Effects of engagement, persistence and adherence on cognitive training outcomes in older adults with and without cognitive impairment: a systematic review and meta-analysis of randomised controlled trials.

Abstract

Limited understanding exists regarding the influences of engagement, persistence and adherence on the efficacy of cognitive training for age-related cognitive decline and neurodegenerative cognitive impairment. This study conducted a meta-analysis of randomised controlled trials (RCTs). We systematically searched MEDLINE, PubMed, Web of Science, Embase and CINAHL databases from 1 January 2012 to 13 June 2023, and included RCTs assessing the effects of cognitive training in older adults, both with and without cognitive impairment. Hedges' g with a 95% confidence interval (CI) was used to synthesise cognitive training effect sizes on various neuropsychological tests. Subgroup analyses were conducted based on variables including engagement, persistence, adherence and cognitive conditions of normal cognition, mild cognitive impairment (MCI) or neurodegenerative dementia. This meta-analysis included 55 RCTs with 4,455 participants with cognitive conditions spanning normal cognition, MCI and neurodegenerative dementia. The mean age of participants was 73.9 (range: 65.7-84.5) years. Overall, cognitive training showed a significant cross-domain effect (Hedges' g = 0.286, 95% CI: 0.224-0.348). Training effects are significant when engagement or persistence rates exceed 60% or when adherence rates exceed 80%. Higher levels of persistence are required to achieve significant training effects in memory, visuospatial ability and reasoning than in executive function and attention and language. Higher persistence is also required for older adults with normal cognition to achieve significant training gains compared to those with cognitive impairment. This systematic review highlights the critical roles of engagement, persistence and adherence in augmenting the efficacy of cognitive training.