Predict Discharge Destination Research Articles

The cutoff values of functional independence measure scores for predicting discharge destination in the early stroke phase.

This study aimed to determine cutoff values of functional independence measure (FIM) scores to predict the discharge destinations of patients with acute stroke. The sample included 318 patients with acute stroke (mean age, 72.0 years; women, 39%). The discharge destination was categorized into three groups: home, postacute rehabilitation (hospital with convalescent rehabilitation wards), and postacute care (institution without convalescent rehabilitation wards). We assessed FIM after lifting bed restriction. Multinomial logistic regression analyses were used to estimate odds ratios (OR) and 95% confidence intervals (CI) of the FIM scores for predicting discharge destinations, with postacute rehabilitation as a reference. Cutoff values of motor and cognitive FIM scores for distinguishing home from postacute rehabilitation and postacute care from postacute rehabilitation were determined using receiver operating characteristic curves. The proportion of home, postacute rehabilitation, and postacute care were 34.6%, 41.8%, and 23.6%, respectively. After adjustments for clinical variables, the ORs (95% CIs) for motor and cognitive FIM scores for home versus postacute rehabilitation were 1.08 (1.04-1.11) and 1.05 (0.98-1.12). Furthermore, those for postacute care versus postacute rehabilitation were 1.01 (0.98-1.04) and 0.92 (0.87-0.98). The cutoff values of the motor and cognitive FIM scores for distinguishing home from postacute rehabilitation were 37.5 (sensitivity: 0.92; specificity: 0.64) and 23.5 (sensitivity: 0.78; specificity: 0.67). Furthermore, those for distinguishing postacute care from postacute rehabilitation were 15.5 (sensitivity, 0.81; specificity, 0.51) and 12.5 (sensitivity, 0.74; specificity, 0.64). The identified cutoff values may serve as early indicators for predicting discharge destinations from acute stroke care.

RAPT and AM-PAC "6-Clicks": Do They Correlate on Predicting Discharge Destination After Total Joint Arthroplasty?

Background: The Risk Assessment and Prediction Tool (RAPT) and the Activity Measure for Post-Acute Care "6-Clicks" Mobility Score (AM-PAC) are validated discharge planning tools for patients undergoing total hip arthroplasty (THA) and total knee arthroplasty (TKA). Planning for discharge with these tools considers very different factors and it is important to determine if they relate. Purpose: We sought to determine whether the preoperative RAPT score would correlate with postoperative AM-PAC score for predicting discharge destination for THA and TKA populations. Secondarily, we sought to examine whether the AM-PAC and RAPT scores would remain statistically significant predictors of discharge destination despite covariates. Methods: A retrospective cohort study was performed for patients who underwent THA or TKA from January 2020 to December 2022 at a specialty orthopedic hospital. Primary variables included the RAPT score, the AM-PAC score, and discharge disposition. Correlation between AM-PAC and RAPT scores was tested using Pearson's correlation coefficient, and association between both scores and discharge destination was tested using chi-square tests and multivariable logistic regression. Results: Our comparison of AM-PAC scores and RAPT scores found a statistically significant, positive correlation in both THA and TKA patients. Regression analysis found that increased RAPT and AM-PAC scores resulted in higher odds of being discharged home for both populations, after adjusting for all other variables. In both cohorts, patients discharged to a facility were more likely to be female, be over the age of 70 years, have Medicare/Medicaid insurance, and have a higher number of preoperative social work visits or any incidence of an intraoperative or hospital complication. Conclusions: This retrospective study found that RAPT score correlated with AM-PAC score for predicting discharge destination for elective THA and TKA populations, suggesting that these scores may be predictors of home discharge destination even when accounting for covariates. Further study is recommended.

Early Prognostication of Critical Patients With Spinal Cord Injury: A Machine Learning Study With 1485 Cases.

