07.05 Association of Greater Social Vulnerability with Increased 30-Day Readmission After Cardiac Surgery

S. Hinks1, S. Sakowitz1, E. Aguayo1, T. Coaston1, S. Mallick1, S.S. Ali1, P. Benharash1,2  2David Geffen School Of Medicine, University Of California At Los Angeles, Center For Advanced Surgical & Interventional Technology, Los Angeles, CA, USA 1David Geffen School Of Medicine, University Of California At Los Angeles, Cardiovascular Outcomes Research Laboratories (CORELAB), Los Angeles, CA, USA

Introduction:
Prior work has considered the impact of patient and hospital factors on unplanned readmission following cardiac surgery. Yet the impact of neighborhood-level socioeconomic vulnerability remains to be elucidated. Using a multi-institutional database from a state-wide collaborative, we analyzed the association of greater social vulnerability with 30-day readmission.

Methods:
The 2018-2023 University of California Consortium Society of Thoracic Surgeons Adult Cardiac Database was queried to identify all adult patients (≥18 years) undergoing elective or emergent coronary artery bypass grafting and/or valve procedures. The Haversine formula was applied to compute the distance between patient residence and operating hospital. Structural vulnerability was assessed using the well-validated social vulnerability index (SVI). Briefly, this index evaluates neighborhood-level vulnerability across four domains (socioeconomic status, housing and transportation, race and ethnicity, and household composition), and ranks communities from 0 (least vulnerable) to 1 (most vulnerable). Following prior methodology, SVI was stratified into quartiles, with those of Least Vulnerable (LV) and Most Vulnerable (MV) compared in a pairwise approach. Kaplan Meier time-to-event analyses and Cox proportional hazard models were utilized to assess non-elective readmission within 30-days.

Results:
Among 1,731 patients, 629 (36%) were Most Vulnerable and 1,102 (64%) Least Vulnerable. On average when compared to LV, the MV cohort was similar in age (63 [53-70] vs. 63 [53-70] years, P=0.66), more often of female sex (32.7 vs 27.4%, P=0.02), and less commonly of white race (56.9 vs. 62.0%, P<0.001). In addition, such patients had a higher direct cost (111,475 vs. $91,685, P=0.03), preprocedural length of stay (LOS) (4.9 vs. 4.2 days, P=0.01), and lower mean distance to the index hospital (179.0 vs. 268.9 kilometers, P<0.001). After multivariable adjustment, MV was significantly associated with an increment in index hospitalization costs (β= + $9,643, 95% CI [$866 – $18,421]) and increased odds of 30-day readmission (adjusted odds ratio 1.70, 95% CI [1.03-2.81]) (Figure).

Conclusion:
Compared to patients with a low social vulnerability index, patients in the highest SVI quartile experienced a higher rate of index hospitalization costs and 30-day readmissions. The incorporation of SVI into risk stratification may improve perioperative outcomes.