Effectiveness of cognitive stimulation for dementia: a systematic review and meta-analysis

Cognitive stimulation (CS) is a nonpharmacological intervention often involving group activities and social interaction used to treat cognitive declines in people with dementia. This preregistered systematic review and meta-analysis evaluated the effectiveness of CS in producing benefits on cognition (primary outcome) and quality of life, activities of daily living, and psychological symptoms (secondary outcomes) across 44 randomized-controlled trials comprising 45 comparisons including 2,444 participants. A medium-sized effect (g = .49) on global cognition was found immediately after the intervention and was supported by decisive Bayesian evidence. Clinical relevance is defined as a reduction of 3 to 4 points on the Alzheimer’s Disease Assessment Scale Cognitive subscale; the average attenuation of cognitive decline observed was 2.41 points (after removing 1 outlier). Therefore, the observed decline was of borderline clinical relevance. CS was also found to significantly improve memory, activities of daily living, depressive symptoms, and dementia ratings; most of these effects were supported by substantial and strong Bayesian evidence. No significant effects were found for global cognition at 1 to 10 months follow-up assessment for language, quality of life, anxiety, and behavior symptoms. However, evidence for the absence of these effects was ambiguous. A review of study bias highlighted that most studies lacked active, double-blinded controls, potentially leading to an overestimation of the effect, and making it difficult to conclusively attribute the observed improvements to the CS intervention. Hence, although effects are promising, the methodological issues highlight there is still a need for better controlled studies that provide more compelling evidence.