Transcultural validation of the Risk Assessment and Predictor Tool (RAPT) to predict discharge outcomes after total hip replacement

Trends in the Adoption of Outpatient Joint Arthroplasties and Patient Risk: A Retrospective Analysis of 2019 to 2021 Medicare Claims Data

INTRODUCTION

Total joint arthroplasties (TJAs) have recently been shifting toward outpatient arthroplasty. This study aims to explore recent trends in outpatient total joint arthroplasty (TJA) procedures and examine whether patients with a higher comorbidity burden are undergoing outpatient arthroplasty.

METHODS

Medicare fee-for-service claims were screened for patients who underwent total hip, knee, or shoulder arthroplasty procedures between January 2019 and December 2022. The procedure was considered to be outpatient if the patient was discharged on the same date of the procedure. The Hierarchical Condition Category Score (HCC) and the Charlson Comorbidity Index (CCI) scores were used to assess patient comorbidity burden. Patient adverse outcomes included all-cause hospital readmission, mortality, and postoperative complications. Logistic regression analyses were used to evaluate if higher HCC/CCI scores were associated with adverse patient outcomes.

RESULTS

A total of 69,520, 116,411, and 41,922 respective total knee, hip, and shoulder arthroplasties were identified, respectively. Despite earlier removal from the inpatient-only list, outpatient knee and hip surgical volume did not markedly increase until the pandemic started. By 2022Q4, 16%, 23%, and 36% of hip, knee, and shoulder arthroplasties were discharged on the same day of surgery, respectively. Both HCC and CCI risk scores in outpatients increased over time (P < 0.001).

DISCUSSION

TJA procedures are shifting toward outpatient surgery over time, largely driven by the COVID-19 pandemic. TJA outpatients' HCC and CCI risk scores increased over this same period, and additional research to determine the effects of this should be pursued.

LEVEL OF EVIDENCE

Level III, therapeutic retrospective cohort study.

Use of The Risk Assessment and Prediction Tool to Predict Same-day Discharge After Primary Hip and Knee Arthroplasty

INTRODUCTION

The Risk Assessment and Prediction Tool (RAPT) is a preoperative screening tool developed to predict discharge disposition after total hip arthroplasty (THA) and total knee arthroplasty (TKA), but its predictive value for same-day discharge (SDD) has not been investigated. The aims of this study were (1) to assess RAPT's ability to predict SDD after primary THA and TKA and (2) to determine a cutoff RAPT score that may recognize patients appropriate for SDD.

METHODS

Data were retrospectively collected from patients undergoing primary THA and TKA at a single tertiary care center between February 2020 and May 2021. A receiver operating characteristic curve was generated to choose a cutoff value to screen for SDD. Logistic regression analysis was done to identify factors including age, BMI, or RAPT score that may be associated with SDD.

RESULTS

Three hundred sixty-one patients with preoperative RAPT scores were included in the analysis of whom 147 (42.6%) underwent SDD. A cutoff of ≥9 was identified for TKA and ≥11 for THA. RAPT had a predictive accuracy of only 66.7% for SDD, whereas the discharge plan documented in the preoperative note was 91.7% accurate.

DISCUSSION

Although there is a positive association between RAPT and SDD, it is not a useful screening tool given its low predictive accuracy.

Six Externally Validated Prognostic Models Have Potential Clinical Value to Predict Patient Health Outcomes in the Rehabilitation of Musculoskeletal Conditions: A Systematic Review

OBJECTIVE

The purpose of this systematic review was to identify and appraise externally validated prognostic models to predict a patient's health outcomes relevant to physical rehabilitation of musculoskeletal (MSK) conditions.

METHODS

We systematically reviewed 8 databases and reported our findings according to Preferred Reporting Items for Systematic Reviews and Meta-Analysis 2020. An information specialist designed a search strategy to identify externally validated prognostic models for MSK conditions. Paired reviewers independently screened the title, abstract, and full text and conducted data extraction. We extracted characteristics of included studies (eg, country and study design), prognostic models (eg, performance measures and type of model) and predicted clinical outcomes (eg, pain and disability). We assessed the risk of bias and concerns of applicability using the prediction model risk of bias assessment tool. We proposed and used a 5-step method to determine which prognostic models were clinically valuable.

