Preoperative prediction model for risk of readmission after total joint replacement surgery: a random forest approach leveraging NLP and unfairness mitigation for improved patient care and cost-effectiveness.

BACKGROUND: The Center for Medicare and Medicaid Services (CMS) imposes payment penalties for readmissions following total joint replacement surgeries. This study focuses on total hip, knee, and shoulder arthroplasty procedures as they account for most joint replacement surgeries. Apart from being a burden to healthcare systems, readmissions are also troublesome for patients. There are several studies which only utilized structured data from Electronic Health Records (EHR) without considering any gender and payor bias adjustments. METHODS: For this study, dataset of 38,581 total knee, hip, and shoulder replacement surgeries performed from 2015 to 2021 at Novant Health was gathered. This data was used to train a random forest machine learning model to predict the combined endpoint of emergency department (ED) visit or unplanned readmissions within 30 days of discharge or discharge to Skilled Nursing Facility (SNF) following the surgery. 98 features of laboratory results, diagnoses, vitals, medications, and utilization history were extracted. A natural language processing (NLP) model finetuned from Clinical BERT was used to generate an NLP risk score feature for each patient based on their clinical notes. To address societal biases, a feature bias analysis was performed in conjunction with propensity score matching. A threshold optimization algorithm from the Fairlearn toolkit was used to mitigate gender and payor biases to promote fairness in predictions. RESULTS: The model achieved an Area Under the Receiver Operating characteristic Curve (AUROC) of 0.738 (95% confidence interval, 0.724 to 0.754) and an Area Under the Precision-Recall Curve (AUPRC) of 0.406 (95% confidence interval, 0.384 to 0.433). Considering an outcome prevalence of 16%, these metrics indicate the model's ability to accurately discriminate between readmission and non-readmission cases within the context of total arthroplasty surgeries while adjusting patient scores in the model to mitigate bias based on patient gender and payor. CONCLUSION: This work culminated in a model that identifies the most predictive and protective features associated with the combined endpoint. This model serves as a tool to empower healthcare providers to proactively intervene based on these influential factors without introducing bias towards protected patient classes, effectively mitigating the risk of negative outcomes and ultimately improving quality of care regardless of socioeconomic factors.

Association of Physician Group Practice Participation in Bundled Payments With Patient Selection, Costs, and Outcomes for Joint Replacement.

Importance: Medicare's Bundled Payments for Care Improvement (BPCI) program, which ran from 2013 to 2018, was an important experiment in physician-focused alternative payment models. However, little is known about whether the program was associated with better quality or outcomes or lower costs. Objective: To determine whether participation in BPCI among physician group practices was associated with advantageous or deleterious changes in costs or patient outcomes. Design Setting and Participants: This cross-sectional study used 2013 to 2017 Medicare files and difference-in-differences (DID) models to compare the change over time in Medicare payments, patient selection, and clinical outcomes between 91 orthopedic groups in BPCI Model 2 and 169 propensity-matched controls for patients undergoing joint replacement. Analyses were performed between December 2019 and February 2021. Exposures: Voluntary participation in BPCI. Main Outcomes and Measures: The primary outcome was 90-day Medicare payments; secondary outcomes were patient selection (volume, comorbidities) and clinical outcomes (30-day and 90-day emergency department visits, readmissions, mortality, and healthy days at home). Results: There were 74 343 patient episodes in the baseline period and 102 790 during the intervention in BPCI practices, and 88 147 patient episodes in the baseline period and 120 253 during the intervention in control practices; 291 214 of 461 598 (63.1%) patients were women, and 419 619 (90.9%) were White. At baseline, mean episode payments among BPCI-participating practices were $18 257, which decreased to $15 320 during the intervention, while control practices decreased from $17 927 to $16 170 (DID, -$1180; 95% CI, -$1565 to -$795; P < .001). Savings were driven by a decrease in postacute care spending. There were no differential changes in volume or comorbidities. The BPCI practices increased the proportion of patients discharged home compared with controls (23.6% to 43.4% vs 22.2% to 31.8%; DID, 10.2% [95% CI, 6.2% to 14.1%]). There were no differential changes in 30-day or 90-day mortality rates or emergency department visits, but 30-day and 90-day readmission rates decreased more among BPCI practices than controls (90 days: 8.7% to 7.5% vs 8.9% to 8.7%; DID, -1.0% [95% CI, -1.4% to -0.5%]), and 90-day healthy days at home increased (BPCI, 82.9 to 84.8, vs controls, 83.1 to 84.4; DID, 0.6 [95% CI, 0.4 to 0.8]). Conclusions and Relevance: Group practice participation in BPCI for joint replacement was associated with reduced Medicare payments and improvements in clinical outcomes.

Low-cost procedure for deformity correction in rheumatoid hands.

BACKGROUND: Surgical correction of ulnar drift of metacarpo-phalangeal joint (MPJ) due to Rheumatoid arthritis (RA) is conventionally done by silicon joint arthroplasty which is expensive and may be associated with many complications. We report the outcome of low-cost autologous interpositional arthroplasty. MATERIAL AND METHODS: Five patients (8 hands, 32 arthroplasties) underwent correction of ulnar drift of MPJ by dorsal capsule interpositional arthroplasty. Results were assessed according to the degree of recovery of movement at the MPJ and correction of ulnar drift. Functional improvement was graded as excellent, good and fair. Pain alleviation was assessed by visual analogue score (VAS) score. RESULTS: Excellent results were seen in 3 patients (5 hands, 20 arthroplasties), good in 1 patient (2 hands, 8 arthroplasties) and fair in 1 patient (1 hand, 4 arthroplasties). VAS score for pain decreased from mean preoperative 8.2/10 to 1/10. On average follow up of 1.4 years there was good hand function, no recurrence of deformities and patients were pain free. CONCLUSION: Interpositional arthroplasty for MPJ using dorsal capsule for correction of post RA ulnar drift is a low-cost option which improves the hand function and cosmesis. Additionally, it avoids all the complications related with the use of silicon joints.

Association of Mandatory Bundled Payments for Joint Replacement With Postacute Care Outcomes Among Medicare and Medicaid Dual Eligible Patients.

IMPORTANCE: The Medicare comprehensive care for joint replacement (CJR) model, a mandatory bundled payment program started in April 2016 for hospitals in randomly selected metropolitan statistical areas (MSAs), may help reduce postacute care (PAC) use and episode costs, but its impact on disparities between Medicaid and non-Medicaid beneficiaries is unknown. OBJECTIVE: To determine effects of the CJR program on differences (or disparities) in PAC use and outcomes by Medicare-Medicaid dual eligibility status. DESIGN, SETTING, AND PARTICIPANTS: Observational cohort study of 2013-2017, based on difference-in-differences (DID) analyses on Medicare data for 1,239,452 Medicare-only patients, 57,452 dual eligibles with full Medicaid benefits, and 50,189 dual eligibles with partial Medicaid benefits who underwent hip or knee surgery in hospitals of 75 CJR MSAs and 121 control MSAs. MAIN OUTCOME MEASURES: Risk-adjusted differences in rates of institutional PAC [skilled nursing facility (SNF), inpatient rehabilitation, or long-term hospital care] use and readmissions; and for the subgroup of patients discharged to SNF, risk-adjusted differences in SNF length of stay, payments, and quality measured by star ratings, rate of successful discharge to community, and rate of transition to long-stay nursing home resident. RESULTS: The CJR program was associated with reduced institutional PAC use and readmissions for patients in all 3 groups. For example, it was associated with reductions in 90-day readmission rate by 1.8 percentage point [DID estimate=-1.8; 95% confidence interval (CI), -2.6 to -0.9; P<0.001] for Medicare-only patients, by 1.6 percentage points (DID estimate=-1.6; 95% CI, -3.1 to -0.1; P=0.04) for full-benefit dual eligibles, and by 2.0 percentage points (DID estimate=-2.0; 95% CI, -3.6 to -0.4; P=0.01) for partial-benefit dual eligibles. These CJR-associated effects did not differ between dual eligibles (differences in above DID estimates=0.2; 95% CI, -1.4 to 1.7; P=0.81 for full-benefit patients; and -0.3; 95% CI, -1.9 to 1.3; P=0.74 for partial-benefit patients) and Medicare-only patients. Among patients discharged to SNF, the CJR program showed no effect on successful community discharge, transition to long-term care, or their persistent disparities. CONCLUSIONS: The CJR program did not help reduce persistent disparities in readmissions or SNF-specific outcomes related to Medicare-Medicaid dual eligibility, likely due to its lack of financial incentives for reduced disparities and improved SNF outcomes.

