Factorial validity of the 12-item general health questionnaire in patients with psychological disorders

The 12-item General Health Questionnaire (GHQ-12) is a self-reporting instrument to assess mental health. The present study was aimed to explore the optimal factor model of the Chinese version of the GHQ-12 in patients with psychological disorders, and examine its measurement invariance across gender. A total of 870 outpatients completed the GHQ-12. Among them, 361 patients also completed the Center for Epidemiologic Studies Depression Scale (CES-D) and the Self-Rating Anxiety Scale (SAS). Confirmatory factor analysis (CFA) was adopted to assess the fit of two models of the GHQ-12. Multi-group confirmatory factor analysis (MGCFA) was performed to verify the measurement invariance of the GHQ-12 across gender. The Cronbach's alpha coefficients, McDonald's omega coefficients (omega), and mean inter-item correlations (M-IC) were calculated to investigate the internal reliability of the GHQ-12. Pearson correlations between GHQ-12 and CES-D/SAS were performed to examine the convergent construct validity. The results showed that the 3-factor model of the GHQ-12 fitted the data better than the 2-factor model, and the 3-factor model of the GHQ-12 was measurement invariant across gender. The Cronbach's alpha coefficients of the GHQ-12 and its factors ranged from 0.63 to 0.83, omega ranged from 0.80 to 0.90, and M-IC ranged from 0.28 to 0.46. The scores of the GHQ-12 and its factors had significantly positive correlations with scores of CES-D and SAS. In conclusion, the good factorial validity and reliability of the Chinese version of the GHQ-12 supported its application in evaluating the general mental health of patients in clinical context.