RESULTS

We found 4896 citations, read 300 full-text articles, and included 46 papers (37 distinct models). Prognostic models were externally validated for the spine, upper limb, lower limb conditions, and MSK trauma, injuries, and pain. All studies presented a high risk of bias. Half of the models showed low concerns for applicability. Reporting of calibration and discrimination performance measures was often lacking. We found 6 externally validated models with adequate measures, which could be deemed clinically valuable [ie, (1) STart Back Screening Tool, (2) Wallis Occupational Rehabilitation RisK model, (3) Da Silva model, (4) PICKUP model, (5) Schellingerhout rule, and (6) Keene model]. Despite having a high risk of bias, which is mostly explained by the very conservative properties of the PROBAST tool, the 6 models remain clinically relevant.

CONCLUSION

We found 6 externally validated prognostic models developed to predict patients' health outcomes that were clinically relevant to the physical rehabilitation of MSK conditions.

IMPACT

Our results provide clinicians with externally validated prognostic models to help them better predict patients' clinical outcomes and facilitate personalized treatment plans. Incorporating clinically valuable prognostic models could inherently improve the value of care provided by physical therapists.

RAPT Scores Predict Inpatient Versus Outpatient Status and Readmission Rates After IPO Changes for Total Joint Arthroplasty: An Analysis of 12,348 Cases

BACKGROUND

Changes to Medicare's Inpatient Only List (IPO) and factors associated with the COVID pandemic have led to more total joint arthroplasty (TJA) patients to be designated as outpatient (OP). This potentially complicates postoperative care for patients with lower functional status and poor social support. These factors make the decision between OP versus inpatient (IP) designation particularly challenging for healthcare teams. The Risk Assessment and Prediction Tool (RAPT) was designed to indicate patient risk for needing posthospital discharge to facility and considers social and functional factors. The purpose of this study is to 1) evaluate the correlation of RAPT as a clinical tool to aid decision-making regarding OP versus IP for Total Hip and Knee Arthroplasty (THA and TKA), 2) assess the impact of recent changes to the IPO and the COVID pandemic on OP TJA readmission rates, and 3) determine whether 90-day readmissions are correlated with RAPT scores after OP TJA.

METHODS

We identified all elective TKA and THA patients from 2015 through 2021 in our electronic health record at our large, urban, academic health system. Fracture patients were excluded. For those patients with available RAPT scores, we determined OP and IP designations, with IP defined as those with length-of-stay 2 midnights or more. We performed subanalysis of OP between same-day and next-day discharge. RAPT scores and readmission rates were compared at time points related to changes in the IPO: before TKA removal in 2018 (period A), from 2018 until THA removal in 2020 (Period B), and after January 1, 2020, inclusive of impact from the COVID pandemic (Period C).

RESULTS

Reviewed were 11,819 elective TKAs and 10,212 elective THAs. RAPT scores were available for 6,759 TKA patients and 5,589 THA patients. For both TKA and THA, RAPT scores between IP, same-day, and next-day discharged OP were significantly different across all time periods (P < .001). The percentage of OP designation increased across all time periods for TKA and THA. Over these same time periods, mean RAPT scores decreased significantly for both OP TKA and OP THA (P < .01). Concurrent with these changes were significant increases in OP THA 90-day readmission rates across Periods A and B (P = .010) as well as A and C (P = .006). Readmitted OP TKA had significantly lower RAPT scores than OP TKA without readmission during Period B (P < .001). Readmitted OP THA had significantly lower RAPT scores than those without readmission for all periods (P < .05). To facilitate clinical utility, median RAPT scores were also analyzed, and showed that RAPT scores for OP THA patients with readmission were 1 to 2 points lower for all time periods.

CONCLUSION

RAPT scores correlate with IP versus OP status for both TKA and THA and vary significantly with same-day versus next-day discharge. OP TJA RAPT scores may also help predict readmission, and counsel some patients away from OP surgery. Average RAPT scores of 10, 9, and 8 appeared to be separators for same day, next day, and inpatient stay. Changes to the IPO and COVID pandemic correlate with decrease in RAPT scores for both TKA and THA patients within all designations. In addition, a shift toward lower RAPT for OP TJA correlates with increased 90-day readmission rates for OP TJA. Taken together, these results suggest that patients with poorer function and worse social support systems are increasingly being driven toward OP surgery by these changes, which may play a role in increasing readmission rates. Social support and functional factors should be considered for OP elective TKA and THA. Further, any OP TJA value-based payment system must account for these variables.

