Screening for diabetes and impaired glucose metabolism in Qatar: models’ development and validation

Abstract

AbstractAimTo establish two scoring models for identifying individuals at risk of developing Impaired Glucose Metabolism (IGM) and Type two Diabetes Mellitus (T2DM) in Qatari.Materials and MethodsA sample of 2000 individuals, from Qatar BioBank, was evaluated to determine features predictive of T2DM and IGM. Another sample of 1000 participants was obtained for external validation of the models. Several scoring models screening for T2DM were evaluated and compared to the model proposed by this study.ResultsAge, gender, waist-to-hip-ratio, history of hypertension and hyperlipidemia, and levels of educational were statistically associated with the risk of T2DM and constituted the Qatari diabetes risk score (QDRISK). Along with, the 6 aforementioned variables, the IGM model showed that BMI was statistically significant. The QDRISK performed well with an area under the curve (AUC) 0.870 and 0.815 in the development and external validation cohort, respectively. The QDRISK showed overall better accuracy and calibration compared to other evaluated scores. The IGM model showed good accuracy and calibration, AUCs 0.796 vs. 0.774 in the development and external validation cohorts, respectively.ConclusionsThis study developed a Qatari-specific risk scores to identify high risk individuals and can guide the development of a nationwide primary prevention program.