A retrospective case-series. The study aims to use machine learning to predict the discharge destination of spinal cord injury (SCI) patients in the intensive care unit. Prognostication following SCI is vital, especially for critical patients who need intensive care. Clinical data of patients diagnosed with SCI were extracted from a publicly available intensive care unit database. The first recorded data of the included patients were used to develop a total of 98 machine learning classifiers, seeking to predict discharge destination (eg, death, further medical care, home, etc.). The microaverage area under the curve (AUC) was the main indicator to assess discrimination. The best average-AUC classifier and the best death-sensitivity classifier were integrated into an ensemble classifier. The discrimination of the ensemble classifier was compared with top death-sensitivity classifiers and top average-AUC classifiers. In addition, prediction consistency and clinical utility were also assessed. A total of 1485 SCI patients were included. The ensemble classifier had a microaverage AUC of 0.851, which was only slightly inferior to the best average-AUC classifier ( P =0.10). The best average-AUC classifier death sensitivity was much lower than that of the ensemble classifier. The ensemble classifier had a death sensitivity of 0.452, which was inferior to the top 8 death-sensitivity classifiers, whose microaverage AUC were inferior to the ensemble classifier ( P <0.05). In addition, the ensemble classifier demonstrated a comparable Brier score and superior net benefit in the DCA when compared with the performance of the origin classifiers. The ensemble classifier shows an overall superior performance in predicting discharge destination, considering discrimination ability, prediction consistency, and clinical utility. This classifier system may aid in the clinical management of critical SCI patients in the early phase following injury. Level 3.

Decision Tree Analysis Accurately Predicts Discharge Destination After Spinal Cord Injury Rehabilitation

ObjectivesTo predict discharge destination after spinal cord injury (SCI) rehabilitation. Study DesignA retrospective, single-center study. We collected the following data from medical charts: age, sex, living arrangement before injury, acute length of stay (LOS), level of injury on admission, American Spinal Injury Association Impairment Scale (AIS) on admission, Upper Extremity Motor Score (UEMS) on admission, Lower Extremity Motor Score on admission (LEMS), Spinal Cord Independence Measure (SCIM) scores on admission and discharge, and discharge destination. A decision tree algorithm was used to establish prediction models in a train-test split manner using features on admission or discharge. SettingA spinal center in Tokyo, Japan. ParticipantsParticipants were individuals with SCI admitted to our hospital from March 2016 to October 2021 for the first rehabilitation after the injury. The study included 210 participants divided into 2 groups: training (n=140) and testing (n=70). Random sampling without replacement was used. InterventionsNot applicable. Main Outcome MeasuresPrediction accuracy was evaluated with sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and area under the receiver operating curve (AUC). ResultsAIS was significantly different between the groups. The prediction model using total SCIM scores on discharge (D-Classification and Regression Tree [CART]) revealed that a cut-off value of 40 accurately predicted the discharge destination. In contrast, the prediction model using features on admission (A-CART) revealed that subtotal SCIM mobility scores of 5, age of 74 years, and UEMS of 23 were significant predictors. Sensitivity, specificity, PPV, NPV, and AUC of D-CART and A-CART were 0.837, 0.810, 0.911, 0.680, and 0.832 and 0.857, 0.810, 0.913, 0.708, and 0.869, respectively. ConclusionsD-CART and A-CART showed comparable prediction accuracies. This suggests that, even during the early stages of rehabilitation, it is possible to predict the discharge destination.

Prediction of Discharge Destination After Inpatient Rehabilitation for Stroke Using Mobility and Self-Care Assessment in Section GG of the Inpatient Rehabilitation Facility – Patient Assessment Instrument

ObjectiveTo determine the ability of Section GG of the Inpatient Rehabilitation Facility – Patient Assessment Inventory (Section GG)’s quantification of mobility and self-care to predict discharge destination for persons with stroke after inpatient rehabilitation. DesignRetrospective, observational cohort study. Setting150-bed inpatient rehabilitation facility within a metropolitan health system. ParticipantsConsecutive sample of adults and older adults with stroke admitted for inpatient rehabilitation from January 2020 to June 2021 (N=1051). Subjects were excluded for discharge to acute care or hospice or if they had COVID-19. InterventionNone. Main Outcome MeasuresSection GG self-care and mobility scores used in reimbursement formulation by Centers for Medicare and Medicaid at admission to inpatient rehabilitation; age; sex; prior living situation; discharge setting. Logistic regression examined binary comparisons of discharge destinations. Receiver operating characteristic (ROC) curves determined cut-off admission Section GG scores for binary comparisons. ResultsLogistic regression demonstrated that presence of a caregiver in the home was consistently the strongest predictor (P<.001) and admission Section GG scores were significant secondary factors in determining the discharge destination. An admission Section GG cut-off score of 33.5 determined home with homecare vs skilled nursing facility and a cut-off of 36.5 determined discharge to home with outpatient care vs skilled nursing facility. ConclusionClinicians responsible for discharge decisions for patients with stroke after inpatient rehabilitation might start by determining the presence of a caregiver in the home and then use Section GG cut-off scores to guide decisions about home (with or without homecare) vs SNF destinations. Such guidance is not advised for the home with outpatient services vs home with homecare decision; clinical judgment is needed to determine the best discharge plan because this ROC had a less robust area under the curve. Sex and race/ethnicity were not determining factors for binary choices of discharge destinations.

