Concurrent Validity and Responsiveness of PROMIS Health Domains Among Patients Presenting for Anterior Cervical Spine Surgery

Abstract

Study design: Prospective cohort study.

Objective: The aim of this study was to determine the validity and responsiveness of Patient-Reported Outcomes Measurement Information System (PROMIS) health domains.

Summary of background data: PROMIS health domains (anxiety, depression, fatigue, pain, physical function, satisfaction with participation in social roles, sleep disturbance) may measure quality of care and determine minimal important differences (MIDs) after spine surgery. We examined concurrent validity of PROMIS domains before and PROMIS domain MIDs after anterior cervical spine surgery.

Methods: We included 148 adults undergoing cervical spine surgery from February 2015 through June 2016. We determined concurrent validity by correlations of preoperative PROMIS domains with legacy measures and responsiveness of PROMIS domains using distribution-based and anchor-based criteria (preoperative to postoperative change, within 6 months) anchored to treatment expectations (assessed using North American Spine Society Patient Satisfaction Index criteria). Statistical significance was accepted as P < 0.05.

Results: All PROMIS domains showed moderate to strong correlations with Neck Disability Index, Short-Form Health Survey, version 2 (SF-12v2), and Brief Pain Inventory pain interference and weak correlations with intensity of arm/neck pain (except between PROMIS pain and neck pain [r = 0.45, P < 0.001] and PROMIS physical function and SF-12v2 physical [r = -0.14, P = 0.138] and mental [r = 0.39, P < 0.001] components). PROMIS domains were well correlated with Generalized Anxiety Disorder-7 and Patient Health Questionnaire-8 except PROMIS physical function (r = -0.29, P = 0.002). Distribution-based PROMIS MID estimates ranged from 2.3 to 3.9 points. Incorporating cross-sectional and longitudinal anchor-based criteria, final PROMIS MID estimates were as follows: anxiety, -5.7; depression, -4.6, fatigue, -5.8; pain, -5.2; physical function, 4.5; satisfaction with participation in social roles, 4.4; and sleep disturbance, -7.4.

Conclusion: PROMIS domains are a valid assessment of health in this population and were responsive to postoperative improvements in symptoms and quality of life.

Level of evidence: 2.