A landmark 20-year analysis of the ACTIVE trial suggests that targeted and enhanced speed-based cognitive training may delay the diagnosis of dementia, providing new insights into how structured mental exercises can support long-term brain health in older populations.
Study: The impact of cognitive training on claims-based diagnosed dementia over 20 years: Evidence from the ACTIVE study. Image credit: Oksana Tkachova / Shutterstock
In a recent study published in the journal Alzheimer’s disease and dementia: translational research and clinical interventionsa group of researchers used Medicare claims data to determine whether domain-specific cognitive training reduces long-term risk of Alzheimer’s disease and related dementias (ADRD) over a 20-year period.
Background and rationale
Nearly half of Americans over the age of 85 are expected to develop dementia in their lifetime, raising the question of whether it is possible to train the brain to resist decline. Cognitive training programs promise sharper memory and faster thinking, but debate continues as to whether such improvements translate into real-world protection against ADRD. Although improvements in thinking, memory, and processing speed have been demonstrated in the short term, long-term prevention of dementia remains uncertain. Research into how psychological training influences clinical diagnosis of dementia has major implications for changes in government policy regarding aging, long-term care, and health care costs. Therefore, further research is needed to identify the types of training that are the basis for long-term protection.
Research design and methods
The Advanced Cognitive Training for Independent and Vital Older Adults Study was a four-arm, multicenter, single-blind, randomized controlled trial that enrolled 2,802 community-dwelling adults aged 65 years and older from 1998 to 1999. Four groups were created and participants were assigned accordingly: memory training, reasoning training, processing speed training, and a no-contact control group.
To be eligible, participants must score 23 or higher on the Mini-Mental Status Examination and be independent in all activities of daily living. Those who had had a stroke within the past 12 months, were undergoing cancer treatment (chemotherapy or radiotherapy), or had sensory impairments that prevented them from participating in training were excluded from the sample population.
For this analysis, participant data were linked to Medicare claims from January 1, 1999 to December 31, 2019. The final analytic sample included 2,021 individuals enrolled in traditional Medicare at baseline. ADRD was identified using the Chronic Disease Warehouse algorithm based on International Classification of Disease codes. Cause-specific Cox proportional hazards models of dementia diagnosis risk with hazard ratio adjustment for competing risks of death were estimated and adjusted for age, sex, race, education, marital status, cardiovascular comorbidities, smoking status, and baseline cognitive score, with additional adjustment for study site and training wave. Booster training sessions (offered at months 11 and 35) were analyzed separately among participants who completed at least 8 of the first 10 training sessions and were eligible for booster randomization.
Long-term dementia risk consequences
After 20 years of follow-up, 48.7% of participants in the control group received a diagnosis of ADRD. Mortality was high in all groups, with 77% dying during follow-up, reflecting the aging of the cohort. Baseline demographic and health characteristics were balanced across intervention groups.
When examining only the first assessment round, none of the three training groups showed a statistically significant reduction in dementia risk compared to the control group after adjusting for covariates. There was some indication of a slight reduction in risk, about 12 to 15%, as suggested by the hazard ratio, but again none was statistically significant.
The most notable discovery emerged when we considered booster sessions. Participants assigned to processing speed training and randomized to receive booster training had a statistically significant 25% lower risk of being diagnosed with ADRD compared to the control group (adjusted hazard ratio 0.75, 95% confidence interval 0.59 to 0.95). In contrast, participants who received speed training without booster sessions showed no protective effect (hazard ratio 1.01, 95% confidence interval 0.81 to 1.27).
Within the speed training group, participants assigned to booster training had a statistically significant borderline lower risk compared to participants eligible for booster training but not assigned (hazard ratio 0.81, 95% confidence interval 0.66 to 1.00). Therefore, although the results of this study suggest that enrichment sessions may enhance or maintain training effects, these findings should not be conclusively interpreted as causal, given that booster eligibility requires completion of a randomization session and may introduce selection bias. We also observed that training focused on memory and reasoning skills did not reduce the risk of dementia, independent of booster participation.
Although age did not significantly change the effect of training, younger participants in the memory group tended to have a lower risk of dementia, and this association was not statistically significant. Similar results were obtained with the Fine-Gray competing risks model.
Real world impact
From a real-world perspective, these findings make sense. Processing speed training focused on visual attention and rapid information processing, especially divided attention, skills closely related to everyday tasks such as driving. Previous analyzes in the same cohort showed reductions in at-fault motor vehicle crashes among participants who received speed training, reinforcing the practical value of this intervention. Our results suggest that sustained, adaptive training targeting attention and processing speed not only improves daily functioning but may also be associated with delays in clinical diagnosis of dementia, although this study relied on claims-based diagnoses rather than adjudicated clinical evaluations, which may underestimate or misclassify true dementia cases depending on health care utilization and diagnostic coding practices.
conclusion
A follow-up study over 20 years found that cognitive training focused on processing speed, especially when reinforced with booster sessions, significantly reduced the risk of ADRD. Memory and reasoning training did not demonstrate comparable long-term protective effects. These findings suggest that adaptive attention-based cognitive training may help delay dementia diagnosis in older adults. Although not a cure, such interventions can extend independence by years and reduce social burden. However, because the results were based on Medicare claims and the analytic sample excluded individuals enrolled in Medicare Advantage at baseline, generalizability may be limited and further confirmation using clinically determined results is warranted.
Reference magazines:
- Coe, N.B., Miller, KEM, Sun, C., Taggart, E., Gross, A.L., Jones, R.N., Felix, C., Albert, M.S., Lebock, G.W., Marciske, M., Ball, K.K., and Willis, S.L. (2026). The impact of cognitive training over 20 years on claims-based diagnosed dementia: evidence from the ACTIVE study. Alzheimer’s disease and dementia: translational research and clinical interventions. 12(1). DOI: 10.1002/trc2.70197, https://alz-journals.onlinelibrary.wiley.com/doi/10.1002/trc2.70197
