Six patient-reported outcome measurement information system short form measures have negligible age- or diagnosis-related differential item functioning in individuals with disabilities

Abstract

Objective: To evaluate the measurement invariance of 6 self-report measures selected for an ongoing longitudinal study of individuals with spinal cord injury, muscular dystrophy, postpolio syndrome, and multiple sclerosis.

Design: Participants completed and returned by mail surveys that included the targeted self-report measures. Ordinal logistic regressions methods were applied to evaluate items for differential item functioning (DIF) by diagnosis and age range.

Setting: Community.

Participants: Participants (N=2479) who had 1 of the 4 target diagnoses.

Interventions: None.

Main outcome measures: Six short-form measures from the Patient-Reported Outcome Measurement Information System (PROMIS) were administered to participants to measure fatigue, pain interference, satisfaction with social roles, sleep disturbance, sleep-related impairment, and depression.

Results: One item of 1 measure (fatigue) exhibited DIF by diagnosis based on a published standard for meaningful DIF. However, scores corrected for this DIF were highly correlated with uncorrected scores (r>.999). No DIF by age range was found for any of the measures.

Conclusions: Study findings support the use of the selected PROMIS short forms for comparing symptoms and quality of life indicators across different diagnoses and age ranges.