Impact of Value-Based Reimbursement on Health-Care Disparities for Total Joint Arthroplasty Candidates.

As the U.S. Centers for Medicare & Medicaid Services (CMS) implements value-based reimbursement models based on predetermined outcome measures, access to total joint arthroplasty (TJA) is jeopardized for patients who are disproportionately affected by conditions that predispose them to higher odds of complications. Obesity, depression, and chronic illness, each of which occur at disproportionately higher rates in minorities or individuals in lower socioeconomic brackets, are individually associated with worse TJA postoperative outcomes, including longer hospital lengths of stay and higher rates of readmission within 90 days. Medicaid may even be considered an independent risk factor for worse outcome measures with TJA as enrollees have higher rates of postoperative mortality and complications and longer lengths of stay than patients on Medicare or with private insurance. As same-day discharge for TJA becomes more common, eligibility requirements for the procedure tighten, and existing disparities in access to the procedure will be further exacerbated. The current CMS uniform quality metrics endanger access to TJA for patients in certain racial and socioeconomic groups and oblige physicians who treat more complex patients to jeopardize their reimbursement.

Association of Skilled Nursing Facility Participation in a Bundled Payment Model With Institutional Spending for Joint Replacement Surgery.

Importance: Medicare recently concluded a national voluntary payment demonstration, Bundled Payments for Care Improvement (BPCI) model 3, in which skilled nursing facilities (SNFs) assumed accountability for patients' Medicare spending for 90 days from initial SNF admission. There is little evidence on outcomes associated with this novel payment model. Objective: To evaluate the association of BPCI model 3 with spending, health care utilization, and patient outcomes for Medicare beneficiaries undergoing lower extremity joint replacement (LEJR). Design, Setting, and Participants: Observational difference-in-difference analysis using Medicare claims from 2013-2017 to evaluate the association of BPCI model 3 with outcomes for 80 648 patients undergoing LEJR. The preintervention period was from January 2013 through September 2013, which was 9 months prior to enrollment of the first BPCI cohort. The postintervention period extended from 3 months post-BPCI enrollment for each SNF through December 2017. BPCI SNFs were matched with control SNFs using propensity score matching on 2013 SNF characteristics. Exposures: Admission to a BPCI model 3-participating SNF. Main Outcomes and Measures: The primary outcome was institutional spending, a combination of postacute care and hospital Medicare-allowed payments. Additional outcomes included other categories of spending, changes in case mix, admission volume, home health use, length of stay, and hospital use within 90 days of SNF admission. Results: There were 448 BPCI SNFs with 18 870 LEJR episodes among 16 837 patients (mean [SD] age, 77.5 [9.4] years; 12 173 [72.3%] women) matched with 1958 control SNFs with 72 005 LEJR episodes among 63 811 patients (mean [SD] age, 77.6 [9.4] years; 46 072 [72.2%] women) in the preintervention and postintervention periods. Seventy-nine percent of matched BPCI SNFs were for-profit facilities, 85% were located in an urban area, and 85% were part of a larger corporate chain. There were no systematic changes in patient case mix or episode volume between BPCI-participating SNFs and controls during the program. Institutional spending decreased from $17 956 to $15 746 in BPCI SNFs and from $17 765 to $16 563 in matched controls, a differential decrease of 5.6% (-$1008 [95% CI, -$1603 to -$414]; P < .001). This decrease was related to a decline in SNF days per beneficiary (from 26.2 to 21.3 days in BPCI SNFs and from 26.3 to 23.4 days in matched controls; differential change, -2.0 days [95% CI, -2.9 to -1.1]). There was no significant change in mortality or 90-day readmissions. Conclusions and Relevance: Among Medicare patients undergoing lower extremity joint replacement from 2013-2017, the BPCI model 3 was significantly associated with a decrease in mean institutional spending on episodes initiated by admission to SNFs. Further research is needed to assess bundled payments in other clinical contexts.

Cost-Effectiveness Analysis of Motion-Preserving Operations for Wrist Arthritis.

