Background: We aimed to define refined body shapes by using multiple anthropometric traits that represent fat distribution, and evaluate their associations with risk of insulin resistance (IR) and cardiometabolic disorders in a Chinese population. Methods: We performed a cross-sectional analysis in 6570 community-based participants aged ≥ 40 years. Four body circumferences (neck, waist, hip and thigh) and their ratios were put simultaneously into an open-source Waikato Environment for Knowledge Analysis platform to select the worthiest indicators in determining IR. The ratio of the top 3 fat distribution indicators were used to define the refined body shapes. Results: We defined 8 distinct body shapes based on sex-specific combinations of waist-to-hip ratio (WHR), waist-to-thigh ratio (WTR) and waist-to-neck ratio (WNR), which differed in participants’ distribution and risk of IR and related cardiometabolic disorders. In women, as compared to the low WHR-low WTR-low WNR shape, all body shapes were significantly associated with IR and related cardiometabolic disorders; while in men, the low WHR-high WTR-high WNR shape and the higher WHR related shapes were significantly associated with IR and related cardiometabolic disorders. Stratified by WHR, the results were consistent in women; however, no significant associations were detected in men. Conclusions: We defined 8 distinct body shapes by taking WHR, WTR, and WNR, simultaneously into account, which differed in association with risk of IR and related cardioembolic disorders in women. This study suggest that body shapes defined by multiple anthropometrics traits could provide a useful, convenient and easily available method for identification cardiometabolic risk.