ABSTRACT
OBJECTIVE: The Patient Health Questionnaire-9 (PHQ-9) is a widely used self-report measure of depression whose psychometric properties were found to be merely acceptable. Insufficient attention has been devoted to its clinimetric validity, while its clinical utility is still debated, particularly for assessing depression severity. This is the first study to test the PHQ-9 construct validity and clinical utility based on clinimetric principles. METHODS: An online survey of 3,398 participants was conducted. Item response theory models (Rasch and Mokken analyses) were used to assess the PHQ-9 validity and determine its clinical utility. RESULTS: Fit to the Rasch model was achieved after adjusting the sample size. Items 2, 4, 6, and 9 over-discriminated, while items 1, 5, and 7 under-discriminated. Local dependency between items 2 and 6 was indicated. The PHQ-9 was not unidimensional. A Loevinger's coefficient of 0.49 was found, indicating an acceptable level of scalability. CONCLUSION: The PHQ-9 is an instrument with potential clinical utility as an overall index of depression, mainly for screening purposes. Substantial revision, particularly in the wording of over- and under-discriminating items, is needed.
Subject(s)
Depression , Patient Health Questionnaire , Psychometrics , Humans , Female , Male , Adult , Reproducibility of Results , Middle Aged , Patient Health Questionnaire/standards , Depression/diagnosis , Depression/psychology , Self Report/standards , Young Adult , Adolescent , Aged , Surveys and Questionnaires/standardsABSTRACT
Objective: The Patient Health Questionnaire-9 (PHQ-9) is a widely used self-report measure of depression whose psychometric properties were found to be merely acceptable. Insufficient attention has been devoted to its clinimetric validity, while its clinical utility is still debated, particularly for assessing depression severity. This is the first study to test the PHQ-9 construct validity and clinical utility based on clinimetric principles. Methods: An online survey of 3,398 participants was conducted. Item response theory models (Rasch and Mokken analyses) were used to assess the PHQ-9 validity and determine its clinical utility. Results: Fit to the Rasch model was achieved after adjusting the sample size. Items 2, 4, 6, and 9 over-discriminated, while items 1, 5, and 7 under-discriminated. Local dependency between items 2 and 6 was indicated. The PHQ-9 was not unidimensional. A Loevinger's coefficient of 0.49 was found, indicating an acceptable level of scalability. Conclusion: The PHQ-9 is an instrument with potential clinical utility as an overall index of depression, mainly for screening purposes. Substantial revision, particularly in the wording of over- and under-discriminating items, is needed.