Predict Hospital Length Of Stay Research Articles

Utility of combining frailty and comorbid disease indices in predicting outcomes following craniotomy for adult primary brain tumors: A mixed-effects model analysis using the nationwide readmissions database

ObjectiveThe escalating healthcare expenditures in the United States, particularly in neurosurgery, necessitate effective tools for predicting patient outcomes and optimizing resource allocation. This study explores the utility of combining frailty and comorbidity indices, specifically the Johns Hopkins Adjusted Clinical Groups (JHACG) frailty index and the Elixhauser Comorbidity Index (ECI), in predicting hospital length of stay (LOS), non-routine discharge, and one-year readmission in patients undergoing craniotomy for benign and malignant primary brain tumors. MethodsLeveraging the Nationwide Readmissions Database (NRD) for 2016–2019, we analyzed data from 645 patients with benign and 30,991 with malignant tumors. Frailty, ECI, and frailty + ECI were assessed as predictors using generalized linear mixed-effects models. Receiver operating characteristic (ROC) curves evaluated predictive performance. ResultsPatients in the benign tumor cohort had a mean LOS of 8.1 ± 15.1 days, and frailty + ECI outperformed frailty alone in predicting non-routine discharge (AUC 0.829 vs. 0.820, p = 0.035). The malignant tumor cohort patients had a mean LOS of 7.9 ± 9.1 days. In this cohort, frailty + ECI (AUC 0.821) outperformed both frailty (AUC 0.744, p < 0.0001) and ECI alone (AUC 0.809, p < 0.0001) in predicting hospital LOS. Frailty + ECI (AUC 0.831) also proved superior to frailty (AUC 0.809, p < 0.0001) and ECI alone (AUC 0.827, p < 0.0001) in predicting non-routine discharge location for patients with malignant tumors. All indices performed comparably to one another as a predictor of readmission in both cohorts. ConclusionThis study highlights the synergistic predictive capacity of frailty + ECI, especially in malignant tumor cases, and further suggests that comorbid diseases may greatly influence perioperative outcomes more than frailty. Enhanced risk assessment could aid clinical decision-making, patient counseling, and resource allocation, ultimately optimizing patient outcomes.

512 Length of Stay Prediction Based on an Artificial Neural Network

Abstract Introduction Hospital length of stay (LOS) is difficult to predict in patients with burn injuries due to the complexity of the injuries and the wide variability in patient outcomes. Predicting hospital LOS in patients with burn injuries can have numerous benefits for patients, physicians, and insurance companies. Artificial neural networks may offer insight into more accurate LOS predictions. Methods We obtained electronic health records (EHR) for burn patients from July 01, 2011 - July 01, 2021. An artificial neural network was used to model the relationship between the input features and LOS. The input features were preprocessed by normalizing each feature to the range [0, 1] and transformed using log normal. The preprocessed data was split into training (70%), validation (15%), and testing (15%) sets. The neural network consisted of one hidden layer with 13 neurons and a linear output layer. Results The neural network achieved an R-squared (R2) value of 0.72 on the training data, indicating that 72% of the variance in the data (n=1347). The validation and testing R2 values were 0.66 and 0.68, respectively. The overall R2 value, computed by combining the predictions on the training, validation, and testing data, was 0.70. The best performance, validation performance, and testing performance achieved by the neural network during the training process were 0.28, 0.27, and 0.30, respectively. These values indicate the average squared difference between the predicted output and the true output for each set, with lower values indicating better performance. Conclusions The results suggest that the neural network can model the relationship between the input features and LOS with reasonable accuracy. The R2 values indicate that the model explains a substantial amount of the variability in the data. Further testing and validation may be needed to assess the robustness and reliability of the model. Applicability of Research to Practice Patient outcomes are largely variable and determining LOS to a closer degree helps planning more robust treatment plans. Additionally, these models have the potential to serve as a more accurate gauge of whether they are following closely to their expected treatment timeline. As survival rates continue to improve, the improvement focus has and will continue to shift towards patient outcomes.

