Analysis of Anthropometric Cut Points for Screening Diabetes Mellitus among Residents Aged 18-69 in Shandong Province

CHEN Xi,GUO Xiao-lei,MA Ji-xiang,TANG Jun-li,LU Zi-long,ZHANG Ji-yu,ZHANG Gao-hui,XU Ai-qiang
DOI: https://doi.org/10.6040/j.issn.1671-7554.2012.04.005
2012-01-01
Abstract:Objective To explore the value and the optimal cut points of the anthropometric indexes to screen diabetes mellitus(DM),and to provide scientific evidence for DM screening and health education of the risk population in Shandong Province.Methods Using the multistage stratified cluster random sampling method,15,350 residents aged 16 to 69 in Shandong Province were screened for DM defined by WHO criteria of 1999,and physically examined on height,weight,WC and hip circumference.The curves of receiver operating characteristic(ROC) were used to calculate the area under the curve(AUC),the sensitivity,the specificity and Youden indexes.Optimal cut points were determined by the maximum of Youden indexes.Optimal cut points and adjacent points were computed by consistency analysis.Results 794 cases were found with DM,and the prevalence was 4.59%.The anthropometric indexes and DM prevalence rates increased with age,and this trend was particularly obvious in females.The AUCs for both males and females had significant meaning(P0.001).The AUC of WHR was the maximum,and had significant difference compared with BMI(P0.05).But there were no significant differences in AUCs between the males and females(P0.05).The cut points of BMI,WC and WHR for men were 25 kg/m2,91 cm and 0.92 respectively.The cut points of BMI,WC and WHR for women were 25 kg/m2,83 cm and 0.87 respectively,and they had the best consistency for screening DM.Conclusion BMI,WC and WHR are effective indicators for DM screening,and have good consistency.The optimal cut points can be used for screening DM and defining the risk population of DM in Shandong Province.
What problem does this paper attempt to address?