Validation of the Canadian Diabetes Risk Questionnaire Tool Among Young African, Caribbean, and Black Adults Living in Canada

OBJECTIVES: The prevalence of type 2 diabetes (T2D) has increased substantially over the last decade among young populations living in Canada, with certain ethnic groups, such as African, Caribbean, and Black (ACB), being disproportionately affected. However, little is known about the suitability of the widely adopted T2D risk assessment tool, Canadian Diabetes Risk Questionnaire (CANRISK), for this population. METHODS: Two analytical samples of individuals aged 18 to 39 years were identified from 3 phases of CANRISK validation data collection: 1) young individuals from ACB ethnicities; and 2) all young individuals. Descriptive analyses were conducted for all predictors from the CANRISK and for prevalent dysglycemia (assessed by oral glucose tolerance test). The discriminatory ability of the CANRISK model was assessed by generating the area under the curve (AUC) and other model metrics. In addition, we assessed several cut points for the ACB subgroup to identify the optimal threshold. RESULTS: The prevalence of dysglycemia was higher among the ACB subgroup (n=715) than the overall young sample (n=3,960) (8.4% vs 7.7%). The CANRISK model performance was comparable for both groups, but slightly more accurate in the ACB subgroup (AUC 71.9% vs AUC 71.5%). The optimal cutoff for young ACB individuals was the same optimal cutoff of 19 as for the general Canadian youth population. CONCLUSIONS: Our findings indicate that the CANRISK tool, with a modified cutoff point, is a suitable tool to identify T2D risk among young adults from ACB ethnicities. Future studies could explore the tool's predictive ability via longitudinal studies to assess long-term diabetes risk.