Validation of the Computerized Adaptive Test for Mental Health in Primary Care

The Article in Brief

Validation of the Computerized Adaptive Test for Mental Health in Primary Care

Andrea K. Graham , and colleagues

Background While screening questionnaires for depression and anxiety exist in primary care settings, electronic health tools, such as computerized adaptive tests based on item response theory, could advance screening practices. This study evaluated the validity of the Computerized Adaptive Test for Mental Health as a tool for screening for major depressive disorder and assessing severity of anxiety and major depressive disorder among adult primary care patients.

What This Study Found Computerized adaptive testing is a valid tool for screening for major depressive disorder in primary care and offers a format that is well received by patients. The research compared computerized adaptive tests, which personalize assessments by adaptively varying questions based on previous responses, with widely used paper screening tools (Patient Health Questionnaire-9, Patient Health Questionnaire-2, and Generalized Anxiety Disorder-7), and semi-structured interview, which is generally considered the gold standard in psychiatric assessment. The diagnostic accuracy of the Computerized Adaptive Diagnostic Test for Major Depressive Disorder was similar to the PHQ-9 and higher than the PHQ-2. Compared to interview, the accuracy of the Computerized Adaptive Test/Anxiety Inventory was similar to the Generalized Anxiety Disorder-7 for assessing anxiety severity. Participants preferred using tablet computers (53 percent), compared to interview (33 percent) and paper-and-pencil questionnaires (14 percent). The majority of participants (64 percent) rated paper-and-pencil questionnaire as their least preferred screening method.

Implications

• The widespread use of electronic health records in primary care presents new opportunities to leverage electronic tools for screening, the authors suggest, while multidimensional item response theory, used in computerized adaptive testing, can increase the efficiency of assessing mental and physical health.