Comparing remote and face-to-face assessments of physical performance in older adults: A reliability study

Abstract

IntroductionOlder people often experience a decline in their physical performance. Tests have been approved to evaluate this performance in person. Yet, the constraints associated with in-person assessments (e.g. lack of medical facilities, pandemic lockdown, and contagion risk) are making us contemplate setting up assessments remotely. ObjectivesTo determine whether remote physical performance measurements of older adults are reliable and valid compared to face-to-face measurements. MethodsForty-five subjects aged 65 and over completed the normal/fast speed test (NWT/FWT), the unipodal balance test (UBT), the normal/fast timed up and go test (NTUG/FTUG), the 5 and 10 rep sit to stand test (5STS and 10STS), the 30 sec chair stand (30CS), the 2 minute step test (2MST) and the flexibility before standing (SAD) once face-to-face and twice remotely, by two different observers. The intraclass correlation coefficients (ICC), the standard errors of measurement (SEM%) and minimum detectable changes (MDC%) were calculated for both intra- and inter-observer conditions, to assess the relative and the absolute reliability. An ICC value exceeding 0.90 indicates a very high reliability, while an ICC between 0.70 and 0.89 signifies a high reliability. In clinical practice, a SEM % of less than 10% is considered acceptable. A smaller MDC % indicates a measurement that is more sensitive to detecting changes. ResultsIntra-observer relative reliability was very high (ICC>0.9) for the UBT, NWT, NTUG, FTUG, 5STS, 10STS, 30CS and the SAD; and high (ICC>0.7) for the 2MST and FWS. SEM% values ranged from 0% to 24.03% and MDC from 0% to 9.93%. Inter-observer relative reliability was considered very high (ICC>0.9) for all tests. SEM% values ranged from 0% to 17.68% and MDC from 0% to 7.32%. ConclusionOur findings demonstrate that remote assessments exhibited consistently high to very high levels of intra- and inter-observer relative reliability when compared to face-to-face assessments. Additionally, certain remote evaluations showed acceptable absolute reliability, making them viable alternatives for healthcare professionals when in-person assessments are not feasible in clinical practice.