Association of ophthalmic disease and physical activity in All of Us Research Program

The association between physical activity (PA) and ophthalmic disease remains uncertain, as most studies relied on self-report. This study aimed to evaluate the associations between long-term, objectively measured PA by wearable devices (Fitbit) and the incidence and prevalence of common ophthalmic diseases in the All of Us Research Program.

We conducted cross-sectional and longitudinal analyses of 14,757 adults with electronic health record (EHR) and Fitbit data. Participants had at least 6 months of valid Fitbit wear (≥10 h/day and ≥100 steps/day). The cohort had a mean age of 55.0 ± 15.7 years (mean ± SD); 70.4% were female, and 79.1% were White. Mean daily step counts and minutes of moderate-to-vigorous physical activity (MVPA) were evaluated in relation to age-related macular degeneration (AMD), diabetic retinopathy (DR), primary open-angle glaucoma (POAG), and senile cataract identified using International Classification of Diseases codes. Logistic regression estimated odds ratios (ORs) for prevalent disease, and Cox proportional hazards models estimated hazard ratios (HRs) for incident disease. Models were adjusted for age, sex, race, and body mass index. DR analyses were also restricted to participants with diabetes mellitus.

In cross-sectional analyses, higher PA was significantly associated with lower odds of DR among participants with DM (steps per 1000: OR = 0.91, 95% confident interval (95%CI): 0.86-0.95; p < 0.001; MVPA per 10 min: OR = 0.87, 95%CI: 0.82-0.93; p < 0.001). No significant cross-sectional associations were observed between PA and AMD, POAG, or senile cataract. In longitudinal analyses, higher PA predicted a lower risk of incident DR among participants with DM (steps: HR = 0.93, 95%CI: 0.87-0.99; p = 0.022; MVPA: HR = 0.86, 95%CI: 0.79-0.93; p < 0.001). No significant longitudinal associations were found between PA and AMD, POAG, or senile cataract.

Our findings suggest a strong inverse association between sustained PA and incident DR and underscore the value of long-term wearable data for ophthalmic research.