Association Between Walking for Exercise and Symptomatic and Structural Progression in Individuals With Knee Osteoarthritis: Data From the Osteoarthritis Initiative Cohort

Lo GH, et al. · Arthritis & Rheumatology

Over 4 years, walkers had lower odds of new frequent knee pain (adjusted OR 0.6, 95% CI 0.4-0.8) and of medial joint-space-narrowing progression (OR 0.8, 95% CI 0.6-1.0). The study's two other outcomes, Kellgren-Lawrence grade worsening and improvement in existing frequent pain, did not separate. The pain result held under both missing-data assumptions; the joint-space result moved toward 1 and lost significance when everyone with missing exposure data was counted as a walker. Stratified by alignment, walkers with varus knees - 48% of the cohort - fared better on all three progression measures, while walkers with neutral knees had more medial joint-space worsening (17% against 11%) and walkers with valgus knees more Kellgren-Lawrence worsening (20% against 15%). Walking looked protective, not destructive, in established knee OA.

Population

1,212 adults 50+ with knee OA in the Osteoarthritis Initiative (73% walked for exercise)

Intervention

Exposure: self-reported walking for exercise

Comparator

Non-walkers

Limitations

Observational, and the walking exposure was collected retrospectively at the 96-month visit for outcomes measured across the first 48 months, so recall bias runs in a direction the design cannot correct; the authors say as much. 1,212 of 2,356 eligible participants had exposure data. Healthy-user bias is possible, the structural result is borderline and does not survive one of the two imputation models, and the alignment strata are exploratory - the authors ask for replication in cohorts with more neutral and valgus knees.