Confirmatory factor analysis of the Patient Health Questionnaire-9: a study of the participants from the spinal cord injury model systems.

OBJECTIVE: To examine the factor structure of the Patient Health Questionnaire-9, a measure of depression, in persons with spinal cord injury (SCI).
DESIGN: Cross-sectional, confirmatory factor analytic study.
SETTING: Community. PARTICIPANTS: Data for 7296 persons with an SCI who had sustained their injury at least 1 year prior to assessment and who had complete Patient Health Questionnaire-9 data collected at a follow-up interview were drawn from the National Spinal Cord Injury Statistical Center Database.
INTERVENTIONS: None. MAIN OUTCOME MEASURES: Factor structure of the Patient Health Questionnaire-9.
RESULTS: Confirmatory factor analysis indicated a marginal fit for the single factor solution (root mean square error of approximation [RMSEA] = 0.086), whereas the solution with 5 somatic items and 4 nonsomatic items had the best fit (RMSEA = 0.054) among 2-factor models that used all 9 items. Of the models that used fewer than 9 items, the best fit was for the 6-item solution with 3 somatic items (sleep, appetite, and fatigue) and 3 nonsomatic items (feeling down, feeling bad about self, and suicidal ideation; RMSEA = 0.043). Similar results were found across the strata except for the Hispanic group (for whom no model fit well).
CONCLUSIONS: Given the results of this analysis that support a 2-factor structure of the Patient Health Questionnaire-9 in persons with SCI, the next step in this line of research is to validate each of these dimensions against other ways of measuring depression.