Multi-component cognitive intervention for older adults with mixed cognitive levels: implementation and preliminary effectiveness in real-world settings

BACKGROUND: In most controlled studies of multi-component cognitive intervention, participants’ cognitive levels are homogenous, which is contrary to real-world settings. There is a lack of research studying the implementation of evidence-based cognitive intervention in communities. This study describes the implementation and preliminary effectiveness of a Multi-component Cognitive Intervention using Simulated Everyday Tasks (MCI-SET) for older adults with different cognitive levels in real-world settings. METHODS: Single group, pre-intervention assessment, post-intervention assessment, and 3-month follow-up research design. MCI-SET consists of 12 two-hour weekly sessions that include motor-cognitive tasks, cognitive training, and cognitive rehabilitation. One hundred and thirty participants, > = 65 and frail, dependence on > = one instrumental daily activity, or with confirmed dementia, from eight community centers were included. The primary outcome is general cognition (Montreal Cognitive Assessment-Taiwan, MoCA-T). Secondary outcomes are memory (Miami Prospective Memory Test, Digits Forward, Digits Backward), attention (Color Trail Test-Part 1), executive function (Color Trail Test-Part 2), and general function (Kihon Checklist-Taiwan). RESULTS: Pre-intervention workshop for group leaders, standardized activity protocols, on-site observation, and ten weekly conferences were conducted to ensure implementation fidelity. MCI-SET had an 85% retention rate and 96% attendance rate. The participants had a mean age of 78.26 ± 7.00 and a mean MoCA-T score of 12.55 ± 7.43. 73% were female. General cognition (Hedges’ g = 0.31), attention (Hedges’ g = 0.23), and general function (Hedges’ g = 0.31), showed significant post-intervention improvement with small effect size. Follow-ups showed maintained improvement in general cognition (Hedges’ g = 0.33), and delayed effect on attention (Hedges’ g = 0.20), short-term memory (Hedges’ g = 0.38), and executive function (Hedges’ g = 0.40). Regression analysis indicated that the intervention settings (day care centers vs neighborhood centers), the pre-intervention cognitive levels, and the pre-intervention general function of the participants were not associated with the outcomes. CONCLUSIONS: MCI-SET is feasible and can improve the cognitive skills and general functions of older adults with heterogeneous cognitive skills or disabilities. It is essential to tailor programs to fit the interests of the participants and the culture of local communities. Group leaders must also have the skills to adjust the cognitive demands of the tasks to meet the heterogeneous cognitive levels of participants. TRIAL REGISTRATION: This study was retrospectively registered at clinicaltrials.gov (Identifier: NCT04615169). SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1186/s12877-021-02489-z.