Annual demand for total knee arthroplasty (TKA) is projected to grow to 3.48 million procedures by the year 2030.1 Effectively addressing this demand requires payors, providers, and policy makers to develop and implement solutions to reduce costs while either maintaining or improving the quality of care provided. Traditional efforts to reduce both payor and provider costs of total joint arthroplasty (TJA) have targeted reducing inpatient length of stay (LOS), incidence of major perioperative complications, and unnecessary readmissions. More recently, however, attention also has turned to shifting the location of care to ambulatory settings for carefully selected subgroups of patients.2
In response to the implementation of bundled payments and the movement of patients out of the inpatient setting, many hospitals have implemented new “rapid recovery” or “fast track” programs to further improve the average LOS from 3 days3–5 to as little as 1 day.6–10 These programs represent the middle ground between traditional inpatient and outpatient settings. However, the identification and management of specific risk factors—the principal challenge of safely shifting TKA patients to the out-patient setting—remains.11,12 Although there is research suggesting specific patient profiles that are most appropriate for out-patient TKA,8,9,11–16 there is little research describing the characteristics of current patients and their outcomes using a multi-center, nationally representative dataset.
The current study examined characteristics of patients who were discharged home after undergoing primary TKA in either an outpatient setting (outpatient hospital or surgical center) or a rapid recovery inpatient setting (defined as a length of stay of 1 day or less) and compared clinical and economic outcomes by setting among a matched subset of patients. Given the clinical protocols required to effectively discharge a patient home in under 24 hours should be similar regardless of the site of care, one would expect similar outcomes regardless of the site of care. If outcomes are similar in the group of patients deemed “eligible” for a rapid recovery protocol, it would follow that there is the potential for substantial cost savings to the US health system with a shift in the surgical setting. Therefore, the current study evaluated differences in outcomes and associated total payor costs attributed to procedure location among matched cohorts of patients discharged home with or without aid.
Materials and Methods
This study was a retrospective analysis of health care claims data from the Truven Health Analytics Marketscan Commercial Research Database (IBM Watson Health, Greenwood Village, Colorado) (2014–2015). The dataset contains more than 180 million unique patients covered with private insurance and Medicare Advantage, and provides nationally representative information on claims for outpatient and inpatient visits, prescription drug fills, demographics, location of service, and total health plan payments. The database is fully deidentified; therefore, this study did not require institutional review board approval.
Patients selected for analysis were 18 years or older and were listed as having either an outpatient or inpatient visit with a primary International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) or International Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM) procedure code for unilateral TKA. Patients were grouped into 2 cohorts: same-day discharge outpatient TKA (performed in an outpatient hospital or ambulatory surgical center setting) or rapid recovery inpatient TKA (performed in an inpatient setting with the number of midnights less than or equal to 1). Patients were excluded from analysis if there was evidence of fracture of the lower limb or if the procedure was bilateral as indicated by Current Procedural Terminology code modifiers.
All patients were required to have continuous health insurance from baseline through the end of follow-up. Patients included both those with commercial-only coverage and those with Medicare Advantage plans. Finally, the study population was limited to patients discharged home under self-care or with home health care services. Patients discharged to post-acute care facilities were excluded to further attribute outcomes differences to the original surgical procedure location, rather than to differences in post-acute care received. The study baseline period was defined as 90 days prior to the index date (date of inpatient admission or outpatient procedure), and follow-up was defined as the day after discharge through 90 days.
Patient clinical histories were evaluated with ICD-9-CM or ICD-10-CM diagnoses codes listed during the baseline or index visit. In addition, the Charlson Comorbidity Index (CCI) score was calculated as an objective aggregate measure of patient comorbidity burden.17
Incidence of complications was evaluated during both the index visit and follow-up periods. Minor complications were defined as blood transfusion (autologous or allogeneic); hemorrhage, hematoma, or seroma (diagnosis complicating a procedure or a procedure for incision and drainage); wound disruption; or urinary tract infection. Major complications were defined as any infection (sepsis or septicemia, bacteremia, bone infection, or other postoperative infection), thromboembolic event (pulmonary, venous, and deep vein), renal failure, renal insufficiency, myocardial infarction, pneumonia, stroke or cerebrovascular accident, unplanned intubation, and reoperation. Implant-related complications included loosening, dislocation, implant failure, periprosthetic fracture, other mechanical complication of an internal orthopedic device (due to internal joint prosthesis, implant, or graft), or pain or stiffness in the joint. The incidence of all-cause readmissions, defined as any inpatient admission occurring during follow-up, also was analyzed.
Total health plan payments were summarized for the index visit and during the 90-day follow-up period. Payments were defined as all medical payments made for a patient, regardless of the site of care or the reason for the visit. All payments were inflation-adjusted to 2015 US dollars using the Medical Care component of the US Consumer Price Index. To address possible miscoding of payments, records with values listed as zero or above the 99th percentile were excluded from analyses.
Due to the nonrandom assignment of patients to each cohort, exact matching was performed to minimize potential confounding effects and selection bias. Exact 1:1 matching allows examination of outcomes across comparable groups of patients to compare differences by procedure setting. Accordingly, each outpatient was matched with one rapid recovery in-patient. Matching covariates were selected based on measures that were significantly different (P<.05) in unmatched analysis and included age group, sex, census region, primary payor, CCI group, coronary artery disease, congestive heart failure, chronic obstructive pulmonary disease, diabetes, hypertension, sleep apnea, diagnosis of overweight or obesity, and history of smoking. In addition, patients were matched by procedure year.
Descriptive analyses were conducted for all study measures outlined above. Statistical significance testing was performed using the chi-square test for comparison of categorical variables and the Wilcoxon–Mann-Whitney test for comparison of continuous variables in unmatched analyses. In matched analyses, significance testing was performed using the McNemar test for categorical variables and the Wilcoxon–Mann-Whitney test for continuous variables.
Sample selection and the creation of analytic variables was performed using the Instant Health Data (IHD) platform (Boston Health Economics, Boston, Massachusetts). Statistical analyses were conducted using R version 3.2.1 software (R Foundation for Statistical Computing, Vienna, Austria) and SAS version 9.4 (SAS Institute Inc, Cary, North Carolina) software.
Patient Demographics and Clinical History
Prior to matching, 969 patients were available for analysis in the outpatient cohort and 8101 patients were available in the rapid recovery inpatient cohort. After matching, 863 patients were assigned to each cohort for analysis.
Before matching, mean patient age was significantly lower in the outpatient cohort vs the rapid recovery inpatient cohort (57.2 vs 60.9 years, respectively), with a greater proportion of females in the outpatient cohort vs the rapid recovery inpatient cohort (56.4% vs 51%, respectively; P=.002) (Table A, available in the online version of this article).
The most common preoperative co-morbid diagnoses were hypertension (44.6% vs 60.6%), obesity (19.7% vs 25.6%), and diabetes without complications (11% vs 16.3%), all of which were significantly less prevalent in the outpatient cohort vs the rapid recovery inpatient cohort (all P<.001; Table B, available in the online version of this article). Following exact matching, all prior significantly different baseline demographic and clinical characteristics for both cohorts were well-balanced (all P>.05; Table A). In both matched cohorts, median age was 58 years, 56.8% were females, and 89.1% of patients had a CCI score of zero.
Co-Occurring comorbidities among unmatched outpatients
Complications and Discharge Destination
In matched analysis, the incidence of minor complications was significantly lower for outpatients compared with rapid recovery inpatients (2.8% vs 5.8%, respectively; P=.002); the incidence of major complications was statistically equivalent (5.2% vs 6.7%; P=.173) (Table C, available in the online version of this article). Use of opiate pain medication was significantly lower in the outpatient cohort compared with the inpatient cohort (80.4% vs 90.7%, respectively; P<.001); this characteristic also was observed in the unmatched analysis. Almost all of the rapid recovery inpatients were discharged home under self-care (99.3%) compared with outpatients (94.6%), and the remaining 0.7% and 5.4%, respectively, were discharged home under care. Consistent with the unmatched analysis, use of physical therapy was statistically equivalent across both cohorts, with 86.2% of outpatients and 85.3% of rapid recovery patients having at least one kind of physical therapy during follow-up (P=.586).
Hospitalization characteristics and incidence of complications from day of admission through 90-day post-discharge follow-up period
Across unmatched cohorts, the incidence of all-cause readmissions during the entire 90-day follow-up period was significantly lower among outpatients compared with rapid recovery inpatients (5.1% vs 6.9%; P=.041). Similarly, in matched analyses, the 90-day readmission rate was lower in the outpatient cohort than in the rapid recovery inpatients; however, the difference was not statistically significant (5.1% vs 7.3%; P=.064). Incidence of readmission during the first 30 days of follow-up also was statistically equivalent between matched cohorts (P=.157); however, readmissions were significantly lower in the outpatient cohort than in the rapid recovery inpatients during days 31 to 90 of follow-up (1.8% vs 5.1%, respectively; P<.001). When stratified by 5-day increments, incidence of readmission was numerically lower in the matched rapid recovery inpatient cohort relative to the outpatient cohort through day 45. Following day 45, an inflection point was observed, with the incidence of readmission remaining lower among the matched outpatient cohort through the end of follow-up (Figure 1).
Graph showing all-cause readmission following total knee arthroplasty, stratified by 5-day increments. *Significance testing included the chi-square test for unmatched and the McNemar test for matched cohorts.
In both unmatched and matched analysis, payments were lower among outpatients (Table D, available in the online version of this article). In matched analysis, the median index visit payment was $6880 (24.1%) lower in the outpatient setting than in the rapid recovery inpatient setting ($21,698 vs $28,578, respectively), and the median 90-day payment was $6824 lower in the outpatient setting than in the rapid recovery inpatient setting ($24,749 vs $31,573, respectively; P<.001).
Total health plan payments (USD)
Given the national trend toward adoption of rapid recovery protocols and a shift to the outpatient setting for TKA procedures, it is critical to understand the value and impact of each setting, as well as the actual profiles of the patients. The current study of clinical and economic outcomes by setting with a nationally representative dataset expands on prior research by focusing on characteristics of outpatient TKA patients relative to those of rapid recovery inpatient programs. In addition, by controlling for the inherent case-selection bias in procedure location by using exact matching, the current study allowed for more accurate attribution of outcomes by procedure location.
As more TJA procedures are shifted to the outpatient setting, understanding the magnitude of risk associated with single and multiple comorbidities will be fundamental to optimize cost and outcomes. Prior research has demonstrated patients who are healthier, with no prior history of smoking, diabetes, hypertension, coronary artery disease, congestive heart failure, and chronic obstructive pulmonary disease, as well as those who are outside the clinical classification of obesity, are at the lowest risk of complications following TKA if surgery is shifted to an outpatient setting.8,11,14,16,18 Objectively, these patients may represent the ideal profile for providers to transition from a rapid recovery inpatient setting to an outpatient setting. Importantly, this population does not need to be in perfect health; more than half (51%) of the patients selected for the outpatient setting in the current study had at least 1 comorbidity compared with 78% in the rapid recovery cohort. In addition, there were relatively few patients with 2 or more major comorbidities in the outpatient cohort; the most common concurrent comorbidities were obesity and hypertension. This is an area for future research, whereby risk scores may be codified for patient selection to specific procedure settings.
Several prior studies have evaluated outcomes following TJA by procedure location. Using the National Surgical Quality Improvement Program database from the American College of Surgeons, Bovonratwet et al8 analyzed patients undergoing primary inpatient and outpatient elective TKA. They found no significant difference by setting in the incidence of minor or severe adverse events, or read-missions during the 30-day postoperative period.
