Back to Annual Meeting Program
Impact of Insurance on Travel Distance and Utilization of Low Volume Hospitals for Pancreatic Cancer Surgery in Massachusetts.
*Zeling Chau1, *Jillian K. Smith1, *Theodore P. McDade1, *Elan R. Witkowski1, *Elizaveta Ragulin-Coyne1, *Sing Chau Ng1, *James K. West2, *Heena Santry1, *Frederick A. Anderson, Jr.1, Jennifer F. Tseng3,1 1University of Massachusetts Medical School, Surgical Outcomes Analysis & Research, Worcester, MA;2MA Department of Public Health, Boston, MA;3Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, MA
Objective: To determine the impact of insurance on travel distance and utilization of low-volume hospitals (LVH) for pancreatic cancer resection. Design: The Massachusetts Division of Health Care Finance and Policy Hospital Inpatient Discharge Database was queried to identify all pancreatic resections performed for cancer during 2005-2009. Travel distance was calculated using zip codes and geodetic methods with SAS. LVHs were defined using Leapfrog criteria(< 11 procedures/year). Setting: Inpatient hospitalizations in Massachusetts. Patients: All pancreatic resections for cancer. Interventions:N/A Main Outcome Measure(s): Travel distance between patient residence and resection hospital; receipt of care at LVH. Results: From 2005-2009 there was a decrease in patients resected at LVHs, however in 2009 18.9% of patients were still resected at a LVH. High-volume hospitals had improved outcomes when compared to LVHs (mortality 1.2% vs. 4.5% (p=0.005), complications 15.1% vs. 27.1% (p=0.0004), median LOS 8 vs.10 days. Insurance-type was associated with care at LVH (32.4% Medicaid, 21.5% Medicare, 14.1% private (p=0.0081)). Median distance to resection was 16.7 miles. Private patients traveled 19.2 miles, Medicare 14.2, Medicaid 13.6. Shorter distance was associated with increased complications (20.1% vs. 14.5%, p=0.0319). Resection at LVH and comorbidities were independent predictors of both complications and mortality. Independent predictors of resection at LVH included low income (OR=2.15, 95%CI=1.16-4.01) and Medicaid (OR=2.42, 95%CI=1.07-5.44). Predictors of shorter travel included: black race (OR=3.26, 95%CI=1.16-9.17) and resection at LVH (OR=3.13 95%CI=2.10-4.66). Conclusions: A significant population still receives care at LVHs despite inferior outcomes. Insurance type is a major predictor of receipt of care at LVHs. In the current era of healthcare reform and increasing scrutiny of surgical outcomes, further investigation of the root causes of these insurance-based disparities is warranted.
Back to Annual Meeting Program
|
|