External validation of two diabetes risk scores in a young UK South Asian population

AimsTo externally validate the Leicester Practice Risk Score (LPRS) and the Leicester Risk Assessment score (LRAS) in a young South Asian population.MethodsSouth Asian participants aged 25–39 years inclusive from a population based screening study were included. The risk scores were calculated and compared to the diagnosis of type 2 diabetes mellitus (T2DM) or T2DM and Impaired Glucose Regulation (IGR, including IFG and IGT) using either an oral glucose tolerance test (OGTT) or a HbA1c (≤48 mmol/mol/6.5% and ≤42 mmol/mol/6.0% respectively). Measures of discrimination and calibration were calculated.ResultsOf the 331 participants 8 (2.4%) had undiagnosed T2DM and 30 (9.1%) had IGR using an OGTT, 11 (3.4%) and 39 (12.1%) were found using HbA1c. Using the LPRS to detect T2DM on an OGTT gives an area under the ROC curve of 0.91 (95% CI 0.86, 0.97), including those with IGR gives an ROC of 0.72 (0.62–0.81), these values are 0.93 (0.88, 0.98) and 0.68 (0.60, 0.77) when using an HbA1c to define outcome. Acceptable levels of calibration were seen. Similar results are found for the LRAS.ConclusionsThese scores can be used to identify those with undiagnosed T2DM and/or IGR in a young South Asian population. This is the first study to externally validate scores developed for prevalent undiagnosed disease in this age group using both OGTT and HbA1c.