Neurology · Medical Literature
[Articles] Multidomain lifestyle intervention for the prevention of cognitive decline in at-risk older adults in Latin America (LatAm-FINGERS): a single-blind, multicentre, randomised controlled trial
Abstract
A culturally adapted multidomain lifestyle intervention was feasible across Latin America and resulted in greater cognitive improvements than a flexible health-advice intervention in older adults at risk of cognitive decline. These findings extend the evidence base for multidomain lifestyle interventions to populations historically under-represented in dementia research, supporting their feasibility and scalability as strategies to reduce cognitive decline risk amid the rapidly growing burden of dementia in low-income and middle-income countries.
Clinical relevance
In older adults aged 60–77 years residing in Latin America who are at elevated risk of dementia, does a culturally adapted, structured, multidomain systematic lifestyle intervention (SLI) produce greater improvement in global cognitive function over 2 years compared with a flexible, general health-advice lifestyle intervention (FLI)? __SNAPSHOT__ {"studyDesign":"Single-blind, multicentre, randomised controlled trial (parallel-group, 1:1 allocation); co-primary endpoints of feasibility and cognitive trajectory","setting":"12 clinical sites across 11 Latin American countries (Argentina, Bolivia, Brazil, Chile, Colombia, Costa Rica, Dominican Republic, Ecuador, Mexico, Peru, Uruguay); mixed academic and community settings","sampleSize":"n=1,065 (SLI: n=539; FLI: n=526); 1,719 individuals assessed for eligibility; recruitment effectiveness (Reach) 62.0%","population":"Adults aged 60–77 years with a Cardiovascular Risk Factors, Aging, and Incidence of Dementia (CAIDE) risk score ≥6 indicating elevated dementia risk, plus suboptimal cognitive performance on standardised screening. Mean age 67.5 years (SD 4.7); 75% women; approximately 70.7% self-identified as mestizo, indigenous, mulatto, mixed, or Black. Participants with established dementia, recent cerebrovascular disease, severe cardiovascular disease, insulin-dependent diabetes mellitus, or significant renal/pulmonary comorbidities were excluded.","intervention":"2-year Systematic Lifestyle Intervention (SLI): structured, coach-led, multidomain programme combining supervised aerobic exercise, nutrition counselling based on a culturally adapted MIND diet (LatAm-MIND, incorporating regional foods such as avocado, quinoa, açaí, chia, pumpkin seeds; exercise incorporating salsa and tango), computerised cognitive training, cardiovascular risk factor monitoring, and 38 group meetings over 2 years to promote social connection and peer accountability","comparator":"2-year Flexible Lifestyle Intervention (FLI): periodic health education and general lifestyle recommendations, plus 4 group meetings covering diet, physical activity, cognitive and social engagement, and vascular risk management—without ongoing coaching or supervision","primaryOutcome":"Co-primary outcomes: (1) Feasibility, assessed via RE-AIM framework metrics (Reach, Implementation, Maintenance); (2) Change in a global cognitive composite score over 2 years","followUp":"24 months (assessments at baseline, 6, 12, 18, and 24 months); enrolment October 2021 – July 2023; last participant completed follow-up November 7, 2025","fundingSource":"Non-profit (Alzheimer's Association); the Alzheimer's Association invested >$81 million across LatAm-FINGERS and U.S. POINTER combined"}
Key findings
Both groups demonstrated improvement in global cognitive composite scores over 2 years; however, the SLI group showed significantly greater improvement than the FLI group. The mean annual change in the global cognitive composite was 0.31 SD (95% CI 0.28–0.34) per year in the SLI group versus 0.20 SD (95% CI 0.17–0.23) per year in the FLI group, yielding a mean between-group difference of 0.11 SD per year (95% CI 0.06–0.15; p<0.0001)—representing approximately 55% greater annual improvement in global cognition in the SLI group (calculated relative difference: [0.31−0.20]/0.20 = 55%, calculated). Domain-specific analyses favoured SLI for episodic memory (between-group difference 0.14 SD/year; p<0.0001), executive function (0.04 SD/year; p=0.006), and processing speed (0.04 SD/year; p=0.020). No significant between-group difference was observed on the Clinical Dementia Rating–Sum of Boxes, likely attributable to minimal baseline functional impairment. No NNT for dementia prevention was calculable from available data (dementia incidence was not a reported endpoint). Feasibility metrics: Reach 62.0%; mean adherence to SLI 71.6% over the full trial; data completeness (Maintenance) at 24 months was 84.8% in SLI vs 79.8% in FLI. Dropout was lower in the SLI group (15.2% vs 20.2%; p=0.042). Adverse events: 478 total adverse events were reported (412 in SLI, 66 in FLI). The most common adverse events were musculoskeletal symptoms (SLI: 21%; FLI: 2%) and upper respiratory infections (SLI: 9%; FLI: <1%), consistent with higher physical activity intensity in SLI. Serious adverse events occurred in 9% (SLI) versus 5% (FLI) of participants; none were adjudicated as intervention-related. Eight deaths occurred (3 SLI, 5 FLI); none were related to the intervention.
