Author(s): BERRY, E., SCULTHORPE, N.F., WARNER, A., SANAL-HAYES, N.E.M., HAYES, L.D., Institution: THE UNIVERSITY OF THE WEST OF SCOTLAND, Country: UNITED KINGDOM, Abstract-ID: 553

INTRODUCTION:
Muscle-strengthening exercise is essential for maintaining health and quality of life among older adults. Yet, few people aged 65+ years meet the guidelines (NHS, 2023). Mobile health (mHealth) shows promise for improving engagement in muscle-strengthening programmes for older adults (Berry et al., 2025), but challenges remain. Remotely assessing strength is difficult, and the validity of in‑app measures of strength and power needs to be established. As part of our ongoing co-design approach, we have developed the Strength Buddy app which targets muscle health in older adults. This study evaluated the validity of the app’s remotely assessed five-times sit-to-stand (5TSTS) test against in-person testing. A secondary aim was to correlate remote and in-person 5TSTS results with gold-standard isokinetic dynamometry.
METHODS:
Twenty-nine participants (67 ± 6 years; 165 ± 89 cm; 82 ± 18 kg) attended the university laboratory once to complete the 5TSTS and dominant-limb knee extensor isokinetic dynamometry (HUMAC NORM). Testing was performed seated with the hip at 90° flexion and an angular velocity of 60°/s. Body-mass–normalised torque (Nm/kg) was recorded, with the best of five repetitions used for analysis. Participants then completed a remote 5TSTS using their smartphone camera, which was scored remotely via video analysis software (DaVinci Resolve). Agreement between in-person and remote measures was assessed using Bland–Altman analysis, with reliability and Pearson correlation also examined.
RESULTS:
Results indicated good agreement between in-person and remote 5TSTS. Reported bias was 0.3s, the upper and lower limits of agreement were 2.84 s and 2.24 s respectively. Reliability analysis revealed a coefficient of variation of 7.35% (good consistency), standard error of measurement of 0.92 s, (good absolute precision), and intraclass correlation coefficient of 0.89 (excellent agreement). Pearson’s correlation between normalised peak torque and in-person 5TSTS (r=-0.519, p=0.004), and remote 5TSTS (r=-0.518, p=0.004) was moderately negative.
CONCLUSION:
Findings indicate the remote 5TSTS test shows good agreement, reliability, and precision compared to in-person assessment. The moderate associations with knee extensor strength further support the construct validity of the remote measure as a valid alternative for assessing lower-limb function which can be used in mHealth muscle assessment for older adults.


REFERENCES

Berry, E.C.J., Sculthorpe, N.F., Warner, A., Mather, J.D., Sanal-Hayes, N.E.M. and Hayes, L.D. (2025) ‘A scoping review of the feasibility, usability, and efficacy of digital interventions in older adults concerning physical activity and/or exercise’, Frontiers in Aging, 6, p. 1516481. Available at: https://doi.org/10.3389/fragi.2025.1516481.

NHS (2023) ‘Health Survey for England, 2021 part 2’. Available at: https://digital.nhs.uk/data-and-information/publications/statistical/health-survey-for-england/2021-part-2.