Investigation of emergency department abandonment rates using machine learning algorithms in a single centre study.

A critical problem that Emergency Departments (EDs) must address is overcrowding, as it causes extended waiting times and increased patient dissatisfaction, both of which are immediately linked to a greater number of patients who leave the ED early, without any evaluation by a healthcare provider (Leave Without Being Seen, LWBS). This has an impact on the hospital in terms of missing income from lost opportunities to offer treatment and, in general, of negative outcomes from the ED process. Consequently, healthcare managers must be able to forecast and control patients who leave the ED without being evaluated in advance. This study is a retrospective analysis of patients registered at the ED of the "San Giovanni di Dio e Ruggi d'Aragona" University Hospital of Salerno (Italy) during the years 2014-2021. The goal was firstly to analyze factors that lead to patients abandoning the ED without being examined, taking into account the features related to patient characteristics such as age, gender, arrival mode, triage color, day of week of arrival, time of arrival, waiting time for take-over and year. These factors were used as process measures to perform a correlation analysis with the LWBS status. Then, Machine Learning (ML) techniques are exploited to develop and compare several LWBS prediction algorithms, with the purpose of providing a useful support model for the administration and management of EDs in the healthcare institutions. During the examined period, 688,870 patients were registered and 39188 (5.68%) left without being seen. Of the total LWBS patients, 59.6% were male and 40.4% were female. Moreover, from the statistical analysis emerged that the parameter that most influence the abandonment rate is the waiting time for take-over. The final ML classification model achieved an Area Under the Curve (AUC) of 0.97, indicating high performance in estimating LWBS for the years considered in this study. Various patient and ED process characteristics are related to patients who LWBS. The possibility of predicting LWBS rates in advance could be a valid tool quickly identifying and addressing "bottlenecks" in the hospital organization, thereby improving efficiency.

A Regressive Model to Study the Hospitalization for Laparoscopic Appendectomy: A Multicenter Study.

Appendicitis is a most common abdominal condition worldwide, and appendectomy especially laparoscopic appendectomy is among the most commonly performed general surgeries. In this study, data were collected from patients who underwent laparoscopic appendectomy surgery at the Evangelical Hospital "Betania" in Naples, Italy. Linear multiple regression was used to obtain a simple predictor that can also assess which of the independent variables considered to be a risk factor. The model with R2 of 0.699 shows that comorbidities and complications during surgery are the main risk factors for prolonged LOS. This result is validated by other studies conducted in the same area.

Effect of Clinical and Demographic Variables on the Hospital Stay of Patients Undergoing Total Knee Arthroplasty.

The knee is the joint most affected by osteoarthritis and in its severe form can significantly affect people's physical and functional abilities. The increased demand for surgery leads to greater attention by health care management to be able to keep costs down. A major expense item for this procedure is Length of Stay (LOS). In this study, several Machine Learning algorithms were tested in order to construct not only a valid predictor of LOS but also to know among the selected variables the main risk factors. To do so, activity data from the Evangelical Hospital "Betania" in Naples, Italy, from 2019-2020 were used. Among the algorithms, the best are the classification algorithms with accuracy values exceeding 90%. Finally, the results are in line with those shown by two other comparison hospitals in the area.

The Classification Algorithms Applied to the Inpatient Stay for Lower Limb Surgery.

The revolutions of recent years in health care have involved several areas ranging from patient treatment to resource management. