Accelerometer-Measured Physical Activity Percentiles Across Age in Adults With Knee Osteoarthritis.

Abstract

To describe age-related percentile distributions of accelerometer-measured physical activity in adults with knee osteoarthritis (OA) and to examine differences according to sex, pain, income, and body mass index (BMI). Cross-sectional secondary analysis. Secondary analysis of data from the accelerometer substudy of the Osteoarthritis Initiative (OAI), a multicenter longitudinal cohort study conducted in [removed for blinded review]. A total of 968 adults with knee OA-related data available for analysis. Not applicable. Weekly minutes of moderate-to-vigorous physical activity (MVPA) and total physical activity (TPA), measured using a uniaxial ActiGraph GT1M accelerometer and classified according to the Freedson cut-point approach. Quantile regression was used to estimate age-related percentile curves. Median weekly MVPA was 13.93 minutes, and median weekly TPA was 293.57 minutes. This median MVPA value is substantially below the WHO. Both MVPA and TPA declined with advancing age. Males accumulated significantly more MVPA than females across percentiles, whereas females accumulated significantly more TPA. Lower physical activity levels were also observed in participants with significant pain and higher BMI, particularly at lower percentiles. No statistically significant differences were detected according to annual income. These findings indicate that percentile position reflects relative activity within this knee OA sample and should not be interpreted as equivalent to meeting guideline-recommended physical activity levels. Adults with knee OA accumulated low levels of device-measured physical activity, with less favorable patterns observed at older ages and among participants with greater pain and higher BMI. These percentile-based distributions may support the interpretation of physical activity patterns in comparable rehabilitation and research settings. These percentile distributions may help contextualize an individual's activity level relative to comparable adults with knee OA in rehabilitation and research settings; however, they are descriptive reference distributions and do not replace guideline-based physical activity targets.