A study by Nelson et al15 of matched patients who underwent outpatient and inpatient (LOS, 1 to 5 days) total hip arthroplasty—a relatively higher acuity procedure—also found procedure setting had little to no effect on outcomes or adverse events. The only exception was transfusion, which was significantly lower among those treated in the outpatient setting. In another study, Courtney et al14 compared patients undergoing elective outpatient and inpatient TJA (LOS greater than 24 hours); they found the outpatient TJA cohort had a significantly lower risk of complications than the inpatient cohort (8% vs 16%, respectively; P<.001), with no difference in reoperation (0.2% vs 0.2%, respectively; P=.796) or readmission (9.3% vs 0.5%, respectively; P=.322) in unadjusted analyses. After controlling for confounding variables, outpatient TJA remained significantly correlated with lower odds of complications (odds ratio, 0.46; 95% confidence interval, 0.37–0.57; P<.001).
Comparatively, the current study included both unmatched and matched analyses to better characterize the real-world profiles of patients selected for outpatient TKA, and outcomes then were further examined by setting during 90 days of follow-up. Similar to prior research, the choice of procedure setting appeared to have no adverse effect on outcomes. In fact, the current study found a significant improvement in the risk of minor complications among patients in the outpatient setting vs the rapid recovery inpatient setting.
Pain management in TJA is critical to patient satisfaction and recovery. Congruently, orthopedic surgeons represent the third-most common prescribers of opiates in US health care today, even when perioperative opiate use has been associated with increased risk of readmission, emergency department visits, infection, and revision TKA.19,20 In the current matched analysis, opiate use in the postoperative period was significantly lower among out-patients (80.4%) relative to rapid recovery inpatients (90.7%; P<.001). Although chart-level detail on specific pain management protocols is unavailable in health care claims, the current findings suggest surgeons operating in the outpatient setting have developed methods that reduce the use of opiate pain medication.
Finally, little has been published regarding the potential savings to the US health care system by shifting TKA location from an inpatient setting (including rapid recovery programs) to an outpatient setting. In a prior analysis of the Medicare 5% sample, Lovald et al21 found the 2-year total incremental Medicare payments after TKA were $6560 lower for the outpatient cohort relative to an inpatient stay of 1 to 2 days. Importantly, the study by Lovald et al21 was completed prior to Medicare reimbursement for outpatient TKA, which was implemented in January 2018.22 Comparatively, the current study examined a commercially insured population where outpatient TKA was reimbursed during the study period. The median 90-day episode cost savings per patient (from the perspective of the health plan payor) associated with the shift to outpatient was $6824, amounting to a per-patient savings of 22% relative to rapid recovery inpatient TKA.
Given the nature of retrospective claims data, details specific to clinical management such as strategies and guidelines for anesthesia, blood product use, and surgical technique (eg, tourniquet-less or direct anterior approach) were unavailable. In addition, factors not coded in insurance claims, such as surgical skill or outcome-influencing factors (eg, tourniquet time) were not evaluated. Potentially confounding factors that involved care setting choice also could not be controlled for in matched analyses (eg, surgeon and patient preference, staffing volume, support services, and geographic convenience). Finally, the current study was performed in a commercial population; further enquiry is required to evaluate whether these findings extend to the Medicare population.
When outcomes were compared by exact matching of patient profiles, those who underwent outpatient TKA demonstrated a significantly lower risk of minor complications, with equivalent risk of major complications and 90-day all-cause readmission. The results of this study suggest that among carefully selected patients, outpatient TKA is a safe and effective option with the opportunity for significant payor cost savings.
- 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. PMID: doi:10.2106/JBJS.F.00222 [CrossRef]17403800
- Huang A, Ryu JJ, Dervin G. Cost savings of outpatient versus standard inpatient total knee arthroplasty. Can J Surg. 2017;60(1):57–62. PMID:28234591
- Accelero Health Partners. Length of stay is critical for total hip and knee replacement cost of care. Retrieved from http://accelero-health.com/wp-content/uploads/2014/04/Length-of-Stay-is-Critical-for-Joint-Replacement-Cost-of-Care.pdf. Accessed October 3, 2017.