BACKGROUND: The authors conducted a cost-effectiveness analysis to answer the question: Which motion-preserving surgical strategy, (1) four-corner fusion, (2) proximal row carpectomy, or (3) total wrist arthroplasty, used for the treatment of wrist osteoarthritis, is the most cost-effective? METHODS: A simulation model was created to model a hypothetical cohort of wrist osteoarthritis patients (mean age, 45 years) presenting with painful wrist and having failed conservative management. Three initial surgical treatment strategies-(1) four-corner fusion, (2) proximal row carpectomy, or (3) total wrist arthroplasty-were compared from a hospital perspective. Outcomes included clinical outcomes and cost-effectiveness outcomes (quality-adjusted life-years and cost) over a lifetime. RESULTS: The highest complication rates were seen in the four-corner fusion cohort: 27.1 percent compared to 20.9 percent for total wrist arthroplasty and 17.4 percent for proximal row carpectomy. Secondary surgery was common for all procedures: 87 percent for four-corner fusion, 57 percent for proximal row carpectomy, and 46 percent for total wrist arthroplasty. Proximal row carpectomy generated the highest quality-adjusted life-years (30.5) over the lifetime time horizon, compared to 30.3 quality-adjusted life-years for total wrist arthroplasty and 30.2 quality-adjusted life-years for four-corner fusion. Proximal row carpectomy was the least costly; the mean expected lifetime cost for patients starting with proximal row carpectomy was $6003, compared to $11,033 for total wrist arthroplasty and $13,632 for four-corner fusion. CONCLUSIONS: The authors' analysis suggests that proximal row carpectomy was the most cost-effective strategy, regardless of patient and parameter level uncertainties. These are important findings for policy makers and clinicians working within a universal health care system.

Are current DRG-based bundled payment models for revision total joint arthroplasty risk-adjusting adequately?

AIMS: Currently, the US Center for Medicaid and Medicare Services (CMS) has been testing bundled payments for revision total joint arthroplasty (TJA) through the Bundled Payment for Care Improvement (BPCI) programme. Under the BPCI, bundled payments for revision TJAs are defined on the basis of diagnosis-related groups (DRGs). However, these DRG-based bundled payment models may not be adequate to account appropriately for the varying case-complexity seen in revision TJAs. METHODS: The 2008-2014 Medicare 5% Standard Analytical Files (SAF5) were used to identify patients undergoing revision TJA under DRG codes 466, 467, or 468. Generalized linear regression models were built to assess the independent marginal cost-impact of patient, procedural, and geographic characteristics on 90-day costs. RESULTS: A total of 9,263 patients (DRG-466 = 838, DRG-467 = 4,573, and DRG-468 = 3,842) undergoing revision TJA from 2008 to 2014 were included in the study. Undergoing revision for a dislocation (+$1,221), periprosthetic fracture (+$4,454), and prosthetic joint infection (+$5,268) were associated with higher 90-day costs. Among comorbidities, malnutrition (+$10,927), chronic liver disease (+$3,894), congestive heart failure (+$3,292), anaemia (+$3,149), and coagulopathy (+$2,997) had the highest marginal cost-increase. The five US states with the highest 90-day costs were Alaska (+$14,751), Maryland (+$13,343), New York (+$7,428), Nevada (+$6,775), and California (+$6,731). CONCLUSION: Under the proposed DRG-based bundled payment methodology, surgeons would be reimbursed the same amount of money for revision TJAs, regardless of the indication (periprosthetic fracture, prosthetic joint infection, mechanical loosening) and/or patient complexity. Cite this article: Bone Joint J 2020;102-B(7):959-964.

Long-term effect of patient decision aids on use of joint replacement and health care costs.