Predictors of discharge to home/community following inpatient-rehabilitation in a US national sample of Guillain-Barre-Syndrome patients.

Guillain-Barre-Syndrome (GBS), an autoimmune polyneuropathy causing acute flaccid paralysis, is a rare condition with1-2 cases per 100,000 annually (approximately 5000 cases/year) in the United States (US). There is a paucity of published data regarding patient outcomes in association with discharge destinations following inpatient-rehabilitation (IR) in this patient population, thus this study. To analyze IR efficacy, and possible predictors of discharge to home/community in a US-national-sample of GBS patients. Retrospective-observational-cohort study of 1304 GBS patients admitted to IR comparing discharge disposition destinations (community/home, skilled-nursing-facility [SNF], or return to acute-care) by demographic (age, gender) and clinical variables (length-of-stay [LOS], case-mix-index [CMI], and Functional-Independence-Measure [FIM] score changes). Multinomial-logistic-regression and discriminant-function-analysis were performed to determine model fit in predicting discharge destination. 81.8% were discharged to home/community- average LOS 19-days, total-FIM-gain 43.2; 9.8% discharged to SNFs- average LOS 27.5-days, total-FIM-gain 27.2; and 8.4% discharged to acute-care- average LOS 15.4-days and total-FIM-gain 16.5, (F = 176, p < .001). Stepwise-linear-regression for prediction of community discharge showed change in FIM-Bed/chair/wheelchair-Transfers was the most significant predictor (Wald = 42.2; p < .001), followed by CMI (Wald = 26.9; p < .001), change in FIM-walking/wheelchair (Wald = 14.9; p < .001), and age (Wald = 9.5; p < .002). Using discriminant-function-analysis to test model validity for predicting discharge disposition, FIM-change for Bed/chair/wheelchair Transfers, Walking, and Self-Care as predictors resulted in a classification rate of 78.1%, 92% of variance explained, and Eigenvalue of .53 (p < .001). Total-FIM scores improved in all groups, and most patients were discharged to home/community suggesting IR efficacy. The ability to transfer bed/chair/wheelchair was the most important predictive factor associated with discharge destination.

The Risk Assessment and Prediction Tool (RAPT) Score Predicts Discharge Destination, Length of Stay, and Postoperative Mobility After Total Joint Arthroplasty

BackgroundPredicting an arthroplasty patient’s discharge disposition, length of stay (LOS), and physical function is helpful because it allows for preoperative patient optimization, expectation management, and discharge planning. The goal of this study was to evaluate the ability of the Risk Assessment and Prediction Tool (RAPT) score to predict discharge destination, LOS, and postoperative mobility in patients undergoing primary total knee arthroplasty (TKA) and total hip arthroplasty (THA). MethodsPrimary unilateral TKAs (n = 9,064) and THAs (n = 8,649) performed for primary osteoarthritis at our institution from 2018 to 2021 (excluding March to June 2020) were identified using a prospectively maintained institutional registry. We evaluated the associations between preoperative RAPT score and (1) discharge destination, (2) LOS, and postoperative mobility as measured by (3) successful ambulation on the day of surgery and (4) Activity Measure for Post-Acute Care (AM-PAC) “6-Clicks” score. ResultsOn multivariable analyses adjusting for multiple covariates, every one-point increase in RAPT score among TKA patients was associated with a 1.82-fold increased odds of home discharge (P < .001), 0.22 days shorter LOS (P < .001), 1.13-fold increased odds of ambulating on postoperative day 0 (P < .001), and 0.25-point higher Activity Measure for Post-Acute Care score (P < .001). Similar findings were seen among THAs. A RAPT score of 8 or higher was the most sensitive and specific cutoff to predict home discharge. ConclusionAmong nearly 18,000 TKA and THA patients, RAPT score was predictive of discharge disposition, LOS, and postoperative mobility. A RAPT score of 8 or higher was the most sensitive and specific cutoff to predict discharge to home. In contrast to prior studies of the RAPT score which have grouped TKAs and THAs together, this study ran separate analyses for TKAs and THAs and found that THA patients seemed to perform better than TKA patients with equal RAPT scores, suggesting that RAPT may behave differently between TKAs and THAs, particularly in the intermediate risk RAPT range.

