PROMIS cluster analysis predicts two-year outcomes in knee surgery patients

Justin E Kung; Samir Kaveeshwar; Michael Rocca; Daniel Rivkin; Jacob Hartline; Jonathan D Packer; Sean J Meredith; Natalie L Leong; R Frank Henn 3rd

doi:10.1016/j.jor.2023.04.001

PROMIS cluster analysis predicts two-year outcomes in knee surgery patients

J Orthop. 2023 Apr 11:39:59-65. doi: 10.1016/j.jor.2023.04.001. eCollection 2023 May.

Authors

Justin E Kung^{1

2}, Samir Kaveeshwar¹, Michael Rocca¹, Daniel Rivkin¹, Jacob Hartline¹, Jonathan D Packer¹, Sean J Meredith¹, Natalie L Leong¹, R Frank Henn 3rd¹

Affiliations

¹ Department of Orthopaedics, University of Maryland School of Medicine, Baltimore, MD, USA.
² Orthopedic Surgery, Prisma Health Midlands/University of South Carolina, Columbia, SC, USA.

Abstract

Aims & objectives: Meaningful clinical interpretation of orthopaedic patient-reported outcome scores remains challenging. Grouping scores may be more meaningful than individual score analysis. The purpose of this study was to determine if grouping knee surgery patients into four preoperative Patient-Reported Outcomes Measurement Information System (PROMIS) clusters would have prognostic value for two-year postoperative outcomes.

Materials & methods: 488 of 697 (70%) patients undergoing elective knee surgery at a single urban institution were enrolled in an orthopaedic registry and completed two-year follow up. Patients were administered questionnaires for PROMIS, International Knee Documentation Committee Score (IKDC), Marx Activity Rating Scale (MARS), and Surgical Satisfaction (SSQ-8). A k-means cluster analysis was performed to identify preoperative PROMIS clusters. Chi-square or Kruskal-Wallis tests were conducted for bivariate analyses. Least-squares multiple linear regression models were performed to identify if cluster group was an independent predictor.

Results: Cluster analysis revealed four clusters of patients. Psychological distress was most significant in determining classification. More impaired clusters were associated with higher rates of arthroplasty, African American race, preoperative opioid use, lower income, higher comorbidity index, and other sociodemographic and operative factors. Worse preoperative cluster status was associated with higher chance of achieving minimally clinically important change (MCID) on all metrics except PROMIS Pain Interference (PI), IKDC, and MARS. Multivariable analysis confirmed better preoperative cluster as predictive of better PROMIS Physical Function (PF), PI, IKDC scores, and satisfaction. Worse preoperative cluster was predictive of greater improvement on PF and PI but not IKDC.

Conclusion: Preoperative PROMIS clusters have prognostic value in predicting outcomes for knee surgery patients. Better preoperative cluster function predicts superior outcomes. While worse preoperative cluster predicts worse outcome, all clusters still significantly improve, so worse preoperative cluster is not a contraindication to surgery.

Keywords: Cluster; Knee; MCID; PROMIS; Patient-reported outcomes; Value-based healthcare.