Relationship between grip strength and minimal clinically important differences in cognitive function in older adults with dementia in a long-term residential facility

Objective. Non-pharmacological approaches effectively improve cognitive function in older adults with dementia in institutionalised settings. We aimed to investigate the physical characteristics of older adults with dementia achieving a minimal clinically important difference (MCID) on the Mini-Mental State Examination (MMSE) following interventions for cognitive rehabilitation based on exercise. Methods. This retrospective analysis included 25 participants with dementia residing in a long-term care facility who underwent group exercise in a quasi-randomised controlled study. We calculated the MCID on the MMSE using a distribution-based method. The rounded values of the standard deviation (SD) of the MMSE at baseline of approximately 0.4 and 0.5 were considered an MCID. Based on intervention outcomes, the participants were divided into MCID achievers and non-achievers. We compared changes in physical function based on grip strength, maximum knee extension strength, maximum 10-m gait time, and 5-m wheelchair driving time. Results. MCID achievers had significantly higher grip strength at baseline than non-achievers for both 0.4 x SD and 0.5 x SD. A multiple logistic regression analysis including age, sex, and MMSE at baseline revealed that grip strength was significantly associated with MCID achievement at 0.4 x SD (odds ratio [OR], 1.614; 95% confidence interval [CI], 1.04-2.51) and 0.5 x SD (OR, 1.585; 95% CI, 1.04-2.42). Conclusions. The importance of measuring grip strength was demonstrated by considering the achievement of an MCID for cognitive function. Assessing objective changes using a distribution-based method may help evaluate rehabilitation outcomes. Higher grip strength at baseline was significantly associated with MCID improvement in the MMSE in institutionalised older adults with dementia.