- Cram P, Lu X, Kates SL, Singh JA, Li Y, Wolf BR. Total knee arthroplasty volume, utilization, and outcomes among Medicare beneficiaries, 1991–2010. JAMA. 2012;308(12):1227–1236. https://doi.org/10.1001/2012.jama.11153 PMID: doi:10.1001/2012.jama.11153 [CrossRef]23011713
- Kelly SJ. Commonly asked questions about recovering from hip or knee replacement surgery. Retrieved from http://ec2-54-209-165-168.compute-1.amazonaws.com/care-for-hips-and-knees/commonly-asked-questions-about-recovering-from-a-hip-or-knee-replacement-surgery. Accessed December 2, 2017.
- Barrington JW. Fast-track recovery and outpatient joint arthroplasty. Am J Orthop (Belle Mead NJ). 2015;44(10)(suppl):S21–S22. PMID:26447429
- Berger RA, Cross MB, Sanders S. Outpatient hip and knee replacement: the experience from the first 15 years. Instr Course Lect. 2016;65:547–551. PMID:27049219
- Bovonratwet P, Ondeck NT, Nelson SJ, Cui JJ, Webb ML, Grauer JN. Comparison of outpatient vs inpatient total knee arthroplasty: an ACS-NSQIP analysis. J Arthroplasty. 2017;32(6):1773–1778. https://doi.org/10.1016/j.arth.2017.01.043 PMID: doi:10.1016/j.arth.2017.01.043 [CrossRef]28237215
- Lovecchio F, Alvi H, Sahota S, Beal M, Manning D. Is outpatient arthroplasty as safe as fast-track inpatient arthroplasty? A propensity score matched analysis. J Arthroplasty. 2016;31(9)(suppl):197–201. https://doi.org/10.1016/j.arth.2016.05.037 PMID: doi:10.1016/j.arth.2016.05.037 [CrossRef]27378634
- Thienpont E, Lavand'homme P, Kehlet H. The constraints on day-case total knee arthroplasty: the fastest fast track. Bone Joint J. 2015;97-B(10)(suppl A):40–44. https://doi.org/10.1302/0301-620X.97B10.36610 PMID: doi:10.1302/0301-620X.97B10.36610 [CrossRef]26430085
- Courtney PM, Rozell JC, Melnic CM, Lee G-C.Who should not undergo short stay hip and knee arthroplasty? Risk factors associated with major medical complications following primary total joint arthroplasty. J Arthroplasty. 2015;30(9)(suppl):1–4. https://doi.org/10.1016/j.arth.2015.01.056 PMID: doi:10.1016/j.arth.2015.01.056 [CrossRef]26105617
- Sher A, Keswani A, Yao DH, Anderson M, Koenig K, Moucha CS. Predictors of same-day discharge in primary total joint arthroplasty patients and risk factors for post-discharge complications. J Arthroplasty. 2017;32:S150–S156.e1. https://doi.org/10.1016/j.arth.2016.12.017 doi:10.1016/j.arth.2016.12.017 [CrossRef]28089186
- Aynardi M, Post Z, Ong A, Orozco F, Sukin DC. Outpatient surgery as a means of cost reduction in total hip arthroplasty: a case-control study. HSS J. 2014;10(3):252–255. https://doi.org/10.1007/s11420-014-9401-0 PMID: doi:10.1007/s11420-014-9401-0 [CrossRef]25264442
- Courtney PM, Boniello AJ, Berger RA. Complications following outpatient total joint arthroplasty: an analysis of a national database. J Arthroplasty. 2017;32(5):1426–1430. https://doi.org/10.1016/j.arth.2016.11.055 PMID: doi:10.1016/j.arth.2016.11.055 [CrossRef]