Deep Learning With Chest Radiographs for Making Prognoses in Patients With COVID-19: Retrospective Cohort Study.

An artificial intelligence (AI) model using chest radiography (CXR) may provide good performance in making prognoses for COVID-19. We aimed to develop and validate a prediction model using CXR based on an AI model and clinical variables to predict clinical outcomes in patients with COVID-19. This retrospective longitudinal study included patients hospitalized for COVID-19 at multiple COVID-19 medical centers between February 2020 and October 2020. Patients at Boramae Medical Center were randomly classified into training, validation, and internal testing sets (at a ratio of 8:1:1, respectively). An AI model using initial CXR images as input, a logistic regression model using clinical information, and a combined model using the output of the AI model (as CXR score) and clinical information were developed and trained to predict hospital length of stay (LOS) ≤2 weeks, need for oxygen supplementation, and acute respiratory distress syndrome (ARDS). The models were externally validated in the Korean Imaging Cohort of COVID-19 data set for discrimination and calibration. The AI model using CXR and the logistic regression model using clinical variables were suboptimal to predict hospital LOS ≤2 weeks or the need for oxygen supplementation but performed acceptably in the prediction of ARDS (AI model area under the curve [AUC] 0.782, 95% CI 0.720-0.845; logistic regression model AUC 0.878, 95% CI 0.838-0.919). The combined model performed better in predicting the need for oxygen supplementation (AUC 0.704, 95% CI 0.646-0.762) and ARDS (AUC 0.890, 95% CI 0.853-0.928) compared to the CXR score alone. Both the AI and combined models showed good calibration for predicting ARDS (P=.079 and P=.859). The combined prediction model, comprising the CXR score and clinical information, was externally validated as having acceptable performance in predicting severe illness and excellent performance in predicting ARDS in patients with COVID-19.

Novel Machine Learning Approach to Predict and Personalize Length of Stay for Patients Admitted with Syncope from the Emergency Department.

Background: Syncope, a common problem encountered in the emergency department (ED), has a multitude of causes ranging from benign to life-threatening. Hospitalization may be required, but the management can vary substantially depending on specific clinical characteristics. Models predicting admission and hospitalization length of stay (LoS) are lacking. The purpose of this study was to design an effective, exploratory model using machine learning (ML) technology to predict LoS for patients presenting with syncope. Methods: This was a retrospective analysis using over 4 million patients from the National Emergency Department Sample (NEDS) database presenting to the ED with syncope between 2016-2019. A multilayer perceptron neural network with one hidden layer was trained and validated on this data set. Results: Receiver Operator Characteristics (ROC) were determined for each of the five ANN models with varying cutoffs for LoS. A fair area under the curve (AUC of 0.78) to good (AUC of 0.88) prediction performance was achieved based on sequential analysis at different cutoff points, starting from the same day discharge and ending at the longest analyzed cutoff LoS ≤7 days versus >7 days, accordingly. The ML algorithm showed significant sensitivity and specificity in predicting short (≤48 h) versus long (>48 h) LoS, with an AUC of 0.81. Conclusions: Using variables available to triaging ED clinicians, ML shows promise in predicting hospital LoS with fair to good performance for patients presenting with syncope.

Role of handgrip strength testing in pre-anaesthetic check-up: A prospective cross-sectional study.

