AUTHOR: Hannah Xu, MD
Retrospective analyses of total hip replacement (THR) surgeries in California, Florida and New York indicate that insurance status is associated with perioperative risks.
According to the findings, Medicaid and other non-private insurance patients had higher unadjusted rates and risk-adjusted odds of in-hospital mortality, postoperative complications, and 30- and 90-day readmission rates than patients with private insurance.
“We were surprised by the extent that health care outcomes were statistically worse for patients with poorer insurance,” said Hannah Xu, MD, an anesthesia resident at NewYork-Presbyterian Hospital/Weill Cornell Medicine, in New York City. “With ongoing discussion of health care reform, this analysis highlights larger socioeconomic and health system–related issues that still need to be addressed to improve surgical outcomes for all patients.”
As Dr. Xu reported at the 2017 spring meeting of the American Society of Regional Anesthesia and Pain Medicine (abstract 3346), research has shown that U.S. citizens who lack access to private health insurance experience disproportionate rates of morbidity and mortality following major surgical operations (J Pediatr Surg 2013;48:81-87; Ann Surg 2010;252:544-550). This study received the Best of Meeting Award.
To update and add to the existing literature, Dr. Xu and her colleague, Robert S. White, MD, examined how insurance status affects health care outcomes of THR, one of the most commonly performed procedures in the United States.
Using the Healthcare Cost and Utilization Project’s State Inpatient Databases of California, Florida and New York, as well as the database from the Agency for Healthcare Research and Quality, the investigators conducted a retrospective study of adults who underwent THR surgeries between 2007 and 2011. They divided primary payor status of the patients into five separate groups: Medicaid, Medicare, private insurance, uninsured and other (i.e., workers compensation, CHAMPUS, CHAMPVA, Title V and auxiliary government programs).
Primary outcomes were inpatient mortality, postoperative complications, and 30- and 90-day readmission rates. Postoperative complications included pulmonary, wound, infectious, urinary, gastrointestinal, cardiovascular, systemic and intraoperative/procedural. A team of statisticians then ran multivariate logistic regression models to test for associations between primary payor status and primary outcomes, adjusting for patient demographics, comorbidities, elective operation status, year of surgery and state of surgery.
Readmission Rates Affected, Too
As Dr. Xu reported, a total of 295,572 patients 18 years of age or older underwent THR from 2007 to 2011, in California, Florida and New York.
Compared with patients who had private insurance, patients having Medicaid and other non-private insurance showed significantly worse postoperative outcomes. Medicaid patients had higher unadjusted rates of:
in-hospital mortality (0.2% vs. 0.0%),
postoperative complications (6.1% vs. 4.1%),
30-day readmission rate (7.7% vs. 3.5%), and
90-day readmission rate (14.5% vs. 7.3%).
After controlling for potentially confounding patient and non-patient variables, poorer outcomes were still associated with underinsured status. Logistic regression models showed that patients with Medicaid were 2.39 times more likely to die in the hospital following THR than those with private insurance. Medicaid patients showed an increased risk for postoperative complications, 30-day readmission and 90-day readmission by a factor of 1.32, 1.71 and 1.63, respectively, compared with patients who had private insurance.
According to Dr. Xu, previous studies have demonstrated a strong association between insurance status and race when investigating health care disparities (Barnett JC. Health Insurance Coverage in the United States: 2015. U.S. Census Bureau Report No. P60-257). Retrospective analysis of claims data, for example, has shown that neuraxial anesthesia was only used in 17% of blacks undergoing total joint replacement compared with 25% of whites (Br J Anaesth 2015;115 Suppl 2:ii57-ii67), and studies have indicated that providers may fail to recognize disparities in their own specialty and practices (J Racial Ethn Health Disparities 2016 Aug 29. [Epub ahead of print]).
“Race plays a part in this discussion, but since we can’t change our race, we can look to other factors to change for health care equality, such as availability of neuraxial versus general anesthesia for joint arthroplasty and optimization of postoperative pain management for minorities,” she said. “Also, educating physicians about implicit racial bias has been shown to change behavior.”
Drs. Xu and White plan to follow up this study by examining the same patient population’s relationship between insurance status, hospital length of stay and total charges. “We’d like to determine hospital resource utilization based on insurance status on this surgical population,” Dr. Xu said.
“We may not be able to change things overnight, but this study provides insight into ways that we can improve, intraoperatively and postoperatively, and that’s encouraging,” she concluded.