Random forest identifies predictors of discharge destination following total shoulder arthroplasty

Machine learning algorithms are finding increasing use in prediction of surgical outcomes in orthopedics. Random forest is one of such algorithms popular for its relative ease of application and high predictability. In the process of sample classification, algorithms also generate a list of variables most crucial in the sorting process. Total shoulder arthroplasty (TSA) is a common orthopedic procedure after which most patients are discharged home. The authors hypothesized that random forest algorithm would be able to determine most important variables in prediction of nonhome discharge. Authors filtered the National Surgical Quality iImprovement Program database for patients undergoing elective TSA (Current Procedural Terminology 23472) between 2008 and 2018. Applied exclusion criteria included avascular necrosis, trauma, rheumatoid arthritis, and other inflammatory arthropathies to only include surgeries performed for primary osteoarthritis. Using Python and the scikit-learn package, various machine learning algorithms including random forest were trained based on the sample patients to predict patients who had nonhome discharge (to facility, nursing home, etc.). List of applied variables were then organized in order of feature importance. The algorithms were evaluated based on area under the curve of the receiver operating characteristic, accuracy, recall, and the F-1 score. Application of inclusion and exclusion criteria yielded 18,883 patients undergoing elective TSA, of whom 1813 patients had nonhome discharge. Random forest outperformed other machine learning algorithms and logistic regression based on American Society of Anesthesiologists (ASA) classification. Random forest ranked age, sex, ASA classification, and functional status as the most important variables with feature importance of 0.340, 0.130, 0.126, and 0.120, respectively. Average age of patients going to facility was 76 years, while average age of patients going home was 68 years. 78.1% of patients going to facility were women, while 52.7% of patients going home were. Among patients with nonhome discharge, 80.3% had ASA scores of 3 or 4, while patients going home had 54% of patients with ASA scores 3 or 4. 10.5% of patients going to facility were considered of partially/totally dependent functional status, whereas 1.3% of patients going home were considered partially or totally dependent (P value < .05 for all). Of various algorithms, random forest best predicted discharge destination following TSA. When using random forest to predict nonhome discharge after TSA, age, gender, ASA scores, and functional status were the most important variables. Two patient groups (home discharge, nonhome discharge) were significantly different when it came to age, gender distribution, ASA scores, and functional status.

Therapists Predict Discharge Destination More Accurately Than the AM-PAC “6 Clicks” at Evaluation and Discharge for Patients With Isolated Coronary Artery Bypass Graft

Background: Clinicians and hospital care coordinators seek effective tools to assist with reducing length of stay, delivering appropriate postacute care services, and reducing readmissions for patients following coronary artery bypass grafting (CABG) surgery. Objective: To evaluate whether the discharge recommendation by the Activity Measure for Post-Acute Care (AM-PAC) “6 Clicks” Inpatient Basic Mobility and Daily Activity Short Forms agreed with therapists' recommendation and actual discharge disposition for patients who underwent isolated CABG. Methods: Physical therapists (PTs) and occupational therapists (OTs) recorded their respective AM-PAC “6 Clicks” score on a cohort of consecutive patients admitted for isolated CABG between March and April 2020. The level of agreement was compared between AM-PAC-projected discharge disposition, therapists' recommendation, and actual discharge disposition. Results: There were 58 patients who received discharge recommendations from PTs and OTs following isolated CABG. Upon PT initial evaluation, the AM-PAC “6 Clicks” Basic Mobility tool recommended postacute placement in all 58 (100%) cases, whereas PTs recommended postacute placement for only 1 (2%). Similarly, the AM-PAC “6 Clicks” Daily Activity tool recommended postacute placement in all 58 (100%) cases, whereas OTs recommended discharge to home for 44 (76%) cases and recommended further assessment of discharge needs on the remaining 14 (24%). At the final session, the AM-PAC “6 Clicks” Basic Mobility tool recommended postacute placement in 56 (98%) cases and PTs recommended discharge to home in all 58 (100%) cases. The AM-PAC “6 Clicks” Daily Activity tool recommended postacute placement for 25 (43%) cases, whereas OTs recommended discharge to home for 55 (95%) cases. Nearly all (56; 97%) patients were discharged to home, with the remaining 2 (3%) discharged to postacute placement. The 30- day readmission rate was 2%. Conclusion: We found the PT/OT recommendations were strongly associated with discharge disposition, but that the AM-PAC “6 Clicks” scores had poor agreement with therapists' recommendations and with actual discharge dispositions for patients with isolated CABG. The utility of this tool should be explored in other cardiac patient populations, such as valve replacements with and without sternotomy, and heart failure, to corroborate the results.

