A systematic review of the prediction of hospital length of stay: Towards a unified framework

Insights from Inputs: Enhancing Revision Total Joint Arthroplasty Resource Allocation with Machine Learning Prediction

BACKGROUND

Revision total knee arthroplasty (rTKA) and revision total hip arthroplasty (rTHA) are among the most resource-intensive orthopaedic procedures. The primary aim of this study was to compare the accuracy of machine learning (ML) models between administrative and institutional datasets for predicting duration of surgery (DOS), length of stay (LOS), and 30-day hospital readmission for rTKA and rTHA based on preoperative factors and identify significant predictive features.

METHODS

A national quality improvement database was queried from 2014 to 2019, and a local institutional arthroplasty database was queried from 2012 to 2022 for rTKAs and rTHAs. Datasets were independently split into training, validation, and testing and normalized based on year. Artificial neural networks (ANNs) for both procedures and each outcome were created, and their performance was compared to multivariable regression models. Models were compared between datasets using buffer accuracy (BA). Feature importance of ANNs was retrieved using Shapley Additive exPlanations (SHAP) values.

RESULTS

A total of 22,851 and 1,025 rTKA and 14,262 and 703 rTHA patients were included from the national and institutional datasets, respectively. For DOS, the institutional ANNs outperformed the national ANNs with a 76.2 versus 45.4%, and 55.4 versus 43.1% 30-minute BA for rTKA and rTHA, respectively. For LOS, the national ANNs yielded a superior 2-day BA than the institutional ANNs of 81.8 versus 67.7%, and 71.4 versus 48.2% for rTKA and rTHA, respectively. The 30-day readmission prediction using the national database had area under the curve (AUC) scores of 0.593 and 0.590 for rTKA and rTHA, respectively.

CONCLUSION

The performance of ANN models varied significantly by dataset, but all models were superior to using historic averages. Future work should consider using accurate, arthroplasty-specific datasets based on the important features identified.

Variability in the Length of Stay and Daily Medical Expenses in Inpatient Care in Korea, 2010-2019: Hypertension and Pneumonia

This study examined the variability in the length of stay (LOS) and daily medical expenses (DME) for hypertension and pneumonia inpatient care. Using 10 years of National Health Insurance Service data (2010-2019), a multilevel analysis assessed variability at the patient and institutional levels. During the study period, the mean LOS decreased, whereas the DME increased for both hypertension and pneumonia. Institutional level variability in the LOS increased during the study period, demonstrating greater variability than that for pneumonia. For both conditions, institutional-level variability was more marked in smaller institutions (hospitals and clinics) than in larger institutions (general and tertiary hospitals). These findings indicate a need for standardized healthcare service protocols to promote consistent and efficient patient care.

Rising rates of hospitalization for dementia in people with type 2 diabetes and the general population in Australia

BackgroundFew recent studies have examined the trends in dementia hospitalization in high-income countries.ObjectiveTo estimate the trends in hospitalization for dementia in people with type 2 diabetes (T2DM) and the general population in Australia using linked, national databases.MethodsAustralians with T2DM and registered on the National Diabetes Services Scheme (n = 438,264), and the general population (n = 8,090,993) from 2010-2011 to 2016-2017 served as the study cohort. Annual rates of hospitalization for dementia were calculated for these individuals aged ≥50 years. Following this, the trends in the rate of hospitalization were estimated using joinpoint regression and summarized as annual percent changes (APCs).ResultsIncreases in hospitalization for dementia over time were observed for the T2DM and the general population; APC 5.2 (95% CI 3.5, 7.3) and 9.4 (95% CI 3.8, 14.3), respectively. The absolute age- and sex-standardized rate of dementia hospitalization was found to be higher in the T2DM than the general population. For vascular dementia, a higher rate of hospitalization was observed for the T2DM population compared to the general population. Conversely, the rate of hospitalization for Alzheimer's disease was higher in the general population than in the T2DM cohort. Further, a higher dementia hospitalization rate was observed among males compared to females in both T2DM, and the general population.ConclusionsDespite the previous studies reporting a decline in dementia incidence in high-income countries, the rate of dementia hospitalization in Australia has risen steadily from 2010-2016 in both T2DM individuals and the general population.

Clinical governance in internal medicine: the challenge of length of stay

Clinical governance highlights the relevance of the combination of "quality" and "safety" with "excellence" and "improvement". In hospitals, as highly complex organizations, the principles and practices of clinical governance are key elements for success. Several angles would be pertinent in the application of clinical governance to the hospital context, and specifically to internal medicine. The length of stay of patients in hospital is frequently used as a quality indicator of clinical activity. Ideally the patient should remain in the hospital during the time in which he has benefit, minimizing inherent risks. With its centrality in the patient, clinical governance, interpreted in the light of the Seven Pillars Model, can provide a comprehensive framework for addressing the management of length of stay. We should try to bring to the day-to-day of our organizations the aggregating vision of clinical governance, whose foundational values need to be reinforced and deepened.

Bayesian Analysis of Length of Stay Determinants in ERAS-Guided Hip Arthroplasty

Background and Objectives: Total hip arthroplasty in China expanded rapidly post-2019. The length of hospital stay in these procedures reflects healthcare quality standards. This study analyzed the correlation between preoperative clinical factors and the length of hospital stay in total hip arthroplasty patients managed via an enhanced recovery after surgery protocol. Methods: Preoperative clinical variables were collected from total hip arthroplasty patients in an accelerated rehabilitation program. One-way ANOVA and other statistical methods analyzed correlations between these data and hospitalization time. Results: A total of 408 patients were included, with a mean length of stay of 12.01 ± 4.281 days. Right lower extremity strength (t = 2.794, p = 0.005), activities of daily living score (t = -3.481, p = 0.001), C-reactive protein (t = -2.514, p = 0.016), thrombin time (t = -2.393, p = 0.019), and prothrombin activity (t = 2.582, p = 0.013) can directly affect the length of stay in patients with total hip arthroplasty. Also, age (F = 1.958, p = 0.006) and erythrocyte sedimentation rate (t = -2.519, p = 0.015) were found to affect the length of hospital stay indirectly. Conclusions: This study demonstrated that right lower extremity strength, activities of daily living score, C-reactive protein, thrombin time, and prothrombin activity significantly influence the length of hospital stay in enhanced recovery after surgery-managed total hip arthroplasty patients. Therefore, early interventions should be made to address the above factors.

Quadriceps muscle thickness as measured by point-of-care ultrasound is associated with hospital length of stay among hospitalised older patients

BACKGROUND

Predicting hospital length of stay (LOS) can potentially improve healthcare resource allocation. Recent studies suggest that point-of-care ultrasound (POCUS), specifically measurements of muscle thickness (MT), may be valuable in assessing patient outcomes, including LOS. This study investigates the hypothesis that quadriceps MT and echo intensity (EI) can predict patient outcomes, particularly LOS.

METHODS

Quadriceps MT and EI were measured using POCUS in patients admitted to a hospital's acute medical unit. Predictor variables included age, sex, MT, EI and the Charlson Comorbidity Index (CCI). The outcome variable was hospital LOS.

RESULTS

One hundred twenty participants were included (average age 76 ± 7, with 64 women and 56 men). The mean LOS was 27 ± 31 days, and the mean MT was 20 ± 6 mm. Sex-based differences in MT were statistically significant (P = .032). Patients with prolonged LOS over 30 days had lower MT (mean 17 mm vs. 21 mm, P < .0001). One unit increase in MT was significantly associated with ~1.5 fewer days of hospital LOS, and one CCI score increase was associated with almost three more days of hospital LOS. Having low MT significantly increased the odds of staying in the hospital longer than 30 days by more than three times in all models.

CONCLUSION

Muscle thickness is a strong predictor of hospital LOS, highlighting the potential of POCUS for assessing patient outcomes.

Leveraging deep neural network and language models for predicting long-term hospitalization risk in schizophrenia

Early warning of long-term hospitalization in schizophrenia (SCZ) patients at the time of admission is crucial for effective resource allocation and individual treatment planning. In this study, we developed a deep learning model that integrates demographic, behavioral, and blood test data from admission to forecast extended hospital stays using a retrospective cohort. By utilizing language models, our developed algorithm efficiently extracts 95% of the unstructured electronic health records data needed for this work, while ensuring data privacy and low error rate. This paradigm has also been demonstrated to have significant advantages in reducing potential discrimination and erroneous dependencies. By utilizing multimodal features, our deep learning model achieved a classification accuracy of 0.81 and an AUC of 0.9. Key risk factors identified included advanced age, longer disease duration, and blood markers such as elevated neutrophil-to-lymphocyte ratio, lower lymphocyte percentage, and reduced albumin levels, validated through comprehensive interpretability analyses and ablation studies. The inclusion of multimodal data significantly improved prediction performance, with demographic variables alone achieving an accuracy of 0.73, which increased to 0.81 with the addition of behavioral and blood test data. Our approach outperformed traditional machine learning methods, which were less effective in predicting long-term stays. This study demonstrates the potential of integrating diverse data types for enhanced predictive accuracy in mental health care, providing a robust framework for early intervention and personalized treatment in SCZ management.

Development of Multiservice Machine Learning Models to Predict Postsurgical Length of Stay and Discharge Disposition at the Time of Case Posting

Objective

Develop machine learning (ML) models to predict postsurgical length of stay (LOS) and discharge disposition (DD) for multiple services with only the data available at the time of case posting.

Background

Surgeries are scheduled largely based on operating room resource availability with little attention to downstream resource availability such as inpatient bed availability and the care needs after hospitalization. Predicting postsurgical LOS and DD at the time of case posting could support resource allocation and earlier discharge planning.

Methods

This retrospective study included 63,574 adult patients undergoing elective inpatient surgery at a large academic health system. We used surgical case data available at the time of case posting and created gradient-boosting decision tree classification models to predict LOS as short (≤1 day), medium (2-4 days), and prolonged stays (≥5 days) and DD as home versus nonhome.

Results

The LOS model achieved an area under the receiver operating characteristic curve (AUC) of 0.81. Adding relative value unit and historical LOS through the similarity cascade increased the accuracy of short and prolonged LOS prediction by 9.0% and 3.9% to 72.9% and 74%, respectively, compared with a model without these features (P = 0.001). The DD model had an AUC of 0.88 for home versus nonhome prediction.

Conclusions

We developed ML models to predict, at the time of case posting, the postsurgical LOS and DD for adult elective inpatient cases across multiple services. These models could support case scheduling, resource allocation, optimal bed utilization, earlier discharge planning, and preventing case cancelation due to bed unavailability.

Optimizing length of hospital stay among inpatients with spinal cord injury: An observational study

INTRODUCTION AND OBJECTIVES

Despite the importance of length of stay (LOS) following spinal cord injury, it remains underexplored in the literature. This study aims to bridge this gap by investigating the association between rehabilitation LOS and functional gains among patients with traumatic (TSCI) or non-traumatic (NTSCI) spinal cord injuries.

METHODS

We conducted a retrospective observational cohort study assessing functional gains using the motor Functional Independence Measure (mFIM) and the Spinal Cord Independence Measure (SCIM III) from rehabilitation admission to discharge. Outcomes were analyzed across four neurological categories based on the American Spinal Injury Association Impairment Scale (AIS): C1-C4 AIS A-C; C5-8 AIS A-C; T1-S5 AIS A-C; and AIS D. Linear regression models estimated changes across rehabilitation LOS quarters (Q1-Q4), adjusting for covariates.

RESULTS

We included 1036 patients admitted for rehabilitation between 2007 and 2023 (46.3% TSCI, 53.7% NTSCI). TSCI: age 42.7, 80.2% male, 41.8% AIS A, LOS 90.5. NTSCI: age 55.7, 54.2% male, 14.2% AIS A, LOS 69.6. For TSCI, mFIM and SCIM III gains increased significantly from Q1 to Q2 (T1-S5-ABC, n=214) and Q2 to Q3 (AIS D, n=129). For NTSCI, gains increased from Q2 to Q3 (T1-S5-ABC, n=195) and from Q1 to Q2 as well as from Q2 to Q3 (AIS D, n=304). Adjusted models showed decreasing gains for Q2 and Q3 vs. Q1 (TSCI) but increasing gains for Q2-Q4 vs. Q1 (NTSCI) for both measures. No significant gains were observed from Q3 to Q4.

CONCLUSIONS

We identified specific neurological categories and LOS quarters yielding to significant functional gains.

A Precision Medicine Model for Targeted Antibiotic Therapy in Urinary Tract Infections: A Valuable Tool to Reduce Hospitalization Stay and the Time to Switch to Oral Treatment

Background/Objectives: The management of urinary tract infections (UTIs) has become an increasingly challenging medical intervention. This study explores whether adoption of a precision medicine model could improve the management of acute uncomplicated pyelonephritis (uAPN) or complicated UTIs (cUTIs) compared with the standard of care approach, in hospitalized patients. Methods: From January 2022 to March 2024, all patients affected by uAPN or cUTIs and attending our urological institution were randomized to receive the following: antibiotic treatment according to guidelines and recommendations (standard of care group) or antibiotic treatment according to the precision medical model (intervention group). The main outcome measures were the rates of clinical success and the length of hospitalization. The time until switching to oral treatment was regarded as a secondary outcome measure. Results: Eighty-three patients were enrolled in the standard of care group, while seventy-nine patients were enrolled in the intervention group. While the overall clinical success rate was similar in the two groups (75 vs. 72; p = 0.97), a statistically significant difference was observed between the two groups in terms of length of hospitalization (8 days vs. 5 days; p = 0.03) and time to switch to oral treatment (96 h vs. 72 h; p = 0.04). A statistically significant difference was found between the two groups regarding the need to change antimicrobial therapy during hospitalization [12 out of 80 vs. 6 out of 77; p = 0.04]. Conclusions: Adoption of the precision medicine model appears as a valuable means to improve the management of patients with uAPN and cUTIs. By reducing the period of hospitalization and the time to switch to oral treatment, the precision medicine model also improves antimicrobial stewardship in the management of UTIs.

Equitable hospital length of stay prediction for patients with learning disabilities and multiple long-term conditions using machine learning

Purpose

Individuals with learning disabilities (LD) often face higher rates of premature mortality and prolonged hospital stays compared to the general population. Predicting the length of stay (LOS) for patients with LD and multiple long-term conditions (MLTCs) is critical for improving patient care and optimising medical resource allocation. However, there is limited research on the application of machine learning (ML) models to this population. Furthermore, approaches designed for the general population often lack generalisability and fairness, particularly when applied across sensitive groups within their cohort.

Method

This study analyses hospitalisations of 9,618 patients with LD in Wales using electronic health records (EHR) from the SAIL Databank. A Random Forest (RF) ML model was developed to predict hospital LOS, incorporating demographics, medication history, lifestyle factors, and 39 long-term conditions. To address fairness concerns, two bias mitigation techniques were applied: a post-processing threshold optimiser and an in-processing reductions method using an exponentiated gradient. These methods aimed to minimise performance discrepancies across ethnic groups while ensuring robust model performance.

Results

The RF model outperformed other state-of-the-art models, achieving an area under the curve of 0.759 for males and 0.756 for females, a false negative rate of 0.224 for males and 0.229 for females, and a balanced accuracy of 0.690 for males and 0.689 for females. Bias mitigation algorithms reduced disparities in prediction performance across ethnic groups, with the threshold optimiser yielding the most notable improvements. Performance metrics, including false positive rate and balanced accuracy, showed significant enhancements in fairness for the male cohort.

Conclusion

This study demonstrates the feasibility of applying ML models to predict LOS for patients with LD and MLTCs, while addressing fairness through bias mitigation techniques. The findings highlight the potential for equitable healthcare predictions using EHR data, paving the way for improved clinical decision-making and resource management.

Prediction of Intensive Care Length of Stay for Surviving and Nonsurviving Patients Using Deep Learning

OBJECTIVES

Length of stay (LOS) models support evaluating ICU care; however, current benchmarking models fail to consider differences in LOS between surviving and nonsurviving patients, which can lead to biased predictions toward the surviving population. We aim to develop a model addressing this as well as documentation bias to improve ICU benchmarking.

DESIGN

The Critical Care Outcomes Prediction Model (CCOPM) LOS uses patient characteristics, vitals, and laboratories during the first 24 hours of ICU admission to predict LOS in the hospital and ICU using a deep learning framework for modeling time to events with competing risk. Data was randomly divided into training, validation, and test (hold out) sets in a 2:1:1 ratio.

SETTING

Electronic ICU Research Institute database from participating tele-critical care programs.

PATIENTS

Six hundred sixty-nine thousand eight hundred seventy-six ICU admissions pertaining to 628,815 patients from 329 ICUs in 194 U.S. hospitals, from 2017 to 2019.

INTERVENTIONS

None.

MEASUREMENTS AND MAIN RESULTS

Model performance was assessed using the coefficient of determination (R2), concordance index, mean absolute error, and calibration. For individual stays in the test set, the ICU LOS model presented R2 = 0.29 and 0.23 for surviving and nonsurviving populations, respectively, at the individual level and R2 = 0.48 and 0.23 at the ICU level. Conversely, hospital LOS model presented R2 = 0.46 and 0.52 at the individual level and R2 = 0.71 and 0.64 at the ICU level. In the subset of the test set containing predictions from Acute Physiology and Chronic Health Evaluation (APACHE) IVb, R2 of ICU LOS for surviving and nonsurviving populations was, respectively, 0.30 and 0.23 for the CCOPM and 0.16 and zero for APACHE IVb. For hospital LOS, the values were R2 = 0.39 and 0.40 for the CCOPM and 0.27 and zero for APACHE IVb.

CONCLUSIONS

This novel LOS model represents a step forward in achieving more equitable benchmarking across diverse ICU settings with varying risk profiles.

Length of Hospital Stay and Its Predictions Among Patients With Exacerbations of Chronic Respiratory Diseases

BACKGROUND

Chronic respiratory disorders such as asthma and chronic obstructive pulmonary disease (COPD) may deteriorate into acute exacerbations requiring hospitalization. Assessing the predictors of prolonged hospital stays could help identify potential interventions to reduce the burden on patients and healthcare systems.

AIM

This study aimed to identify the risk factors attributed to prolonged hospital stays among patients admitted with acute exacerbations of chronic respiratory disorders in Jordan.

METHODS

A retrospective cohort study was conducted by reviewing the demographic and clinical characteristics of hospitalized patients with asthma and COPD exacerbations between January 2017 and July 2021. The recorded variables were checked for their independence. Simple and stepwise multivariate linear regressions were then performed to identify variables associated significantly with a longer hospital length of stay (LOS).

RESULTS

A total of 896 cases were evaluated. The mean ± SD stay was 5.66 ± 3.40 days, whereas the median (IQR) was 5.00 (4.00) days. Variables associated significantly with prolonged LOS in the multivariate analysis were female gender (β = 0.089, p = 0.011), pulmonary hypertension (β = 0.093, p = 0.004), allergic rhinitis (β = 0.086, p = 0.007), ICU admission (β = 0.096, p = 0.003), requirement for mechanical ventilation (β = 0.102, p = 0.002), higher total number of medications (β = 0.281, p < 0.001) and the number of exacerbation-related medications (β = 0.200, p < 0.001). However, smoking (β = -0.091, p = 0.008) was significantly associated with a shorter LOS.

CONCLUSIONS

Gender, pulmonary hypertension, allergic rhinitis, ICU admission, mechanical ventilation, the number of medications and smoking were significantly related to LOS. These findings emphasize the importance of patients' demographics and their clinical status in determining LOS, hence providing protective interventions to shorten it.

Immunotherapy-Related Adverse Events and Clinical Outcomes in Adult Solid-Tumor Patients Admitted to an Onco-Hospitalist Medicine Service

Background/Objectives: Few studies have focused on patients with immune-related adverse events (irAEs) after immune checkpoint inhibitor (ICI) treatment who were cared for primarily by hospitalists. The objective of our study was to describe the patterns and outcomes of adult solid-tumor cancer patients admitted to our onco-hospital medicine service. Methods: We retrospectively reviewed patients with solid tumors who received ICIs and were admitted to our service in 2021-2022 with an irAE and compared them to a control group (IOTOX vs. NO IOTOX, respectively). The primary outcome was the patterns of irAEs requiring hospitalization; secondary outcomes included 30-day emergency room visit, readmission, and 30-day mortality. Results: There were 144 patients in the IOTOX group and 286 controls. The most common tumor type was lung and thoracic malignancies (62, 43.1%). The most common ICI causing the irAEs was pembrolizumab (66, 45.8%). The most common irAEs were pneumonitis (49, 34%), colitis (28, 19.4%), hepatitis (18, 12.5%), and myocarditis (16, 11.1%). Of the 144 patients, eight (6%) died from the hospitalization irAE. Fifteen (15.6%) had an ER visit within 30 days due to the same irAE, and thirteen (13.7%) were readmitted. Survival at 30 days after discharge did not differ significantly between groups. Conclusions: Despite many patients having severe irAEs and irAEs associated with higher mortality, they generally had a favorable outcome compared to the literature.

Transformer-based model for predicting length of stay in intensive care unit in sepsis patients

Introduction

Sepsis, a life-threatening condition with a high mortality rate, requires intensive care unit (ICU) admission. The increasing hospitalization rate for patients with sepsis has escalated medical costs due to the strain on ICU resources. Efficient management of ICU resources is critical to addressing this challenge.

Methods

This study utilized the dataset collected from 521 patients with sepsis at Chungbuk National University Hospital between July 2020 and August 2023. A transformer-based deep learning model was developed to predict ICU length of stay (LOS). The model incorporated global and local input data analysis through classification and feature-wise tokens, based on sequential organ failure assessment (SOFA) criteria. Model performance was evaluated using four-fold cross-validation.

Results

The proposed model achieved a mean absolute error (MAE) of 2.05 days for predicting ICU LOS. The result demonstrates the ability of the proposed model to provide accurate and reliable predictions.

Discussion

The proposed model offers valuable insights for healthcare resource management by optimizing ICU resource allocation and potentially reducing medical expenses. These findings highlight the applicability of the proposed model to efficient healthcare cost management.

Interpretable machine learning for time-to-event prediction in medicine and healthcare

Time-to-event prediction, e.g. cancer survival analysis or hospital length of stay, is a highly prominent machine learning task in medical and healthcare applications. However, only a few interpretable machine learning methods comply with its challenges. To facilitate a comprehensive explanatory analysis of survival models, we formally introduce time-dependent feature effects and global feature importance explanations. We show how post-hoc interpretation methods allow for finding biases in AI systems predicting length of stay using a novel multi-modal dataset created from 1235 X-ray images with textual radiology reports annotated by human experts. Moreover, we evaluate cancer survival models beyond predictive performance to include the importance of multi-omics feature groups based on a large-scale benchmark comprising 11 datasets from The Cancer Genome Atlas (TCGA). Model developers can use the proposed methods to debug and improve machine learning algorithms, while physicians can discover disease biomarkers and assess their significance. We contribute open data and code resources to facilitate future work in the emerging research direction of explainable survival analysis.

The application of machine learning algorithms for predicting length of stay before and during the COVID-19 pandemic: evidence from Wuhan-area hospitals

Objective

The COVID-19 pandemic has placed unprecedented strain on healthcare systems, mainly due to the highly variable and challenging to predict patient length of stay (LOS). This study aims to identify the primary factors impacting LOS for patients before and during the COVID-19 pandemic.

Methods

This study collected electronic medical record data from Zhongnan Hospital of Wuhan University. We employed six machine learning algorithms to predict the probability of LOS.

Results

After implementing variable selection, we identified 35 variables affecting the LOS for COVID-19 patients to establish the model. The top three predictive factors were out-of-pocket amount, medical insurance, and admission deplanement. The experiments conducted showed that XGBoost (XGB) achieved the best performance. The MAE, RMSE, and MAPE errors before and during the COVID-19 pandemic are lower than 3% on average for household registration in Wuhan and non-household registration in Wuhan.

Conclusions

Research finds machine learning is reasonable in predicting LOS before and during the COVID-19 pandemic. This study offers valuable guidance to hospital administrators for planning resource allocation strategies that can effectively meet the demand. Consequently, these insights contribute to improved quality of care and wiser utilization of scarce resources.

A Multimodal Machine Learning Model in Pneumonia Patients Hospital Length of Stay Prediction

Hospital overcrowding, driven by both structural management challenges and widespread medical emergencies, has prompted extensive research into machine learning (ML) solutions for predicting patient length of stay (LOS) to optimize bed allocation. While many existing models simplify the LOS prediction problem to a classification task, predicting broad ranges of hospital days, an exact day-based regression model is often crucial for precise planning. Additionally, available data are typically limited and heterogeneous, often collected from a small patient cohort. To address these challenges, we present a novel multimodal ML framework that combines imaging and clinical data to enhance LOS prediction accuracy. Specifically, our approach uses the following: (i) feature extraction from chest CT scans via a convolutional neural network (CNN), (ii) their integration with clinically relevant tabular data from patient exams, refined through a feature selection system to retain only significant predictors. As a case study, we applied this framework to pneumonia patient data collected during the COVID-19 pandemic at two hospitals in Naples, Italy—one specializing in infectious diseases and the other general-purpose. Under our experimental setup, the proposed system achieved an average prediction error of only three days, demonstrating its potential to improve patient flow management in critical care environments.

Identifying Markov Chain Models from Time-to-Event Data: An Algebraic Approach

Many biological and medical questions can be modeled using time-to-event data in finite-state Markov chains, with the phase-type distribution describing intervals between events. We solve the inverse problem: given a phase-type distribution, can we identify the transition rate parameters of the underlying Markov chain? For a specific class of solvable Markov models, we show this problem has a unique solution up to finite symmetry transformations, and we outline a recursive method for computing symbolic solutions for these models across any number of states. Using the Thomas decomposition technique from computer algebra, we further provide symbolic solutions for any model. Interestingly, different models with the same state count but distinct transition graphs can yield identical phase-type distributions. To distinguish among these, we propose additional properties beyond just the time to the next event. We demonstrate the method's applicability by inferring transcriptional regulation models from single-cell transcription imaging data.

