Background: Waist-to-height ratio (WHtR) is an easy and inexpensive adiposity index that reflects central obesity. In this study, we examined the association of various baseline adiposity indices, including WHtR, with the development of diabetes over 4 years of follow-up in apparently healthy Korean individuals. Methods: A total of 2,900 nondiabetic participants (mean age, 44.3 years; 2,078 men) in a health screening program, who repeated the medical check-up in 2005 and 2009, were recruited. Subjects were divided into two groups according to development of diabetes after 4 years. The cut-off values of baseline body mass index (BMI), waist circumference (WC), and WHtR for the development of diabetes over 4 years were calculated. The sensitivity, specificity, and mean area under the receiver operator characteristic curve (AUROC) of each index were assessed. The odds ratio (OR) for diabetes development was analyzed for each of the three baseline adiposity indices. Results: During the follow-up period, 101 new cases (3.5%) of diabetes were diagnosed. The cut-off WHtR value for diabetes development was 0.51. Moreover, WHtR had the highest AUROC value for diabetes development among the three adiposity indices (0.716, 95% confidence interval [CI], 0.669 to 0.763; 0.702, 95% CI, 0.655 to 0.750 for WC; 0.700, 95% CI, 0.651 to 0.750 for BMI). After adjusting for confounding variables, the ORs of WHtR and WC for diabetes development were 1.95 (95% CI, 1.14 to 3.34) and 1.96 (95% CI, 1.10 to 3.49), respectively. No significant differences were observed between the two groups regarding BMI. Conclusion: Increased baseline WHtR and WC correlated with the development of diabetes after 4 years. WHtR might be a useful screening measurement to identify individuals at high risk for diabetes.