Comparative Analysis of the Ability of Machine Learning Models in Predicting In-hospital Postoperative Outcomes After Total Hip Arthroplasty

BACKGROUND

Machine learning (ML) methods have shown promise in a wide range of applications including the development of patient-specific predictive models before surgical interventions. The purpose of this study was to develop, test, and compare four distinct ML models to predict postoperative parameters after primary total hip arthroplasty.

METHODS

Data from the Nationwide Inpatient Sample were used to identify patients undergoing total hip arthroplasty from 2016 to 2017. Linear support vector machine (LSVM), random forest (RF), neural network (NN), and extreme gradient boost trees (XGBoost) predictive of mortality, length of stay, and discharge disposition were developed and validated using 15 predictive patient-specific and hospital-specific factors. Area under the curve of the receiver operating characteristic (AUCROC) curve and accuracy were used as validity metrics, and the strongest predictive variables under each model were assessed.

RESULTS

A total of 177,442 patients were included in this analysis. For mortality, the XGBoost, NN, and LSVM models all had excellent responsiveness during validation while RF had fair responsiveness. LSVM had the highest responsiveness with an AUCROC of 0.973 during validation. For the length of stay, the LSVM and NN models had fair responsiveness while the XGBoost and random forest models had poor responsiveness. LSVM had the highest responsiveness with an AUCROC of 0.744 during validation. For the discharge disposition outcome, LSVM had good responsiveness while the XGBoost, NN, and RF models all had fair responsiveness. LSVM had the highest responsiveness with an AUCROC of 0.801.

DISCUSSION

The ML methods tested demonstrated a range of poor-to-excellent responsiveness and accuracy in the prediction of the assessed metrics, with LSVM being the best performer. Such models should be further developed, with eventual integration into clinical practice to inform patient discussions and management decision making, with the potential for integration into tiered bundled payment models.

RAPT score and preoperative factors to predict discharge location following adult spinal deformity surgery

PURPOSE

To assess factors, including RAPT score, predictive of non-home discharges following adult spinal deformity (ASD) operations.

METHODS

Adults who underwent thoracolumbar instrumented fusions to the pelvis for ASD (1/2019-1/2020) were reviewed. Patient demographics, RAPT metrics, hospital length of stay (LOS), operative details, and complications were compared between patients discharged home and non-home. Univariate and multivariate analyses were performed using logistic regression to determine the relative risk of non-home discharge. Area Under the Receiver Operating Characteristic curve (AUROC) for RAPT score and non-home discharge was also determined.

RESULTS

Ninety-nine patients (average age 68 ± 9 years; female-64; average RAPT 8.6 ± 2.2) were analyzed. Operations had the following characteristics: average # levels fused 11 ± 3, revisions 54%, anterior-posterior 70%, 3-column osteotomies 23%. Average LOS was 8.5 ± 3.6 days. The majority of patients (75.8%) had non-home discharges. Non-home discharges had significantly lower RAPT scores (8.3 vs. 9.6; p = 0.02), more advanced age (70 vs. 63 years; p = 0.01), and higher Charlson Comorbidity Index (CCI) scores (3.6 vs. 2.5; p < 0.01) compared to home discharges. On univariate analysis, factors significantly associated with non-home discharge were older age [relative risk (RR) 1.09, p < 0.01], higher CCI (RR 1.73, p = 0.01), total # levels fused (RR 1.24, p = 0.04), and lower RAPT scores (RR 0.71, p = 0.01). RAPT score < 8 was most predictive of non-home discharge (RR 4.87, p = 0.04). An AUROC relating RAPT scores and non-home discharge was 0.7.

CONCLUSIONS

Non-home discharges after ASD operations are common. Of the four factors associated with non-home discharges (elderly age, higher CCI, total number of levels fused, RAPT score), a RAPT score < 8 was most predictive. The RAPT score holds promising utility for pre-operative patient counseling and discharge planning for adults undergoing operations for spinal deformity.

Predictive Value of the Risk Assessment and Prediction Tool (RAPT) Score for Primary Hip and Knee Arthroplasty Patients: A Single-Center Study

The Risk Assessment and Prediction Tool (RAPT) was developed to predict patient discharge destination for arthroplasty operations. However, since Enhanced Recovery After Surgery (ERAS) programs have been utilized in the UK, the RAPT score has not been validated for use. The aim of the current study was to evaluate the predictive validity of the RAPT score in an ERAS environment with short length of stay. Data were compiled from 545 patients receiving a primary elective total hip or total knee arthroplasty in a district general hospital over 12 months. RAPT scores, length of stay, and discharge destinations were recorded. Patients were classified as low, intermediate, or high risk as per their RAPT score. Length of stay was significantly different between groups (p = 0.008), with low-risk patients having shorter length of stay. However, RAPT scores did not predict discharge destination; the overall correct prediction was only 31.9%. Furthermore, the most likely discharge destination was directly home in ≤3 days in all groups (68.5%, 60.2%, and 40% for the low-, intermediate-, and high-risk groups, respectively). The RAPT score is not an adequate tool to predict the discharge disposition following primary total knee and hip replacement surgery in a UK hospital with a standardized modern ERAS program. Alternative predictive tools are required.

