Selecting appropriate candidates for monogenic diabetes testing remains a key challenge in clinical practice. A study published in Diabetes Care describes an updated probability calculator that combines clinical variables and biomarkers to improve diagnostic precision in adults with early-onset diabetes.

The model integrates clinical parameters with C-peptide and islet autoantibody data to predict the probability of monogenic diabetes. Two models were developed: one for early insulin-treated patients (proxy for type 1 diabetes) and another for not-early insulin-treated individuals (proxy for type 2 diabetes). Model development used case-control data (n = 1,191) and recalibration with the UNITED population study (n = 1,299). The model was externally validated in a population-based cohort (n = 1,025).

In early insulin-treated patients, the integrated model demonstrated superior performance (ROCAUC 0.98; 95% CrI 0.95–0.98) compared with models based on clinical features alone (0.80) or biomarkers alone (0.96). Among not-early insulin-treated participants, discrimination remained strong (ROCAUC 0.86). External validation confirmed high predictive accuracy (ROCAUC 0.98 and 0.92, respectively). Applying a ≥5% probability threshold identified suitable candidates for genetic testing with 16–19% positivity for monogenic diabetes.

This online calculator provides immediate clinical utility by enabling targeted and data-driven genetic testing in adults diagnosed with diabetes at or before 35 years of age. The model represents a meaningful advancement toward precision-based screening and management of monogenic diabetes.