Abstract
Introduction
Knee arthroplasty is one of the most common inpatient surgeries procedures performed every year however complications do occur. Prior studies have examined the impact of insurance status on complications after TJA in small or focused cohorts. The purpose of our study was to utilize a large all-payer inpatient healthcare database to evaluate the effect of patient insurance status on complications following knee arthroplasty.
Methods
Data was obtained from the Nationwide Inpatient Sample between 2004 and 2011. Analysis included patients undergoing knee arthroplasty procedures determined by ICD-9 procedure codes. Patient demographics and comorbidities were analyzed and stratified by insurance type. The primary outcome was medical complications, surgical complications and mortality during the same hospitalization. A secondary analysis was performed using a matched cohort comparing patients with Medicare vs private insurance using the coarsened exact matching algorithm. Pearson's chi-squared test and multivariate regression were performed.
Results
Overall, 1,352,505 (57.8% Medicare, 35.6% private insurance, 2.6% Medicaid or uninsured, 3.3% Other) patients fulfilled criteria for inclusion into the study. Most were primary total knee arthroplasties (96.1%) with 3.9% revision knee arthroplasties. Multivariate regression analysis showed that patients with private insurance had fewer complications (OR 0.82, p=<0.001) compared to Medicare patients. Similar trends were found for surgical complications and mortality. Patients with Medicare or no insurance had more surgical complications but equivalent rates of medical complications and mortality. The matched cohort showed Medicare and private insurance patients had overall low mortality rates and complication. The most common complication was postoperative anemia, occurring in 16.2% of Medicare patients and 15.3% of patients with private insurance (RR=1.06, p<0.001). Mortality (RR 1.34), wound dehiscence (RR 1.32), CNS, GI complications, although rare, were all statistically more common in Medicare patients (p<0.05) while cardiac complications (RR 0.93, p=0.003) was more common in patients with private insurance.
Discussion and Conclusion
This data reveals that patients with Medicare insurance have higher risk of medical complications, surgical complications and mortality following knee arthroplasty. Using a matched cohort to directly compare Medicare and private insurance patients, the risk of postoperative complications were low overall (with the exception of postoperative anemia), but in general were more common in Medicare patients.
