Evaluating the Correlation and Performance of PROMIS to SRS Questionnaires in Adult and Pediatric Spinal Deformity Patients

Abstract

Study design: Retrospective cross-sectional cohort analysis.

Objectives: 1) To assess the correlation of Patient-Reported Outcomes Management Information System (PROMIS) domains with SRS-22r/SRS-30 domains in all scoliosis patients; 2) to assess the correlation of PROMIS domains with SRS-30 domains in adult scoliosis patients; 3) to assess the correlation of PROMIS domains with SRS-22r/SRS-30 domains in pediatric scoliosis patients; and 4) to assess ceiling and floor effects of PROMIS and SRS-22r/SRS-30 domains.

Summary of background data: Studies evaluating correlations between PROMIS and a number of legacy PRO tools have been conducted. To our knowledge, no literature exists examining the correlation of PROMIS and SRS questionnaires in adult and pediatric spinal deformity patients.

Methods: Outpatient visits from July 2015 to December 2017 with concurrent PROMIS and SRS questionnaires were analyzed. Pediatric patients completed the SRS-22r, whereas adults completed the SRS-30. PROMIS measured Physical Function/Mobility, Pain Interference, and Depression domains. Spearman correlation coefficients (ρ) were calculated. Ceiling and floor effects were calculated and compared.

Results: 227 (164 adult; 64 pediatric) patient visits representing 173 patients were included. Moderate to strong correlation existed between PROMIS Physical Function/Mobility and SRS Function/Activity (F/A) domains (ρ, range 0.59-0.84; p < .001). PROMIS Pain Interference and SRS Pain domains showed strong-moderate to strong correlation (ρ, range -0.68 to -0.83; p < .001). PROMIS Depression and SRS Mental Health (MH) domains demonstrated strong-moderate to strong correlation (ρ, range -0.67 to -0.80; p < .001). Ceiling and floor effects were all less in PROMIS domains (range, 0.44% to 0.88%) compared with SRS domains (range, 0.88% to 17.62%).

Conclusions: PROMIS Physical Function/Mobility, Pain Interference, and Depression domains correlate well with SRS F/A, Pain, and MH. SRS SI/A and Satisfaction are not as well captured. PROMIS showed better ceiling and floor effects than SRS.

Level of evidence: Level III.

Keywords: PROMIS; Patient-reported outcomes; SRS questionnaire; Spinal deformity; Value-based healthcare.