Validation of the Patient Health Questionnaire for Depression Screening Among the Elderly Patients in Taiwan

Background This study aimed to assess the reliability, validity, and factor structure of the Patient Health Questionnaire-9 (PHQ-9) and its two subscales (PHQ-2, PHQ-1) for screening major depression disorder (MDD) in primary care elderly patients in Taiwan. Method A total of 634 participants aged ≥60 years were enrolled from the primary care settings. They completed PHQ-9, and Short Form of Quality of Life Enjoyment and Satisfaction Questionnaire (Q-LES-Q-SF). After that, they were interviewed by using Schedule for Clinical Assessments in Neuropsychiatry (SCAN) and 17-item of Hamilton Rating Scale (HAMD-17). Result The PHQ-9 demonstrated acceptable internal consistency (α = 0.77), test-retest reliability (intraclass correlation coefficient = 0.79), and concurrent validity with HAMD-17 (r = 0.66; p < 0.001). Exploratory factor analysis yielded 2-factor structure consisted of non-somatic (13.3% variance) and somatic (38.8% variance) dimensions. The correlations between PHQ-9 with Q-LES-Q-SF (whole and two subscales) were moderate (r range −0.47 to −0.53, all ps < 0.001). Using SCAN as gold standard, the optimal cutoff score of ≥6 for PHQ-9 demonstrated good sensitivity and specificity for detecting MDD (1.00 and 0.85, respectively). The screening accuracy of the two items version was also acceptable (PHQ-2 score ≥ 2: sensitivity 0.93; specificity 0.90; PHQ-1 score ≥ 1: sensitivity 1.00; specificity 0.77). Conclusion The Chinese versions of the PHQ-9 and its two subscales (PHQ-2 and PHQ-1) are valid and reliable to screen MDD in primary care elderly patients in Taiwan and could be proper alternatives to Geriatric Depression Scale 15-item version.