Validity of the Patient Health Questionnaire, PHQ-9, administered through interactive-voice-response technology

We compared the administration of the 9-item Patient Health Questionnaire (PHQ-9) to assess depressive symptoms using interactive voice response (IVR) technology with the method of administration using paper-and-pencil. Data were collected from 51 veterans participating in an 8-week randomized controlled trial of an illness management programme for heart failure. To counter possible bias in answering questions via IVR technology, the anchoring responses of the PHQ-9 questionnaire were reversed so that lower numbers corresponded to more severe depression. The mean for the pencil-and-paper administered PHQ-9 was 4.1 (SD = 4.5) and the mean for IVR administration was 2.8 (SD = 3.1). The internal consistency (Cronbach's alpha) of the PHQ-9 was 0.76 for IVR administration and 0.82 for paper administration. The intraclass correlation coefficient for the two modes of administration was 0.65, indicating moderate agreement. IVR administration of the PHQ-9 produces similar results to pencil-and-paper administration, but the former is not as sensitive to higher levels of depressive symptom severity. This suggests that a lower threshold for probable depression is warranted when assessing depressive symptoms with IVR.