OBJECTIVES : Few outcome measures were validated for assessing depressive symptoms in AD. Patient Health Questionnaire-9 (PHQ9) and the abridged PHQ2 are established patient-reported outcome measures of depressive symptoms. We sought to examine the measurement properties of PHQ9 and PHQ2 in adult AD.

METHODS : A prospective dermatology-practice based study of 458 AD patients (age 18-97 years).

RESULTS : PHQ9 moderately correlated with Dermatology Life Quality Index, Patient-Reported Outcomes Measurement Information System (PROMIS) Sleep-Disturbance and Sleep-Related Impairment, and PROMIS Itch Questionnaire Mood and Sleep (PIQ-MS), and weakly correlated with Patient-Oriented Eczema Measure, Numeric Rating Scale (NRS) average-itch, NRS-sleep, Eczema Area and Severity Index, Scoring AD and Rajka-Langeland scores. PHQ2 had significantly weaker correlations than PHQ9 with PROMIS SD, SRI and PIQ-MS, but similar correlations with other outcomes. PHQ9 and PHQ2 had good discriminant validity. Changes from baseline in PHQ9 and PHQ2 were poorly or weakly correlated with changes of the other outcome measures. There was no differential item functioning of PHQ items. PHQ9 showed good reliability (intraclass correlation coefficient range: 0.80-0.87). PHQ2 had slightly lower reliability (0.76-0.82).

CONCLUSIONS : PHQ9 and PHQ2 had similar measurement properties, but PHQ2 was more feasible to assess depressive symptoms in AD.