OBJECTIVE To evaluate if type 2 diabetes is an independent risk predictor for severe osteoarthritis (OA). RESEARCH DESIGN AND METHODS Population-based cohort study with an age- and sex-stratified random sample of 927 men and women aged 40–80 years and followed over 20 years (1990–2010). RESULTS Rates of arthroplasty (95% CI) were 17.7 (9.4–30.2) per 1,000 person-years in patients with type 2 diabetes and 5.3 (4.1–6.6) per 1,000 person-years in those without (P < 0.001). Type 2 diabetes emerged as an independent risk predictor for arthroplasty: hazard ratios (95% CI), 3.8 (2.1–6.8) (P < 0.001) in an unadjusted analysis and 2.1 (1.1–3.8) (P = 0.023) after adjustment for age, BMI, and other risk factors for OA. The probability of arthroplasty increased with disease duration of type 2 diabetes and applied to men and women, as well as subgroups according to age and BMI. Our findings were corroborated in cross-sectional evaluation by more severe clinical symptoms of OA and structural joint changes in subjects with type 2 diabetes compared with those without type 2 diabetes. CONCLUSIONS Type 2 diabetes predicts the development of severe OA independent of age and BMI. Our findings strengthen the concept of a strong metabolic component in the pathogenesis of OA.