Brain HealthResearch PaperOpen Access

Social Engagement as Dementia Prevention Tested in Canada and Botswana RCT

A pilot RCT tests whether participatory community research activities can preserve executive function and reduce dementia risk in socially isolated older adults.

Thursday, October 1, 2026 1 view
Published in BMJ Open
An elderly man and woman seated at a community table outdoors with neighbors, reviewing papers and engaged in animated discussion in a village setting

Summary

Drugs and standard lifestyle programs have not reliably prevented dementia, and most people with dementia live in lower-income countries. This paper describes a pilot trial of a different approach. Socially isolated adults aged 65 and older in rural Botswana and urban Montreal will form small groups around shared concerns. They will analyze the problem, share it with local stakeholders, carry out solutions, and evaluate the results. The idea is that shared decision-making, using evidence, and innovating exercise the brain's executive function. About 200 people will take part, with half receiving the program immediately and half later. The main goal is to learn whether the approach is feasible and acceptable. Cognitive effects are a secondary, exploratory outcome. No results are reported yet.

Detailed Summary

Dementia is projected to affect 152 million people by 2050, mostly in low- and middle-income countries. Pharmacology has not delivered: a 2025 systematic umbrella review found insufficient evidence for repurposing any systemic drug to reduce dementia risk. Multidomain lifestyle trials such as preDIVA, MAPT and HATICE did not reduce dementia incidence. In 14 exercise studies reviewed by Gates and colleagues, 92% of cognitive outcomes did not differ from controls. Social isolation, meanwhile, is a known risk factor.

This paper is a trial protocol and reports no results. The authors propose that authentic participatory research could be a prevention strategy in itself. Its three features are shared governance, active use of evidence, and innovation. They map onto the proposed components of executive function: inhibition (effortful control), active memory, and innovation. Older adults act as authors and decision-makers rather than passive recipients of pills or exercise instructions.

The pilot randomized trial will enroll adults aged 65 and older without dementia, prioritizing those who are socially isolated. In southern Botswana, participants are recruited at old-age pension payouts in two rural communities, and one community is randomized to immediate intervention. In Montreal, participants come from families of long-term care residents with advanced dementia or from community organizations, and individuals are randomized. The target is at least 50 per arm, or 200 in total. Botswana's sample is 30 women and 20 men per community.

The intervention has six steps: individuals name a priority concern; groups of 3 to 10 form around shared concerns; groups analyze causes and solutions through narratives or fuzzy cognitive mapping; they share findings using photovoice, participatory video or cellphilms; they implement actions; and they self-evaluate. Each cycle involves about 12 weekly or fortnightly meetings of 1 to 2 hours, and groups are expected to tackle three to four issues. In Botswana, male facilitators work with men and female facilitators with women.

The primary outcome is feasibility: recruitment, retention, adherence, data quality, fidelity, an economic evaluation, and sustainability 18 months after the intervention ends. Secondary outcomes are the Frontal Assessment Battery (maximum score 18), the Integral Brain Health Index, social isolation, personal goal attainment, and survival. These are measured at baseline, 18 and 33 months. The delayed group crosses over at roughly 14 months. Analysis is blinded and uses generalized linear mixed models with site as a random effect, difference-in-differences comparisons, and intra-class correlation estimates. A hybrid thematic analysis and the most significant change technique will capture participants' experience.

The authors acknowledge that the intervention period may be too short to change executive function. The pilot is also not powered to detect moderate effects. Using the Whitehead framework, 50 per arm suits a future definitive trial aimed at effect sizes of 0.1 to 0.2. Recruitment runs from April 2025 to December 2026, and the pilot ends in March 2028 (ISRCTN14273485).

Key Findings

  • Protocol only, with no efficacy results yet. The pilot targets at least 200 adults aged 65 and older (about 100 in rural Botswana and 100 in Montreal).
  • Dementia is projected to reach 152 million people by 2050, mostly in low- and middle-income countries. A 2025 umbrella review found insufficient evidence to repurpose any systemic drug for dementia risk reduction.
  • Cited prior trials (preDIVA, MAPT, HATICE) showed no reduction in dementia incidence, and 92% of cognitive outcomes in 14 exercise studies reviewed by Gates et al. did not differ from controls.
  • The six-step participatory cycle involves groups of 3 to 10 and about 12 meetings of 1 to 2 hours per cycle, with each group expected to address three to four self-chosen issues.
  • The primary outcome is feasibility (recruitment, retention, adherence, fidelity, cost). The secondary outcome is executive function on the Frontal Assessment Battery (max 18) at baseline, 18 and 33 months.
  • At 50 per arm, the pilot is judged adequate to inform a future definitive trial powered for small effect sizes of 0.1 to 0.2 (Whitehead framework), but it is not powered to detect moderate effects.
  • Recruitment runs from April 2025 to December 2026, with study completion in March 2028.

Methodology

This is a two-site pilot randomized trial with an immediate versus delayed (crossover) intervention design. In Botswana, one of two rural communities is randomized to start first. In Montreal, individuals are randomized before being grouped by shared concern. The minimum sample is 200, with at least 50 per arm. Assessments occur at baseline, 18 and 33 months, with the delayed group crossing over at about 14 months. Primary feasibility outcomes are analyzed descriptively with an economic evaluation. Secondary outcomes (Frontal Assessment Battery, Integral Brain Health Index) are analyzed blinded using generalized linear mixed models with site as a random effect, difference-in-differences comparisons, and intra-class correlation estimates. Qualitative analysis uses hybrid thematic analysis and the most significant change technique.

Study Limitations

The authors state that the intervention period may be too short to produce a meaningful change in executive function, and the pilot is not powered to detect moderate effects with statistical significance. Recruitment and sampling differ greatly between sites (pension payout points in Botswana versus family members of long-term care residents and community organizations in Montreal). Conflict-of-interest and funding details were not available in the supplied text.

Enjoyed this summary?

Get the latest longevity research delivered to your inbox every week.

Enter your email to subscribe: