Assessing measurement invariance of three depression scales between neurologic samples and community samples.

PURPOSE

Measurement invariance is necessary for meaningful group comparisons. The purpose of this study was to test measurement invariance of three patient-reported measures of depressive symptoms between neurologic and community samples.

METHODS

The instruments tested included the center for epidemiologic studies depression scale (CESD-20), the patient health questionnaire-9 (PHQ-9), and the patient-reported outcome measurement information system depression short form (PROMIS-D-8). Responses from a community sample were compared to responses from samples with two neurologic conditions: multiple sclerosis and spinal cord injury. Multi-group confirmatory factor analysis was used to evaluate successive levels of measurement invariance: (a) configural invariance, i.e., equivalent item factor structure between groups; (b) metric invariance, i.e., equivalent unstandardized factor loadings between groups; and (c) scalar invariance, i.e., equivalent item intercepts between groups.

RESULTS

Results of this study supported metric invariance for the CESD-20, PHQ-9, and PROMIS-D-8 scores between the community sample and the samples with neurologic conditions. The most rigorous form of invariance (i.e., scalar) also holds for the CESD-20 and the PROMIS-D-8.

CONCLUSIONS

The current study suggests that depressive symptoms as measured by three different outcome measures have the same meaning across clinical and community samples. Thus, the use of these measures for group comparisons is supported.