Psychometric properties of three depression scales in people with multiple sclerosis.
|Title||Psychometric properties of three depression scales in people with multiple sclerosis.|
|Publication Type||Journal Article|
|Authors||Kim J, Amtmann D, Cook KF, Johnson KL, Bamer AM, Chung H, Askew RL|
|Journal||International Journal of MS Care|
|Quality of Life in MS|
Background: A number of self-reported depression instruments are available. Some of the symptoms of depression are similar to symptoms of multiple sclerosis (MS), so it is important to examine how the scales function in this population. Objectives: The purpose of this study was to compare psychometric properties of the Patient Health Questionnaire-9 (PHQ-9), the Center for Epidemiological Studies Depression Scale-10 (CESD-10), and the 8-item depression short form from the Patient Reported Outcome Measurement Information System (PROMIS-D-8). Methods: The sample consisted of 455 participants with MS responding to a mailed paper survey. Data were analyzed to assess factor structure, internal reliability, and convergent/discriminant validity of the PHQ-9, CESD-10, and PROMIS-D-8. In addition, severity analyses of the PHQ-9 and CESD-10 were conducted. Results: Bifactor analysis suggested that one general factor explained the majority of the common variance for each measure. Cronbach alpha values for each measure were all equal to or greater than 0.85. The corrected item-total correlation showed that all items in each measure except one item in both the PHQ-9 and CESD-10 were adequately correlated. The three scales were more correlated with one another than with measures of pain, sleep, and fatigue, supporting convergent/discriminant validity. Based on CESD-10 scores, more people were categorized as having significant depressive symptoms compared to PHQ-9. Conclusions: Psychometrically, the three scales were found to be unidimensional and showed acceptable internal reliability and convergent/ discriminant validity in participants with MS. All scales are appropriate for use in people with MS. Researchers and clinicians should choose the most appropriate measure based on their needs. For example, the PHQ-9 is the best choice for those who are interested in diagnosing clinical depression. The CESD-10 and PROMIS-D-8 could be used to assess depressive symptoms for epidemiological purposes. PROMISD-8 allows for comparison of depression level to national norms.