Frailty has been known to be associated with postoperative adverse events and longer hospital length of stay (LOS). Hand grip strength (HGS) is one of the parameters of measuring frailty. The aim of the study was to correlate preoperative handgrip strength and 30-day outcome of patients undergoing major abdominal surgery. It also aimed to evaluate the role of the standard preoperative variables like metabolic equivalents, revised cardiac risk index (RCRI), serum albumin, and serum creatinine along with their association with HGS testing in determining the postoperative outcome in surgical patients. This prospective observational study included 149 American Society of Anesthesiologists class III/IV patients presenting for major abdominal surgery. A mean of three measurements of dominant HGS using Camry hand dynamometer was measured. The patients were divided into groups: weak, normal, and strong depending on grip strength. Patients were followed for 30 days and postoperative outcome in terms of ventilatory support, admission to intensive care unit, cardiac complications, in-hospital mortality, and LOS were recorded. Observational data obtained were reported as mean value and analyzed using Student's t-test or Wilcoxon/Mann-Whitney Rank test. Associations between RCRI, serum albumin, and LOS with HGS were evaluated using logistic regression. The hospital LOS was significantly longer in patients with weak HGS (15.11 ± 11.03 days versus 10 ± 5.71 days, P = 0.001). Patients with weak HGS had significantly lower mean serum albumin levels compared to normal HGS (P = 0.0001) and a statistically significant RCRI score (P = 0.013). HGS can be used as a preoperative test in predicting hospital LOS after major surgery.

Use of Machine Learning and Statistical Algorithms to Predict Hospital Length of Stay Following Colorectal Cancer Resection: A South African Pilot Study.

The aim of this pilot study was to develop logistic regression (LR) and support vector machine (SVM) models that differentiate low from high risk for prolonged hospital length of stay (LOS) in a South African cohort of 383 colorectal cancer patients who underwent surgical resection with curative intent. Additionally, the impact of 10-fold cross-validation (CV), Monte Carlo CV, and bootstrap internal validation methods on the performance of the two models was evaluated. The median LOS was 9 days, and prolonged LOS was defined as greater than 9 days post-operation. Preoperative factors associated with prolonged LOS were a prior history of hypertension and an Eastern Cooperative Oncology Group score between 2 and 4. Postoperative factors related to prolonged LOS were the need for a stoma as part of the surgical procedure and the development of post-surgical complications. The risk of prolonged LOS was higher in male patients and in any patient with lower preoperative hemoglobin. The highest area under the receiving operating characteristics (AU-ROC) was achieved using LR of 0.823 (CI = 0.798–0.849) and SVM of 0.821 (CI = 0.776–0.825), with each model using the Monte Carlo CV method for internal validation. However, bootstrapping resulted in models with slightly lower variability. We found no significant difference between the models across the three internal validation methods. The LR and SVM algorithms used in this study required incorporating important features for optimal hospital LOS predictions. The factors identified in this study, especially postoperative complications, can be employed as a simple and quick test clinicians may flag a patient at risk of prolonged LOS.

Use of heart rate variability to predict hospital length of stay for COVID-19 patients: A prospective observational study.

Background:As the COVID-19 pandemic continues, determining hospital demands has become a vital priority. Heart rate variability (HRV) has been linked to both the presence of viral infection and its severity. We investigate the possibility of using HRV parameters in comparison to other clinical parameters for predicting the hospital length of stay (LOS) for COVID-19 patients.Methods:This was a population-based cohort study. Measurements were performed in a specialized hospital for respiratory disease, dedicated to COVID-19. Patients were polymerase chain reaction positive for COVID-19 and on their 1st day of admission. Heart period, respiratory sinus arrhythmia (RSA), low frequency (LF) HRV, and vagal efficiency were calculated from electrocardiogram signals. This study investigated the correlation of HRV, demographic, and laboratory parameters with hospital LOS.Results:Forty-one participants were recruited, with a significant relationship, observed between hospital LOS and some demographic and clinical parameters such as lymphocyte count, age, and oxygen saturation of arterial blood. There was a negative relationship between LF and hospital LOS (r = −0.53, 95% confidence interval: −0.73, −0.24). Higher vagal efficiency predicted shorter hospital LOS in patients younger than 40 years of age (19.27% shorter hospital LOS was associated with a one SD higher value of VE, P = 0.007).Conclusion:HRV measurement is a non-invasive, inexpensive, and scalable procedure that produces several metrics, some of which are useful for predicting hospital LOS and managing treatment resources during COVID-19 pandemic.

Surgical Factors and Treatment Severity for Perioperative Complications Predict Hospital Length of Stay in Adult Spinal Deformity Surgery.

Retrospective review of prospectively collected multicenter registry data. The aim of this study was to determine whether surgical variables and complications as graded by treatment severity impact postoperative hospital length of stay (LOS). Surgical treatment can substantially improve quality of life for patients with adult spinal deformity (ASD). However, surgical treatment is associated with high complication rates, which may impact hospital LOS. Classifying complications by severity of subsequent treatment may allow surgeons to better understand complications and predict their impact on important outcome metrics, including LOS. Patients enrolled in a multicenter, prospectively enrolled database for ASD were assessed for study inclusion. Complications were graded based on intervention severity. Associations between LOS, complication intervention severity, and surgical variables (fusion length, use of interbody fusion, use of major osteotomy, primary versus revision surgery, same day vs. staged surgery, and surgical approach), were assessed. Two multivariate regression models were constructed to assess for independent associations with LOS. Of 1183 patients meeting inclusion criteria, 708 did not and 475 did experience a perioperative complication during their index hospitalization, with 660 and 436 included in the final cohorts, respectively. Among those with complications, intervention severities included 14.9% with no intervention, 68.6% with minor, 8.9% with moderate, and 7.6% with severe interventions. Multivariate regression modeling demonstrated that length of posterior fusion, use of major osteotomy, staged surgery, and severity of intervention for complications were significantly associated with LOS. Careful selection of surgical factors may help reduce hospital LOS following surgery for ASD. Classification of complications by treatment severity can help surgeons better understand and predict the implications of complications, in turn assisting with surgical planning and patient counseling.Level of Evidence: 4.

Can the Pediatric Early Warning Score (PEWS) Predict Hospital Length of Stay?

Background: Limited studies have evaluated the utility of scoring systems in the pediatric emergency department (PED) and no studies have evaluated their ability to predict hospital length of stay (LOS) and the usage of Observation units (OUs).Objective: To evaluate the utility of the Pediatric Early Warning Score (PEWS) in predicting LOS in pediatric patients and thus anticipate admission to an OU versus the pediatric ward.Methods: A retrospective study of pediatric inpatients (0 to 18 years) at an inner-city community hospital between January 2014 and December 2014. Patients with psychiatric illness, non-medical reasons for hospital stay, and those not discharged to ‘home’ were excluded. Demographic data, PEWS in the ED, and LOS for each patient were recorded and analyzed.Results: A total of 719 patients were analyzed. PEWS range was 0 to 8. The mean LOS was 56.8 hours for patients with PEWS 0-1 compared to 62.7 hours for patients with PEWS ≥2 (p=0.02). There was a significant difference in PEWS for LOS ≤24 and ≤36 hours in comparison to those with LOS >24 hours and >36 hours, respectively (p<0.001). Overall, the PEWS correlated with LOS (r=0.11, p=0.002). Age correlated inversely with LOS (r=-0.16, p<0.001), without correlation to PEWS (r=-0.002, p= 0.96).Conclusions: PEWS correlated weakly with LOS. A statistically significant lower PEWS was observed for patients who had short stays (both ≤24 and ≤36 hours) in comparison to those requiring longer inpatient care. Therefore, the PEWS is a useful tool to predict LOS and aid ED physicians to determine disposition, although further prospective studies in centers with OUs would better characterize its ability to suggest admission to an OU compared to the wards.

68 Application of the Modified Hospital Length of Stay Predictor

Abstract Introduction Developing formulas to predict accurate length of stay (LOS) is challenging in burn patients as many complex variables can effect LOS. A recently published article in 2016 provided new formulas (NFs) to predict hospital LOS. Our goal is to compare our patients actual LOS (ALOS) to the NFs published, the traditional 1day/%TBSA rule, and to the hospital expected LOS (ELOS) based on patients CMI/DRGs. Methods Data was collected from Jan ‘16 to Dec ‘18 on all patients &amp;gt;18 years of age using NTRACS burn registry. We excluded deaths, non-burns, readmissions, any records with missing information, and any patients transferred to another acute care hospital. We performed multiple regression to examine the relationship between LOS with age, inhalational injury (INHINJ), and TBSA on all patients. We divided this population into three groups per the NFs recommendations: 1) INHINJ; 2) without INHINJ &amp;lt; 40 years old; 3) without INHINJ &amp;gt;40 years old. Using these three groups, we calculated the difference between ALOS to 1) the traditional LOS formula, 2) the NFs as established in the 2016 article, and to 3) the hospitals ELOS metrics. For acuity comparison, we also calculated average CMI in each study group. The NFs published are as follows: Results The multiple regression model with all three variables (Age, % TBSA, and INHINJ) produced R² = 0.321, F(3,883) = 139.23, p &amp;lt; .05. INHINJ, age, and TBSA had significant positive regression, indicating an increase or presence of all three variables will lengthen LOS. The regression equation is: LOS = (-3.93) +1.71(AGE) +1.11(TBSA) +15.81 (Inhalation). Every year increase in age increased LOS by 1.71 days. Every 1% increase in TBSA increased LOS by 1.11 days. Presence of INHINJ (0= no INHINJ; 1= INHINJ) increased LOS by 15.81 days. Table 1 depicts the mean ± SD of the difference in LOS (predicted - actual) for the three predictive models of LOS. Negative values indicate underestimation of LOS and positive values indicate overestimate of LOS. Conclusions Application of the predictive LOS formulas has underestimated LOS in this population, with the exception of the NFs to predict LOS on patients &amp;gt;40 years old without INHINJ. Furthermore, an R² value of 0.32 indicates that the variables utilized in this study do not account for 68% of the variation we see. This is further exemplified by the wide standard deviations, particularly in the INHINJ group. Applicability of Research to Practice Directly Applicable.

Comparison of Measures to Predict Mortality and Length of Stay in Hospitalized Patients.

Patient risk adjustment is critical for hospital benchmarking and allocation of healthcare resources. However, considerable heterogeneity exists among measures. The performance of five measures was compared to predict mortality and length of stay (LOS) in hospitalized adults using claims data; these include three comorbidity composite scores (Charlson/Deyo age-comorbidity score, V W Elixhauser comorbidity score, and V W Elixhauser age-comorbidity score), 3 M risk of mortality (3 M ROM), and 3 M severity of illness (3 M SOI) subclasses. Binary logistic and zero-truncated negative binomial regression models were applied to a 2-year retrospective dataset (2013-2014) with 123,641 adult inpatient admissions from a large hospital system in New York City. All five measures demonstrated good to strong model fit for predicting in-hospital mortality, with C-statistics of 0.74 (95% confidence interval [CI] [0.74, 0.75]), 0.80 (95% CI [0.80, 0.81]), 0.81(95% CI [0.81, 0.82]), 0.94 (95% CI [0.93, 0.94]), and 0.90 (95% CI [0.90, 0.91]) for Charlson/Deyo age-comorbidity score, V W Elixhauser comorbidity score, V W Elixhauser age-comorbidity score, 3 M ROM, and 3 M SOI, respectively. The model fit statistics to predict hospital LOS measured by the likelihood ratio index were 0.3%, 1.2%, 1.1%, 6.2%, and 4.3%, respectively. The measures tested in this study can guide nurse managers in the assignment of nursing care and coordination of needed patient services and administrators to effectively and efficiently support optimal nursing care.

Use of a comprehensive frailty assessment to predict morbidity in patients with multiple myeloma undergoing transplant.