Activity Measure for Post-Acute Care “6-Clicks” Basic Mobility Scores Predict Discharge Destination After Acute Care Hospitalization in Select Patient Groups: A Retrospective, Observational Study

ObjectivesTo establish cutoff scores for the Activity Measure for Post-Acute Care “6-Clicks” standardized Basic Mobility scores (sBMSs) for predicting discharge destination after acute care hospitalization for diagnostic subgroups within an acute care population and to evaluate the need for a second score to improve predictive ability. DesignRetrospective, observational design. SettingMajor medical center in metropolitan area. ParticipantsElectronic medical records of 1696 adult patients (>18 years) admitted to acute care from January to October 2018. Records were stratified by orthopedic, cardiac, pulmonary, stroke, and other neurological diagnoses (N=1696). Interventions: None Main Outcome MeasurePhysical therapists scored patients’ sBMSs after referral for physical therapy and prior to discharge. Receiver operating characteristic curves delineated sBMS cutoff scores distinguishing various pairings of home, home with services, inpatient rehabilitation, or skilled nursing facility discharges. First and second sBMSs were compared with percentage change of the area under the curve and inferential statistics. ResultsHome vs institution cutoff score was 42.88 for combined sample, pulmonary and neurological cases. The cutoff score for orthopedic diagnoses score was 41.46. Cardiac and stroke model quality invalidated cutoff scores. Home without services vs skilled nursing discharges and home with services vs skilled nursing discharges were predicted with varying cutoff scores per diagnosis. sBMS cutoff scores collected closer to discharge were either the same or higher than first cutoffs, with varying effects on predictive ability. ConclusionssBMSs can help decide institution vs home discharge and finer distinctions among discharge settings for some diagnostic groups. A single sBMS may provide sufficient assistance with discharge destination decisions but timing of scoring and diagnostic group may influence cutoff score selection.

Predicting Elective Surgical Patient Outcome Destination Based on the Preoperative Modified Frailty Index and Laboratory Values.

To determine the accuracy of preoperative modified frailty index (mFI) with or without laboratory values (mFI-labs or labs-continuous) in predicting postoperative discharge destination. Discharge destination is important to providers and patients. The ability to accurately predict discharge destination preoperatively can improve hospital resource utilization and help set patient and family expectations. Cohort analysis of the 2018 American College of Surgeon National Surgical Quality Improvement Project (ACS-NSQIP) Participant Use File of patients undergoing operations with complete data point sets: age, sex, operation work relative-value units; mFI-clinical based on 12 clinical findings, mFI-labs based on seven laboratory values. The nine hierarchical destinations: home, home with assistance, multi-level community, unskilled-care facility, rehabilitation facility, skilled-nursing facility, acute care hospital, hospice, or death, from best to worst outcome. Data were analyzed using univariate analysis, multiple logistic regression and supervised learning artificial neural networks. Univariate and multivariate in general showed that patients with higher mFI-clinical and mFI-lab scores, as well as older age and more complex operations were more likely to be discharged to facilities other than home. However, these statistical techniques could not predict the exact destination. An artificial neural network analysis demonstrated perfect location prediction in 64.9% of cases and within one level of prefect prediction is 87.4%. Using a limited number of preoperative factors, combining the mFI-clinical with laboratory values significantly improves the destination prediction performance significantly better than using the values separately. Preoperative knowledge of the likely discharge destination can benefit postoperative care planning and delivery.

265 Use of support vector machine to predict discharge destination on an acute stroke ward

BackgroundPredicting the discharge destination of stroke patients at admission is important for stream- lining of services, management of patient and carer expectations, and the planning and facilitating of discharge to appropriate destinations.MethodsWe reviewed data from the Sentinel Stroke National Audit Programme (SSNAP) from a single acute stroke ward in South Wales. We analysed age, modified Rankin Scale (mRS) and National Institute of Health Stroke Scale (NIHSS) data using a multi-class Support Vector Machine (SVM) to train and test a model to predict discharge destination.ResultsData was extracted from 1131 patients with acute and subacute stroke admitted to an acute stroke unit between January 1st, 2015, and December 31st, 2016. Per-class sensitivity of the model for final destination or outcome was 0.89 for home, 0.6 community hospital, 0.75 inpatient rehabilitation, 0.45 death. Per-class positive predictive value (PPV) was 0.84 for home, 0.62 community hospital, 0.72 inpatient rehabilitation, 0.55 death.ConclusionFrom data that is available from simple assessments done in routine clinical pratice it is possible to identify patients with acute stroke for whom discharge home or inpatient rehabilitation is highly likely. Collaborations between clinicians and data scientists may help to maximise the value of clinical assessment tools that are already in widespread use.Tom.Hughes2@wales.nhs.uk

Utilization of comorbidity indices to predict discharge destination and complications following total shoulder arthroplasty.

