Cross-sectional data from the 2007 to 2008 Canadian Community Health Survey were used to investigate the joint effect of obesity and female on the prevalence of arthritis. A total of 94,758 subjects (44,282 men and 50,476 women) aged 18 years or over who answered the question concerning arthritis were included, and sampling weights and design effects were taken into account in the analysis. The prevalence of arthritis among obese (body mass index ≥ 30 kg/m(2)) men was 18.0 % and 29.8 % among obese women. The adjusted prevalence ratio for obesity associated with arthritis was 1.45 (95 % confidence interval 1.39, 1.52) for women and 1.38 (confidence interval 1.29, 1.47) for men. The combined effect of obesity and female sex was more than additivity, with the relative excess risk due to the interaction being 0.21 (95 % confidence interval 0.09, 0.32). Obesity is likely to have a larger effect on women with arthritis than men.