A Prediction Model of the Incidence of Type 2 Diabetes in Individuals with Abdominal Obesity: Insights from the General Population.

Abstract

Background:

This study aimed to distinguish the risk factors for type 2 diabetes mellitus (T2DM) and construct a predictive model of T2DM in Japanese adults with abdominal obesity.

Methods:

This study was a post hoc analysis. A total of 2012 individuals with abdominal obesity were included and randomly divided into training and validation groups at 70% (n = 1518) and 30% (n = 494), respectively. The LASSO method was used to screen for risk variables for T2DM, and to construct a nomogram incorporating the selected risk factors in the training group. We used the C-index, calibration plot, decision curve analysis, and cumulative hazard analysis to test the discrimination, calibration and clinical significance of the nomogram.

Results:

In the training cohort, the C-index and receiver operating characteristic were 0.819 and the 95% CI was 0.776-0.858, with a specificity and sensitivity of 77% and 74.68%, respectively. In the validation cohort, the C-index was 0.853; sensitivity and specificity were 77.6% and 88.1%, respectively. The decision curve analysis showed that the model's prediction was effective and cumulative hazard analysis demonstrated that the high-risk score group was more likely to develop T2DM than the low-risk score group.

Conclusion:

This nomogram may help clinicians screen abdominal obesity at a high risk for T2DM.

• Authors: Caixia Tan, Bo Li, Lingzhi Xiao, Yun Zhang, Yingjie Su, Ning Ding
• Journal: Diabetes, metabolic syndrome and obesity : targets and therapy (Diabetes Metab Syndr Obes) Vol. 15 Pg. 3555-3564 ( 2022) ISSN: 1178-7007 [Print] New Zealand
• PMID: 36411787 (Publication Type: Journal Article)
• Copyright: © 2022 Tan et al.