Ultra-processed food consumption is associated with knee osteoarthritis: Data from the Osteoarthritis Initiative

Akkaya Z, Sims WM, Lynch JA, Löffler MT, Gassert F, Nevitt M, McCulloch CE, Lane NE, Pedoia V, Ziegeler K, Link TM, Joseph GB · Osteoarthritis and Cartilage

The rare dietary analysis with a quantified cartilage endpoint. In women, greater ultra-processed intake was associated with worse WOMAC pain (β = 0.17, 95% CI 0.093-0.242), worse activities of daily living (β = 0.59, 0.365-0.832), worse total score (β = 0.81, 0.483-1.13), **thinner cartilage on 3T MRI** (β = −0.013, −0.02 to −0.006) and slower gait (β = −0.035, −0.042 to −0.027), all p < 0.001. Sex interactions were significant for those outcomes (p 0.006 to < 0.001). Chair-stand performance was worse with more UPF in both sexes; the stiffness association did not survive Bonferroni correction. Adjusted for age, race, BMI, daily calories, physical activity and insurance.

Population

4,403 Osteoarthritis Initiative participants with adequate dietary and sociodemographic data

Intervention

Higher proportion of NOVA-4 (ultra-processed) servings in the daily diet

Comparator

Lower proportion

Limitations

Cross-sectional, so cartilage thickness and diet are measured at the same moment and neither can be shown to precede the other. Food-frequency questionnaires classify processing crudely. Adjusting for BMI may over-adjust if body weight is on the causal path between ultra-processed diet and joint loading. Two authors declare industry consultancies unrelated to food.