The revised-FINDRISC: A tool for type 2 diabetes risk screening across diverse populations incorporating sociodemographic indicators

AIM: Τo develop a revised version of the FINDRISC tailored to improve the identification of individuals at high risk for type 2 diabetes (T2D) across diverse populations. METHODS: The revised-FINDRISC was developed using pooled harmonized baseline data from the multinational Feel4Diabetes (NCT02393872; N = 3526) and DigiCare4You (NCT05648383; N = 2156) studies. Validation utilized data from the 20-year ATTICA cohort study in Greece. Original FINDRISC components were analyzed using Fisher's Linear Discriminant Analysis followed by supervised stepwise logistic regression to identify significant T2D cross-sectional predictors. The revised-FINDRISC was developed by re-scoring or excluding original items and incorporating new predictors, generating an integer-based score to enhance usability. RESULTS: The revised-FINDRISC significantly improved accuracy in identifying current T2D risk over the original, with an overall AUC of 0.911(95%CI: 0.901,0.922) versus 0.832(0.815,0.846). When stratified by country economic classification, the revised score outperformed the original across all groups: LMICs: 0.881(0.864,0.898) vs. 0.815(0.793,0.837), South European-HICs: 0.891(0.867,0.916) vs. 0.839(0.811,0.869), and North European-HICs: 0.921(0.872,0.971) vs. 0.844(0.742,0.949). A clinical threshold score of 13/24 in the revised-FINDRISC offers optimal sensitivity and specificity for identifying individuals at risk of T2D. CONCLUSION: Enhanced with demographic, socio-economic, and clinical factors, the revised-FINDRISC potentially represents a substantive advancement in T2D screening, supporting early detection and targeted prevention.