Predicting hospital bed utilisation for post-surgical care by means of the Monte Carlo method with historical data

Objective This study aim was to develop a predictive model of bed utilisation to support the decision process of elective surgery planning and bed management to improve post-surgical care. Methods This study undertook a retrospective analysis of de-identified data from a tertiary metropolitan hospital in Southeast Queensland, Australia. With a reference sample from 2years of historical data, a model based on the Monte Carlo method has been developed to predict hospital bed utilisation for post-surgical care of patients who have undergone surgical procedures. A separate test sample from comparable data of 8weeks of actual utilisation was employed to assess the performance of the prediction model. Results Applying the developed prediction model to an 8-week period test sample, the mean percentage error of the prediction was 1.5% and the mean absolute percentage error 5.4%. Conclusions The predictive model developed in this study may assist in bed management and the planning process of elective surgeries, and in so doing also reduce the likelihood of Emergency Department access block.

A need for systems thinking and the appliance of (complexity) science in healthcare

Hospitals represent complex adaptive systems where interactions and relationships of different components both affect and shape the way they work simultaneously. Pressures on hospitals determine how they behave and many of the problems seen in the NHS and indeed other health services can be viewed through the lens of complexity science and systems thinking. 'Flow' of patients through the hospital can be seen as an indicator of how well the hospital 'system' is working. The better flow is, the more patients can be treated and the less time is spent waiting in the various queues that accrue around the hospital, In this article, we explore the impact of these disciplines on patient flow and examine how short-term and overly simple solutions can exacerbate problems in the health service, despite the best intentions of those working in it. Many of today's problems can be described in terms of 'system archetypes' and 'game theory'. Understanding this may lead to improvement in how services are redesigned to solve these problems.

Hospital Length of Stay Prediction for Planned Admissions Using Observational Medical Outcomes Partnership Common Data Model: Retrospective Study

BACKGROUND

Accurate hospital length of stay (LoS) prediction enables efficient resource management. Conventional LoS prediction models with limited covariates and nonstandardized data have limited reproducibility when applied to the general population.

OBJECTIVE

In this study, we developed and validated a machine learning (ML)-based LoS prediction model for planned admissions using the Observational Medical Outcomes Partnership Common Data Model (OMOP CDM).

METHODS

Retrospective patient-level prediction models used electronic health record (EHR) data converted to the OMOP CDM (version 5.3) from Seoul National University Bundang Hospital (SNUBH) in South Korea. The study included 137,437 hospital admission episodes between January 2016 and December 2020. Covariates from the patient, condition occurrence, medication, observation, measurement, procedure, and visit occurrence tables were included in the analysis. To perform feature selection, we applied Lasso regularization in the logistic regression. The primary outcome was an LoS of 7 days or longer, while the secondary outcome was an LoS of 3 days or longer. The prediction models were developed using 6 ML algorithms, with the training and test set split in a 7:3 ratio. The performance of each model was evaluated based on the area under the receiver operating characteristic curve (AUROC) and the area under the precision-recall curve (AUPRC). Shapley Additive Explanations (SHAP) analysis measured feature importance, while calibration plots assessed the reliability of the prediction models. External validation of the developed models occurred at an independent institution, the Seoul National University Hospital.

RESULTS

The final sample included 129,938 patient entry events in the planned admissions. The Extreme Gradient Boosting (XGB) model achieved the best performance in binary classification for predicting an LoS of 7 days or longer, with an AUROC of 0.891 (95% CI 0.887-0.894) and an AUPRC of 0.819 (95% CI 0.813-0.826) on the internal test set. The Light Gradient Boosting (LGB) model performed the best in the multiclassification for predicting an LoS of 3 days or more, with an AUROC of 0.901 (95% CI 0.898-0.904) and an AUPRC of 0.770 (95% CI 0.762-0.779). The most important features contributing to the models were the operation performed, frequency of previous outpatient visits, patient admission department, age, and day of admission. The RF model showed robust performance in the external validation set, achieving an AUROC of 0.804 (95% CI 0.802-0.807).

CONCLUSIONS

The use of the OMOP CDM in predicting hospital LoS for planned admissions demonstrates promising predictive capabilities for stays of varying durations. It underscores the advantage of standardized data in achieving reproducible results. This approach should serve as a model for enhancing operational efficiency and patient care coordination across health care settings.

Effect of goal-directed mobilisation versus standard care on physical functioning among medical inpatients: the GoMob-in randomised, controlled trial

OBJECTIVE

To assess the effect of goal-directed mobilisation (GDM) on physical functioning in medical inpatients.

DESIGN

Randomised, controlled, single-centre, parallel, superiority trial with a 3-month follow-up and blinded outcome assessment.

SETTING

General internal medicine wards of a Swiss tertiary acute hospital, September 2021 to April 2023.

PARTICIPANTS

Adults with expected hospitalisation of ≥5 days, physiotherapy prescription and ability to follow study procedures.

INTERVENTION

GDM during hospitalisation, which includes personal goal setting and a short session of patient education through a physiotherapist (experimental group), versus standard care (control group).

OUTCOME MEASURES

The primary outcome was the change in physical activity between baseline and day 5 (De Morton Mobility Index (DEMMI)). Secondary outcomes included in-hospital accelerometer-measured mobilisation time; in-hospital falls; delirium; length of stay; change in independence in activities of daily living, concerns of falling and quality of life; falls, readmission and mortality within 3 months.

RESULTS

The study was completed by 123 of 162 (76%) patients enrolled, with the primary outcome collected at day 5 in 126 (78%) participants. DEMMI Score improved by 8.2 (SD 15.1) points in the control group and 9.4 (SD 14.2) in the intervention group, with a mean difference of 0.3 (adjusted for the stratification factors age and initial DEMMI Score, 95% CI -4.1 to 4.8, p=0.88). We did not observe a statistically significant difference in effects of the interventions on any secondary outcome.

CONCLUSIONS

The patient's physical functioning improved during hospitalisation, but the improvement was similar for GDM and standard of care. Improving physical activity during an acute medical hospitalisation remains challenging. Future interventions should target additional barriers that can be implemented without augmenting resources.

TRIAL REGISTRATION NUMBER

NCT04760392.

New horizons in hospital-associated deconditioning: a global condition of body and mind

Hospital-associated deconditioning is a broad term, which refers non-specifically to declines in any function of the body secondary to hospitalisation. Older people, particularly those living with frailty, are known to be at greatest risk. It has historically been most commonly used as a term to describe declines in muscle mass and function (i.e. acute sarcopenia). However, declines in physical function do not occur in isolation, and it is recognised that cognitive deconditioning (defined by delayed mental processing as part of a spectrum with fulminant delirium at one end) is commonly encountered by patients in hospital. Whilst the term 'deconditioning' is descriptive, it perhaps leads to under-emphasis on the inherent organ dysfunction that is associated, and also implies some ease of reversibility. Whilst deconditioning may be reversible with early intervention strategies, the long-term effects can be devastating. In this article, we summarise the most recent research on this topic including new promising interventions and describe our recommendations for implementation of tools such as the Frailty Care Bundle.

Challenges and opportunities in enhanced recovery after surgery programs: An overview

Enhanced Recovery After Surgery (ERAS) programs were developed as evidence-based, multi-disciplinary interventions in all the perioperative phases to minimise the surgical stress response, reduce complications, and enhance outcomes. The results across various surgical procedures have been positive, with a reduction in medical complications, a reduction in length of hospital stay, and a reduction in care costs without increased re-admission rates. However, implementation for many institutions has not been easy and suboptimal at best. The robust and pervasive adoption of these programs should be based on effective change management, dynamic and engaged clinical leadership, adherence to the principles of continuous quality improvement programs, and the adoption of evidence-based and data-driven changes in pathway development and implementation. Rapid cycle, randomised/quasi-randomised quality improvement projects must be the core foundation of an ERAS program. Finally, research methodologies should focus on controlling for adherence to the core elements of the pathways and testing for the effectiveness of an individual intervention in a randomised controlled trial.

Does the level of mobility on ICU discharge impact post-ICU outcomes? A retrospective analysis

PURPOSE

Mobilisation is a common intervention in Intensive Care (ICU). However, few studies have explored the relationship between mobility levels and outcomes. This study assessed the association of the level of mobility on ICU discharge with discharge destination from the hospital and hospital length of stay.

MATERIALS AND METHODS

A retrospective analysis of data from 522 patients admitted to a single UK general ICU who were ventilated for ≥5 days was performed. The level of mobility was assessed using the Manchester Mobility Score (MMS). Multivariable regression analysed demographic and clinical variables for the independence of association with discharge destination and hospital length of stay.

RESULTS

MMS ≥5 on ICU discharge was independently associated with discharge destination and hospital LOS (p < 0.001). Patients achieving MMS ≥5 on ICU discharge were more likely to be discharged home (OR 3.86 95% CI 2.1 to 6.9, p < 0.001), and had an 11.8 day shorter hospital LOS (95% CI -17.6 to -6.1, p < 0.001).

CONCLUSIONS

The ability to step transfer to a chair (MMS ≥5) before ICU discharge was independently associated with discharge to usual residence and hospital LOS, irrespective of preadmission morbidity. Increasing the level of patient mobility at ICU discharge should be a key focus of rehabilitation interventions.

