Evaluating the psychometric properties and responsiveness to change of 3 depression measures in a sample of persons with traumatic spinal cord injury and major depressive disorder

Abstract

OBJECTIVES: To compare the measurement properties and responsiveness to change of the Patient Health Questionnaire-9 (PHQ-9), the Hopkins Symptom Checklist (HSCL-20) and the Hamilton Depression Rating Scale (HAM-D) in people with spinal cord injury (SCI) diagnosed with major depressive disorder (MDD). DESIGN: Secondary analysis of depression symptoms measured at 6 occasions over 12 weeks as part of a randomized control trial of venlafaxine XR for MDD in persons with SCI. SETTING: Outpatient and community settings. PARTICIPANTS: 133 individuals consented and completed the drug trial. Eligibility criteria were age at least 18 years old; traumatic SCI; and diagnosis of MDD. INTERVENTION: venlafaxine XR. MAIN OUTCOME MEASURES: Patients completed the PHQ-9 and the HSCL-20 depression scales; clinical investigators completed the HAM-D and the Structured Clinical Interview for DSM-IV Depression Module (SCID-D), which was used as a diagnostic criterion measure. RESULTS: All 3 instruments were improved with rating scale analysis. The HSCL-20 and HAM-D contained items that misfit the underlying construct and that correlated weakly with the total measures. Removing these items improved the internal consistency with floor effects increasing slightly. The HAM-D correlated most strongly with SCID-D diagnoses. Depression improvement was similar on all measures in both treatment and control groups. CONCLUSIONS: The psychometric properties of the revised depression instruments are more than adequate for routine use in adults with SCI and are responsive to clinical improvement. The PHQ-9 is the simplest instrument, with measurement properties as good as or better than those of the other instruments and requiring the fewest modifications.