Factors associated with readmission to index vs. non-index hospitals after major cancer surgery: Does centralization play a role?

Abstract

11073 Background: While centralization of complex cancer surgery at regional referral centers improves perioperative outcomes, many vulnerable patients face barriers in accessing these hospitals. When these patients do manage to undergo surgery at referral centers, it remains unclear where they are readmitted to receive postoperative care when unplanned complications arise. Patients with cancer surgery who are readmitted to the hospital where the surgery was performed (index readmission) often have improved outcomes compared with those readmitted to a different hospital (non-index readmission). This study examined whether factors associated with readmission to index versus non-index hospitals differ for patients undergoing surgery at referral versus non-referral centers. Methods: We used data from the Pennsylvania Cancer Registry linked to all-payer statewide inpatient discharge records to identify patients who had surgery for bladder, brain, esophageal, liver, lung and pancreatic cancers between 2013-2019 and were subsequently readmitted within 90 days. We fit a multivariable logistic regression model to identify factors associated with 90-day readmission to an index versus non-index hospital. We included an interaction term between referral center status and cancer type in this model. We defined referral centers as National Cancer Institute-designated cancer centers or American College of Surgeons Commission on Cancer-accredited academic comprehensive cancer programs. Results: Of the 28,951 patients with major cancer surgery, 28% (N=8215) were readmitted within 90 days of cancer surgery. Of all patients readmitted, 57% (N=4671) were originally treated at referral centers and 78% (N=6388) were readmitted to the index hospital. On multivariable analysis, factors associated with lower odds of index versus non-index readmission included older age (≥70 years: odds ratio (OR)=0.61; 95% confidence interval (CI), 0.49-0.77, relative to <55 years), high Elixhauser comorbidity scores (>16: OR=0.74; 95% CI, 0.63-0.88; relative to <8), longer travel times (>60 minutes: OR=0.12; 95% CI, 0.10-0.15; relative to <15 minutes), and Medicaid insurance (OR=0.68; 95% CI, 0.54-0.86; relative to commercial insurance). There was no significant difference in odds of index readmission when patients were treated at referral versus non-referral centers (OR=0.77; 95% CI, 0.50-1.20). When assessing interactions, patients with lung cancer had lower odds of index readmission when treated at referral versus non-referral centers, relative to other cancers (OR=0.59; 95% CI, 0.40-0.89). Conclusions: Ongoing centralization may be significantly increasing care fragmentation for patients with lung cancer surgery. Future interventions to improve care coordination after surgery should target patients with higher clinical complexity and greater travel burdens.