We prospectively evaluated 100 MM patients for frailty before and after ASCT using a Geriatric Assessment (GA) and collected T-cells for analysis of p16 using a custom nanostring codeset. Pre-transplant physical function was predicative of hospital length of stay (LOS). Each one-unit increase in physical function score, the average LOS decreased by 0.52 days (95% CI, -1.03-0.02); p = .04). Similarly, higher self-report of ADL/IADL (Human Activity Profile was associated with shorter LOS (0.65 less days (95% CI -1.15 to -0.15), p = .01). Patients with anxiety/depression (OR = 1.10 (95% CI 1.00-1.22), p = .056), lower handgrip strength (OR = 0.90 (95% CI 0.82-0.98), p = .02), falls (OR = 1.60 (95% CI 1.07-2.38), p = .02), or weight loss (OR = 5.65 (95% CI 1.17-25.24), p = .03) were more likely to be re-admitted. The estimated EFS at 1-year was 85% (95% CI, 75-91) with median follow-up of 15.7 months. Weight loss was a significant predictor of EFS (HR = 3.13 (95% CI 1.15-8.50), p = .03). Frailty assessment by self-reported fatigue minimally correlated with T-cell p16 expression (r = 0.28; p = .02). Age, Karnofsky Performance Status (KPS), or Hematopoietic cell transplantation-specific Co-Morbidity Index (HCT-CI) did not predict hospital LOS or readmissions. Our data illustrate that a GA can identify individuals with MM who are at greater risk for morbidity following ASCT.

The American College of Surgeons National Surgical Quality Improvement Program Surgical Risk Calculator Does Not Accurately Predict Risk of 30-Day Complications Among Patients Undergoing Microvascular Head and Neck Reconstruction

The American College of Surgeons (ACS) National Surgical Quality Improvement Program (NSQIP) Surgical Risk Calculator (SRC) is a novel universal risk calculator designed to aid in risk stratification of patients undergoing various types of major surgery. The purpose of this study was to assess the validity of the ACS NSQIP SRC in predicting postoperative complications in patients undergoing microvascular head and neck reconstruction. A retrospective cohort study of patients undergoing head and neck microvascular reconstruction with fibular free flaps at a single institution was completed. The NSQIP SRC was used to compute complication risk estimates and length of stay (LOS) estimates for all patients under study. Associations between complication risk estimates generated by the SRC and actual rates of observed complications were evaluated using logistic regression models. Logistic regression models also were used to evaluate the SRC estimates for LOS duration compared with the actual observed LOS after surgery. Of 153 patients under study, 46 (30%) developed a postoperative complication corresponding to those defined by NSQIP SRC. Thirty-eight patients (25%) developed a postoperative complication categorized as severe in the parameters of the NSQIP SRC. None of the SRC complication estimates showed a statistically relevant association with the corresponding observed rates of complications. The mean LOS predicted by the SRC was 8.0days (median, 7.5days; interquartile range [IQR], 6.5 to 9; range, 5.0 to 18.5days). The mean observed LOS for the study group was 9.6days (median, 7.0days; IQR, 6 to 9; range, 5 to 67days). Lin's (Biometrics 45:255, 1989) concordance correlation coefficient to measure agreement between observed and predicted LOS was 0.10, indicating only slight agreement between the 2 values. The ACS NSQIP SRC is not a useful risk-stratifying metric for patients undergoing major head and neck reconstruction with microvascular fibular free flaps. The SRC also does not accurately predict hospital LOS for this same patient cohort.

Factors Affecting Hospital Length of Stay for Drivers with Traumatic Brain Injuries from Motor Vehicle Crashes: South Carolina Case Study

Traumatic brain injuries (TBIs) are a major cause of disability and mortality, and many TBIs are caused by motor vehicle crashes. Hospital length of stay (LOS) is often regarded as a representation of consumption of hospital resources and may also represent injury severity. This study investigated the effects of crash factors and driver characteristics on hospital LOS for drivers sustaining a TBI in traffic crashes. The South Carolina Crash Outcome Data Evaluation System for 2005 to 2007 was used to build a zero-inflated negative binomial regression to predict hospital LOS for drivers experiencing a TBI as a result of a crash. Not wearing a seat belt, ejection from the vehicle, and a crash on a roadway with a high speed limit were factors associated with a greater likelihood of being admitted to the hospital and having longer hospital LOS for drivers sustaining TBIs in crashes. Intoxicated drivers were more likely to be admitted to the hospital but tended to have shorter LOS when they sustained a TBI in a crash. Hospital LOS may be a surrogate measurement of injury severity as well as a representation of health care resource consumption. Understanding the effects of crash factors on hospital LOS for drivers sustaining a TBI in crashes can help policy makers, roadway designers, and health care providers develop specific interventions for understanding traffic crash–related health care resource consumption.