An Electronic Medical Record-Based Discharge Disposition Tool Gets Bundle Busted: Decaying Relevance of Clinical Data Accuracy in Machine Learning

Background

Determining discharge disposition after total joint arthroplasty (TJA) has been a challenge. Advances in machine learning (ML) have produced computer models that learn by example to generate predictions on future events. We hypothesized a trained ML algorithm’s diagnostic accuracy will be better than that of current predictive tools to predict discharge disposition after primary TJA.

Methods

This study was a retrospective cohort study from a single, tertiary referral center for primary TJA. We trained and validated an artificial neural network (ANN) based on 4368 distinct surgical encounters between 1/1/2013 and 6/28/2016. The ANN’s ability to identify discharge disposition was then tested on 1452 distinct surgical encounters between 1/3/17 and 11/30/17.

Results

The area under the curve and accuracy achieved during model validation were 0.973 and 91.7%, respectively, with 25% of patients being discharged to skilled nursing facilities (SNFs). Within our testing data set, 6.7% of patients went to SNFs. The performance in the testing set included an area under the curve of 0.804, accuracy of 61.3%, sensitivity of 28.9%, and specificity of 93.8%.

Conclusions

This is the first prediction tool using an electronic medical record–integrated ANN to predict discharge disposition after TJA based on locally generated data. Dramatically reduced numbers of patients discharged to SNFs due to implementation of a bundled payment model lead to poor recall in the testing model. This model serves as a proof of concept for developing an ML prediction tool using a relatively small data set and subsequent integration into the electronic medical record.

The Risk Assessment and Prediction Tool (RAPT) after Hip and Knee Replacement: A Systematic Review

Purpose  The Risk Assessment and Prediction Tool (RAPT) is an interesting instrument for predicting the discharge destination and length of stay (LOS) for patients after hip or knee arthroplasty. The aim of this review is to describe its predictive ability, current utilization, and future prospects through the analysis of scientific literature.

Methods  The databases of PubMed, Web of Sciences, Cochrane Library, and Pedro were searched for English studies on RAPT prediction capacity. Only original prospective or retrospective articles that analyze specifically the use of RAPT were included, whereas those concerned with other preoperative prediction tools or those only considering other aspects of recovery after joint replacements were excluded.

Results  A total of 27 references were retrieved, and 8 studies were selected. All analyzed studies demonstrated that RAPT could reduce LOS and accurately predict discharge disposition especially for high- and low-risk patients. In the intermediate risk category, a targeted intensive postoperative rehabilitation program has demonstrated good results in reducing the uncertain outcome.

Conclusion  Although contrarily to many of the other scores, the RAPT has been validated in multiple countries with relatively similar results between different institutions; however, its validity has yet to be tested and adapted in every nation context. Further studies confirming the predictive accuracy of RAPT at other institutions are needed as well as studies assessing the effect of using RAPT to identify patients for targeted interventions in terms of LOS, discharge disposition, clinical outcomes, and financial impact.

Level of Evidence  This is a level IV, systematic review of level III and IV study.

Are the Cumulated Ambulation Score and Risk Assessment and Prediction Tool useful for predicting discharge destination and length of stay following total knee arthroplasty?

BACKGROUND

Postoperative management of patients undergoing total knee arthroplasty (TKA) is continually changing. Costs related to TKA are driven by implant cost, operating room cost, hospital length of stay (LOS), and rehabilitation approach. Discharges to rehabilitation centers have declined significantly in recent years.

AIM

We evaluated the usefulness of the Cumulated Ambulation Score (CAS) and Risk Assessment and Prediction Tool (RAPT) to predict discharge destination and estimate hospital LOS of patients undergoing TKA.

DESIGN

Prospective cohort study.

SETTING

University hospital inpatients.

POPULATION

Patients undergoing elective primary TKA.

