Supplementary Table 6. Threshold-gradient direct comparison of weekend warriors with regularly active participants.

Top-two-day threshold	Covariate-adjusted HR (95% CI); P	Additionally adjusted for weekly MVPA, HR (95% CI); P
≥ 50%	1.004 (0.945, 1.067); 0.901	1.031 (0.968, 1.099); 0.344
≥ 55%	1.006 (0.950, 1.065); 0.844	1.028 (0.969, 1.091); 0.362
≥ 60%	0.968 (0.912, 1.028); 0.291	0.988 (0.929, 1.051); 0.699
≥ 65%	0.971 (0.908, 1.038); 0.382	0.989 (0.923, 1.059); 0.752
≥ 70%	0.998 (0.925, 1.077); 0.965	1.019 (0.943, 1.101); 0.633
≥ 75%	1.016 (0.930, 1.111); 0.719	1.038 (0.949, 1.136); 0.413
≥ 80%	0.975 (0.873, 1.088); 0.649	0.996 (0.891, 1.114); 0.948
≥ 85%	0.920 (0.792, 1.068); 0.271	0.939 (0.808, 1.091); 0.409
≥ 90%	0.827 (0.680, 1.006); 0.057	0.846 (0.695, 1.030); 0.095
≥ 95%	0.978 (0.744, 1.286); 0.873	1.001 (0.761, 1.316); 0.997

HRs were estimated using Cox proportional hazards models restricted to active participants accumulating >=150 min/week of MVPA. At each top-two-day threshold, the exposure contrasted WW participants (at least the stated percentage of total weekly MVPA accumulated on the two most active days) with regularly active participants (below the stated percentage); regularly active participants were the reference group. The outcome was incident overall OA, defined as the first recorded ICD-10 M15-M19 diagnosis during follow-up among participants without OA at baseline. The covariate-adjusted model included age at accelerometer assessment, sex, ethnicity, education, Townsend deprivation index, smoking status, alcohol intake frequency, and BMI; the second model additionally included continuous total weekly MVPA (min/week). Abbreviations: BMI, body mass index; CI, confidence interval; HR, hazard ratio; ICD-10, International Classification of Diseases, 10th Revision; MVPA, moderate-to-vigorous physical activity; OA, osteoarthritis; WW, weekend warrior.