110 Initial Urine Output and Pro-BNP Independently Predict Hospital Length of Stay in Acute Heart Failure

110 Initial Urine Output and Pro-BNP Independently Predict Hospital Length of Stay in Acute Heart Failure

Systematic review of risk adjustment models of hospital length of stay (LOS).

Policy decisions in health care, such as hospital performance evaluation and performance-based budgeting, require an accurate prediction of hospital length of stay (LOS). This paper provides a systematic review of risk adjustment models for hospital LOS, and focuses primarily on studies that use administrative data. MEDLINE, EMBASE, Cochrane, PubMed, and EconLit were searched for studies that tested the performance of risk adjustment models in predicting hospital LOS. We included studies that tested models developed for the general inpatient population, and excluded those that analyzed risk factors only correlated with LOS, impact analyses, or those that used disease-specific scales and indexes to predict LOS. Our search yielded 3973 abstracts, of which 37 were included. These studies used various disease groupers and severity/morbidity indexes to predict LOS. Few models were developed specifically for explaining hospital LOS; most focused primarily on explaining resource spending and the costs associated with hospital LOS, and applied these models to hospital LOS. We found a large variation in predictive power across different LOS predictive models. The best model performance for most studies fell in the range of 0.30-0.60, approximately. The current risk adjustment methodologies for predicting LOS are still limited in terms of models, predictors, and predictive power. One possible approach to improving the performance of LOS risk adjustment models is to include more disease-specific variables, such as disease-specific or condition-specific measures, and functional measures. For this approach, however, more comprehensive and standardized data are urgently needed. In addition, statistical methods and evaluation tools more appropriate to LOS should be tested and adopted.

Factors associated with length of stay following an emergency medical admission

BackgroundHospitals are under pressure to use resources in the most efficient manner. We have examined the factors predicting Length of Stay (LOS) in one institution, using a database of all episodes of emergency medical admissions prospectively collected over 12years. AimTo examine the ability to predict hospital LOS following an emergency medical hospital admission. MethodsAll emergency admissions (66,933 episodes; 36,271 patients) to St. James's Hospital, Dublin, Ireland over a 12-year period (2002–2013) were evaluated in relation to LOS. Predictor variables (identified univariately) were entered into a multiple logistic regression model to predict a longer or shorter LOS (bivariate at the median). The data was also modelled as count data (absolute LOS), using zero truncated Poisson regression methodology. Appropriate post-estimation techniques for model fit were then applied to assess the resulting model. ResultsThe major predictors of LOS included Acute Illness Severity (biochemical laboratory score at admission), Charlson co-morbidity, Manchester Triage Category at admission, Diagnosis Related Group, sepsis status (based on blood culture result), and Chronic Disease Score Indicator. The full model to predict a LOS above or below the median had an Area Under Receiver Operating Characteristic (AUROC) of 0.71 (95% CI: 0.70, 0.71). The truncated Poisson model appeared to achieve a good model fit (R2 statistic=0.76). ConclusionPredictor variables strongly correlated with LOS; there were linear increases within categories and summation between variables. More predictor variables may improve model reliability but predicting LOS ranges or quantiles may be more realistic, based on these results.

The SOAR stroke score predicts hospital length of stay in acute stroke: an external validation study.

