Patient Outcomes Following Total Joint Replacement Surgery: A Comparison of Hospitals and Ambulatory Surgery Centers

Background

CMS recently announced that it would begin paying for total knee replacement surgeries (arthroplasty; TKA) performed in outpatient settings in January 2018, and that it is considering reimbursing for outpatient total hip arthroplasty (THA) in the future. This policy change has stirred standing concerns over delivery of care in physician-owned specialty facilities, including ambulatory surgery centers (ASCs). ASCs are self-described as a way to reduce Medicare and other costs and it is possible that comparable outcomes following TKA and THA can be achieved in ASCs. Currently, total joint replacements are being performed on commercially insured patients in over 200 ambulatory surgery (ASCs), up from approximately 25 ASCs in 2014. However, despite claims by physician-owners of ASCs that outpatient total joint replacement is appropriate and safe for a large portion of candidates, evidence to date is anecdotal and lacks a rigorous scientific base.

Research Questions

We are comparing health outcomes and costs following TKA and THA procedures performed on patients treated in ASCs with patients treated as inpatients in order to 1) better understand whether claims regarding the patient safety of TKA and THA performed in ASCs are justified 2) measure relative costs across settings 3) identify priority areas for TKA and THA patient safety interventions, and 4) add to the evidence base regarding the delivery of services in physician-owned specialty facilities.

Methods

We are conducting quantitative data analyses using the Truven Health Analytics MarketScan^® Commercial Claims and Encounters database. MarketScan, which contains data on approximately 50 million covered lives per year, is one of the few large commercial insurance claims databases available to researchers. Data include detailed information on place of service, procedures, diagnoses, as well as actual payments to providers, and allows tracking unique patients over time.

To date we have identified over 5,000 patients aged 55-64 who underwent TKA or THA in an ASC during the period 2014-2016 and who were continuously enrolled for 90 days following surgery. We are selecting a sample of TKA/THA inpatients numbering three times that of ASC patients. The matched sample is being drawn using propensity score matching, adjusting risk by incorporating scores generated from the HHS Hierarchical Condition Categories risk adjustment model which was designed under the ACA for application to the commercially-insured population. Risk-adjustment also accounts for AHRQ chronic condition and comorbidity indicators. Outcomes are cases of related 30-day and 90-day readmission, and cases within 90-days of revision surgery, surgical site infection, deep-vein thrombosis, pulmonary embolism, and dislocation/loosening/breakage of prosthesis. We are testing differences across sites using chi-square and other statistical techniques. Finally, we are measuring the relative cost of episodes of care including related expenditures incurred during the 90-day post-inpatient stay or ASC encounter.

Relevance

This project is informative to CMS on potential health consequences of removing total joint arthroplasty from the inpatient only list. For providers, it identifies areas of potential patient safety lapses. Finally, it facilitates patient decision making for “gray area” candidates for outpatient joint replacement.