Discriminant Ability, Concurrent Validity, and Responsiveness of PROMIS Health Domains Among Patients With Lumbar Degenerative Disease Undergoing Decompression With or Without Arthrodesis.

STUDY DESIGN: A prospective cohort study.

OBJECTIVE: The aim of this study was to investigate the ability of Patient-Reported Outcomes Measurement Information System (PROMIS) health domains to discriminate between levels of disease severity and to determine the concurrent validity and responsiveness of PROMIS relative to "legacy" measures.

SUMMARY OF BACKGROUND DATA: PROMIS may measure recovery after lumbar spine surgery. Concurrent validity and responsiveness have not been compared with legacy measures in this population.

METHODS: We included 231 adults undergoing surgery for lumbar degenerative disease. Discriminant ability of PROMIS was estimated for adjacent categories of disease severity using the Oswestry Disability Index (ODI). Concurrent validity was determined through correlation between preoperative legacy measures and PROMIS. Responsiveness was estimated using distribution-based and anchor-based criteria (change from preoperatively to within 3 months postoperatively) anchored to treatment expectations (North American Spine Society Patient Satisfaction Index) to determine minimal important differences (MIDs). Significance was accepted at P < 0.05.

RESULTS: PROMIS discriminated between disease severity levels, with mean differences between adjacent categories of 3 to 8 points. There were strong to very strong correlations between Patient Health Questionnaire-8, Generalized Anxiety Disorder-7, and PROMIS anxiety, depression, fatigue, and sleep disturbance; between ODI and PROMIS fatigue, pain, and physical function; between the 12-Item Short-Form Health Survey physical component and PROMIS pain and physical function; and between the Brief Pain Inventory (BPI) pain interference and PROMIS depression and pain. BPI back pain and leg pain intensity showed weak or no correlation with PROMIS. Distribution-based MIDs ranged from 3.0 to 3.5 points. After incorporating longitudinal anchor-based estimates, final PROMIS MID estimates were anxiety, -4.4; depression, -6.0; fatigue, -5.3; pain, -5.4; physical function, 5.2; satisfaction with participation in social roles, 6.0; and sleep disturbance, -6.5.

CONCLUSION: PROMIS discriminated between disease severity levels, demonstrated good concurrent validity, and was responsive to changes after lumbar spine surgery.

LEVEL OF EVIDENCE: 2.