Comorbidity indices can help identify patients at risk for postoperative complications. Purpose of this study was to compare different comorbidity indices to predict discharge destination and complications after shoulder arthroplasty. Retrospective review of institutional shoulder arthroplasty database of primary anatomic (TSA) and reverse (RSA) shoulder arthroplasties. Patient demographic information was collected in order to calculate Modified Frailty Index (mFI-5), Charlson Comorbidity Index (CCI), age adjusted CCI (age-CCI), and American Society of Anesthesiologists physical status classification system (ASA). Statistical analysis performed to analyze length of stay (LOS), discharge destination, and 90-day complications. There were 1365 patients included with 672 TSA and 693 RSA patients. RSA patients were older and had higher CCI, age adjusted CCI, ASA, and mFI-5 (p < 0.001). RSA patients had longer lengths of stay (LOS), more likely to have an adverse discharge (p < 0.001), and higher reoperation rate (p = 0.003). Age-CCI was most predictive of adverse discharge (AUC 0.721, 95% CI 0.704-0.768). Patients undergoing RSA had more medical comorbidities, experienced greater LOS, higher reoperation rate, and were more likely to have an adverse discharge. Age-CCI had the best ability to predict which patients were likely to require higher-level discharge planning.

678 Can patients have their old life back? Using PM:L3 ratio to predict discharge destination of emergency laparotomy survivors

Abstract Aim Psoas muscle:L3 vertebra (PM:L3) ratio is a relatively new risk assessment tool for emergency laparotomy(1,2) based upon the proven concept that sarcopenia – as diagnosed by low skeletal muscle index(3–6), psoas muscle density(7–9) or total psoas area(10–13) in a single axial slice CT image – correlates with increasing risks of bad outcomes in surgery. This study looks into the association between PM:L3 ratio of emergency laparotomy patients from home and their discharge destination. Method We performed an analysis on patients in the Hunter Emergency Laparotomy Audit (HELA) database, from January 2016 to December 2017. HELA is a retrospective review of all emergency laparotomy undertaken in a discrete area in NSW, Australia. All patients admitted from home, survived to be discharged from hospital and had an available CT abdomen were included (N = 433/562). A single slice axial CT image at the L3 endplate level was analysed using ImageJ® software to measure the area of L3 and bilateral psoas muscles. Results PM:L3 ratio is significantly lower in the group of patient discharged to a care facility than the group discharging back to their previous home residence (mean 0.951 vs. mean 1.128, p &amp;lt; 0.001). Upon further analysis, the PM:L3s are divided into quartiles and stratified by sex. There is an association between lower PM:L3 and risk of discharging into a care facility (Q1 22.45%, Q2 19.59%, Q3 19.10%, Q4 5.71%). Conclusions PM:L3 ratio can predict the discharge destination of patients undergoing emergency laparotomy.

Examination of GG-Mobility Items and Select Core Set Outcome Measures on Predicting Discharge Destination and Length of Stay From an Inpatient Rehab Facility in Stroke: A Pilot Study