- Nelson SJ, Webb ML, Lukasiewicz AM, Varthi AG, Samuel AM, Grauer JN. Is out-patient total hip arthroplasty safe?J Arthroplasty. 2017;32(5):1439–1442. https://doi.org/10.1016/j.arth.2016.11.053 PMID: doi:10.1016/j.arth.2016.11.053 [CrossRef]28065622
- Parcells BW, Giacobbe D, Macknet D, et al. Total joint arthroplasty in a stand-alone ambulatory surgical center: short-term outcomes. Orthopedics. 2016;39(4):223–228. https://doi.org/10.3928/01477447-20160419-06 PMID: doi:10.3928/01477447-20160419-06 [CrossRef]27111079
- 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. https://doi.org/10.1016/0021-9681(87)90171-8 PMID: doi:10.1016/0021-9681(87)90171-8 [CrossRef]3558716
- Klein GR, Posner JM, Levine HB, Hartzband MA. Same day total hip arthroplasty performed at an ambulatory surgical center: 90-day complication rate on 549 patients. J Arthroplasty. 2017;32(4):1103–1106. https://doi.org/10.1016/j.arth.2016.10.013 PMID: doi:10.1016/j.arth.2016.10.013 [CrossRef]
- Ben-Ari A, Chansky H, Rozet I. Preoperative opioid use is associated with early revision after total knee arthroplasty: a study of male patients treated in the Veterans Affairs system. J Bone Joint Surg Am. 2017;99(1):1–9. https://doi.org/10.2106/JBJS.16.00167 PMID: doi:10.2106/JBJS.16.00167 [CrossRef]28060227
- Cancienne JM, Patel KJ, Browne JA, Werner BC. Narcotic use and total knee arthroplasty. J Arthroplasty. 2018;33(1):113–118. PMID: doi:10.1016/j.arth.2017.08.006 [CrossRef]
- Lovald ST, Ong KL, Malkani AL, et al. Complications, mortality, and costs for out-patient and short-stay total knee arthroplasty patients in comparison to standard-stay patients. J Arthroplasty. 2014;29(3):510–515. https://doi.org/10.1016/j.arth.2013.07.020 PMID: doi:10.1016/j.arth.2013.07.020 [CrossRef]
- Centers for Medicare & Medicaid Services. January2018update of the Hospital Outpatient Prospective Payment System (OPPS). Retrieved from https://www.cms.gov/Out-reach-and-Education/Medicare-Learning-Network-MLN/MLNMattersArticles/downloads/MM10417.pdf. Accessed May 18, 2018.
Patient demographics and clinical characteristics
|Age Group||< 0.0001||1.000|
|Census Region||< 0.0001||1.000|
|Primary Payor||< 0.0001||1.000|
| Medicare Advantage||4.6%||27.9%||4.5%||4.5%|
|Charlson Score (CCI)||< 0.0001||0.999|
|CCI Grouping||< 0.0001||1.000|
|Baseline Through Index Clinical Diagnoses|
| Hypertension||44.6%||60.6%||< 0.0001||44.4%||44.4%||1.000|
| Obesec||19.7%||25.6%||< 0.0001||18.0%||18.0%||1.000|
| Diabetes||11.5%||16.5%||< 0.0001||8.8%||8.8%||1.000|
| Chronic Obstructive Pulmonary Disease||7.4%||12.9%||< 0.0001||5.9%||5.9%||1.000|
| History of Smoking||5.2%||17.0%||< 0.0001||3.9%||3.9%||1.000|
| Obstructive Sleep Apnea||7.6%||13.3%||< 0.0001||5.3%||5.3%||1.000|
| Coronary Artery Disease||2.8%||8.9%||< 0.0001||2.3%||2.3%||1.000|
