INTRODUCTION: The grip strength test is often used during geriatric assessment (GA) to assess muscle strength in older adults. However, it is unclear which grip strength cutoffs are most relevant to older adults in the context of GA. Physical performance during GA is often assessed via the Short Physical Performance Battery (SPPB). Whether the SPPB is superior to two of its individual components (4-m gait speed and the 5-chair stand test) for identifying GA abnormalities is unknown. The objectives of this study were (i) to identify which grip strength thresholds are associated with impairments in GA domains and with an abnormal GA overall and (ii) to examine whether total SPPB score is a stronger indicator of an abnormal GA and each of its domains than 4-m gait speed and the 5-chair stand test. MATERIALS AND METHODS: This was a retrospective cohort study of older adults with cancer aged ≥65 years who had undergone a GA prior to treatment. Grip strength and the SPPB were completed during GA. We examined three different grip strength cutoffs: (i) European Working Group on Sarcopenia in Older People 2 (EWGSOP2)
(ii) the Foundation for the National Institutes of Health (FNIH)
and (iii) the Sarcopenia Definitions and Outcomes Consortium (SDOC). Low SPPB was defined as a score of ≤9 out of 12 points. A score of ≤3 out of 4 points was used to identify abnormalities in the 4-m gait speed and 5-chair stand test. Multivariable logistic regression was used to address the study objectives. RESULTS: A total of 475 participants (mean age: 80.7 years, 42.9 % female) were included. The FNIH grip strength criteria had a higher discriminative ability of an abnormal GA (area under the curve [AUC] = 0.646) than the EWGSOP2 and the SDOC criteria. Compared to the SPPB and the 5-chair stand test, the 4-m gait speed was the strongest indicator of an abnormal GA (AUC = 0.737). The addition of low grip strength improved the performance of the SPPB (AUC Δ = +0.05) and gait speed (AUC Δ = +0.04) for identifying an abnormal GA. DISCUSSION: Low grip strength per the FNIH and slow gait speed are of clinical relevance during GA.