Validating the CANRISK prognostic model for assessing diabetes risk in Canada's multi-ethnic population

Area of Science:

• Endocrinology
• Public Health
• Epidemiology

Background:

• High prevalence of undiagnosed diabetes and prediabetes in Canada necessitates effective risk assessment tools.
• Existing tools may not be suitable for Canada's diverse ethnic population.
• Lack of validated instruments for personal diabetes risk estimation in the Canadian context.

Purpose of the Study:

• To evaluate the accuracy and discrimination of the Canadian Diabetes Risk Assessment Questionnaire (CANRISK).
• To assess CANRISK's effectiveness in detecting diabetes and prediabetes (dysglycemia) in a multi-ethnic Canadian population.
• To compare CANRISK's performance against alternative risk-scoring models.

Main Methods:

• Cross-sectional screening study involving 6223 adults of diverse ethnicities.
• Glycemic status confirmed using the oral glucose tolerance test (OGTT).
• Development and validation of electronic and paper-based CANRISK scores using logistic regression and ROC analyses (AUC).

Main Results:

• Electronic and paper-based CANRISK scores achieved an Area Under the Curve (AUC) of 0.75.
• CANRISK demonstrated significantly higher accuracy than the Finnish FINDRISC score (AUC: 0.66) and a simple Obesity model (AUC: 0.69).
• The 95% confidence intervals for CANRISK AUC were 0.73-0.78.

Conclusions:

• CANRISK is a statistically valid tool for assessing diabetes risk in Canada's multi-ethnic population.
• CANRISK offers superior accuracy in identifying individuals with dysglycemia compared to FINDRISC and an Obesity model.
• The questionnaire provides a promising solution for diabetes risk screening in Canada.