<h3>Research Objectives</h3> To determine which mobility assessment measures at admission are predictive of IRF length of stay and IRF discharge destination in patients post stroke. <h3>Design</h3> A clinical prediction pilot study was carried out on a cohort sample utilizing a retrospective audit. <h3>Setting</h3> The study took place in an IRF. <h3>Participants</h3> Inpatients with a diagnosis of acute/subacute stroke were included if they received IRF-PAI assessment, Berg Balance Scale, 10 Meter Walk Test, and Five Times Sit-to-Stand on admission (n = 65). <h3>Interventions</h3> N/A. <h3>Main Outcome Measures</h3> Subjects involved in this analysis completed the Berg Balance Scale, 10-Meter Walk Test scores, Five Times Sit-to-Stand Test, and GG Mobility items at admission including Mobility composite score, Walk composite score, Walk 150ft score, and Transfers score. Additionally, length of stay and discharge destination were examined. <h3>Results</h3> Examination of relationships between selected Core Set measures and selected GG-items revealed that all were significantly correlated at admission (p < .001). Analyses revealed that discharge destination (home versus skilled nursing facility) significantly correlated with Berg Balance Scale (p < .05), Ten-Meter Walk Test (p < .05), Five Times Sit-to-Stand (p < .05), GG Mobility Composite (p < .01), and GG Transfer score (p < .01) at admission; however, only GG Transfer score was a significant predictor of discharge destination when examining with logistic regression (p = .005). Analyses revealed that length of stay significantly correlated with Berg Balance Scale (p < .001), Ten-Meter Walk Test (p < .001), Five Times Sit-to-Stand (p < .01), GG Walk Composite (p < .001), GG Walk 150 feet (p < .01), GG Mobility Composite (p < .001), and GG Transfer score (p < .01) at admission; however, Berg Balance Scale score at admission was the only significant predictor of length of stay when examining with linear regression (p < .001). <h3>Conclusions</h3> In this pilot study, selected Core Set measures were strongly correlated with selected GG mobility items at admission and possessed significant predictive capabilities in regards to discharge destination and length of stay in IRF. <h3>Author(s) Disclosures</h3> None.

Application of Machine Learning Techniques to Identify Data Reliability and Factors Affecting Outcome After Stroke Using Electronic Administrative Records.

Aim: To use available electronic administrative records to identify data reliability, predict discharge destination, and identify risk factors associated with specific outcomes following hospital admission with stroke, compared to stroke specific clinical factors, using machine learning techniques.Method: The study included 2,531 patients having at least one admission with a confirmed diagnosis of stroke, collected from a regional hospital in Australia within 2009–2013. Using machine learning (penalized regression with Lasso) techniques, patients having their index admission between June 2009 and July 2012 were used to derive predictive models, and patients having their index admission between July 2012 and June 2013 were used for validation. Three different stroke types [intracerebral hemorrhage (ICH), ischemic stroke, transient ischemic attack (TIA)] were considered and five different comparison outcome settings were considered. Our electronic administrative record based predictive model was compared with a predictive model composed of “baseline” clinical features, more specific for stroke, such as age, gender, smoking habits, co-morbidities (high cholesterol, hypertension, atrial fibrillation, and ischemic heart disease), types of imaging done (CT scan, MRI, etc.), and occurrence of in-hospital pneumonia. Risk factors associated with likelihood of negative outcomes were identified.Results: The data was highly reliable at predicting discharge to rehabilitation and all other outcomes vs. death for ICH (AUC 0.85 and 0.825, respectively), all discharge outcomes except home vs. rehabilitation for ischemic stroke, and discharge home vs. others and home vs. rehabilitation for TIA (AUC 0.948 and 0.873, respectively). Electronic health record data appeared to provide improved prediction of outcomes over stroke specific clinical factors from the machine learning models. Common risk factors associated with a negative impact on expected outcomes appeared clinically intuitive, and included older age groups, prior ventilatory support, urinary incontinence, need for imaging, and need for allied health input.Conclusion: Electronic administrative records from this cohort produced reliable outcome prediction and identified clinically appropriate factors negatively impacting most outcome variables following hospital admission with stroke. This presents a means of future identification of modifiable factors associated with patient discharge destination. This may potentially aid in patient selection for certain interventions and aid in better patient and clinician education regarding expected discharge outcomes.

AM-PAC “6-clicks” Predicts Discharge Destination for Patients Hospitalized with COVID-19