Characterization of Telecare Conversations on Lifestyle Management and Their Relation to Health Care Utilization for Patients with Heart Failure: Mixed Methods Study

BACKGROUND

Telehealth interventions where providers offer support and coaching to patients with chronic conditions such as heart failure (HF) and type 2 diabetes mellitus (T2DM) are effective in improving health outcomes. However, the understanding of the content and structure of these interactions and how they relate to health care utilization remains incomplete.

OBJECTIVE

This study aimed to characterize the content and structure of telecare conversations on lifestyle management for patients with HF and investigate how these conversations relate to health care utilization.

METHODS

We leveraged real-world data from 50 patients with HF enrolled in a postdischarge telehealth program, with the primary intervention comprising a series of telephone calls from nurse telecarers over a 12-month period. For the full cohort, we transcribed 729 English-language calls and annotated conversation topics. For a subcohort (25 patients with both HF and T2DM), we annotated lifestyle management content with fine-grained dialogue acts describing typical conversational structures. For each patient, we identified calls with unusually high ratios of utterances on lifestyle management as lifestyle-focused calls. We further extracted structured data for inpatient admissions from 6 months before to 6 months after the intervention period. First, to understand conversational structures and content of lifestyle-focused calls, we compared the number of utterances, dialogue acts, and symptom attributes in lifestyle-focused calls to those in calls containing but not focused on lifestyle management. Second, to understand the perspectives of nurse telecarers on these calls, we conducted an expert evaluation where 2 nurse telecarers judged levels of concern and follow-up actions for lifestyle-focused and other calls (not focused on lifestyle management content). Finally, we assessed how the number of lifestyle-focused calls relates to the number of admissions, and to the average length of stay per admission.

RESULTS

In comparative analyses, lifestyle-focused calls had significantly fewer utterances (P=.01) and more dialogue acts (Padj=.005) than calls containing but not focused on lifestyle management. Lifestyle-focused calls did not contain deeper discussions on clinical symptoms. These findings indicate that lifestyle-focused calls entail short, intense discussions with greater emphasis on understanding patient experience and coaching than on clinical content. In the expert evaluation, nurse telecarers identified 24.2% (29/120) of calls assessed as concerning enough for follow-up. For these 29 calls, nurse telecarers were more attuned to concerns about symptoms and vitals (19/29, 65.5%) than lifestyle management concerns (4/29, 13.8%). The number of lifestyle-focused calls a patient had was modestly (but not significantly) associated with a lower average length of stay for inpatient admissions (Spearman ρ=-0.30; Padj=.06), but not with the number of admissions (Spearman ρ=-0.03; Padj=.84).

CONCLUSIONS

Our approach and findings offer novel perspectives on the content, structure, and clinical associations of telehealth conversations on lifestyle management for patients with HF. Hence, our study could inform ways to enhance telehealth programs for self-care management in chronic conditions.

Streamlining management in thoracic trauma: radiomics- and AI-based assessment of patient risks

BACKGROUND

In blunt chest trauma, patient management is challenging because clinical guidelines miss tools for risk assessment. No clinical scale reliably measures the severity of cases and the chance of complications.

AIM

The objective of the study was to optimize the management of patients with blunt chest trauma by creating models prognosticating the transfer to the intensive care unit and in-hospital length of stay (LOS).

METHODS

The study cohort consisted of 212 cases. We retrieved information on the cases from the hospital's trauma registry. After segmenting the lungs with Lung CT Analyzer, we performed volumetric feature extraction with data-characterization algorithms in PyRadiomics.

RESULTS

To predict whether the patient will require intensive care, we used the three groups of findings: ambulance, admission, and radiomics data. When trained on the ambulance data, the models exhibited a borderline performance. The metrics improved after we retrained the models on a combination of ambulance, laboratory, radiologic, and physical examination data (81.5% vs. 94.4% Sn). Radiomics data were the top-accurate predictors (96.3% Sn). Age, vital signs, anthropometrics, and first aid time were the best-performing features collected by the ambulance service. Laboratory findings, AIS scores for the lower extremity, abdomen, head, and thorax constituted the top-rank predictors received on admission to the hospital. The original first-order kurtosis had the highest predictive value among radiomics data. Top-informative radiomics features were derived from the right hemithorax because the right lung is larger. We constructed regression models that can adequately reflect the in-hospital LOS. When trained on different groups of data, the machine-learning regression models showed similar performance (MAE/ROV ≈ 8%). Anatomic scores for the body parts other than thorax and laboratory markers of hemorrhage had the highest predictive value. Hence, the number of injured body parts correlated with the case severity.

CONCLUSION

The study findings can be used to optimize the management of patients with a chest blunt injury as a specific case of monotrauma. The models we built may help physicians to stratify patients by risk of worsening and overcome the limitations of existing tools for risk assessment. High-quality AI models trained on radiomics data demonstrate superior performance.

Exploring the Association between Complexity of Care, Medical Complexity, and Length of Stay in the Paediatric Setting Using a Nursing Minimum Data Set: A Study Protocol

BACKGROUND/OBJECTIVES

The complexity of care requires systematic documentation to fully understand its relationship with medical complexity and its impact on patient outcomes. The Nursing Minimum Data Set (NMDS) plays a crucial role by capturing essential nursing data, enabling a detailed analysis of care and its impact on outcomes, such as length of stay (LOS). However, despite its potential, the use of NMDS in paediatric care remains limited. This study aims to explore the association between nursing and medical complexities and LOS in paediatric patients.

METHODS

A descriptive, retrospective, monocentric study will be conducted. The data will be collected through a nursing information system (Professional Assessment Instrument (PAIped)) and the hospital discharge register of patients admitted to the paediatric department in 2022 in an Italian university hospital. Conclusions and Expected Results: The use of PAIped will allow for the description of the complexity of care and enable an analysis of its relationship with medical complexity and LOS.

The Gap Between the Actual Cost and Tariffs of Global Surgical Procedures: A Retrospective Cross-sectional Study in Qazvin Province, Iran

BACKGROUND

Iran's healthcare system has a significant discrepancy between the national tariff and the cost of global surgical procedures (GSPs). This study aimed to compare the actual costs of GSPs with national tariffs in Iran's public hospitals.

METHODS

This retrospective cross-sectional study was conducted in 2017. Using the census method, 6126 GSPs performed in three public hospitals were investigated in this study. Additionally, national tariffs from the Supreme Council of Health Insurance were obtained. The tariff-cost gap was the discrepancy between a GSP's actual costs and tariff. Multiple linear regression analysis determined factors affecting the tariff-cost gap.

RESULTS

The average actual cost of GSPs was 637 USD, while the average tariff was 495 USD. The reimbursement covered only 78% of the costs. The gap was higher in older (B=1.05, 95% CI: 0.76-1.35, P<0.001), females (B=26.7, 95% CI: 15.5-37.9, P<0.001), patients with a longer stay (B=81.2, 95% CI: 77.5-84.8, P<0.001), and procedures performed by full-time surgeons (B=67.3, 95% CI: 56.9-77.5, P<0.001). Furthermore, neurosurgery had the highest effect on forecasting the gap between actual costs and tariffs among surgical specialties (B=346.9, 95% CI: 214.3-479.5, P<0.001).

CONCLUSION

Public hospitals suffer from large financial losses due to the national tariff for many GSPs not covering their actual costs. It is suggested that tariffs be increased for certain customer segments that can bear higher costs and global tariffs be adjusted to match actual service delivery costs.

Embedding Multimodal Rehabilitation Within Routine Cancer Care in Sheffield-The Active Together Service Evaluation Protocol

BACKGROUND

Approximately 3 million people in the United Kingdom are currently living with or beyond cancer. People undergoing treatment for cancer are at risk of complications following treatment. Increasing evidence supports the role of rehabilitation (including prehabilitation) in enhancing psychological and physical well-being in patients with cancer and improving outcomes. Active Together is an evidence-based, multimodal rehabilitation service for patients with cancer, providing support to help patients prepare for and recover from treatment. This paper presents the evaluation protocol for the Active Together service, aiming to determine its impact on patient-reported outcomes and clinical endpoints, as well as understand processes and mechanisms that influence its delivery and outcomes.

METHODS

This evaluation comprises an outcome and process evaluation, with service implementation data integrated into the analysis of outcome measures. The outcome evaluation will assess changes in outcomes of patients that attend the service and compare health care resource use against historical data. The process evaluation will use performance indicators, semistructured interviews, and focus groups to explore mechanisms of action and contextual factors influencing delivery and outcomes. Integrating psychological change mechanisms with outcome data might help to clarify complex causal pathways within the service.

CONCLUSIONS

Evidence to support the role of multimodal rehabilitation before, during, and after cancer treatment is increasing. The translation of that evidence into practice is less advanced. Findings from this evaluation will contribute to our understanding of the real-world impact of cancer rehabilitation and strengthen the case for widespread adoption of rehabilitation into routine care for people with cancer.

