Examining the classification accuracy of GAD-7 and PHQ-9 against the QuickSCID-5 in Canadian undergraduates: Improving post-secondary student mental health screening

Background: The classification accuracy of the General Anxiety Disorder Scale (GAD-7) and Patient Health Questionnaire (PHQ-9) in post-secondary students remains uncertain. Methods: Undergraduate students completed a survey (n = 159) and QuickSCID-5 (n = 130). Sensitivity, specificity, and receiver operating characteristics were calculated, and alternative scoring strategies were explored. Results: Most students identified as first-year, Canadian, and female. Among survey respondents, 42% (GAD-7) and 35% (PHQ-9) scored >10. Most students with high stress and GAD-7 scores did not meet the criteria for GAD. At a screening cut-off of >10, sensitivity and specificity were: 80.6/72.3% (GAD-7) and 92.9/72.4% (PHQ-9), respectively. Preliminary evidence supported moderate classification accuracy of GAD-7 (AUC = 0.78) for detecting GAD, but poor accuracy for detecting other disorders. PHQ-9 showed high classification accuracy (AUC = 0.92) for depression; preliminary evidence supported improved accuracy with alternative scoring. Conclusions: While PHQ-9 appears to be a useful screening measure, findings suggest the GAD-7 may capture stress rather than clinical anxiety.