Assessing Correlations and Demographic Interactions Between PROMIS25 and SRS30 in Pediatric Patients With Spinal Deformity.

INTRODUCTION

Historically, patient-reported outcome measures (PROMs) have been measured using anatomy-specific instruments. However, these instruments may be influenced by the presence of concomitant illnesses not associated with the orthopaedic condition. As such, standardized PROM tool, such as PROMIS25, have been developed to assess patient outcomes. In this study, we aim to compare the correlation and association between 2 common PROMs used for pediatric patients with spinal deformity.

METHODS

This study included patients younger than 21 years who were indicated for spinal deformity surgery at our institution. All patients were invited to complete SRS-30 and PROMIS-25 instruments through an automated, electronic patient-reported outcome platform before surgery. Patient demographics were recorded and Spearman correlations were calculated between the various PROMIS and SRS domains. Correlations were compared between different baseline characteristics and demographics using a rho-to-z transformation method. Ceiling and floor effects were also investigated.

RESULTS

A total of 207 patients were enrolled in this study. The majority of PROMIS and SRS domains had moderate to strong correlations, with higher correlations observed when each instrument tried to capture the same construct. For example, there was strong correlation between PROMIS raw pain and SRS pain (ρ=-0.86, P <0.01), and PROMIS Depression and SRS Mental Health (ρ=-0.76, P <0.01). Ceiling effects ranged from 1.0 to 16.9% in SRS and 0.5 to 28.5% is PROMIS. Floor effects ranged from 0.5 to 1.5% in SRS and 0.5 to 29.0% in PROMIS. There were significant differences in the correlation between SRS Mental Health and PROMIS Depression when comparing between different sexes (male: ρ=-0.74, female: ρ=-0.59, P =0.04). In addition, there were significant differences in the correlation between SRS Mental Health and PROMIS Anxiety when comparing between different insurance payors (commercial: ρ=-0.86, female: ρ=-0.75, P =0.03).

CONCLUSIONS

There is strong correlation between SRS and PROMIS domains in pediatric patients with spinal deformity. These correlations can differ by baseline characteristics and demographics. Providers should consider these differences when interpreting PROMs in diverse patient populations.

LEVEL OF EVIDENCE: Level III; Retrospective comparative study.