PHQ-9: One factor or two?

• Published in: Psychiatry research Vol. 271; pp. 532 - 534
• Main Authors: Boothroyd, Luke, Dagnan, David, Muncer, Steven
• Format: Journal Article
• Language: English
• Published: Ireland Elsevier B.V 01.01.2019
• Subjects: Adolescent; Adult; Aged; Aged, 80 and over; Depressive Disorder - diagnosis; Factor Analysis, Statistical; Female; Humans; Male; Middle Aged; Patient Health Questionnaire; Psychometrics; Reproducibility of Results; Retrospective Studies; Severity of Illness Index; Young Adult
• ISSN: 0165-1781; 1872-7123; 1872-7123
• DOI: 10.1016/j.psychres.2018.12.048

•Confirmatory Factor Analysis revealed a two-factor solution of the PHQ9 to be a better fit.•The two factors represented a somatic factor and a cognitive-affective factor of depression but they are very highly correlated.•Mokken analysis suggested a viable one factor structure.•Given the high correlation and Mokken analysis, it is recommended that the PHQ9 is considered to be one scale. The PHQ-9 has been found to be reliable and valid, but with both a single and two-factor structure suggested in the literature. This measure has not yet been subject to psychometric investigation using Mokken scale analysis. Confirmatory Factor Analysis (CFA) and Mokken analysis were performed on retrospective data from an NHS Trust. CFA found a two-factor structure was a significantly better fit; however, the factors were highly correlated. Mokken analysis suggested that a single scale was viable. The findings support recent research by Gonzalez-Blanch et al. (2018) in suggesting a one factor model is most appropriate for the PHQ-9