Remote administration of physical performance tests among persons with and without a cancer history: Establishing reliability and agreement with in-person assessment

Objectives: To assess the reliability of using videoconference technology to remotely administer the Short Physical Performance Battery (SPPB), including the 5-time sit-to-stand (5XSTS) and usual 4-m walk (4mWT), and the Timed Up and Go (TUG) tests and agreement with in-person administration among adults with and without cancer. Methods: Participants from two ongoing clinical exercise trials in cancer survivors, one that included partners without cancer, comprised the available sample (n = 176; mean age 62.5 ± 11.5 years.). Remote tests were administered on two separate days by either the same or a different assessor to determine intra-rater and inter-rater reliability, respectively. We also compared tests conducted remotely and in-person using the same assessor and the same participant. Intraclass correlation coefficients (ICC) and 95% confidence intervals (95% CI) were used for all comparisons, except for the SPPB score, which used Cohen's kappa and Krippendorf's alpha for intra- and inter-rater reliability, respectively. Results: Remote assessment of the TUG test had excellent intra-rater reliability (0.98, 95% CI 0.93–0.99), inter-rater reliability (ICC = 0.96, 95% CI 0.90–0.99), and good agreement with in-person tests (ICC = 0.88, 95% CI 0.74–0.94). The 5XSTS and 4mWT showed excellent (ICC = 0.92, 95% CI 0.84–0.96) and good (ICC = 0.87, 95% CI 0.71–0.94) intra-rater reliability, respectively, but somewhat lower inter-rater reliability (5XSTS: ICC = 0.65, 95% CI 0.34–0.83 and 4mWT: ICC = 0.62, 95% CI 0.30–0.81). Remote 5XSTS had moderate agreement (ICC = 0.72, 95% CI 0.62–0.80) and 4mWT had poor agreement (ICC = 0.48, 95% CI -0.07-0.76) with in-person tests. Conclusions: Remote assessment of common physical function tests in older adults, including those who have cancer, is feasible and highly reliable when using the same assessor. TUG may be the most methodologically robust measure for remote assessment because it is also highly reliable when using different assessors and correlates strongly with in-person testing. Adapting administration of objective measures of physical function for the remote environment could significantly expand the reach of research and clinical practice to assess populations at risk of functional decline.