| Nutritional Anemia||1.9%||2.6%||0.183||1.9%||2.7%||0.262|
| Congestive Heart Failure||0.5%||2.5%||0.000||0.1%||0.1%||1.000|
| Liver Disease or Cirrhosis||1.4%||1.6%||0.788||0.6%||0.8%||0.527|
| Bleeding Disorderd||0.6%||1.2%||0.165||0.5%||0.9%||0.248|
Co-Occurring comorbidities among unmatched outpatients
|Incidence||Comorbidities||Hypertension||Obesity||Diabetes without Complications||Chronic Pulmonary Disease||Obstructive Sleep Apnea||History of Smoking|
|11.0%||Diabetes without Complications||7.5%||3.2%||n/a||1.0%||2.0%||0.7%|
|7.4%||Chronic Pulmonary Disease||4.5%||1.9%||1.0%||n/a||1.2%||0.9%|
|7.6%||Obstructive Sleep Apnea||4.2%||2.7%||2.0%||1.2%||n/a||0.8%|
|5.2%||History of Smoking||3.5%||1.4%||0.7%||0.9%||0.8%||n/a|
|12.2%||Hypertension & Obesity||n/a||n/a||2.6%||1.3%||1.8%||1.1%|
|7.5%||Hypertension & Diabetes w out CC||n/a||2.6%||n/a||0.8%||1.3%||0.4%|
|4.5%||Hypertension & Chronic pulmonary disease||n/a||1.3%||0.8%||n/a||0.7%||0.8%|
|4.2%||Hypertension & Sleep Apnea||n/a||1.8%||1.3%||0.7%||n/a||0.7%|
|3.5%||Hypertension & Smoking||n/a||1.1%||0.4%||0.8%||0.7%||n/a|
Hospitalization characteristics and incidence of complications from day of admission through 90-day post-discharge follow-up period
|First Discharge Destination||< 0.0001||< 0.0001|
| Home Under Self-Care||94.5%||99.4%||94.6%||99.3%|
| Home with Health Aid||5.5%||0.6%||5.5%||0.7%|
|≥1 Physical Therapy Visit During Follow-upc||85.6%||84.6%||0.462||86.2%||85.3%||0.586|
| Opiate Pain Medication Use||80.0%||82.4%||0.076||80.4%||90.7%||< 0.0001|
|Minor Complicationsd||3.0%||6.2%||< 0.0001||2.8%||5.8%||0.002|
|Major Complicationse||5.4%||9.9%||< 0.0001||5.2%||6.7%||0.173|
| Nausea and Vomiting||2.9%||3.4%||0.464||3.0%||3.6%||0.500|
| Urinary Tract Infection||1.9%||3.7%||0.004||1.9%||2.8%||0.182|
| Any Infectionf||2.0%||1.6%||0.415||2.0%||1.4%||0.353|
| Transfusion (autologous or allogeneic)||0.2%||1.7%||< 0.0001||0.2%||1.7%||0.002|
| Deep Vein Thrombosis||0.9%||1.5%||0.175||1.0%||1.4%||0.513|
| Urinary or renal complications||0.5%||0.8%||0.550||0.6%||0.4%||0.480|
| Myocardial Infarction||0.1%||1.7%||< 0.0001||0.1%||0.9%||0.020|
| Hemorrhage, hematoma, seroma||0.8%||1.1%||0.617||0.6%||1.3%||0.134|
| Cardiac complications||0.1%||0.4%||0.360||0.1%||0.6%||0.103|
| Pulmonary Embolism||0.6%||0.6%||1.000||0.0%||0.0%||na|
| Wound disruption||0.1%||0.0%||0.288||0.1%||0.0%||0.317|
Total health plan payments (USD)
|Total Index Visit Paymentc|
| Mean (SD)||$23,929 ($15,332)||$26,811 ($12,363)||< 0.0001||$24,058 ($15,186)||$28,4856 ($11,170)||< 0.0001|
|Among those readmitted, Total Readmission Payment|
| Mean (SD)||$25,836 ($19,409)||$24,259 ($14,478)||0.779||$24,246 ($17,854)||$25,629 ($13,217)||0.273|
|Total 90-Day Coste|
| Mean (SD)||$28,548 ($18,068)||$30,780 ($14,939)||< 0.0001||$28,440 ($17,566)||$32,635 ($14,143)||< 0.0001|
|Post-Acute Care as % of total 90-day cost||15%||12%||0.007||15%||12%||0.007|