BMC Psychiatry (Jan 2021)

Psychometric properties of the Norwegian version of the Patient Health Questionnaire-9 (PHQ-9) in a large female sample of adults with and without eating disorders

  • Line Wisting,
  • Sverre Urnes Johnson,
  • Cynthia M. Bulik,
  • Ole A. Andreassen,
  • Øyvind Rø,
  • Lasse Bang

DOI
https://doi.org/10.1186/s12888-020-03013-0
Journal volume & issue
Vol. 21, no. 1
pp. 1 – 11

Abstract

Read online

Abstract Background Internationally, the Patient Health Questionnaire-9 (PHQ-9) is commonly used to assess the frequency and severity of depressive symptoms. However, psychometric properties of the Norwegian version of the PHQ-9 have only been assessed in adolescents. We present normative data for women and an evaluation of the psychometric properties (internal consistency, convergent validity, and factor structure) of the Norwegian PHQ-9 among women with and without eating disorders (ED). Methods In this case-control study, a total of 793 females aged 18–78 years (mean 30.39; SD 9.83) completed an online self-report assessment. Measures included the ED100K and Eating Disorder Examination Questionnaire (EDE-Q) to assess ED psychopathology, and the Generalized Anxiety Disorder (GAD) scale and Difficulties in Emotion Regulation Scale Short Form (DERS-SF) to assess symptoms of anxiety and emotion regulation deficits. Participants were categorized into three groups, i.e., previous ED (19.7%, n = 148), current ED (36.3%, n = 272), and no history of ED (44.0%, n = 330), based on self-reported scores on the ED 100 K and the EDE-Q. Results Mean PHQ-9 total score for those with a previous history of ED was 10.67 (SD 6.33), for those with a current ED 16.61 (SD 5.84), and for those with no lifetime history of ED 6.83 (SD 5.58). Excellent internal consistency was demonstrated by Cronbach’s alpha’s for individuals with a previous ED (.88), for individuals with a current ED (.86), and for individuals with no history of ED (.88). Acceptable convergent validity was indicated based on significant correlations between the PHQ-9 and GAD-7 and DERS-SF. Confirmatory Factor Analyses revealed a mediocre fit for a one-factor structure of the PHQ-9, regardless of diagnostic status. Conclusions The psychometric properties of the Norwegian version of the PHQ-9 are acceptable across females with and without ED, and the PHQ-9 can be recommended for use in clinical ED settings and for people without mental disorders.

Keywords