Skip to the content

Research summaryStrong evidence

Cognitive Stimulation Therapy: The Best-Supported Group Activity in Dementia Care

A Cochrane review of 37 trials, the founding randomized trial, and a national guideline recommendation

4 min read4 sourcesReviewed September 30, 2026

Written for: Medical directors, Activity directors, Administrators and boards

Reading settings
Text size
Line spacing
Colors

Settings are kept in this browser, for every Research Center page.

Cite this page

Key takeaways

  • The 2023 Cochrane review of 37 trials found that cognitive stimulation probably brings a small benefit to cognition at the end of treatment, roughly equal to a six-month delay in the decline usually expected in mild to moderate dementia.1
  • Benefits were larger with sessions twice a week or more, and in mild rather than moderate dementia.1
  • Self-reported quality of life probably improved slightly, and no harms were reported.1
  • The founding trial, 14 sessions over 7 weeks with 201 people, improved cognition and quality of life2, and an economic analysis found a high probability that it is cost-effective.3
  • NICE, England’s clinical guideline body, recommends offering group cognitive stimulation therapy to people living with mild to moderate dementia.4

At a glance

37 trials

in the 2023 Cochrane review of cognitive stimulation Source: Woods et al., 20231

About 6 months

is the delay in expected cognitive decline that the average benefit roughly equals Source: Woods et al., 20231

14 sessions

of 45 minutes, twice a week for 7 weeks, in the founding trial of 201 people Source: Spector et al., 20032

What cognitive stimulation is1 min

In brief: Themed, sociable group sessions that engage thinking, memory and language, without testing.

Cognitive stimulation brings a small group together for themed sessions that engage thinking, memory and language through activities and discussion, such as word games, sorting and categorizing, current events and reminiscence. The emphasis is on opinions and enjoyment rather than right answers.

Cognitive Stimulation Therapy (CST) is the manualized version tested in the founding trial: 14 sessions of 45 minutes, twice a week for seven weeks, led by trained facilitators.2

The evidence1 min

In brief: Consistent small benefits across 37 trials, a positive founding trial, and a favorable cost analysis.

An economic evaluation alongside the founding trial found health and social care costs slightly higher over the eight weeks, but a high probability that cognitive stimulation is cost-effective, particularly given the gains in quality of life.3 NICE recommends offering a range of activities tailored to the person’s preferences, and offering group cognitive stimulation therapy to people living with mild to moderate dementia.4

What makes it work1 min

In brief: Frequency, the right stage, a consistent group, a skilled facilitator and a respectful atmosphere.

  • Frequency: benefits were larger with sessions twice a week or more.1
  • Stage: benefits were larger in mild than in moderate dementia.1
  • A consistent small group and facilitator, so members come to know each other.
  • Opinions over facts: no one is put on the spot, and there are no wrong answers.

How Cognicopia supports it1 min

In brief: Materials matched to each stage, calendars for twice-weekly sessions, and staff training.

  • The Packet Builder’s word, number and letter activities print at three support tiers, so a group’s materials can match each member’s stage.
  • The Facility Portal schedules group sessions across the month and balances the six activity pillars.
  • The Academy masterclass Matching Activities to Cognitive Stage covers how to pitch an activity at the right level.

Running CST as tested

Cognicopia is not the CST program. Facilities that want to run CST as it was tested should use the published CST manual and facilitator training. Cognicopia’s pages can supply materials for the sessions.

What this means for your community

  • For residents with mild to moderate dementia, plan themed group sessions at least twice a week.
  • Keep groups small and consistent, and use materials matched to each member’s stage.
  • Track attendance and engagement, and review the program at 90 days.
  • If you adopt CST itself, follow its manual and train its facilitators.

What this evidence does not show

  • The benefit is small, and it was measured at the end of the course. Longer-term effects are less certain.
  • Most trials were in mild to moderate dementia. Evidence for advanced dementia is limited.
  • The trials tested structured programs with trained facilitators. Generic “brain games” are not the same thing.

References

  1. Woods B, Rai HK, Elliott E, Aguirre E, Orrell M, Spector A. Cognitive stimulation to improve cognitive functioning in people with dementia. Cochrane Database of Systematic Reviews. 2023;1:CD005562.
  2. Spector A, Thorgrimsen L, Woods B, et al. Efficacy of an evidence-based cognitive stimulation therapy programme for people with dementia: randomised controlled trial. British Journal of Psychiatry. 2003;183:248–254.
  3. Knapp M, Thorgrimsen L, Patel A, et al. Cognitive stimulation therapy for people with dementia: cost-effectiveness analysis. British Journal of Psychiatry. 2006;188(6):574–580.
  4. National Institute for Health and Care Excellence. Dementia: assessment, management and support for people living with dementia and their carers (NICE guideline NG97). NICE; 2018. Accessed September 30, 2026. https://www.nice.org.uk/guidance/ng97 View source (opens in a new tab)

How to cite this page

Cognicopia. Cognitive Stimulation Therapy: The Best-Supported Group Activity in Dementia Care. Cognicopia Clinical Science & Research Center. Updated September 30, 2026. Accessed [date]. https://cognicopia.org/resources/research/cognitive-stimulation/

An educational summary for professionals, not medical advice. It reports each source’s findings as the source states them, and was reviewed against its sources on September 30, 2026. Cognicopia makes the tools this page mentions. Found an error? Tell us.