The Study of Psychometric Properties of the Persian Version of the Patient Health Questionnaire-9 (PHQ-9) Among Students of Guilan University of Medical Sciences in the Academic Year 2017-18

Abstract

Objectives: The Patient Health Questionnaire-9 (PHQ-9) is one of the most commonly used measures for screening, diagnosing and monitoring of depression treatment. So far, four studies in Iran have examined some of its psychometric properties in the population of medical and psychiatric patients - however, some aspects of its psychometrics and in the student population have not yet been adequately evaluated. The aim of this study is to examine more aspects of psychometric characteristics of PHQ-9 in Iran with a focus on the student population. Methods: In a descriptive study, 463 students of Guilan University of Medical Sciences in the academic year 2017-18were selected using convenience sampling and completed the research measures. The reliability of the Persian version of PHQ-9 was assessed using internal consistency and test-retest with two week interval. Its construct validity in the sample of students was evaluated by exploratory and confirmatory factor analyses. To evaluate its convergent validity, the second version of the Beck Depression Inventory (BDI-II), the depression subscale of the 21-item Depression Anxiety Depression Inventory (DASS-21), the neuroticism subscale of NEO Five Factor Inventory (NEO-FFI) and the negative affect subscale of the Positive Affect and Negative Affect Scale (PANAS) were used. The openness subscale of NEO-FFI and the positive affect subscale of PANAS were also used to assess the discriminant validity. Results: According to the findings, Cronbachchr('39')s alpha coefficient of 0.856 and intraclass correlation coefficient of 0.869 indicated, respectively, internal consistency and high test-retest reliability of PHQ-9. Exploratory factor analysis found that all questions were in one factor, accounting for 47.59% of the total variance. Confirmatory factor analysis showed that all questions were related to one factor. Positive correlation of PHQ-9 with BDI-II (r = 0.769), the neuroticism subscale of NEO-FFI (r = 0.508), the depression subscale of DASS-21 (r = 0.647) and the negative affect subscale of PANAS (r = 0.430) represented the convergent validity and its negative correlation with the positive affect subscale PANAS (r = -0.444) indicated its discriminant validity, while the openness subscale of NEO indicates NEO-FFI (r = 0.116) did not show a significant relationship. Conclusion: Overall, the findings of the present study suggest that PHQ-9 as a short-term measure has the potential to be used in clinical and research domains to screen, diagnose, and monitor clinical and/or subclinical depression in the student population.