METHODS

Consecutive patients were prospectively evaluated. Outcome measures were discharge destination and LOS dichotomized at the median (LOS<8 versus LOS≥8). Patients completed five outcome questionnaires and knee range of motion was measured preoperatively. RAPT was considered continuous, and also dichotomized (RAPT≤9 versus RAPT>9; RAPT9). CAS was dichotomized (CAS<11 versus CAS≥11; CAS11). Surgical technique and aftercare were similar for all patients.

RESULTS

Sixty-four patients (37 females), mean age 69.3±10.2 years were evaluated. CAS11 and discharge destination were strongly associated: 75.9% of patients with CAS≥11 were discharged home; 85.7% of patients with CAS<11 were discharged to a rehabilitation center (P<0.001). 80.7% of patients with RAPT≤9 were discharged to a rehabilitation center, versus 36.4% of patients with RAPT>9 (P=0.002). Odds ratios for discharge home were 18.9 (CAS11) and 7.3 (RAPT). CAS11 and RAPT were not related to LOS.

CONCLUSIONS

The CAS and RAPT can assist clinicians in estimating the discharge destination and developing patient care plans following TKA. However, predicting LOS with such tools alone was inaccurate.

CLINICAL REHABILITATION IMPACT

Use of the CAS and RAPT can inform discharge destination and patient care plans following TKA and has the potential to optimise resources and costs. However, due to social and organizational constraints on discharge, predicting LOS with such tools alone revealed to be inaccurate.

The Risk Assessment and Prediction Tool Is Less Accurate in Extended Length of Stay Patients Following Total Joint Arthroplasty

BACKGROUND

Although preoperative risk assessment tools have been effective in predicting discharge disposition after total joint arthroplasty (TJA), studies reporting on discharge planning in extended length of stay (ELOS), >3 days, patients are lacking. The purpose of this study was to describe the predictive utility of the Risk Assessment and Prediction Tool (RAPT) for discharge disposition in ELOS patients.

METHODS

Our study included 260 patients with LOS >3 days who underwent primary TJA between 2014 and 2016. Patients were separated into 3 cohorts, based on their RAPT score: low risk (9-12), medium risk (6-9), and high risk for discharge to a facility (1-6). Scores were compared among cohorts and correlated with discharge disposition for patients who stayed beyond 3 days.

RESULTS

In ELOS, RAPT had a higher utility in predicting discharge disposition in the low-risk (76.5% to home) and high-risk (62.9% to facility) patient cohorts, while medium-risk patients (56.5% to home) were the least accurate. Responses that significantly correlated with discharge home included male gender (odds ratio [OR], 1.81; P < .05), ambulation without walking aids (OR, 2.94; P < .01) or a single-point cane (OR, 2.95; P < .0001), <1 community support visit per week preoperatively (OR, 1.86; P < .05), and having support from someone at home (OR, 3.43; P < .0001).

CONCLUSION

The RAPT score in ELOS patients is better correlated with the low-risk and high-risk cohorts than in medium-risk patients. Conversely, medium-risk ELOS patients constituted 56.8% of our sample size, but only predicted 56.5% of discharge dispositions correctly. Future discharge disposition risk assessment tools are needed to stratify medium-risk patients.

Mixed qualitative and quantitative approach for validating an information booklet before total hip arthroplasty

OBJECTIVE

Providing patients with validated information before total hip arthroplasty may help lessen discrepancies between patients' expectations and the surgical result. This study sought to validate an information booklet for candidates for hip arthroplasty by using a mixed qualitative and quantitative approach based on a panel of patients and a sample of healthcare professionals.

METHODS

We developed a booklet in accordance with the standard methods and then conducted focus groups to collect the opinions of a sample of multidisciplinary experts involved in the care of patients with hip osteoarthritis. The number of focus groups and experts was determined according to the data saturation principle. A panel of patients awaiting hip arthroplasty or those in the immediate post-operative period assessed the booklet with self-reporting questionnaires (knowledge, beliefs, and expectations) and semi-structured interviews.

RESULTS

All experts and both patient groups validated the booklet in terms of content and presentation. Semi-structured interviews were uninformative, especially for post-operative patients. Reading the booklet significantly (P<0.001) improved the knowledge scores in both groups, with no intergroup differences, but did not affect beliefs in either patient group. Only pre-operative patients significantly changed their expectations.

CONCLUSION

Our mixed qualitative and quantitative approach allowed us to validate a booklet for patients awaiting hip arthroplasty, taking into account the opinions of both patients and healthcare professionals.