Clinical and socioeconomic factors predicting return-to-work times after cholecystectomy

BACKGROUND

Cholecystectomy, a type of surgery commonly performed globally, has possible mutual effects on the socioeconomic conditions of different countries due to various postoperative recovery times.

AIMS

This study evaluated the medical and socioeconomic factors affecting delayed return-to-work (RTW) time after elective cholecystectomy.

METHODS

This retrospective study analysed patients who underwent elective cholecystectomy for benign gallbladder diseases from January 2022 to April 2023. The patients' medical and socioeconomic data were collected to investigate the clinical and socioeconomic factors correlated with RTW time of >30 days after surgery.

RESULTS

This study included 180 consecutive patients. Significant correlations were found between delayed RTW time (>30 days) and age (odds ratio [OR]: 1.059, 95% confidence interval [CI] 1.008-1.113, P = 0.024), lack of medical insurance (OR: 2.935, 95% CI 1.189-7.249, P = 0.02) and high-intensity labour jobs (OR: 3.649, 95% CI 1.495-8.909, P = 0.004). Patients without medical insurance (26.6 versus 18.9 days) and those with high-intensity labour jobs (23.9 versus 18.8 days) had a higher mean RTW time than those with insurance and a less-intense labour job (P < 0.001).

CONCLUSIONS

After cholecystectomy, older age, lack of medical insurance and high-intensity labour job were correlated with a delayed RTW time. Informing patients about their expected RTW time after surgery can help reduce costs.

Tracing In-Hospital COVID-19 Outcomes: A Multistate Model Exploration (TRACE)

This study aims to develop and apply multistate models to estimate, forecast, and manage hospital length of stay during the COVID-19 epidemic without using any external packages. Data from Bellvitge University Hospital in Barcelona, Spain, were analyzed, involving 2285 hospitalized COVID-19 patients with moderate to severe conditions. The implemented multistate model includes transition probabilities and risk rates calculated from transitions between defined states, such as admission, ICU transfer, discharge, and death. In addition to examining key factors like age and gender, diabetes, lymphocyte count, comorbidity burden, symptom duration, and different COVID-19 waves were analyzed. Based on the model, patients hospitalized stay an average of 11.90 days before discharge, 2.84 days before moving to the ICU, or 34.21 days before death. ICU patients remain for about 24.08 days, with subsequent stays of 124.30 days before discharge and 35.44 days before death. These results highlight hospital stays' varying durations and trajectories, providing critical insights into patient flow and healthcare resource utilization. Additionally, it can predict ICU peak loads for specific subgroups, aiding in preparedness. Future work will integrate the developed code into the hospital's Health Information System (HIS) following ISO 13606 EHR standards and implement recursive methods to enhance the model's efficiency and accuracy.

Development of a machine learning model to estimate length of stay in coronary artery bypass grafting

OBJECTIVE

To develop and validate a predictive model utilizing machine-learning techniques for estimating the length of hospital stay among patients who underwent coronary artery bypass grafting.

METHODS

Three machine learning models (random forest, extreme gradient boosting and neural networks) and three traditional regression models (Poisson regression, linear regression, negative binomial regression) were trained in a dataset of 9,584 patients who underwent coronary artery bypass grafting between January 2017 and December 2021. The data were collected from hospital discharges from 133 centers in Brazil. Algorithms were ranked by calculating the root mean squared logarithmic error (RMSLE). The top performing algorithm was validated in a never-before-seen database of 2,627 patients. We also developed a model with the top ten variables to improve usability.

RESULTS

The random forest technique produced the model with the lowest error. The RMLSE was 0.412 (95%CI 0.405-0.419) on the training dataset and 0.454 (95%CI 0.441-0.468) on the validation dataset. Non-elective surgery, admission to a public hospital, heart failure, and age had the greatest impact on length of hospital stay.

CONCLUSIONS

The predictive model can be used to generate length of hospital stay indices that could be used as markers of efficiency and identify patients with the potential for prolonged hospitalization, helping the institution in managing beds, scheduling surgeries, and allocating resources.

Oculomic stratification of COVID-19 patients' intensive therapy unit admission status and mortality by retinal morphological findings

To investigate if retinal thickness has predictive utility in COVID-19 outcomes by evaluating the statistical association between retinal thickness using OCT and of COVID-19-related mortality. Secondary outcomes included associations between retinal thickness and length of stay (LoS) in hospital. In this retrospective cohort study, OCT scans from 230 COVID-19 patients admitted to the Intensive Care Unit (ITU) were compared with age and gender-matched patients with pneumonia from before March 2020. Total retinal, GCL + IPL, and RNFL thicknesses were recorded, and analysed with systemic measures collected at the time of admission and mortality outcomes, using linear regression models, Pearson's R correlation, and Principal Component Analysis. Retinal thickness was significantly associated with all-time mortality on follow up in the COVID-19 group (p = 0.015), but not 28-day mortality (p = 0.151). Retinal and GCL + IPL layer thicknesses were both significantly associated with LoS in hospital for COVID-19 patients (p = 0.006 for both), but not for patients with pneumonia (p = 0.706 and 0.989 respectively). RNFL thickness was not associated with LoS in either group (COVID-19 p = 0.097, pneumonia p = 0.692). Retinal thickness associated with LoS in hospital and long-term mortality in COVID-19 patients, suggesting that retinal structure could be a surrogate marker for frailty and predictor of disease severity in this group of patients, but not in patients with pneumonia from other causes.

Impact of robotic-assisted surgery on length of hospital stay in Paris public hospitals: a retrospective analysis

The number of available hospital beds is decreasing in many countries. Reducing the length of hospital stay (LOS) and increasing bed turnover could improve patient flow. We evaluated whether robot-assisted surgery (RAS) had a beneficial impact on the LOS in a French hospital trust with a long-established robotic program (Assistance Publique-Hôpitaux de Paris, AP-HP). We extracted data from "Programme de Médicalisation des Systèmes d'Information" to determine the median LOS for adults in our trust after RAS versus laparoscopy and open surgery in 2021-2022 for eight target procedures, and compared data nationally and at similar academic centres (same database). We also calculated the number of hospitalisation days 'saved' using RAS. Overall, 9326 target procedures were performed at AP-HP: 3864 (41.4%) RAS, 2978 (31.9%) laparoscopies, and 2484 (26.6%) open surgeries. The median LOS for RAS was lower than laparoscopy and open surgery for all procedures, apart from hysterectomy and colectomy (equivalent to laparoscopy). Results for urological procedures at AP-HP reflected national values. The equivalent of 5390 hospitalisation days was saved in 2021-2022 using RAS instead of open surgery or laparoscopy at AP-HP; of these, 86% represented hospitalisation days saved using RAS in urological procedures. Using RAS instead of open surgery or laparoscopy (particularly in urological procedures) reduced the median LOS and may save thousands of hospitalisation days every year. This should help to increase patient turnover and facilitate patient flow.

Utilizing a comprehensive machine learning approach to identify patients at high risk for extended length of stay following spinal deformity surgery in pediatric patients with early onset scoliosis

PURPOSE

Early onset scoliosis (EOS) patient diversity makes outcome prediction challenging. Machine learning offers an innovative approach to analyze patient data and predict results, including LOS in pediatric spinal deformity surgery.

METHODS

Children under 10 with EOS were chosen from the American College of Surgeon's NSQIP database. Extended LOS, defined as over 5 days, was predicted using feature selection and machine learning in Python. The best model, determined by the area under the curve (AUC), was optimized and used to create a risk calculator for prolonged LOS.

RESULTS

The study included 1587 patients, mostly young (average age: 6.94 ± 2.58 years), with 33.1% experiencing prolonged LOS (n = 526). Most patients were female (59.2%, n = 940), with an average BMI of 17.0 ± 8.7. Factors influencing LOS were operative time, age, BMI, ASA class, levels operated on, etiology, nutritional support, pulmonary and neurologic comorbidities. The gradient boosting model performed best with a test accuracy of 0.723, AUC of 0.630, and a Brier score of 0.189, leading to a patient-specific risk calculator for prolonged LOS.

CONCLUSIONS

Machine learning algorithms accurately predict extended LOS across a national patient cohort and characterize key preoperative drivers of increased LOS after PSIF in pediatric patients with EOS.

Development of a System for Predicting Hospitalization Time for Patients With Traumatic Brain Injury Based on Machine Learning Algorithms: User-Centered Design Case Study

Background

Currently, the treatment and care of patients with traumatic brain injury (TBI) are intractable health problems worldwide and greatly increase the medical burden in society. However, machine learning-based algorithms and the use of a large amount of data accumulated in the clinic in the past can predict the hospitalization time of patients with brain injury in advance, so as to design a reasonable arrangement of resources and effectively reduce the medical burden of society. Especially in China, where medical resources are so tight, this method has important application value.

Objective

We aimed to develop a system based on a machine learning model for predicting the length of hospitalization of patients with TBI, which is available to patients, nurses, and physicians.