Practice implications
LatAm-FINGERS established that a culturally adapted, structured multidomain lifestyle intervention is feasible and superior to a general health-advice comparator for improving global cognition and several cognitive subdomains over 2 years in at-risk older adults across diverse Latin American LMIC settings. This represents the first large-scale RCT of this design conducted in Latin America and provides an independent replication—in a distinct continental, cultural, and socioeconomic context—of earlier findings from the Finnish FINGER trial and the U.S. POINTER study. The effect magnitude (0.11 SD/year between-group difference in a composite score) is statistically robust but the clinical significance is uncertain, as minimal clinically important differences for this composite in this population have not been established, and the outcome is a surrogate measure rather than dementia incidence or functional disability. Notably, the CDR–Sum of Boxes did not differ between groups, indicating no demonstrable effect on functional cognition, likely because participants were not functionally impaired at baseline. The study does not establish whether the intervention modifies Alzheimer's disease pathology (no biomarker data reported) or acts through cognitive reserve and resilience mechanisms. Participants were aged 60–77 with CAIDE score ≥6 and suboptimal cognition; those with severe comorbidities were excluded, limiting generalisability to frailer or older populations. The 75% female composition limits inference for men. Because the FLI comparator included some structured elements (4 group meetings), the effect of SLI is benchmarked against an active—not inert—control, which may underestimate the true benefit relative to no intervention but enhances real-world applicability. The higher rate of adverse events (particularly musculoskeletal) in the SLI group warrants attention in implementation, particularly in populations with pre-existing musculoskeletal conditions. Only the abstract was available for this summary, limiting assessment of full analytic detail, subgroup analyses, and adjustment methodology; clinicians should review the full publication. __LIMITATIONS__ Active rather than inert comparator: the FLI provided some structured health advice and 4 group meetings, which may partially attenuate the observed between-group effect; a true no-intervention arm was not included.,Single-blind design: participants and intervention staff were unmasked to group allocation, introducing performance and reporting biases, particularly for subjective outcomes.,Cognitive composite as a surrogate endpoint: dementia incidence and functional disability (CDR–SoB was non-significant) were not the primary endpoints; the clinical meaning of a 0.11 SD/year difference on a composite measure in this population is not established.,Demographic imbalance: 75% of participants were women, limiting extrapolation to men; ethnic diversity was substantial but the cohort predominantly comprised individuals with above-average education (mean ~12.9 years), potentially underrepresenting the lowest-resource populations in Latin America.,Practice-effect confounding: both groups improved over time, which could reflect test familiarity with repeated cognitive assessments across 5 time-points, inflating within-group estimates.,No biomarker data: absence of amyloid PET, tau, or CSF/blood biomarkers means mechanisms of cognitive improvement (neuroprotective vs. resilience vs. practice effect) cannot be determined.,Differential adverse event reporting: the higher rate of musculoskeletal and respiratory adverse events in SLI (likely due to increased physical contact/exercise) raises the question of surveillance intensity bias between arms.,Reach of 62% and adherence of 71.6% in SLI may not reflect real-world population-level implementation capacity in lower-resource public health systems.,Non-profit funding (Alzheimer's Association) reduces commercial conflict-of-interest risk, though grant-funded investigators may have publication pressure; the Alzheimer's Association was both funder and an institutional author affiliation.,Follow-up limited to 2 years: durability of cognitive benefit beyond the intervention period is unknown. __BOTTOM_LINE_CATEGORY__ potentially_practice_informing __BOTTOM_LINE__ LatAm-FINGERS demonstrates that a structured, culturally adapted multidomain lifestyle intervention is feasible and produces statistically significant, domain-broad improvements in global cognition (0.11 SD/year advantage over active comparator; p<0.0001) in at-risk older adults across 11 Latin American countries. The clinical magnitude remains uncertain, dementia incidence was not measured, and no biomarker data were reported; however, convergent results across FINGER, U.S. POINTER, and LatAm-FINGERS substantially strengthen the case that structured lifestyle programmes combining exercise, dietary modification, cognitive training, and cardiovascular risk management represent a safe, scalable strategy for brain health in diverse populations. Clinicians in Latin American and other LMIC settings should consider these findings relevant to population-level prevention planning, while awaiting longer-term efficacy data and cost-effectiveness analyses before widespread health-system implementation.