Objectives: To determine the performance of the Patient Health Questionnaire 2 (PHQ-2) against the criterion standard for diagnosing major depression, the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV), and to examine its performance across age, sex, and racial/ethnic groups.
Design: Cross-sectional observational study.
Setting: The National Epidemiologic Survey on Alcohol and Related Conditions (NESARC) (2001/02), a nationally representative survey of the noninstitutionalized U.S. household population.
Participants: The 8,205 adults aged 65 and older who participated in NESARC.
Measurements: The PHQ-2's criterion validity (sensitivity, specificity, and area under the receiver operating characteristic curve (AUC) against the DSM-IV) and construct validity (Spearman correlations between the PHQ-2 and the six scales of the Medical Outcomes Study 12-item Short Form Questionnaire (SF-12)) were calculated.
Results: The PHQ-2's criterion validity for major depression was good (sensitivity=100%, specificity=77%, AUC=0.88). Its sensitivity was 100% for each subgroup. Specificity increased with age, was higher for men than for women, and differed across racial and ethnic groups. For the total sample and each sex, all six SF-12 scales were significantly lower in people who tested positive using the PHQ-2. For each age and racial or ethnic group, most of the six scales differed significantly between those who tested positive and tested negative.
Conclusion: The PHQ-2 is a valid screening tool for major depression in older people but should be followed by a more-comprehensive diagnostic process. Although its specificity differs by age, sex, and racial and ethnic groups, these differences appear to be of little clinical significance.