Methods

We collected information on 1128 patients who received treatment at the Neurosurgery Center of the Second Affiliated Hospital of Anhui Medical University from May 2017 to May 2022, and we trained and tested the machine learning model using 5 cross-validations to avoid overfitting; 28 types of independent variables were used as input variables in the machine learning model, and the length of hospitalization was used as the output variables. Once the models were trained, we obtained the error and goodness of fit (R2) of each machine learning model from the 5 rounds of cross-validation and compared them to select the best predictive model to be encapsulated in the developed system. In addition, we externally tested the models using clinical data related to patients treated at the First Affiliated Hospital of Anhui Medical University from June 2021 to February 2022.

Results

Six machine learning models were built, including support vector regression machine, convolutional neural network, back propagation neural network, random forest, logistic regression, and multilayer perceptron. Among them, the support vector regression has the smallest error of 10.22% on the test set, the highest goodness of fit of 90.4%, and all performances are the best among the 6 models. In addition, we used external datasets to verify the experimental results of these 6 models in order to avoid experimental chance, and the support vector regression machine eventually performed the best in the external datasets. Therefore, we chose to encapsulate the support vector regression machine into our system for predicting the length of stay of patients with traumatic brain trauma. Finally, we made the developed system available to patients, nurses, and physicians, and the satisfaction questionnaire showed that patients, nurses, and physicians agreed that the system was effective in providing clinical decisions to help patients, nurses, and physicians.

Conclusions

This study shows that the support vector regression machine model developed using machine learning methods can accurately predict the length of hospitalization of patients with TBI, and the developed prediction system has strong clinical use.

Predicting Postoperative Length of Stay in Patients Undergoing Laparoscopic Right Hemicolectomy for Colon Cancer: A Machine Learning Approach Using SICE (Società Italiana di Chirurgia Endoscopica) CoDIG Data

The evolution of laparoscopic right hemicolectomy, particularly with complete mesocolic excision (CME) and central vascular ligation (CVL), represents a significant advancement in colon cancer surgery. The CoDIG 1 and CoDIG 2 studies highlighted Italy's progressive approach, providing useful findings for optimizing patient outcomes and procedural efficiency. Within this context, accurately predicting postoperative length of stay (LoS) is crucial for improving resource allocation and patient care, yet its determination through machine learning techniques (MLTs) remains underexplored. This study aimed to harness MLTs to forecast the LoS for patients undergoing right hemicolectomy for colon cancer, using data from the CoDIG 1 (1224 patients) and CoDIG 2 (788 patients) studies. Multiple MLT algorithms, including random forest (RF) and support vector machine (SVM), were trained to predict LoS, with CoDIG 1 data used for internal validation and CoDIG 2 data for external validation. The RF algorithm showed a strong internal validation performance, achieving the best performances and a 0.92 ROC in predicting long-term stays (more than 5 days). External validation using the SVM model demonstrated 75% ROC values. Factors such as fast-track protocols, anastomosis, and drainage emerged as key predictors of LoS. Integrating MLTs into predicting postoperative LOS in colon cancer surgery offers a promising avenue for personalized patient care and improved surgical management. Using intraoperative features in the algorithm enables the profiling of a patient's stay based on the planned intervention. This issue is important for tailoring postoperative care to individual patients and for hospitals to effectively plan and manage long-term stays for more critical procedures.

Synergistic patient factors are driving recent increased pediatric urgent care demand

OBJECTIVES

We aimed to use the high fidelity urgent care patient data to model the factors that have led to the increased demand at our local pediatric urgent care centre.

METHODS

The dataset for this retrospective cohort study was obtained from our local healthcare centre's national reporting data for pediatric urgent care visits from 2006 to 2022. Variables analyzed included: basic patient demographics, chief complaint, triage urgency, date and time of registration/discharge, discharge diagnosis, and discharge destination. Statistical analysis of non-linear trends was summarized by locally estimated scatterplot smoothing splines. For machine learning, we used the tidymodels R package. Models were validated in training using k-fold cross validation with k = 5. We used univariate linear regression as a baseline model. After the data was standardized, correlation and homoscedasticity were evaluated between all parameter permutations.

RESULTS

This dataset consisted of 164,660 unique visits to our academic centre's pediatric urgent care. Over the study period, there was an overall substantial increase in the number of urgent care visits per day, with a rapid increase beyond previous levels in 2021 and further in 2022. The increased length of stay trend was consistent across presenting complaint categories. The proportion of patients without primary care in 2022 was 2.5 times higher than in 2013. A random forest machine learning model revealed the relative importance of features to predicting a visit in 2022 were: longer stay, later registration in the day, diagnosis of an infectious illness, and younger age.

CONCLUSIONS

This study identified a combination of declining primary care access, circulating viral infections, and shifting chief complaints as factors driving the recent increase in frequency and duration of visits to our urgent care service.

Predicting hospital length of stay using machine learning on a large open health dataset

BACKGROUND

Governments worldwide are facing growing pressure to increase transparency, as citizens demand greater insight into decision-making processes and public spending. An example is the release of open healthcare data to researchers, as healthcare is one of the top economic sectors. Significant information systems development and computational experimentation are required to extract meaning and value from these datasets. We use a large open health dataset provided by the New York State Statewide Planning and Research Cooperative System (SPARCS) containing 2.3 million de-identified patient records. One of the fields in these records is a patient's length of stay (LoS) in a hospital, which is crucial in estimating healthcare costs and planning hospital capacity for future needs. Hence it would be very beneficial for hospitals to be able to predict the LoS early. The area of machine learning offers a potential solution, which is the focus of the current paper.

METHODS

We investigated multiple machine learning techniques including feature engineering, regression, and classification trees to predict the length of stay (LoS) of all the hospital procedures currently available in the dataset. Whereas many researchers focus on LoS prediction for a specific disease, a unique feature of our model is its ability to simultaneously handle 285 diagnosis codes from the Clinical Classification System (CCS). We focused on the interpretability and explainability of input features and the resulting models. We developed separate models for newborns and non-newborns.

RESULTS

The study yields promising results, demonstrating the effectiveness of machine learning in predicting LoS. The best R2 scores achieved are noteworthy: 0.82 for newborns using linear regression and 0.43 for non-newborns using catboost regression. Focusing on cardiovascular disease refines the predictive capability, achieving an improved R2 score of 0.62. The models not only demonstrate high performance but also provide understandable insights. For instance, birth-weight is employed for predicting LoS in newborns, while diagnostic-related group classification proves valuable for non-newborns.

CONCLUSION

Our study showcases the practical utility of machine learning models in predicting LoS during patient admittance. The emphasis on interpretability ensures that the models can be easily comprehended and replicated by other researchers. Healthcare stakeholders, including providers, administrators, and patients, stand to benefit significantly. The findings offer valuable insights for cost estimation and capacity planning, contributing to the overall enhancement of healthcare management and delivery.

Defining the correlates of lymphopenia and independent predictors of poor clinical outcome in adults hospitalized with COVID-19 in Australia

Lymphopenia is a common feature of acute COVID-19 and is associated with increased disease severity and 30-day mortality. Here we aim to define the demographic and clinical characteristics that correlate with lymphopenia in COVID-19 and determine if lymphopenia is an independent predictor of poor clinical outcome. We analysed the ENTER-COVID (Epidemiology of hospitalized in-patient admissions following planned introduction of Epidemic SARS-CoV-2 to highly vaccinated COVID-19 naïve population) dataset of adults (N = 811) admitted for COVID-19 treatment in South Australia in a retrospective registry study, categorizing them as (a) lymphopenic (lymphocyte count < 1 × 109/L) or (b) non-lymphopenic at hospital admission. Comorbidities and laboratory parameters were compared between groups. Multiple regression analysis was performed using a linear or logistic model. Intensive care unit (ICU) patients and non-survivors exhibited lower median lymphocyte counts than non-ICU patients and survivors respectively. Univariate analysis revealed that low lymphocyte counts associated with hypertension and correlated with haemoglobin, platelet count and negatively correlated with urea, creatinine, bilirubin, and aspartate aminotransferase (AST). Multivariate analysis identified age, male, haemoglobin, platelet count, diabetes, creatinine, bilirubin, alanine transaminase, c-reactive protein (CRP) and lactate dehydrogenase (LDH) as independent predictors of poor clinical outcome in COVID-19, while lymphopenia did not emerge as a significant predictor.

Adult psychiatric inpatient admissions and length of stay before and during the COVID-19 pandemic in a large urban hospital setting in Vancouver, British Columbia

Introduction

During the COVID-19 pandemic individuals with mental illnesses faced challenges accessing psychiatric care. Our study aimed to describe patient characteristics and compare admissions and length of stay (LOS) for psychiatric-related hospitalizations before and during the COVID-19 pandemic.

Methods

We conducted a retrospective analysis using health administrative data comparing individuals with an acute psychiatric admission between two time periods: 1st March 2019 to 31st December 2019 (pre-COVID) and 1st March 2020 to 31st December 2020 (during-COVID). Multivariable negative binomial regression was used to model the association between most responsible diagnosis type and the two-time periods to hospital LOS, reporting the Rate Ratio (RR) as the measure of effect.

