Key result
Every additional hour of weekly wearable-derived moderate-to-vigorous physical activity was associated with 6% lower odds of incident type 2 diabetes (OR 0.94; 95% CI 0.89-1.00; p=0.046).
Why the study?
Diabetes risk scores rely on self-reported physical activity, which is prone to bias, so researchers evaluated whether wearable-derived physical activity predicts incident type 2 diabetes.
Does wearable-derived moderate-to-vigorous physical activity reduce the incidence of type 2 diabetes in adults without a history of the disease?
Cohort (n=13,482)
Yes
Does wearable-derived moderate-to-vigorous physical activity reduce the incidence of type 2 diabetes in adults without a history of the disease?
Odds Ratio: 0.94 (95% CI 0.89–1)
p-value: p=0.046
Objectively measured moderate-to-vigorous physical activity using wearables is independently associated with a lower risk of developing type 2 diabetes, supporting its potential to refine diabetes risk scores.
Introduction and Objective: Many diabetes risk scores including the ADA’s risk assessment tool incorporate self-reported physical activity, which is prone to bias and coarse categorization. We evaluated whether objectively measured, wearable-derived moderate-to-vigorous physical activity (MVPA) predicts incidence of type 2 diabetes (T2D) in the All of Us (AoU) cohort. Methods: We analyzed AoU participants with linked EHR and Fitbit data. Weekly MVPA was derived from the first year of Fitbit wear among participants with ≥180 valid wear days (>100 steps/day). Incidence of T2D was identified using validated ICD-10 codes (E11) over a 10-year follow-up period. Weekly MVPA was computed using Fitbit-derived daily values. Multivariable logistic regression estimated odds ratios (ORs) for T2D adjusting for sex, race (White vs non-White), ethnicity, hypertension history, age (per 10 years), and BMI (per 5 kg/m²). Model’s T2D prediction performance was assessed using area under the ROC curve (AUROC). Results: 13,482 participants (51±16 years; 72% female; 79% White) without a history of T2D were analyzed, of whom 189 developed T2D over the 10-year follow-up. The cohort had 327±50 days of wearable data and 249±207 minutes of weekly MVPA. In adjusted models, every additional hour of weekly MVPA was associated with 6% lower odds of T2D (OR:0.94, 95% CI 0.89-1.00, p=.046). Traditional risk factors remained significant, including male sex (OR 1.94, p<.001), non-White race (OR 1.84, p=.005), hypertension history (OR 1.54, p=.02), older age (OR 1.22 per 10 years, p=.002), and higher BMI (OR 1.52 per 5 kg/m², p<.001). The model achieved an AUROC of 0.75 (95% CI 0.69-0.80) for 10-year T2D risk prediction. Conclusion: In a large national cohort, objective wearable-derived weekly MVPA was independently associated with lower odds of T2D after adjustment for well-known risk factors. These findings support the potential value of wearable-derived activity to complement or refine self-reported measures in diabetes risk scores. Disclosure D. Upadhyay: None. A. Soumahoro: None. B. Dorcely: None. M. Bergman: Advisory Panel; Ended; Novo Nordisk. Advisory Panel; Current; Abbott. S. Barua: None.
No takes yet. Share an insight, caveat, or question.
Upadhyay et al. (2026) conducted a cohort in Type 2 diabetes (n=13,482). Wearable-derived moderate-to-vigorous physical activity (MVPA) was evaluated on Incidence of type 2 diabetes (OR 0.94, 95% CI 0.89-1.00, p=0.046). Every additional hour of weekly wearable-derived moderate-to-vigorous physical activity was associated with 6% lower odds of incident type 2 diabetes (OR 0.94; 95% CI 0.89-1.00; p=0.046).
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: