Orthopedics

Feature Article 

Insurance Status Affects In-Hospital Complication Rates After Total Knee Arthroplasty

David R. Veltre, MD; Paul H. Yi, MD; David C. Sing, MD; Emily J. Curry, BA; Atsushi Endo, MD, MPH; Eric L. Smith, MD; Xinning Li, MD

Abstract

Insurance status has been shown to be a predictor of patient morbidity and mortality. The purpose of this study was to evaluate the effect of patient insurance status on the in-hospital complication rates following total knee arthroplasty. Data were obtained from the Nationwide Inpatient Sample (2004 through 2011). Patient demographics and comorbidities were analyzed and stratified by insurance type. Analysis was performed with a matched cohort comparing complication rates between patients with Medicare vs private insurance using the coarsened exact matching algorithm and multivariable logistic regression. A total of 1,352,505 patients (Medicare, 57.8%; private insurance, 35.6%; Medicaid/uninsured, 3.1%; other, 3.3%; unknown, 0.2%) fulfilled the inclusion criteria. The matched cohort analysis comparing Medicare with privately insured patients showed significantly higher risk of mortality (relative risk [RR], 1.34; P<.001), wound dehiscence (RR, 1.32; P<.001), central nervous system complications (RR, 1.16; P=.030), and gastrointestinal complications (RR, 1.13; P<.001) in Medicare patients, whereas privately insured patients had a higher risk of cardiac complications (RR, 0.93; P=.003). Independent of insurance status, older patients and patients with an increased comorbidity index were also associated with a higher complication rate and mortality following total knee arthroplasty. These data suggest that insurance status may be considered as an independent risk factor for increased complications when stratifying patients preoperatively for total knee replacement. Further research is needed to investigate the disparities in these findings to optimize patient outcomes following total knee arthroplasty. [Orthopedics. 2018; 41(3):e340–e347.]

Abstract

Insurance status has been shown to be a predictor of patient morbidity and mortality. The purpose of this study was to evaluate the effect of patient insurance status on the in-hospital complication rates following total knee arthroplasty. Data were obtained from the Nationwide Inpatient Sample (2004 through 2011). Patient demographics and comorbidities were analyzed and stratified by insurance type. Analysis was performed with a matched cohort comparing complication rates between patients with Medicare vs private insurance using the coarsened exact matching algorithm and multivariable logistic regression. A total of 1,352,505 patients (Medicare, 57.8%; private insurance, 35.6%; Medicaid/uninsured, 3.1%; other, 3.3%; unknown, 0.2%) fulfilled the inclusion criteria. The matched cohort analysis comparing Medicare with privately insured patients showed significantly higher risk of mortality (relative risk [RR], 1.34; P<.001), wound dehiscence (RR, 1.32; P<.001), central nervous system complications (RR, 1.16; P=.030), and gastrointestinal complications (RR, 1.13; P<.001) in Medicare patients, whereas privately insured patients had a higher risk of cardiac complications (RR, 0.93; P=.003). Independent of insurance status, older patients and patients with an increased comorbidity index were also associated with a higher complication rate and mortality following total knee arthroplasty. These data suggest that insurance status may be considered as an independent risk factor for increased complications when stratifying patients preoperatively for total knee replacement. Further research is needed to investigate the disparities in these findings to optimize patient outcomes following total knee arthroplasty. [Orthopedics. 2018; 41(3):e340–e347.]

Total knee arthroplasty (TKA) is one of the most commonly performed inpatient procedures in the United States. In 2010, 693,400 total knee replacements were performed in patients older than 45 years, and the demand is projected to rise dramatically during the next decade in response to the aging “baby boomer” population.1,2 Both medical and surgical complications can occur, particularly because patients undergoing these procedures tend to be older and have preexisting medical comorbidities. Due to this increased demand, surgeons and patients must identify and understand risk factors responsible for the development of complications following TKA.

Factors that may predispose a patient to complications after TKA have been well described.3,4 Previous studies have shown that a wide range of conditions and factors, including psychiatric disorders, obesity, advanced age, smoking, and length of hospitalization, are associated with an increased complication rate after TKA.5–10 One variable that has become particularly relevant is that of insurance status. Despite the Affordable Care Act, passed in 2010 and allowing for Medicaid expansion in the United States, insurance status has been implicated in not only patient inaccessibility to both hip and knee arthroplasty treatment but also worse outcomes after total joint arthroplasty.11–16

Although prior studies have examined the impact of insurance status on complications after total joint arthroplasty, they have either involved small cohorts or focused on limited insurance types, such as Medicaid and Medicare populations.14,17 The purpose of this study was to evaluate the effect of patient insurance status (Medicare vs private) and other patient factors on the in-hospital complication rate following TKA using a large all-payer inpatient health care database with a matched cohort. The authors hypothesized that patients with public insurance (Medicare or Medicaid) would have more complications than patients with private insurance.

Materials and Methods

This study was deemed not human subjects research by the authors' institutional review board. This study used the data collected in the Healthcare Cost and Utilization Project Nationwide Inpatient Sample from 2004 through 2011. The Nationwide Inpatient Sample is the largest publicly available all-payer inpatient health care database in the United States, with data from more than 7 million hospital stays each year. Patients were identified using the International Classification of Diseases, 9th Revision, Clinical Modification procedural codes (total knee replacement, 81.54; revision total knee replacement, 81.55).

Patient demographics (sex, age, race/ethnicity, and admission type) and comorbidities (Charlson comorbidity index) were analyzed and stratified based on insurance type (Medicare, private insurance, Medicaid/uninsured, and other).18 “Other” insurance types included workers' compensation, Civilian Health and Medical Program of the Uniformed Services, Civilian Health and Medical Program of the Department of Veterans Affairs, Maternal and Child Health Services Block Grant (Title V of the Social Security Act), and other governmental programs.19 Age was divided into 4 ordinal categories (younger than 40, 40 to 64, 65 to 79, and 80 years and older). Medical complications, surgical complications, and mortality during the same hospitalization period were the primary outcomes. Medical complications included acute cardiac event, pulmonary edema, venous thromboembolic event, cerebrovascular event, acute kidney injury, pneumonia, sepsis, and urinary tract infection. Surgical complications included shock, hematoma, accidental puncture, wound dehiscence, retained foreign body, postoperative injection, and nonhealing surgical wound. Pearson's chi-square test was used for comparison of baseline characteristics on each insurance cohort. Multivariable logistic regression was also used to assess the influence of insurance type on the aforementioned complications.

A matched-cohort comparison of patients with Medicare (public insurance) vs private insurance was performed as a secondary analysis. To control for different baseline demographics, each Medicare patient was matched one-to-one with a repeated sampling of private insurance patient on the basis of age, sex, and other demographics/comorbidities using the coarsened exact matching algorithm. By reducing the imbalance in demographics between groups, matching improved the estimation of covariates in predicting outcomes. The matching algorithm worked by sorting patients based on matching demographics, then discarding patients who did not have any matching demographics with a corresponding Medicare patient. All analysis was performed using R version 3.0.3 software (R Foundation, www.r-project.org).

Results

Demographics

Overall, 1,352,505 patients were included in this study. Demographic information is presented in Table 1. For race/ethnicity, most patients were white, followed by black and then Hispanic. Predictably, more than 87% of the patients with Medicare who had a knee arthroplasty were 65 years or older.

Demographics of the Total Knee Arthroplasty Patients

Table 1:

Demographics of the Total Knee Arthroplasty Patients

The payer mix included 57.8% (781,255) of patients with Medicare, 35.6% (481,054) with private insurance, 3.1% (42,580) with Medicaid or no insurance, and 3.3% (44,725) from other insurance programs. Approximately 0.2% (2891) of the patients had unknown payer status. Most patients with Medicaid and private insurance were younger than 65 years (83.1% and 85%, respectively). In all insurance groups, between 89% and 93% of surgeries were elective in nature. Patients with government-sponsored insurance (Medicare or Medicaid) tended to be less healthy than those who were privately insured, as indicated by a higher comorbidity index (comorbidity index ≥3: Medicare, 3.5%; Medicaid, 3.3%; private, 1.6%).

Most of the knee arthroplasties performed were primary TKAs (1,296,469; 96.1%), with 3.9% (53,145) being revision TKAs. In all insurance groups, primary TKA was the most common procedure, accounting for more than 90% of the surgeries.

Summary of Complications

Among all reported surgical and medical complications, cardiac events were the most common and occurred in 8.0% of the Medicare patients, which is more than twice the rate of patients with private insurance (3.5%; Figure). Urinary tract infections and surgery-related complications were the next two most frequent complications overall, and both were more commonly seen in those patients with government-sponsored insurance (Medicaid/Medicare). Urinary tract infection was seen in 3.2% of Medicare patients and 2.8% of Medicaid/uninsured patients, compared with 1.8% of the privately insured patients. The overall highest complication rates were found in the Medicare population, followed by the Medicaid/uninsured, other insurance, and privately insured patient populations.

Types of medical and surgical complications for total knee arthroplasty.

Figure:

Types of medical and surgical complications for total knee arthroplasty.

Medical Complications

The multivariable logistic regression analysis revealed that white patients had significantly more medical complications than black, Hispanic, or Asian patients (P<.001). Increasing age and comorbidity index were also substantially associated with medical complications. Compared with patients younger than 40 years, patients 40 to 64 years old had 2 times (odds ratio [OR], 2.68; 95% confidence interval [CI], 2.39–3.01; P<.001), patients 65 to 79 years old had 4 times (OR, 4.06; 95% CI, 3.62–4.56; P<.001), and patients 80 years and older had 6 times (OR, 6.25; 95% CI, 5.56–7.03; P<.001) higher odds of having postoperative medical complications. The trend was similar for those with a higher comorbidity index. Patients with a comorbidity index of 3 or greater had 22 times higher odds of having a medical complication postoperatively compared with patients without any preoperative comorbidity. Revision knee arthroplasty also had increased odds of medical complications (OR, 1.19; 95% CI, 1.16–1.23; P<.001) compared with primary total knee replacements.

In terms of differences related to insurance status, privately insured patients were associated with lower risk of overall medical complications compared with Medicare patients (OR, 0.82; 95% CI, 0.81–0.84; P<.001) (Table 2). Patients with Medicaid or no insurance also had fewer complications compared with Medicare patients; however, this difference was not statistically significant (OR, 0.96; 95% CI, 0.93–1.00; P=.060).

Medical Complications, Surgical Complications, and Mortality After Total Knee Arthroplasty on Multivariable Logistic Regression Analysis

Table 2:

Medical Complications, Surgical Complications, and Mortality After Total Knee Arthroplasty on Multivariable Logistic Regression Analysis

Surgical Complications

In contrast to medical complications, black, Hispanic, and Asian patients had significantly more surgical complications than white patients. There was also a significant increase in surgical complications in patients who were younger than 40 years (P<.001) and male patients (P<.001). Revision knee arthroplasty cases also had almost a 2-fold greater odds (OR, 1.87; 95% CI, 1.79–1.96; P<.001) of surgical complications compared with primary TKA surgery cases. Patients with preoperative comorbidities had an increased risk of surgical complications, although this was only significant if the comorbidity index was greater than 2. On comparison of reported surgical complications (Table 2) across insurance types, patients with private insurance had a statistically significant lower risk of surgical complications (OR, 0.93; 95% CI, 0.90–0.96; P<.001), whereas patients who had Medicaid or no insurance had an increased risk (OR, 1.23; 95% CI, 1.15–1.32; P<.001).

Mortality

The overall mortality rate for patients undergoing knee arthroplasty was 0.12% (n=1577). Women (OR, 0.59; 95% CI, 0.53–0.66; P<.001) had a lower mortality risk compared with men. Older patients, patients with a higher comorbidity index, and those undergoing revision TKAs all had a significantly greater risk of mortality (Table 2). Compared with Medicare patients, patients with Medicaid or no insurance had no significant difference in mortality (OR, 0.95; 95% CI, 0.66–1.37; P=.779). Those with private insurance had the lowest mortality risk (OR, 0.67; 95% CI, 0.56–0.80; P<.001) compared with all other insurance types.

Matched-Cohort Analysis of Medicare Versus Private Insurance

Demographics for the matched cohort of Medicare and private insurance patients are listed in Table 3, revealing similar baseline characteristics and comorbidities. Table 4 provides a comparison of the risks of individual complications between the 2 groups when controlling for other risk factors. Patients with Medicare had statistically significant higher risks of central nervous system complications (relative risk [RR], 1.16; 95% CI, 1.01–1.33; P=.030), gastrointestinal complications (RR, 1.13; 95% CI, 1.08–1.19; P<.001), wound dehiscence (RR, 1.32; 95% CI, 1.16–1.49; P<.001), postoperative anemia (RR, 1.06; 95% CI, 1.05–1.07; P<.001), and mortality (RR, 1.34; 95% CI, 1.16–1.54; P<.001). On the contrary, the risk of cardiac complications was lower in Medicare patients (RR, 0.93; 95% CI, 0.89–0.97; P=.003) compared with privately insured patients. The authors found no statistical difference in the risk of other medical or surgical complications.

Baseline Demographics and Comorbidities Between the Matched Medicare and Private Insurance Total Knee Arthroplasty PatientsBaseline Demographics and Comorbidities Between the Matched Medicare and Private Insurance Total Knee Arthroplasty Patients

Table 3:

Baseline Demographics and Comorbidities Between the Matched Medicare and Private Insurance Total Knee Arthroplasty Patients

Comparison of In-Hospital Complications After Total Knee Arthroplasty Between the Matched Medicare and Private Insurance Patients

Table 4:

Comparison of In-Hospital Complications After Total Knee Arthroplasty Between the Matched Medicare and Private Insurance Patients

Discussion

To the authors' knowledge, this study is the largest and most comprehensive assessment of medical and surgical complications after TKA and of specifically how patients' insurance status affects overall complication rates. This multivariable regression analysis showed that increased postoperative complications after TKA were associated with having nonprivate or Medicare insurance status. After both demographics and comorbidities were matched and controlled for between Medicare and private insurance patients, patients with Medicare or public insurance had significantly higher risks of developing central nervous system complications, gastrointestinal complications, wound dehiscence, and postoperative anemia and also had an increased rate of mortality compared with patients with private insurance. There was no difference in both the medical and the surgical complications between publicly insured patients with Medicare vs Medicaid.

Multivariable regression analysis in this study included more than 1 million patients and indicated that patients with Medicare, Medicaid, or no insurance were more likely to have medical complications, increased mortality, and surgical complications compared with those with private insurance. These findings are similar to those of previous studies that found worse clinical outcomes in patients with Medicaid insurance.13–15 A study performed by Browne et al16 using the Nationwide Inpatient Sample database evaluated complications in a matched group of Medicaid vs non-Medicaid patients who underwent primary hip and knee arthroplasty. Consistent with the current study's findings, they found a higher rate of wound dehiscence, infection, and anemia but a lower risk of cardiac complications in the Medicaid patient population.16 Additionally, no statistical differences were found between the various insurance types in all other complications. Similar large database studies in the general surgery trauma literature show a similar increased complication rate among uninsured patients after trauma, suggesting that this association is not exclusive to orthopedic surgery and knee arthroplasty patients.20,21

The current authors also found that social factors such as race/ethnicity contributed to increased risk of developing a surgical complication. Their analysis showed that black and Hispanic patients had an increased risk of surgical complications compared with white patients. Previous literature in the area of joint arthroplasty has primarily focused on the use of total hip and knee arthroplasty, specifically highlighting the low utilization rates of both for black and Hispanic patients.14,22–28 This may be the result of limited access to high-volume private surgeons because of government insurance status or socioeconomic factors. Despite the fact that these patient populations have a greater number of surgical complications while also using arthroplasty procedures less frequently, little is known about why these disparities exist. Perhaps more important, it is not known how to address this health care disparity. Paradoxically, although black and Hispanic patients had an increased risk of surgical complications, the current analysis also showed that white patients had an increased risk of medical complications. This finding was surprising, and this study highlights that further investigation into these unanswered questions of the relationship of race to complications is prudent.

Independent of insurance status, the current analysis also found that both increased age and increased comorbidity index were risk factors for medical complications. Patients 80 years and older had 6 times the complication rate of patients younger than 40 years. This finding is consistent with previous studies that have found a higher complication rate in patients 80 years and older compared with younger patients.7,8,29,30 Furthermore, a comorbidity index of 3 or greater resulted in 22 times the odds of a postoperative medical complication developing, compared with patients with a comorbidity index of 0. Previous studies have also found that high comorbidity index is a risk factor for compromised patient outcomes after TKA; therefore, optimizing patient health status should be an important preoperative consideration.31–33 Also, as expected, compared with primary knee arthroplasty, revision knee arthroplasty will increase the odds of medical complications. The risk factors mentioned above were also associated with increased mortality after TKA.

The authors analyzed a large number of patient records to find statistical trends that might not otherwise be discernable in a smaller retrospective study. This study had limitations. The database is a nationwide sampling of all knee arthroplasties, so the findings can be broadly applicable on a national level. However, due to the nature of a retrospective study, the authors are limited in their ability to analyze causality. The data also only have information on a single hospital discharge. The authors were unable to assess patient outcomes, readmission rates, or delayed complications that occurred after a patient was discharged. Additionally, the authors examined both primary and revision TKAs in these results, which contributed some heterogeneity to their findings. However, the authors did address this by using multivariable regression to adjust for the above variables along with age, sex, and comorbidities. Additionally, the authors performed a matched-cohort analysis of patients with Medicare insurance compared with private insurance status using the coarsen exact matching algorithm to confirm their findings. Although statistical significance can be seen in a large database study such as this, the clinical relevance of small but significant differences is an area of debate and many of these findings should not be evaluated based on statistical significance but rather clinical significance.

Conclusion

These findings showed that, in general, patients with government-sponsored (Medicare and Medicaid) or no insurance were at a higher risk of developing medical complications, including having higher mortality. Additionally, older patients and patients with an increased comorbidity index were also associated with a higher complication rate and mortality following knee arthroplasty. When a matched cohort was used to directly compare Medicare patients with private insurance patients on individual complications, Medicare patients had a significantly higher risk of mortality, complications related to the central nervous system and gastrointestinal system, wound dehiscence, and post-operative anemia. These data suggest that insurance status may be considered as an independent risk factor for increased complications when stratifying patients preoperatively for total knee replacement. Further research is needed to investigate the disparities in these findings to optimize patient outcomes following TKA.

References

  1. Williams SN, Wolford ML, Bercovitz A. NCHS Data Brief. No. 210. Hospitalization for total knee replacement among inpatients aged 45 and over: United States, 2000–2010. http://www.cdc.gov/nchs/products/databriefs/db210.htm. Accessed April 10, 2017.
  2. Kurtz S, Ong K, Lau E, Mowat F, Halpern M. Projections of primary and revision hip and knee arthroplasty in the United States from 2005 to 2030. J Bone Joint Surg Am. 2007; 89(4):780–785.
  3. Bozic KJ, Lau E, Ong K, et al. Risk factors for early revision after primary TKA in Medicare patients. Clin Orthop Relat Res. 2014; 472(1):232–237. doi:10.1007/s11999-013-3045-0 [CrossRef]
  4. Jain NB, Guller U, Pietrobon R, Bond TK, Higgins LD. Comorbidities increase complication rates in patients having arthroplasty. Clin Orthop Relat Res. 2005; 435:232–238. doi:10.1097/01.blo.0000156479.97488.a2 [CrossRef]
  5. Klement MR, Nickel BT, Penrose CT, et al. Psychiatric disorders increase complication rate after primary total knee arthroplasty. Knee. 2016; 23(5):883–886. doi:10.1016/j.knee.2016.05.007 [CrossRef]
  6. Kerkhoffs GM, Servien E, Dunn W, Dahm D, Bramer JA, Haverkamp D. The influence of obesity on the complication rate and outcome of total knee arthroplasty: a meta-analysis and systematic literature review. J Bone Joint Surg Am. 2012; 94(20):1839–1844. doi:10.2106/JBJS.K.00820 [CrossRef]
  7. Kennedy JW, Johnston L, Cochrane L, Boscainos PJ. Total knee arthroplasty in the elderly: does age affect pain, function or complications?Clin Orthop Relat Res. 2013; 471(6):1964–1969. doi:10.1007/s11999-013-2803-3 [CrossRef]
  8. Memtsoudis SG, González Della Valle A, Besculides MC, Gaber L, Sculco TP. Inhospital complications and mortality of unilateral, bilateral, and revision TKA: based on an estimate of 4,159,661 discharges. Clin Orthop Relat Res. 2008; 466(11):2617–2627. doi:10.1007/s11999-008-0402-5 [CrossRef]
  9. Duchman KR, Gao Y, Pugely AJ, Martin CT, Noiseux NO, Callaghan JJ. The effect of smoking on short-term complications following total hip and knee arthroplasty. J Bone Joint Surg Am. 2015; 97(13):1049–1058. doi:10.2106/JBJS.N.01016 [CrossRef]
  10. Otero JE, Gholson JJ, Pugely AJ, Gao Y, Bedard NA, Callaghan JJ. Length of hospitalization after joint arthroplasty: does early discharge affect complications and readmission rates?J Arthroplasty. 2016; 31(12):2714–2725. doi:10.1016/j.arth.2016.07.026 [CrossRef]
  11. Schwarzkopf R, Phan DL, Hoang M, Ross SD, Mukamel D. Do patients with income-based insurance have access to total joint arthroplasty?J Arthroplasty. 2014; 29(6):1083–1086. doi:10.1016/j.arth.2013.11.022 [CrossRef]
  12. Kim CY, Wiznia DH, Hsiang WR, Pelker RR. The effect of insurance type on patient access to knee arthroplasty and revision under the Affordable Care Act. J Arthroplasty. 2015; 30(9):1498–1501. doi:10.1016/j.arth.2015.03.015 [CrossRef]
  13. Martin CT, Callaghan JJ, Liu SS, Gao Y, Warth LC, Johnston RC. Disparity in total joint arthroplasty patient comorbidities, demographics, and postoperative outcomes based on insurance payer type. J Arthroplasty. 2012; 27(10):1761–1765. doi:10.1016/j.arth.2012.06.007 [CrossRef]
  14. Rosenthal BD, Hulst JB, Moric M, Levine BR, Sporer SM. The effect of payer type on clinical outcomes in total knee arthroplasty. J Arthroplasty. 2014; 29(2):295–298. doi:10.1016/j.arth.2013.06.010 [CrossRef]
  15. Hinman A, Bozic KJ. Impact of payer type on resource utilization, outcomes and access to care in total hip arthroplasty. J Arthroplasty. 2008; 23(6)(suppl 1):9–14. doi:10.1016/j.arth.2008.05.010 [CrossRef]
  16. Browne JA, Novicoff WM, D'Apuzzo MR. Medicaid payer status is associated with inhospital morbidity and resource utilization following primary total joint arthroplasty. J Bone Joint Surg Am. 2014; 96(21):e180. doi:10.2106/JBJS.N.00133 [CrossRef]
  17. Browne JA, Sandberg BF, D'Apuzzo MR, Novicoff WM. Depression is associated with early postoperative outcomes following total joint arthroplasty: a nationwide database study. J Arthroplasty. 2014; 29(3):481–483. doi:10.1016/j.arth.2013.08.025 [CrossRef]
  18. Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis. 1987; 40(5):373–383. doi:10.1016/0021-9681(87)90171-8 [CrossRef]
  19. Vishnevsky KA, Zemchenkov AY, Korosteleva NY, Smirnov AV. Use of the Charlson comorbidity index and the Barthel disability index in the integrated assessment of the sociomedical status of patients receiving continuous renal replacement therapy with hemodialysis [in Russian]. Ter Arkh. 2015; 87(6):62–67. doi:10.17116/terarkh201587662-67 [CrossRef]
  20. Havens JM, Olufajo OA, Cooper ZR, Haider AH, Shah AA, Salim A. Defining rates and risk factors for readmissions following emergency general surgery. JAMA Surg. 2016; 151(4):330–336. doi:10.1001/jamasurg.2015.4056 [CrossRef]
  21. Haider AH, Chang DC, Efron DT, Haut ER, Crandall M, Cornwell EE III, . Race and insurance status as risk factors for trauma mortality. Arch Surg. 2008; 143(10):945–949. doi:10.1001/archsurg.143.10.945 [CrossRef]
  22. Nelson CL. Disparities in orthopaedic surgical intervention. J Am Acad Orthop Surg. 2007; 15(suppl 1):S13–S17. doi:10.5435/00124635-200700001-00005 [CrossRef]
  23. McBean AM, Gornick M. Differences by race in the rates of procedures performed in hospitals for Medicare beneficiaries. Health Care Financ Rev. 1994; 15(4):77–90.
  24. Escalante A, Barrett J, del Rincón I, Cornell JE, Phillips CB, Katz JN. Disparity in total hip replacement affecting Hispanic Medicare beneficiaries. Med Care. 2002; 40(6):451–460. doi:10.1097/00005650-200206000-00002 [CrossRef]
  25. Jones A, Kwoh CK, Kelley ME, Ibrahim SA. Racial disparity in knee arthroplasty utilization in the veterans health administration. Arthritis Rheum. 2005; 53(6):979–981. doi:10.1002/art.21596 [CrossRef]
  26. Wilson MG, May DS, Kelly JJ. Racial differences in the use of total knee arthroplasty for osteoarthritis among older Americans. Ethn Dis. 1994; 4(1):57–67.
  27. Skinner J, Weinstein JN, Sporer SM, Wennberg JE. Racial, ethnic, and geographic disparities in rates of knee arthroplasty among Medicare patients. N Engl J Med. 2003; 349(14):1350–1359. doi:10.1056/NEJMsa021569 [CrossRef]
  28. Escalante A, Espinosa-Morales R, del Rincón I, Arroyo RA, Older SA. Recipients of hip replacement for arthritis are less likely to be Hispanic, independent of access to health care and socioeconomic status. Arthritis Rheum. 2000; 43(2):390–399. doi:10.1002/1529-0131(200002)43:2<390::AID-ANR20>3.0.CO;2-R [CrossRef]
  29. Clement ND, MacDonald D, Howie CR, Biant LC. The outcome of primary total hip and knee arthroplasty in patients aged 80 years or more. J Bone Joint Surg Br. 2011; 93(9):1265–1270. doi:10.1302/0301-620X.93B9.25962 [CrossRef]
  30. Hosick WB, Lotke PA, Baldwin A. Total knee arthroplasty in patients 80 years of age and older. Clin Orthop Relat Res. 1994; 299:77–80.
  31. Wasielewski RC, Weed H, Prezioso C, Nicholson C, Puri RD. Patient comorbidity: relationship to outcomes of total knee arthroplasty. Clin Orthop Relat Res. 1998; 356:85–92. doi:10.1097/00003086-199811000-00014 [CrossRef]
  32. Singh JA, O'Byrne M, Harmsen S, Lewallen D. Predictors of moderate-severe functional limitation after primary total knee arthroplasty (TKA): 4701 TKAs at 2-years and 2935 TKAs at 5-years. Osteoarthritis Cartilage. 2010; 18(4):515–521. doi:10.1016/j.joca.2009.12.001 [CrossRef]
  33. Gordon SM, Culver DH, Simmons BP, Jarvis WR. Risk factors for wound infections after total knee arthroplasty. Am J Epidemiol. 1990; 131(5):905–916. doi:10.1093/oxfordjournals.aje.a115580 [CrossRef]

Demographics of the Total Knee Arthroplasty Patients

CharacteristicNo. (%)P

MedicareMedicaid/UninsuredPrivate InsuranceOther
Total781,25542,580481,05444,725
Procedure type<.001
  Revision total knee replacement33,374 (4.3)1948 (4.6)14,918 (3.1)2905 (6.5)
  Total knee replacement747,881 (95.7)40,632 (95.4)466,136 (96.9)41,820 (93.5)
Age groupa<.001
  <40 y1220 (0.2)1898 (4.5)5286 (1.1)896 (2.0)
  40–64 y98,708 (12.6)33,444 (78.6)403,095 (83.9)36,703 (82.3)
  65–79 y540,623 (69.2)6180 (14.5)64,539 (13.4)6241 (14.1)
  ≥80 y140,346 (18.0)1006 (2.4)7707 (1.6)734 (1.6)
Racea<.001
  White523,025 (66.9)18,221 (42.8)306,347 (63.7)26,583 (59.4)
  Black37,010 (4.7)6027 (14.2)27,017 (5.6)3761 (8.4)
  Hispanic28,313 (3.6)5543 (13.0)15,372 (3.2)2913 (6.5)
  Asian/Pacific Islander6108 (0.8)994 (2.3)3400 (0.7)357 (0.8)
Sex
  Male271,772 (34.8)10,523 (24.8)188,910 (39.3)22,422 (50.1)
Comorbidity index<.001
  0478,594 (61.3)23,884 (56.1)329,062 (68.4)30,479 (68.1)
  1212,060 (27.1)13,193 (31.0)118,689 (24.7)11,132 (24.9)
  263,092 (8.1)4101 (9.6)25,410 (5.3)2411 (5.4)
  ≥327,509 (3.5)1402 (3.3)7893 (1.6)703 (1.6)
Hospital annual volume<.001
  1–173267,262 (34.2)20,017 (47.0)142,984 (29.7)17,745 (39.7)
  174–370257,402 (32.9)13,328 (31.3)156,671 (32.6)14,004 (31.3)
  ≥371256,591 (32.8)9235 (21.7)181,399 (37.7)12,976 (29.0)

Medical Complications, Surgical Complications, and Mortality After Total Knee Arthroplasty on Multivariable Logistic Regression Analysis

CharacteristicMedical ComplicationsSurgical ComplicationsMortality



Adjusted OR (95% CI)PaPbAdjusted OR (95% CI)PaPbAdjusted OR (95% CI)PaPb
Payer<.001<.001<.001
Reference=Medicare
  Medicaid/uninsured0.96 (0.93–1.00).0601.23 (1.15–1.32)<.0010.95 (0.66–1.37).779
  Private insurance0.82 (0.81–0.84)<.0010.93 (0.90–0.96)<.0010.67 (0.56–0.80)<.001
  Other0.86 (0.82–0.89)<.0011.08 (1.01–1.16).0260.77 (0.51–1.17).223
Female: F vs M0.84 (0.83–0.86)<.001<.0010.79 (0.77–0.81)<.001<.0010.59 (0.53–0.66)<.001<.001
Race<.001<.001.041
  Reference=white
  Black0.90 (0.88–0.92)<.0011.56 (1.49–1.64)<.0011.09 (0.87–1.37).436
  Hispanic0.74 (0.72–0.77)<.0011.37 (1.29–1.47)<.0010.87 (0.62–1.23).434
  Asian/Pacific Islander0.67 (0.61–0.73)<.0011.29 (1.11–1.51).0011.49 (0.80–2.79).211
Age group<.001<.001<.001
  Reference=<40 y
  40–64 y2.68 (2.39–3.01)<.0010.70 (0.63–0.79)<.0011.49 (0.55–4.02).430
  65–79 y4.06 (3.62–4.56)<.0010.76 (0.67–0.86)<.0012.64 (0.98–7.15).055
  ≥80 y6.25 (5.56–7.03)<.0010.83 (0.73–0.94).0036.32 (2.33–17.14)<.001
Procedure type
  Reference=total knee replacement
  Revision total knee replacement1.19 (1.16–1.23)<.001<.0011.87 (1.79–1.96)<.001<.0011.68 (1.39–2.04)<.001<.001
Comorbidity Index<.001<.001
  Reference=0
  14.50 (4.43–4.56)<.0011.03 (1.00–1.06).0562.24 (1.97–2.55)<.001
  212.31 (12.09–12.53)<.0011.11 (1.06–1.16)<.0014.65 (4.00–5.41)<.001
  ≥321.91 (21.38–22.45)<.0011.11 (1.04–1.19).00210.22 (8.75–11.92)<.001

Baseline Demographics and Comorbidities Between the Matched Medicare and Private Insurance Total Knee Arthroplasty Patients

CharacteristicMedicarePrivate InsuranceP
Total No.595,440594,842
Age, mean (SD), y66.2 (17.3)65.6 (17.5)<.001
Sex, No. (%).999
  Male221,776 (37.2)221,554 (37.2)
  Female373,664 (62.8)373,288 (62.8)
Race, No. (%).923
  White383,601 (64.4)383,001 (64.4)
  Black32,168 (5.4)32,201 (5.4)
  Hispanic19,633 (3.3)19,529 (3.3)
  Other160,038 (26.9)160,111 (26.9)
Procedure, No. (%).924
  Revision total knee replacement16,137 (2.7)16,104 (2.7)
  Total knee replacement579,303 (97.3)578,738 (97.3)
Hospital teaching status, No. (%).416
  Nonteaching367,108 (61.7)366,308 (61.6)
  Teaching228,332 (38.3)228,534 (38.4)
Comorbidities, No. (%)
  Congestive heart failure3553 (0.6)3564 (0.6).862
  Cardiac arrhythmia27,639 (4.6)27,611 (4.6).999
  Vascular disease8598 (1.4)8589 (1.4).998
  Pulmonary circulation disorder157 (0)158 (0).948
  Peripheral vascular disease2460 (0.4)2441 (0.4).813
  Hypertension380,301 (63.9)380,054 (63.9).797
  Paralysis40 (0)38 (0).824
  Neurologic disorders7291 (1.2)7323 (1.2).743
  Chronic pulmonary disease61,445 (10.3)61,397 (10.3).967
  Diabetes, uncomplicated105,014 (17.6)104,860 (17.6).907
  Diabetes, with chronic complication2442 (0.4)2435 (0.4).948
  Hypothyroidism63,160 (10.6)63,163 (10.6).843
  Renal failure4476 (0.8)4443 (0.7).762
  Liver disease350 (0.1)344 (0.1).830
  Peptic ulcer disease971 (0.2)963 (0.2).873
  AIDS20 (0)20 (0).823
  Lymphoma307 (0.1)313 (0.1).997
  Metastatic cancer14 (0)15 (0).800
  Solid tumor without metastasis193 (0)200 (0).851
  Rheumatoid arthritis/collagen vascular disease17,602 (3)17,657 (3).717
  Coagulopathy1805 (0.3)1786 (0.3).694
  Obesity80,609 (13.5)80,576 (13.5).774
  Weight loss19 (0)19 (0).898
  Fluid and electrolyte disorder20,472 (3.4)20,473 (3.4).998
  Chronic blood loss anemia3342 (0.6)3309 (0.6).914
  Deficiency anemia1036 (0.2)1025 (0.2).715
  Alcohol abuse669 (0.1)682 (0.1).826
  Psychosis2702 (0.5)2676 (0.4).710
  Depression43,985 (7.4)43,939 (7.4).750
  Smoking22,104 (3.7)22,128 (3.7).995

Comparison of In-Hospital Complications After Total Knee Arthroplasty Between the Matched Medicare and Private Insurance Patients

ComplicationNo. (%)PRelative Risk (95%Confidence Interval)

MedicarePrivate Insurance
Central nervous system463 (0.1)399 (0.1).030a1.16 (1.01–1.33)
Cardiac3287 (0.6)3532 (0.6).003a0.93 (0.89–0.97)
Peripheral vascular1151 (0.2)1144 (0.2).9031.01 (0.93–1.09)
Respiratory3900 (0.7)3828 (0.6).4371.02 (0.97–1.06)
Gastrointestinal3308 (0.6)2924 (0.5)<.001a1.13 (1.08–1.19)
Genitourinary3742 (0.6)3799 (0.6).4830.98 (0.94–1.03)
Hematoma/seroma4563 (0.8)4397 (0.7).0871.04 (0.99–1.08)
Wound dehiscence573 (0.1)434 (0.1)<.001a1.32 (1.16–1.49)
Postoperative infection1062 (0.2)1123 (0.2).1910.94 (0.87–1.02)
Deep venous thrombosis1542 (0.3)1614 (0.3).1900.95 (0.89–1.02)
Pulmonary embolism598 (0.1)652 (0.1).1220.92 (0.82–1.02)
Postoperative anemia96,717 (16.2)91,277 (15.3)<.001a1.06 (1.05–1.07)
Mortality429 (0.1)321 (0.1)<.001a1.34 (1.16–1.54)
Authors

The authors are from the Department of Orthopaedic Surgery, Boston University School of Medicine (DRV, EJC, ELS, XL) and the Harvard T.H. Chan School of Public Health (AE), Boston, Massachusetts; the University of Wisconsin School of Medicine and Public Health (PHY), Madison, Wisconsin; and the University of California San Francisco School of Medicine (DCS), San Francisco, California.

Dr Veltre, Dr Yi, Dr Sing, Ms Curry, Dr Endo, and Dr Li have no relevant financial relationships to disclose. Dr Smith is a paid consultant for Arthrocare, Conformis, and DePuy and has received research support from Conformis and DePuy.

Correspondence should be addressed to: Xinning Li, MD, Department of Orthopaedic Surgery, Boston University School of Medicine, 850 Harrison Ave, Dowling 2 North, Boston, MA 02118 ( xinning.li@gmail.com).

Received: June 25, 2017
Accepted: December 15, 2017
Posted Online: March 02, 2018

10.3928/01477447-20180226-07

Sign up to receive

Journal E-contents