Comparison of the association and discriminatory ability of CVAI, LAP, CI, and AVI for type 2 diabetes in Chinese adults aged ≥ 50 years: a sex-specific analysis

Objective The aim of this study was to compare the associations of four visceral adiposity indices—Chinese visceral adiposity index (CVAI), lipid accumulation product (LAP), conicity index (CI), and abdominal volume index (AVI)—with prevalent type 2 diabetes mellitus (T2DM) in Chinese adults aged ≥ 50 years and to evaluate their discriminatory ability within the study sample. We further investigated whether sex modifies these associations and discriminatory ability. Methods In this community-based case-control study with frequency matching by sex and age conducted in Anhui Province, China (1, 088 participants; 491 T2DM cases, 597 controls), we assessed associations using unconditional multivariable logistic regression (fully adjusted model) and discriminatory ability using receiver operating characteristic (ROC) curve analysis, with prespecified sex-stratified analyses. Results After full adjustment, all four indices were positively associated with prevalent T2DM. CVAI and CI showed significant associations across Q2-Q4, while LAP and AVI were significant only in Q3-Q4. In terms of discriminatory ability, CVAI had the numerically highest AUC in the overall population (0.634), followed by LAP (0.633), AVI (0.612), and CI (0.591). Sex differences were also observed. CVAI, CI, and AVI showed stronger associations in males, while LAP showed a stronger association in females. Discriminatory ability followed the same pattern, with CVAI showing the numerically highest area under the curve (AUC) in males (0.658) and LAP in females (0.642). Conclusion This study suggests that CVAI, LAP, CI, and AVI are independently associated with prevalent T2DM in adults aged ≥50 years, with significant sex modification in both association and discriminatory ability within the study sample. Findings suggest sex-specific differences: CVAI had stronger associations and higher discriminatory ability in males, whereas LAP had stronger associations and higher discriminatory ability in females. As a case-control study from a single province, these findings do not establish causality or predictive utility. Prospective studies are needed to validate these findings.