2130 Can Clustering of Sociodemographic Components Predict 30-Day Complications and Resource Utilization for Adult Patients Undergoing Lumbar Spinal Fusion Surgery?

View at publisher Summarise with VIRAT
Authors: Chiemela Izima, Eunice Yang, Travis Stanley CreveCoeur, Nathan A. Shlobin, Dean Chou, Andrew T. Chan
Cited by: 0

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

INTRODUCTION: Social determinants of health (SDoH) influence postoperative outcomes in adult lumbar fusion surgery, however, specific profiles incorporating multiple SDoH are limited. METHODS: A tertiary academic center’s database was retrospectively reviewed. K-modes clustering grouped patients by race/ethnicity, age, marital status, gender, and zip code–linked socioeconomic status using two validated indices (SVI, DCI). Baseline/intraoperative characteristics, 30-day complications, 30-/90-day readmissions, and 2-year reoperations were compared between clusters. RESULTS: Two clusters resulted from 4,678 patients, cluster 1 represented 77.6% of patients. Cluster 1 patients were predominantly female, white, married, and lived in economically “prosperous” zip codes. Inversely, cluster 2 patients were younger (50.4 ± 17.1 vs 56.9 ± 17, p < 0.001), male, non-white, single, and lived in economically “at-risk” areas. Cluster 2 exhibited less osteoporosis (1.1% vs 2.9%, p = 0.001), lumbar stenosis (22.4% vs 26.3%, p = 0.012), scoliosis (11.3% vs 14.3%, p = 0.001), and spondylolisthesis (18.4% vs 22.5%, p = 0.005). No differences existed for smoking history, diabetes, cardiovascular, or chronic kidney disease. Cluster 2 underwent more posterior-only approaches (82.3% vs 79.1%, p = 0.026), with no difference in levels fused. Multivariate analysis controlling for cluster-based comorbidities and perioperative parameters demonstrated cluster 2 identity was independently associated with increased odds (OR = 1.93, 95% CI [1.30, 2.85], p = 0.001) of AKI within 30 days. No difference remained in 30-/90-day readmission or 2-year reoperation. CONCLUSIONS: SDoH clustering of multiple sociodemographic domains identified two clusters with different baseline and surgical characteristics among adult lumbar fusion patients. Cluster 1 patients were often female, white, married, lived in more economically prosperous areas, and had more spinal pathology, in contrast to Cluster 2 patients who were younger, male, non-white, single, and overall more socioeconomically disadvantaged. SDoH profiles can identify at-risk patient phenotypes more likely to experience specific 30-day complications like AKI and contextualize resource utilization across groups.

Fields of Study & Subjects

Spine and Intervertebral Disc Pathology Health disparities and outcomes Musculoskeletal pain and rehabilitation

Publication details

Journal / source
Neurosurgery
Volume
72
Issue
Supplement_1
Pages
280–281
Publication date
26 March 2026
Language
EN