Key result
Montoye actigraphy closely tracks self-reported MVPA while Freedson overestimates daily activity by ~71 minutes.
Why the study?
Most physical activity evidence in DPP studies relies on potentially biased self-report, and how actigraphy-derived moderate-to-vigorous physical activity compares with self-report is poorly understood.
How do actigraphy-derived MVPA estimates using different algorithms compare to self-reported MVPA in older adults?
Cross-Sectional (n=39)
How do actigraphy-derived MVPA estimates using different algorithms compare to self-reported MVPA in older adults?
Mean Difference: -11 (95% CI -30–8)
Absolute Event Rate: 48% vs 37%
p-value: p=0.49
In older adults, the Montoye algorithm for wrist-worn actigraphy closely approximates self-reported physical activity, whereas the widely used Freedson algorithm substantially overestimates it.
Introduction and Objective: Physical activity (PA) is a core component of the Diabetes Prevention Program (DPP), yet most evidence on PA in DPP studies relies on self-report, which may be subject to bias. Wrist-worn actigraphy may provide more objective PA assessment, but how actigraphy-derived moderate-to-vigorous physical activity (MVPA) compares with self-report is poorly understood. Methods: We analyzed baseline PA data from adults aged ≥65 years enrolled in BRIDGE. Participants wore a wrist ActiGraph GT9x for 7 days and concurrently completed the CDC My Physical Activity diary. Daily minutes of activities labeled ‘moderate’ or ‘vigorous’ were averaged to obtain self-reported MVPA. Daily MVPA from actigraphy was estimated using two algorithms: Freedson vertical-axis (1998) and Montoye (2020). The three MVPA estimates were compared using one-way repeated-measures ANOVA with Bonferroni-adjusted post hoc tests. Participants (%) meeting ≥30 minutes/day MVPA by each method was also assessed. Results: Among 39 participants (age 73±5 years; 77% female; 74% White), mean daily MVPA was 37±31 minutes by self-report, 108±60 minutes using Freedson, and 48±38 minutes using the Montoye algorithm (overall P<0.001). Freedson MVPA significantly exceeded self-reported MVPA by 71 minutes/day (95% CI: 43-98; P<0.001) and Montoye MVPA by 60 minutes/day (95% CI: 49-70; P<0.001). There was no difference between self-reported and Montoye MVPA (−11 minutes/day; 95% CI: −30 to 8; P=0.49). Those getting ≥30 minutes/day of MVPA was 54% by self-report, 95% by Freedson, and 67% by Montoye. Conclusion: In older adults participating in the DPP, the Montoye algorithm closely approximated self-reported MVPA, whereas the widely used Freedson algorithm substantially overestimated MVPA. These findings suggest that algorithm choice may meaningfully influence conclusions about PA levels in older adults and that the Montoye algorithm may be more appropriate for wrist-based PA assessment in this population. Disclosure S. Barua: None. E.A. Johnston: None. D. Upadhyay: None. S. Pradhan: None. A. Kulkarni: None. J. Chodosh: None. J.M. Beasley: None. Funding National Institutes of Health (R01DK12796)
No takes yet. Share an insight, caveat, or question.
BARUA et al. (2026) conducted a cross-sectional in Diabetes Prevention Program (DPP) (n=39). Actigraphy-derived MVPA (Montoye algorithm) vs. Self-reported MVPA was evaluated on Daily moderate-to-vigorous physical activity (MVPA) minutes (MD -11, 95% CI -30 to 8, p=0.49). The Montoye actigraphy algorithm closely approximated self-reported MVPA (difference -11 mins/day; P=0.49), whereas the Freedson algorithm overestimated it by 71 mins/day (P<0.001).
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: