Association between grip strength and cognitive impairment in older American adults.

Background and aims: Exponential population aging has led to an increased prevalence of cognitive impairment worldwide. Hand grip strength, which may be associated with physical activity, could be a useful predictor of cognitive impairment. However, few studies have reported the association, if any, between hand grip strength and cognitive function. Methods: We used data obtained from the National Health and Nutrition Examination Survey between 2011-2012 and 2013-2014 to investigate the association between hand grip strength and cognitive impairment. Cognitive impairment was assessed using the Consortium to Establish a Registry for Alzheimer's Disease (CERAD), animal fluency (AF), and digit symbol substitution test (DSST) scores. Cutoff values of CERAD < 5, AF < 14, and DSST < 34 were used to define cognitive impairment. In this cross-sectional study, we used odds ratios to determine the potential usefulness of hand grip strength for the prediction of cognitive impairment. Results: This study included 2,623 participants aged ≥60 years. The DSST results showed that hand grip strength was associated with a low risk of cognitive impairment and that subgroup analysis showed that male sex, 60-69 years of age, and the Non-Hispanic (NH)-White, NH Black, and Asian were associated with a significantly low risk of cognitive impairment. The CERAD test results showed that 70-79 years of age and the NH White were significantly associated with a low risk of cognitive impairment. By following full adjustment, we did not observe statistically significant differences between hand grip strength and cognitive impairment based on the CERAD test. The AF test results showed that >80 years of age, female sex, and the NH White were associated with a significantly low risk of cognitive impairment. The most important finding is that a linear association lies between grip strength and cognitive impairment, as well as a sex-based linear association. Machine learning of the XGBoost model suggests that grip strength is one of the top two most important negative predictor variables. Conclusion: We observed an inverse relationship between hand grip strength and cognitive impairment, which might suggest a shared underlying mechanism that needs to be further investigated using a large-scale prospective clinical trial to validate our findings.