Handgrip strength but not SARC-F score predicts cognitive impairment in older adults with multimorbidity in primary care: a cohort study

BACKGROUND: Assessing motor function is a simple way to track cognitive impairment. We analysed the associations between cognitive and motor function and assessed the predictive value of two motor function measuring tools for cognitive impairment in older adults with multimorbidity in primary care settings. METHODS: We conducted a prospective cohort study with a 1 year follow-up. Patients aged ≥60 years with ≥2 morbidities were recruited from four primary care clinics. Motor function was assessed using handgrip strength and a sarcopenia screening scale (SARC-F). Cognitive function was measured using the Hong Kong Montreal Cognitive Assessment (HK-MoCA). We defined cognitive impairment as an HK-MoCA score < 22. The associations between cognitive and motor functions were examined from a bidirectional perspective. RESULTS: We included 477 participants (mean age 69.4, 68.6% female) with a mean (SD) HK-MoCA score of 25.5 (3.38), SARC-F score of 1.1 (1.36), and handgrip strength of 21.2 (6.99) kg at baseline. Multivariable linear regression models showed bidirectional cross-sectional associations of the HK-MoCA score and cognitive impairment with SARC-F score and handgrip strength at baseline and 1 year. Cox regression revealed a longitudinal association between baseline handgrip strength and cognitive impairment at 1 year (hazard ratio: 0.48, 95% CI 0.33–0.69) but no longitudinal association between SARC-F and cognitive impairment. Variation in the SARC-F score increased with decreasing HK-MoCA score (Brown–Forsythe test F statistic = 17.9, p < 0.001), while variability in the handgrip strength remained small (modified signed-likelihood ratio test, p < 0.001). CONCLUSIONS: Primary healthcare providers may use handgrip strength to track cognitive function decline in older adults with multimorbidity. However, the SARC-F scale may not have the same predictive value. Further research is needed to evaluate the performance and variability of the SARC-F score in individuals with poor cognitive function. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1186/s12877-022-03034-2.