Comparative performance of FINDRISC and IDRS in screening for type 2 diabetes risk among Yemeni university students

BACKGROUND: Type 2 diabetes mellitus (T2DM) is a leading cause of death worldwide.To address the rise of T2DM, it might be more effective to create and validate a targeted risk scoring system for specific populations. This study aimed to evaluate and compare the diagnostic accuracy of the Finnish Diabetes Risk Score (FINDRISC) and the Indian Diabetes Risk Score (IDRS) in screening for dysglycemia (i.e. the presence of prediabetes or T2DM) and to identify the factors associated with T2DM risk among young medical students at the Faculty of Medicine, Taiz University, Yemen. METHODS: A cross-sectional study was conducted among 200 medical students without a prior history of diabetes at Taiz University. The FINDRISC and IDRS questionnaires were used along with fasting blood sugar to estimate the efficacy of both risk scores in screening dysglycemia. Descriptive statistics and the chi-square test were used, with P < 0.05 regarded as statistically significant. The diagnostic accuracy of FINDRISC and IDRS was compared using the area under the receiver operating characteristic curve (AUC-ROC). Sensitivity, specificity, and the best cut-off to detect dysglycemia were calculated for each risk score. The best cutoffs were determined by maximizing the Youden Index. RESULTS: of the 200 participants, 10.5% and 1.5% were diagnosed with prediabetes and T2DM, respectively, where females had a higher prevalence than males for both outcomes (P < 0.001). The screening performance for dysglycemia differed significantly between the two tools (P < 0.001). The FINDRISC achieved a larger AUC-ROC (0.782; 95% CI: 0.68–0.88) with ≥ 9 as an optimal cut-off (sensitivity: 67.0%, specificity: 80.1%, and Youden index: 0.44) compared to that of IDRS (0.671; 95% CI: 0.56–0.78) with ≥ 45 as an optimal cut-off (sensitivity: 46.0%, specificity: 80.0%, Youden Index: 0.25). A Bland-Altman plot indicated adequate agreement at lower risk scores; however, significant divergence at higher risk scores, suggesting the tools are not interchangeable. CONCLUSION: The FINDRISC demonstrated acceptable diagnostic utility and significantly outperformed the IDRS in this young Yemeni population;with its effectiveness increasing at higher risk scores. This makes it the preferred initial screening tool for identifying individuals with dysglycemia, especially in resource-constrained environments, such as the Yemeni setting. However, due to its modest standalone sensitivity, we recommend that individuals with a FINDRISC score ≥ 9 should undergo definitive laboratory testing to confirm the diagnosis and ensure early detection of dysglycemia.