancet)

13. Working memory training does not reliably transfer to general intelligence, per the Melby-Lervåg et al. 2016 meta-analysis covering 87 studies. Near transfer (improvement on the trained task) is robust; far transfer is weak to absent. The Simons et al. 2016 consensus review reached the same conclusion. (Simons et al., 2016, PSPI)

Sleep, learning, and consolidation

14. Spaced practice produces 2-3x better long-term retention than massed practice of the same total duration, per Cepeda et al.'s 2008 meta-analysis covering 317 experiments. The spacing effect is one of the most replicated findings in cognitive science. (Cepeda et al., 2008, Psychological Science)

15. Active retrieval (testing) produces substantially better retention than re-reading the same material, per Roediger and Karpicke's 2006 Psychological Science paper. The testing effect generalizes across material types and ages. (Roediger & Karpicke, 2006)

16. Memories show approximately 10-20% better retention after a night of sleep than after an equivalent period of wake across declarative-memory tasks, per the Diekelmann and Born 2010 Nature Reviews Neuroscience synthesis. The effect is largest for material consolidated during deep slow-wave sleep. (Diekelmann & Born, 2010)

Diet

17. The 2023 NEJM trial of the MIND diet was null on cognitive outcomes over three years. Both the MIND-diet group and a mild caloric-restriction control improved modestly with no significant between-group difference. The earlier observational evidence (Morris 2015) had suggested much stronger protection. (Barnes et al., 2023, NEJM)

18. Mediterranean and DASH dietary patterns are associated with modest reductions in cognitive decline rate across multiple large cohorts. The mechanism is plausibly cardiovascular and metabolic; the cognitive effect is real but smaller than the strongest interventions. The Lancet Commission's 2024 framework treats diet primarily through cardiovascular pathways rather than as a standalone factor.

Industry and policy

19. The FTC settled with Lumos Labs for $2 million in January 2016 for advertising that the games could prevent or delay age-related cognitive decline, dementia, and Alzheimer's. The original suspended judgment was $50 million. The case set the regulatory line for what consumer brain-training apps can claim. (FTC, 2016)

20. The global brain-training app market reached approximately $9 billion in 2024, projected to grow at roughly 12-15% annually through the end of the decade per industry reports. The market remains fragmented; no single product holds majority share.

Methodology and caveats

21. Effect sizes for any single brain-training intervention sit in the 0.1-0.4 standardized-mean-difference range in the most rigorous published meta-analyses. These are real effects but small. Combining cognitive training with the rest of the evidence-based brain-health stack consistently outperforms any single intervention.

How to use this list

A few caveats worth keeping in mind when citing any of these:

Sources

The full citation list with PubMed links is in the metadata of this page and surfaced in the "Sources & further reading" block below. For the wider context on the science behind these numbers:

If you cite this page, link directly here. If you cite a specific statistic, use the primary-source link in the citations block, not the URL of this page.