In at-risk older adults, a structured, higher-intensity multidomain lifestyle program produced a modest but statistically significant greater improvement in global cognition over 2 years versus a self-guided approach; both groups improved and interventions were safe.

Background

Nonpharmacological, multidomain lifestyle interventions targeting modifiable risk factors are promising, scalable strategies to prevent or slow cognitive decline. USPOINTER tested whether a structured, higher-intensity lifestyle program improves global cognition more than a lower-intensity, self-guided program in a diverse US cohort at elevated risk for cognitive decline and dementia.

Patients

• N = 2111 randomized; mean age 68.2 years (SD 5.2); 68.9% female; 30.8% racial/ethnic minority representation.
• At risk for cognitive decline: ages 60–79, sedentary, suboptimal MIND diet adherence, plus ≥2 additional risk factors (family history, cardiometabolic risk, race/ethnicity, older age, male sex).
• Cognitively largely intact at baseline (MMSE median 29); <5% adjudicated mild cognitive impairment.
• Five US clinical sites; 89% completed 24-month assessment.

Intervention

Structured multidomain lifestyle program (n = 1056)

• 38 facilitated peer-team meetings over 2 years with education, goal-setting, and accountability delivered by trained interventionists.
• Exercise: aerobic (4 days/week, 30–35 min), resistance (2 days/week, 15–20 min), flexibility/balance (2 days/week, 10–15 min), primarily in community facilities.
• Nutrition: encouragement and support to follow the MIND diet (e.g., provision of tools, monthly rebates for select foods such as blueberries).
• Cognitive/social engagement: home-based computerized cognitive training (3×/week, 15–20 min) and regular intellectually/socially engaging activities.
• Health monitoring: medical advisor appointments every 6 months to review blood pressure and laboratory results; guideline-based coaching.
• Team-based support with Alzheimer’s Association navigators and certified interventionists.

Control

Self-guided multidomain lifestyle program (n = 1055)

• 6 facilitated peer-team meetings over 2 years providing education and encouragement (no goal-directed coaching).
• Access to publicly available tools to support physical activity, diet, and cognitive/social stimulation; gift cards provided at meetings to facilitate behavior change.
• Annual guideline-based health monitoring during clinic visits; Alzheimer’s Association navigators provided general support.

Outcome

• Primary outcome: Annual rate of change in a global cognitive composite (executive function, episodic memory, processing speed) over 2 years.
• Key secondary outcomes (cognitive): Annual rate of change in domain composites for executive function, episodic memory, and processing speed.
• Safety: Serious and nonserious adverse events (ascertained at 6-month visits and through spontaneous reports).

Study Design

• Single-blind, multicenter randomized clinical trial (5 US sites), 1:1 allocation to structured vs self-guided interventions.
• Stratified randomization by site; outcome assessors masked; participants and intervention teams unmasked.
• Assessments at baseline and every 6 months for 2 years.
• Primary analysis: mixed-effects models for repeated measures with covariate adjustment (site, age, test version, practice effects).
• Intention-to-treat analysis; prespecified subgroup analyses (baseline cognition, APOE ε4 status) and exploratory subgroups (sex, age, cardiometabolic health).
• Trial registration: ClinicalTrials.gov NCT03688126.

Level of Evidence

Level I (multicenter randomized clinical trial, single-blind).

Follow up period

24 months (2 years); final follow-up May 14, 2025.

Results

• Primary outcome (global cognitive composite):
  • Structured: mean increase 0.243 SD per year (95% CI, 0.227–0.258).
  • Self-guided: mean increase 0.213 SD per year (95% CI, 0.198–0.229).
  • Between-group difference in slope: 0.029 SD per year (95% CI, 0.008–0.050) favoring structured.
• Secondary cognitive outcomes (domain composites):
  • Executive function: difference 0.037 SD/year (95% CI, 0.010–0.064) favoring structured.
  • Processing speed: difference 0.023 SD/year (95% CI, −0.004 to 0.050) favoring structured (not statistically significant).
  • Episodic memory: difference 0.009 SD/year (95% CI, −0.019 to 0.037); no group difference.
• Subgroups (primary outcome):
  • Effect consistent by APOE ε4 status (no interaction); appeared greater among participants with lower baseline cognition (interaction present).
  • Effects broadly consistent across sex, age (<70 vs ≥70), and cardiometabolic risk strata.
• Safety:
  • Ascertainment-based serious adverse events: 151 (structured) vs 190 (self-guided).
  • Deaths: 11 (structured) vs 5 (self-guided).
  • Intervention-related serious adverse events: 9 (structured) vs 2 (self-guided).
  • Nonserious adverse events (ascertained): 1091 (structured) vs 1225 (self-guided); volunteered nonserious events were more frequent in the structured group, likely due to more staff contact.
  • COVID-19 positivity was the most common adverse event overall, more frequent in the structured group; no clustering linked to team meetings.
• NNT: Not applicable (no dichotomous primary outcomes reported).

Limitations

• Only 5 sites; selected inclusion criteria may limit generalizability.
• Not powered for clinical endpoints (cognitive impairment/dementia) and no true no-intervention control arm.
• Participants unmasked to assignment; practice effects may contribute to improvements despite adjustments.
• Durability, scalability, and long-term clinical significance require ongoing follow-up.
• Potential bias from missing data cannot be entirely excluded.

Funding

Alzheimer’s Association (POINTER-19-611541). The sponsor participated in study design and conduct; data collection, management, analysis, and interpretation; manuscript preparation, review, and approval; and the decision to submit for publication.

Citation

Baker LD, Espeland MA, Whitmer RA, et al. Structured vs Self-Guided Multidomain Lifestyle Interventions for Global Cognitive Function: The US POINTER Randomized Clinical Trial. JAMA. Published online July 28, 2025. doi:10.1001/jama.2025.12923. ClinicalTrials.gov: NCT03688126.