Aims Patients with type 2 diabetes have a high risk of cardiovascular disease (CVD). Central obesity has been particularly associated with this risk relationship. We aimed to evaluate waist to height ratio (WHtR) as a predictor of risk in such patients. Methods WHtR was evaluated as a predictor of the risk of CVD and mortality amongst 11125 participants with type 2 diabetes in the ADVANCE and ADVANCE-ON studies, and was compared with body mass index (BMI), waist circumference and waist hip ratio (WHR). Primary outcome was a composite of death from CVD, non-fatal myocardial infarction or non-fatal stroke. Secondary outcomes were myocardial infarction, stroke, cardiovascular death and death from any cause. Cox models were used, with bootstrapping to compare associations between anthropometric measures for the primary outcome. Results Median follow-up time was 9.0 years. There was a positive association between WHtR and adverse outcomes. The hazard ratio (HR) (confidence interval), per SD higher WHtR, was 1.16 (1.11-1.22) for the primary endpoint, with no heterogeneity by sex or region, but a stronger effect in individuals aged 66 years or older. The other 3 anthropometric measurements showed similar associations, although there was evidence that WHtR marginally outperformed BMI and WHR. Based on commonly used BMI cut-points, the equivalent WHtR cut-points were estimated to be 0.55 and 0.6, with no evidence of a difference across subgroups. Conclusions In patients with diabetes, WHtR is a useful indicator of future adverse risk, with similar effects in different population subgroups.