We will be performing site maintenance on AAOS.org on February 8th from 7:00 PM – 9:00 PM CST which may cause sitewide downtime. We apologize for the inconvenience.


Published 8/31/2021
Ariel DeMaio

Do THA Patients Have Better Outcomes at Public or Private Hospitals?

As the number of total hip arthroplasties (THAs) performed in the United States each year continues to grow, federal agencies have implemented new payment models to lower costs and improve the quality of patient care associated with these procedures. According to a retrospective analysis, these models are affecting public and private hospital systems differently.

“This study found each hospital system occupying a specific sector: government hospitals catering to the underserved, for-profits maximizing cost efficiency, and nonprofits associating with the lowest complication rates,” lead author Wayne Wilkie, DO, MHSA, told AAOS Now Daily Edition.

Dr. Wilkie, a clinical research fellow in the Clinical Research Division at the Rubin Institute for Advanced Orthopedics in Baltimore, will present the data today at 11:10 a.m. in Ballroom 20A.

For the analysis, the researchers collected data on patients who underwent THA between Jan. 1 and Sept. 30 in 2016 and 2017 and were included in the National Readmission Database (NRD). Selecting those two time periods allowed for 90-day follow-up, because “readmissions in the NRD are linked only within one calendar year,” Dr. Wilkie explained.

A total of 303,633 patients were identified, then stratified into three groups according to the type of institution where they underwent THA:

  • government non-federal (n = 24,606)
  • private for-profit (n = 39,346)
  • private nonprofit (n = 239,681)

The researchers also collected information on demographics, including age, sex, adjusted Charlson comorbidity index, BMI, primary payer, and median household income. Hospital characteristics analyzed included bed size, location, and teaching status.

Of the for-profit systems identified, nearly half (49.1 percent) were located in urban areas and were nonteaching organizations. Comparatively, 21.8 percent of government institutions and 23.4 percent of nonprofit institutions could be considered urban, nonteaching organizations.

The authors noted that patient demographics were largely equivalent among the different kinds of institutional systems.

Median ages were 65, 66, and 67 years at government, for-profit, and nonprofit institutions, respectively. Medicare was the most common primary insurance type, at approximately 56 percent across all cohorts, followed by private insurance, at approximately 36 percent.

However, regarding patient income, private nonprofit organizations treated most patients within the two highest income quartiles.

Discharge disposition was also similar among patients treated at the different types of institutions. Most patients were discharged with home health care (48.2 percent for government institutions versus 46.7 percent for for-profit institutions versus 44.9 percent for nonprofit institutions [P <0.001]). approximately one-third of patients in all groups had routine discharge.>

However, costs of THA differed significantly among institutions: $21,675 for government institutions versus $16,887 for for-profit institutions versus $19,148 for nonprofit institutions (P <0.001).>

Length of stay was longest in government institutions (3.19 days) compared to the for-profit (3.04 days) and nonprofit (2.60 days) groups.

Regarding complications, incidence of the following events was significantly different among the government, for-profit, and nonprofit cohorts:

  • cerebrovascular accidents: 0.5 percent versus 0.7 percent versus 0.7 percent (P <0.001)>
  • myocardial infarctions: 0.4 percent versus 0.3 percent versus 0.3 percent (P <0.001)>
  • pneumonia: 0.2 percent versus 0.1 percent versus 0.1 percent (P = 0.001)
  • respiratory failures: 0.9 percent versus 0.8 percent versus 0.7 percent
    (P <0.001)>
  • sepsis: 1.4 percent versus 1.4 percent versus 1.2 percent (P <0.001)>
  • transfusions: 5.1 percent versus
    3.9 percent versus 4.0 percent
    (P <0.001)>
  • urinary tract infections: 1.9 percent versus 2.1 percent versus 1.6 percent (P <0.001)>
  • periprosthetic joint infections: 1.0 percent versus 0.8 percent versus 0.7 percent (P <0.001)>
  • dislocations: 0.6 percent versus 0.8 percent versus 0.7 percent (P = 0.025)

These statistical differences were reflected in the multinomial and multiple regressions, the authors reported. Patients treated at private institutions had lower odds ratios of transfusion
(P <0.006) and discharge to home health care (p><0.003) but higher odds ratios of cerebrovascular accidents (>P <0.03). the analysis did not account for surgical case complexity among the different kinds of institutions, which may have affected complication rates.>

“The growth of THA toward half a million cases annually, alterations in payment models, and access to health care all have significant implications for the future of orthopaedics,” the authors concluded. “More research should be performed to extend and analyze any trends that exist in this data over large periods of time.”

Dr. Wilkie’s coauthors of “Do Total Hip Arthroplasty Patients Fare Better Under Publicly or Privately Owned Hospitals? A Comparison of 90-Day Outcomes” are Scott Douglas, MD; Ethan Remily, DO; Nequesha Mohamed, MD; Sahir Pervaiz, MD, MS; Oliver Sax, DO, MS; James Nace, DO; and Ronald Emilio Delanois, MD, FAAOS.

Ariel DeMaio is the managing editor of AAOS Now. She can be reached at ademaio@aaos.org.