Reliability and validity of the Chinese version of the Patient Health Questionnaire (PHQ-9) in the general population

Objective: Depression is one of the most common mental illnesses. The reliability and the validity of the Patient Health Questionnaire (PHQ)-9, a depression screening tool, have not been examined in the general population in China. Thus, this study evaluated the reliability and the validity of the Chinese version of the PHQ-9 in detecting major depression in residents of a Chinese community.

Methods: A total of 1045 participants from a Shanghai community were enrolled in our study. Participants completed the Chinese versions of the PHQ-9, the Self-Rating Depression Scale (SDS), the 36-item Short Form Health Survey (SF-36), and the Mini International Neuropsychiatric Interview. One hundred participants were randomly selected to complete the PHQ-9 again 2 weeks after the initial assessment. The reliability, the validity and the receiver operating characteristic (ROC) curve of the PHQ-9 were analyzed.

Results: Cronbach's alpha for the internal consistency reliability of the Chinese version of the PHQ-9 was 0.86 for the entire scale. The correlation coefficient for the 2-week test-retest of the total score was 0.86. The PHQ-9 scale correlated positively with the SDS (r=0.29, p<0.001) and correlated negatively with all subscale scores of the SF-36 (correlation coefficients ranged from -0.11 to -0.47, p<0.001). The area under the curve of the ROC was 0.92 (95% confidence interval: 0.86-0.97). A cutoff score of 7 or higher on the PHQ-9 had a sensitivity of 0.86 and a specificity of 0.86.

Conclusions: In the general Chinese population, the Chinese version of the PHQ-9 is a valid and efficient tool for screening depression, with a recommended cutoff score of 7 or more.