OBJECTIVE: Shared decision-making supported by patient decisions aids may improve care and reduce healthcare costs for persons considering total joint replacement. Observational studies and randomized controlled trials (RCTs) have evaluated the short-term impact of decision aids on uptake of surgery and costs, however the long-term effects are unclear. This analysis aimed to evaluate the effect of patient decision aids on 1) use of joint replacement up to 7-years of follow-up, and 2) osteoarthritis-related health system costs. METHODS: 324 participants in a Canadian RCT with 2-years follow-up who were randomized to either a decision aid (n = 161) or usual care (n = 163) had their trial and health administrative data linked. The proportion undergoing surgery up to 7-years were compared using cumulative incidence plots and competing risk regression. Mean per-patient costs were compared using two sample t-tests. RESULTS: At 2-years, 119 of 161 (73.9%) patients in the decision aid arm and 129 of 163 (79.1%) patients in the usual care arm had surgery. Between two and 7-years, 17 additional patients in both the decision aid (of 42, 40.4%) and usual care (of 34, 50.0%) arms underwent surgery. At 7-years, patients exposed to decision aids had a similar likelihood of undergoing surgery (HR = 0.92, 95% CI:0.73 to 1.17, p = 0.49) and mean per-patient costs ($21,965 vs $23,681, incremental cost: -$1,717, 95% CI:-$5,631 to $2,198) compared to those in usual care. CONCLUSIONS: This is the first study to assess the long-term impact of decision aids on use of joint replacement and healthcare costs. These results are not conclusive but can inform future trial design. CLINICAL TRIAL REGISTRATION: The full trial protocol is available at ClinicalTrials.Gov (NCT00911638).

Challenges to Implementing Total Joint Replacement Programs in Developing Countries.

In this review article, the authors present the many challenges that orthopedic surgeons in developing countries face when implementing arthroplasty programs. The issues of cost, sterility, and patient demographics are specifically addressed. Despite the many challenges, developing countries are beginning to offer hip and knee reconstructive surgery to respond to the increasing demand for such elective operations as the prevalence of osteoarthritis continues to increase. The authors shed light on these nascent arthroplasty programs.

International Collaboration in Total Joint Arthroplasty: A Framework for Establishing Meaningful International Alliances.

Despite the increase in utilization of total joint arthroplasty (TJA) throughout high-income countries, there is a lack of access to basic surgical care, including TJA, in low- and middle-income countries (LMICs). Multiple strategies, including short-term surgical trips, establishment of local TJA centers, and education-based international academic collaborations, have been used to bridge the gap in access to quality TJA. The authors review the obstacles to providing TJA in LMICs, the outcomes of the 3 strategies in use to bridge gaps, and a framework for the establishment and maintenance of meaningful international collaborations.

Assessing Preoperative Risk Factors With Sex Disparities in Total Joint Arthroplasty Patients and Financial Outcomes From the National Inpatient Sample Database.

INTRODUCTION: Disparities in the healthcare system imply potential risks for vulnerable groups whose needs are not appropriately met. Total joint arthroplasty (TJA) is successful in treating end-stage arthritis, resulting in increased demand for the procedure, however remains underused in both sexes, especially in women. Although multiple studies assessed the differences in postoperative morbidities between sexes, there remains a lack in understanding patients' preoperative clinical profile and nonclinical demographics. The aim of this study is to provide a population-based epidemiologic assessment of preoperative risk factors and sex disparities and assess differences in outcomes following TJA. METHODS: The National Inpatient Sample database from 2006 to 2011 was analyzed. Patients who underwent primary total knee and hip arthroplasty were identified and stratified into two cohorts of male and female, and demographic data and comorbidities were collected. Postoperative complications, length of stay, total charges, and discharge destination were measured for matched cohorts. RESULTS: Female patients present for TJAs at an older average age, are less likely to present with AIDS, alcohol abuse, coagulopathy, congestive heart failure, drug abuse, liver disease, peripheral vascular disease, and renal failure, and are more likely to present with anemia, autoimmune disorders, chronic obstructive pulmonary disease, depression, obesity, and valvular disease. Postoperatively, the average length of stay for female patients was markedly higher (3.52 versus 3.39) and a lower percentage went home (59% versus 73%). Overall, female patients experience greater odds of any complication while in-patient. DISCUSSION: This study highlighted sex differences in areas that could account for the underuse of the procedure in both sexes, with women affected to a greater extent. Understanding these factors will help address the unmet needs of both sexes after TJA by encouraging future studies and provider education to ensure that all patients are able to access the necessary procedures for pain relief and functional improvement.

Payments for outpatient joint replacement surgery: A comparison of hospital outpatient departments and ambulatory surgery centers.

