Extensive research documents disparities in health outcomes for vulnerable populations. Safety-net hospitals-those that serve a greater proportion of vulnerable patients with Medicaid or no insurance-may yield better outcomes for these vulnerable patients because of their expertise with this population. National Inpatient Sample data from 2005-2007 show that predicted rates of complications following colorectal cancer surgery are approximately 20% lower for vulnerable patients in safety-net than in non-safety-net hospitals (0.273 versus 0.340; 95% CI for the difference: -0.11, -0.001). Differences by safety-net status for Medicare and privately-insured patients were smaller and not statistically significant. The lower complication rates are not due to earlier discharge, as vulnerable patients had significantly longer stays at safety-net hospitals (1.79 days; 95% CI 0.13, 3.40). Additional research is needed to determine whether improvement in post-operative outcomes is due to the expertise of safety-net hospitals in caring for patients with Medicaid or no insurance.