Back to 2018 Posters

---

DISPARITIES IN CARE: IMPACT OF SOCIOECONOMIC FACTORS AND FACILITY TYPE ON PANCREATIC SURGERY IN THE UNITED STATES: EXPLORING THE NATIONAL CANCER DATABASE
Michael Makar^*, Joseph Blansfield, Kasondra Hartman, Tania Arora, James Dove, Marie Hunsinger, Jacqueline Oxenberg, Ericha L. Worple
Surgery, Geisinger, Rutherford, NJ
Background: Pancreaticoduodenectomy (PD) remains the cornerstone of treatment in patients with pancreatic cancer. Prior studies have shown that patients undergoing surgery at high-volume institutions have decreased mortality and increased survival. However, not all patients are able to have surgery at high volume centers and there may be socioeconomic factors which influence where a patient is treated. The goal of this study is to examine socioeconomic factors that determine where a patient is treated and how that location affects outcome.

Methods: Socioeconomic factors and location of pancreatectomy were examined using the National Cancer Database (NCDB) from 2004-2014. Locations of resection were grouped according to NCDB categories. Community programs (CP) included Community Cancer and Comprehensive Community Cancer Programs while Academic Programs (AP) included Academic Research and Integrated Network Cancer Programs.

Results:
A total of 17,632 patients were included in this study; 13,344 patients (76%) underwent PD at APs while 4,288 patients (24%) underwent PD at CPs. Seventy-seven percent of APs were high volume (15 or more pancreatectomies per year) versus 26% of CPs. Ninety percent of patients receiving preoperative radiation had surgery at APs. Eighty-eight percent of patients receiving preoperative chemotherapy had surgery at an AP. Patients treated at APs had to travel a mean distance of 80.9 miles while patients treated at CP had to travel a mean distance of 31.7 miles (p<.0001).

There were several socioeconomic factors that affected whether a patient had surgery at an AP. Spanish and Hispanic patients were less likely to travel to an AP (69% had surgery at an AP versus 76% of non-Hispanic patients, p<0.001). This difference remained on multivariate (MV) analysis (OR 1.42, 95% CI: 1.21-1.67). Patients with higher comorbidities were also more likely to have care at CPs. Seventy-six percent of patients with a Charlson Deyo score 0 had surgery at an AP versus only 73% of those with a score of 2 and 68% of those with a score of 3. This difference remained on MV analysis.

Surgery at an AP was shown to improve outcomes for this patient population. After propensity matching, patients with surgery at AP had fewer positive margins (22.6% vs 25.8%) and a shorter length of stay (8 vs 10 days). In addition, AP had lower 30 day (2.8% vs 5.4%) and 90 day mortalities (6% vs 9.8%). In addition, patients who had surgery at an AP had better overall survival.

Conclusions:
Patients who had pancreatic cancer surgery at CP were more likely to be Hispanic or with higher medical comorbidities. Those who had surgery at AP traveled further distances but had better peri-operative outcomes and had an improvement in overall survival. Further studies are needed to explore the decision making for this patient population.

---

Back to 2018 Posters