The objective of this study is to externally validate the SOAR stroke score (Stroke subtype, Oxfordshire Community Stroke Project Classification, Age and prestroke modified Rankin score) in predicting hospital length of stay (LOS) following an admission for acute stroke. We conducted a multi-centre observational study in eight National Health Service hospital trusts in the Anglia Stroke & Heart Clinical Network between September 2008 and April 2011. The usefulness of the SOAR stroke score in predicting hospital LOS in the acute settings was examined for all stroke and then stratified by discharge status (discharged alive or died during the admission). A total of 3596 patients (mean age 77 years) with first-ever or recurrent stroke (92% ischaemic) were included. Increasing LOS was observed with increasing SOAR stroke score (p < 0.001 for both mean and median) and the SOAR stroke score of 0 had the shortest mean LOS (12 ± 20 days) while the SOAR stroke score of 6 had the longest mean LOS (26 ± 28 days). Among patients who were discharged alive, increasing SOAR stroke score had a significantly higher mean and median LOS (p < 0.001 for both mean and median) and the LOS peaked among patients with score value of 6 [mean (SD) 35 ± 31 days, median (IQR) 23 (14-48) days]. For patients who died as in-patient, there was no significant difference in mean or median LOS with increasing SOAR stroke score (p = 0.68 and p = 0.79, respectively). This external validation study confirms the usefulness of the SOAR stroke score in predicting LOS in patients with acute stroke especially in those who are likely to survive to discharge. This provides a simple prognostic score useful for clinicians, patients and service providers.

Preoperative Pain Level and Patient Expectation Predict Hospital Length of Stay After Total Hip Arthroplasty

The purpose of this study was to identify preoperative predictors of length of stay after primary total hip arthroplasty in a patient population reflecting current trends toward shorter hospitalization and using readily obtainable factors that do not require scoring systems. A retrospective review of 112 consecutive patients was performed. High preoperative pain level and patient expectation of discharge to extended care facilities (ECFs) were the only significant multivariable predictors of hospitalization extending beyond 2 days (P=0.001 and P<0.001 respectively). Patient expectation remained significant after adjusting for Medicare's 3-day requirement for discharge to ECFs (P<0.001). The study was adequately powered to analyze the variables in the multivariable logistic regression model, which had a concordance index of 0.857.

Revised Atlanta and Determinant-Based Classification: Application in a Prospective Cohort of Acute Pancreatitis Patients

Atlanta classification (Atlanta 1992) of acute pancreatitis (AP) has several limitations. Two new classification systems were recently proposed: the Atlanta reclassification (Atlanta 2012) and the determinant-based classification (DBC). The aim of our study was to: (i) determine the association between different severity categories and clinical outcomes and (ii) perform a head-to-head comparison between Atlanta 1992, Atlanta 2012, and DBC in predicting these clinical outcomes. A total of 256 prospectively enrolled patients were assigned a severity category for all three classifications. Five clinical outcomes were evaluated: mortality, intensive care unit (ICU) admission and length of stay (LOS), need for interventions, and hospital LOS. Pairwise testing between severity grades within a classification system was performed using Fisher's exact and Kruskal-Wallis tests. Predictive accuracies were evaluated using area under the ROC curve (AUC) and Somer's D co-efficient. Overall, higher grades of severity were associated with worse clinical outcomes for all three classification systems. Atlanta 2012 and DBC performed better than Atlanta 1992 and were comparable in predicting mortality (AUC 0.89 for both vs. 0.76, P<0.001), ICU admission (AUC 0.91 for both vs. 0.80, P<0.001), and ICU LOS (Somer's D 0.21 and 0.28 vs. 0.07, P<0.05). DBC performed better in predicting need for interventions (AUC 0.93 vs. 0.85, P<0.001), whereas Atlanta 2012 performed better in predicting hospital LOS (Somer's D 0.43 vs. 0.37, P=0.04). Atlanta 2012 and DBC severity categories accurately reflected clinical outcomes in our cohort and were superior to Atlanta 1992. These novel classification systems can guide the selection of homogeneous patient populations for clinical research and provide an accurate spectrum of disease severity categories in the clinical setting.