Research ObjectivesTo investigate the ability of the initial Activity Measure for Post-Acute Care (AM-PAC) “6-clicks” mobility and activity scores to predict discharge disposition for patients hospitalized with COVID-19.DesignRetrospective cohort using data from a COVID-19 registry.SettingFive inpatient hospitals in a single academic Health System within the United States.ParticipantsAdults (> 17 years) who were urgently or emergently admitted to one of the five acute hospitals between March 1st and July 31st, 2020, tested positive for COVID-19 during their hospitalization, and had at least 1 AM-PAC “6-clicks” score in their medical record.InterventionsN/A.Main Outcome MeasuresDischarge destination, dichotomized as home vs. facility.ResultsOf the 2565 records in the registry, 1486 included AM-PAC “6-clicks” mobility scores and 1200 included activity scores. Median age was 64 years (IQR=26), COVID-19 was primary or admitting diagnosis for 47% (n = 700), and 61.6% (n = 915) were discharged home. Initial AM-PAC was assessed 1 (IQR=3) days after admission, and the median for both mobility and activity scores was 18 (IQR=14). Multivariate logistic regression analyses revealed that, after controlling for the influence of covariates, initial mobility and activity scores were both independent predictors of discharge destination. Each point decrease in initial AM-PAC score increased the odds of discharge to a faculty by 1.15 (95 CI% 1.12 - 1.19; p < 0.001) and 1.16 (95 CI% 1.11 - 1.22; p < 0.001) fold for mobility and activity scores, respectively. Receiver operating characteristic (ROC) curve analysis revealed that initial AM-PAC scores were strong predictors of discharge destination, with area under the curve (AUC) of 0.806 (95 CI% 0.781 - 0.831; p < 0.001) and 0.796 (95% CI 0.767 - 0.826; p < 0.001) for mobility and activity scores, respectively. A score of 17.5 predicted discharge with a sensitivity of .687 and specificity of 0.81 for mobility and 0.724 and 0.770 for activity.ConclusionsInitial AM-PAC “6-clicks” activity and mobility scores were both predictive of discharge destination in individuals hospitalized with COVID-19.Author(s) DisclosuresDr. Adler is a paid consultant for MedBridge Education.

Is the Parker Mobility Score in the older patient with a traumatic hip fracture associated with discharge disposition after surgery? A retrospective cohort study.

The research questions for this study were as follows: (1) is the Parker Mobility Score (PMS) associated with discharge disposition and hospital length of stay (HLOS) of geriatric traumatic hip fracture patients? (2) Can the PMS be incorporated in a decision tree for the prediction of discharge disposition of geriatric traumatic hip fracture patients upon admittance. A dual-center retrospective cohort study was conducted at two level II trauma centers. All patients aged 70years and older with traumatic hip fractures undergoing surgery in 2018 and 2019 were included consecutively (n = 649). A χ2 automatic interaction detection analysis was performed to determine the association of the PMS (and other variables) with discharge disposition and HLOS and predict discharge destination. The decision tree for discharge disposition classified patients with an overall accuracy of 82.1% and a positive predictive value of 91% for discharge to a rehabilitation facility. The PMS had the second most significant effect on discharge disposition (χ2 = 22.409, p < 0.001) after age (χ2 = 79.094, p < 0.001). Regarding the tree analysis of HLOS, of all variables in the analysis, PMS had the most significant association with HLOS (F = 14.891, p < 0.001). Patients who were discharged home had a mean HLOS of 6.5days (SD 8.0), whereas patients who were discharged to an institutional care facility had a mean HLOS of 9.7days (SD 6.4; p < 0.001). This study shows that the PMS was strongly associated with discharge disposition and HLOS. The decision tree for the discharge disposition of geriatric traumatic hip fracture patients offers a practical solution to start discharge planning upon admittance which could potentially reduce HLOS. Level III, diagnostic.

Predicting Inpatient Flow at a Major Hospital Using Interpretable Analytics

Problem definition: Translate data from electronic health records (EHR) into accurate predictions on patient flows and inform daily decision making at a major hospital. Academic/practical relevance: In a constrained hospital environment, forecasts on patient demand patterns could help match capacity and demand and improve hospital operations. Methodology: We use data from 63,432 admissions at a large academic hospital (50% female, median age 64 years old, median length of stay 3.12 days). We construct an expertise-driven patient representation on top of their EHR data and apply a broad class of machine learning methods to predict several aspects of patient flows. Results: With a unique patient representation, we estimate short-term discharges, identify long-stay patients, predict discharge destination, and anticipate flows in and out of intensive care units with accuracy in the 80%+ range. More importantly, we implement this machine learning pipeline into the EHR system of the hospital and construct prediction-informed dashboards to support daily bed placement decisions. Managerial implications: Our study demonstrates that interpretable machine learning techniques combined with EHR data can be used to provide visibility on patient flows. Our approach provides an alternative to deep learning techniques that is equally accurate, interpretable, frugal in data and computational power, and production ready. History: This paper has been accepted for the Manufacturing &amp; Service Operations Management Special Section on Responsible Research in Operations Management. Funding: The research was funded by Beth Israel Deaconess Medical Center. Supplemental Material: The e-companion is available at https://doi.org/10.1287/msom.2021.0971 .

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.