Comparing the diagnostic value of different obesity indices in prediabetes: a cross-sectional study in China

Abstract

Diabetes, particularly type 2 diabetes, has emerged as a significant public health burden globally, and prediabetes is a high-risk state for type 2 diabetes. This study aimed to compare the diagnostic performance of body mass index (BMI), waist circumference (WC), waist-to-hip ratio (WHR), waist-to-height ratio (WHtR), lipid accumulation product index (LAP), visceral adiposity index (VAI), Chinese visceral adiposity index (CVAI) and relative fat mass (RFM) in identifying prediabetes in a Chinese population. A cross-sectional study was conducted with 7167 individuals aged 18–65 years in Jiangsu Province, China by using multistage stratified sampling method. The eligible participants completed a structured questionnaire, and anthropometric and clinical biochemical indicators were measured according to standardized protocols. Multivariate logistic regression models were used to assess the associations between obesity indices and prediabetes. Receiver operating characteristic analysis was applied to compare the diagnostic abilities of eight obesity indices, and the Delong test was used to compare the areas under the curve (AUCs). The prevalence of prediabetes was 17.9% in Jiangsu Province. All obesity indices were significantly associated with prediabetes after adjustment for potential confounders in both men and women. The AUC ranges of the eight indices in men and women were 0.555–0.627 and 0.620–0.692, respectively. In men, WHR, WHtR, CVAI and RFM had larger AUCs than BMI, WC, LAP and VAI, but there was no significant difference in AUCs of WHR, WHtR, CVAI and RFM after the Delong test. The optimal cut-off values of WHR, WHtR, CVAI and RFM for diagnosing prediabetes were 0.88, 0.52, 80.32 and 25.37, respectively. CVAI in women had the largest AUC, significantly greater than the other indices. The optimal cut-off value of CVAI for diagnosing prediabetes in female was 86.48, where the sensitivity and specificity were 62.2% and 67.5%, respectively. In the male population, WHR, WHtR, CVAI, and RFM were superior to the other four indicators in detecting prediabetes. CVAI had the greatest ability to identify prediabetes among the female population.