Results

The cohort comprised 939 individuals who were predominately male (60.3%) with a severe mental illness (schizophrenia or mood-affective disorder) (72.7%) and a median age of 38 (IQR: 28.0, 52.0) years. In the multivariable analysis, anxiety disorders (RR: 0.63, CI: 0.4, 0.99) and personality disorders (RR: 0.52, CI: 0.32, 0.85) were significantly associated with a shorter LOS when compared to individuals without those disorders. Additionally, when compared to hospital admissions for non-substance related disorders the LOS for patients with substance-related disorders were significantly shorter during the COVID period (RR: 0.45, CI: 0.30, 0.67) and pre-COVID period (RR: 0.31, CI: 0.21, 0.46).

Conclusions

We observed a significant difference in the type and length of admissions for various psychiatric disorders during the COVID-19 period. These findings can support systems of care in adapting to utilization changes during pandemics or other global health events.

Applying models of care for total hip and knee arthroplasty: External validation of a published predictive model to identify extended stay risk prior to lower-limb arthroplasty

OBJECTIVE

This study aimed to externally validate a reported model for identifying patients requiring extended stay following lower limb arthroplasty in a new setting.

DESIGN

External validation of a previously reported prognostic model, using retrospective data.

SETTING

Medium-sized hospital orthopaedic department, Australia.

PARTICIPANTS

Electronic medical records were accessed for data collection between Sep-2019 and Feb-2020 and retrospective data extracted from 200 randomly selected total hip or knee arthroplasty patients.

INTERVENTION

Participants received total hip or knee replacement between 2-Feb-16 and 4-Apr-19. This study was a non-interventional retrospective study.

MAIN MEASURES

Model validation was assessed with discrimination, calibration on both original and adjusted forms of the candidate model. Decision curve analysis was conducted on the outputs of the adjusted model to determine net benefit at a predetermined decision threshold (0.5).

RESULTS

The original model performed poorly, grossly overestimating length of stay with mean calibration of -3.6 (95% confidence interval -3.9 to -3.2) and calibration slope of 0.52. Performance improved following adjustment of the model intercept and model coefficients (mean calibration 0.48, 95% confidence interval 0.16 to 0.80 and slope of 1.0), but remained poorly calibrated at low and medium risk threshold and net benefit was modest (three additional patients per hundred identified as at-risk) at the a-priori risk threshold.

CONCLUSIONS

External validation demonstrated poor performance when applied to a new patient population and would provide limited benefit for our institution. Implementation of predictive models for arthroplasty should include practical assessment of discrimination, calibration and net benefit at a clinically acceptable threshold.

Predictors of recovery from severe acute malnutrition among 6-59 months children admitted to a hospital

Background and aim

Severe acute malnutrition is a threat to child survival as mortality rates in children with severe malnutrition are nine times higher. Globally, about 19 million children are severely malnourished. This study looked at children aged 6-59 months admitted to hospital to see how quickly they recovered from severe acute malnutrition as well as what factors predicted their recovery.

Methods

The study included 543 systematically chosen children with severe acute malnutrition who were admitted to the stabilization center of a hospital. Data from the patient registry were gathered using a retrospective follow-up study design. In order to find predictors of recovery, the Cox proportional hazard model was applied.

Results

From 543 children, 425 (78.27%) were recovered. The median survival time was 8 days. Having grade II edema, grade III edema, and pneumonia were negatively associated with recovery. Similarly, taking ceftriaxone, cloxacillin, and being on a nasogastric tube were associated with poor recovery. Conversely, better recovery rates were linked to exclusive breastfeeding and vitamin A supplementation.

Conclusion

Both the recovery rate and the median survival time fell within acceptable bounds. To boost the recovery rate, efforts are needed to lessen comorbidities.

Clinical Utility of the Tokyo Guidelines 2018 for Acute Cholangitis in the Emergency Department and Comparison with Novel Markers (Neutrophil-to-Lymphocyte and Blood Nitrogen Urea-to-Albumin Ratios)

Introduction: The Tokyo Guidelines 2018 (TG2018) is a scoring system used to recommend the clinical management of AC. However, such a scoring system must incorporate a variety of clinical outcomes of acute cholangitis (AC). In an emergency department (ED)-based setting, where efficiency and practicality are highly desired, clinicians may find the application of various parameters challenging. The neutrophil-to-lymphocyte ratio (NLR) and blood urea nitrogen-to-albumin ratio (BAR) are relatively common biomarkers used to assess disease severity. This study evaluated the potential value of TG2018 scores measured in an ED to predict a variety of clinical outcomes. Furthermore, the study also compared TG2018 scores with NLR and BAR scores to demonstrate their usefulness. Methods: This retrospective observational study was performed in an ED. In total, 502 patients with AC visited the ED between January 2016 and December 2021. The primary endpoint was to evaluate whether the TG2018 scoring system measured in the ED was a predictor of intensive care, long-term hospital stays (≥14 days), percutaneous transhepatic biliary drainage (PTBD) during admission care, and endotracheal intubation (ETI). Results: The analysis included 81 patients requiring intensive care, 111 requiring long-term hospital stays (≥14 days), 49 requiring PTBD during hospitalization, and 14 requiring ETI during hospitalization. For the TG2018 score, the adjusted OR (aOR) using (1) as a reference was 23.169 (95% CI: 9.788-54.844) for (3) compared to (1). The AUC of the TG2018 for the need for intensive care was 0.850 (95% CI: 0.815-0.881) with a cutoff of >2. The AUC for long-term hospital stays did not exceed 0.7 for any of the markers. the AUC for PTBD also did not exceed 0.7 for any of the markers. The AUC for ETI was the highest for BAR at 0.870 (95% CI: 0.837-0.899) with a cutoff value of >5.2. Conclusions: The TG2018 score measured in the ED helps predict various clinical outcomes of AC. Other novel markers such as BAR and NLR are also associated, but their explanatory power is weak.

Machine learning for healthcare that matters: Reorienting from technical novelty to equitable impact

Despite significant technical advances in machine learning (ML) over the past several years, the tangible impact of this technology in healthcare has been limited. This is due not only to the particular complexities of healthcare, but also due to structural issues in the machine learning for healthcare (MLHC) community which broadly reward technical novelty over tangible, equitable impact. We structure our work as a healthcare-focused echo of the 2012 paper "Machine Learning that Matters", which highlighted such structural issues in the ML community at large, and offered a series of clearly defined "Impact Challenges" to which the field should orient itself. Drawing on the expertise of a diverse and international group of authors, we engage in a narrative review and examine issues in the research background environment, training processes, evaluation metrics, and deployment protocols which act to limit the real-world applicability of MLHC. Broadly, we seek to distinguish between machine learning ON healthcare data and machine learning FOR healthcare-the former of which sees healthcare as merely a source of interesting technical challenges, and the latter of which regards ML as a tool in service of meeting tangible clinical needs. We offer specific recommendations for a series of stakeholders in the field, from ML researchers and clinicians, to the institutions in which they work, and the governments which regulate their data access.

Effect of sarcopenia and frailty on outcomes among patients with brain metastases

PURPOSE

Sarcopenia and frailty have been associated with increased mortality and duration of hospitalization in cancer. However, data investigating these effects in patients with brain metastases remain limited. This study aimed to investigate the effects of sarcopenia and frailty on clinical outcomes in patients with surgically treated brain metastases.

METHODS

Patients who underwent surgical resection of brain metastases from 2011 to 2019 were included. Psoas cross-sectional area and temporalis thickness were measured by two independent radiologists (Cronbach's alpha > 0.98). Frailty was assessed using the Clinical Frailty Scale (CFS) pre-operatively and post-operatively. Overall mortality, recurrence, and duration of hospitalization were collected. Cox regression was performed for mortality and recurrence, and multiple linear regression for duration of hospitalization.

RESULTS

145 patients were included, with median age 60.0 years and 52.4% female. Psoas cross-sectional area was an independent risk factor for overall mortality (HR = 2.68, 95% CI 1.64-4.38, p < 0.001) and recurrence (HR = 2.31, 95% CI 1.14-4.65, p = 0.020), while post-operative CFS was an independent risk factor for overall mortality (HR = 1.88, 95% CI 1.14-3.09, p = 0.013). Post-operative CFS (β = 15.69, 95% CI 7.67-23.72, p < 0.001) and increase in CFS (β = 11.71, 95% CI 3.91-19.51, p = 0.004) were independently associated with increased duration of hospitalization.

CONCLUSION

In patients with surgically treated brain metastases, psoas cross-sectional area was an independent risk factor for mortality and recurrence, while post-operative CFS was an independent risk factor for mortality. Post-operative frailty and increase in CFS significantly increased duration of hospitalization. Measurement of psoas cross-sectional area and CFS may aid in risk stratification of surgical candidates for brain metastases.