RESUMO
BACKGROUND: Handgrip strength (HGS) testing is a highly recommended method for screening for sarcopenia in older adults. However, there is no consensus on the optimal protocol and number of trials for screening sarcopenia in older adults with cognitive impairment. OBJECTIVE: To investigate the use of the first trial (FT), the mean of three trials (MT), and the highest value (HT) from three trials of the HGS test to screen for sarcopenia in older adults with cognitive impairment. Additionally, to analyze the consistency, agreement, and measurement error in the diagnosis of muscle weakness. METHODS: 176 older adults with cognitive impairment were evaluated. The HGS test was repeated three times. Analyses were performed using the Friedman repeated measures test with Wilcoxon post-hoc, intraclass correlation coefficient (ICC), Standard Error of Measurement (SEM), Minimal Detectable Change (MDC95), and Kappa index tests. RESULTS: There was no significant difference between the first trial (FT) and the mean of three trials (MT) (d = 0.17 [95 % CI: -0.08, 0.42]), but both differed significantly from the highest value (HT) (p < 0.001). The ICC indicated a reliability of 0.97 (95 % CI: 0.95, 0.98) across all participants, while the kappa index demonstrated over 80 % agreement. The SEM for the first measure of HGS ranged from 0.59 to 2.12 kgf. The MDC95 ranged from 1.64 to 5.87 kgf. CONCLUSION: For HGS testing, there was excellent consistency between the FM and MT. All three testing methods demonstrated excellent agreement in diagnosing muscle weakness. The measurement errors confirm that FT can be reliably used to monitor changes during rehabilitation.