OBJECTIVE: To compare commercial insurance payments for outpatient total knee and hip replacement surgeries performed in hospital outpatient departments (HOPDs) and in ambulatory surgery centers (ASCs). DATA SOURCES: A large national claims database that contains information on actual prices paid to providers over the period 2014-2017. DATA COLLECTION: We identified all patients receiving total knee replacement surgery and total hip replacement surgery in HOPDs and in ASCs for each of the 4 years. STUDY DESIGN: For each year, we conducted descriptive and statistical patient-level analyses of the facility component of payments to HOPDs and to ASCs. PRINCIPAL FINDINGS: For each procedure and for each year, ASC payments exceeded HOPD payments by a wide margin; however, the gap across settings declined over time. In 2014, knee replacement payments to HOPDs (n = 67) were $6016 compared to $23 244 in ASCs (n = 68). By 2017, payments to HOPDs (n = 223) had grown to $10 060 compared to $18 234 in ASCs (n = 602). Similarly, for hip replacements, HOPD payments (n = 43) rose from $6980 in 2014 to $11 139 in 2017 (n = 206) and in ASCs fell from $28 485 in 2014 (n = 82) to $18 595 in 2017 (n = 465). CONCLUSIONS: Results suggest that for total joint replacement, common perceptions of cost savings from transition of services from hospitals to ASCs may be misguided.

Do Bundled Payment Programs in Joint Replacement Care Hold Promise for Improving Patient Outcomes?

The Centers for Medicare and Medicaid Services (CMS) Innovation Center offers two alternative payment models for joint replacement: the voluntary Bundled Payment for Care Improvement (BPCI) model and the mandatory Comprehensive Care for Joint Replacement (CJR) model. As CMS considers methods for cost reduction, research is needed to understand patient-level outcomes and organizational-level success factors. A retrospective cross-sectional study of hospitals was performed, using regression models to evaluate an aggregate patient satisfaction score, complication rates, and operational differences among BPCI, CJR, and nonparticipating hospitals. Results show that BPCI hospitals received significantly better patient satisfaction scores (88.6) than CJR hospitals (86.0), but complication rates were not significantly different between CJR and BPCI hospitals (2.83 and 2.77, respectively). Factors associated with BPCI participation include academic affiliation, a Northeast region locale, and having a higher CMS efficiency score. Thus, requiring more hospitals to participate in CMS-bundled payment programs as a federal policy may not be the optimal way to improve patient satisfaction and outcomes. Rather, the CJR and BPCI programs should be further studied, and the results generalized for use by nonparticipating hospitals to encourage preparation and participation in CMS value-based initiatives.

Variation in the Cost of Care for Different Types of Joint Arthroplasty.

BACKGROUND: Lower-extremity arthroplasty constitutes the largest burden on health-care spending of any Medicare diagnosis group. Demand for upper extremity arthroplasty also continues to rise. It is necessary to better understand costs as health care shifts toward a bundled-payment accounting approach. We aimed (1) to identify whether variation exists in total cost for different types of joint arthroplasty, and, if so, (2) to determine which cost parameters drive this variation. METHODS: The cost of the episode of inpatient care for 22,215 total joint arthroplasties was calculated by implementing time-driven activity-based costing (TDABC) at a single orthopaedic specialty hospital from 2015 to 2018. Implant price, supply costs, personnel costs, and length of stay for total knee, total hip, anatomic total shoulder, reverse total shoulder, total elbow, and total ankle arthroplasty were analyzed. Individual cost parameters were compared with total cost and volume. RESULTS: Higher implant cost appeared to correlate with higher total costs and represented 53.8% of the total cost for an inpatient care cycle. Total knee arthroplasty was the least-expensive and highest-volume procedure, whereas total elbow arthroplasty had the lowest volume and highest cost (1.65 times more than that of total knee arthroplasty). Length of stay was correlated with increased personnel cost but did not have a significant effect on total cost. CONCLUSIONS: Total inpatient cost at our orthopaedic specialty hospital varied by up to a factor of 1.65 between different fields of arthroplasty. The highest-volume procedures-total knee and hip arthroplasty-were the least expensive, driven predominantly by lower implant purchase prices. CLINICAL RELEVANCE: We are not aware of any previous studies that have accurately compared cost structures across upper and lower-extremity arthroplasty with a uniform methodology. The present study, because of its uniform accounting process, provides reliable data that will allow clinicians to better understand cost relationships between different procedures.