Measuring depression and anxiety in autistic college students: A psychometric evaluation of the PHQ-9 and GAD-7

Anxiety, depression, and suicidality are major concerns among college students, though less is known about these constructs in autistic college students. Given the rising number of autistic individuals entering post-secondary education, adequate measurement and diagnosis of anxiety, depression, and suicidality is necessary to provide clinical care to this population. While the Generalized Anxiety Disorder-7 (GAD-7) and Patient Health Questionnaire-9 (PHQ-9) are commonly used in university settings, these instruments have not been psychometrically evaluated in autistic college students. The current study examines the differential psychometric properties of the GAD-7 and PHQ-9 utilizing invariance testing via structural equation modeling in a sample of self-identified autistic (n = 477) and non-autistic (n = 429) university students. Results showed that although the GAD-7 is invariant, the PHQ-9 was mostly invariant, with one item displaying non-invariance. Specifically, the PHQ-9 item assessing suicidality was found to be non-invariant at the metric level, indicating that autistic and non-autistic college students interpret this question differently. Future investigators should continue to assess the appropriateness of using common screening measures, originally created for non-autistic people, in autistic populations.Lay abstract Anxiety, depression, and suicidality are major concerns among college students, though less is known about these constructs in autistic college students. Given the rising number of autistic individuals entering college, adequate screening and diagnosis of mental health difficulties is necessary to provide care to this population. For example, despite widespread usage of the Patient Health Questionnaire-9 (PHQ-9) and Generalized Anxiety Disorder-7 (GAD-7), it is unknown if the two instruments accurately measure depression and anxiety the same way between non-autistic and autistic individuals. Thus, the current study was the first to examine how these instruments measure depression and anxiety symptoms in autistic versus non-autistic college students. Utilizing tests of construct measurement in a sample of autistic (n = 477) and non-autistic (n = 429) university students aged 18-29, results showed that while the GAD-7 and PHQ-9 appear to be accurately capturing anxiety and depression in autistic college students, the PHQ-9 item assessing suicidality was found to be measuring different things. This indicates that autistic college students are not interpreting this question the way non-autistic students are, and thus, the measure is failing to capture suicidality in autistic people. Future investigators should continue to assess the appropriateness of using common screening measures, originally created by non-autistic people, in autistic populations.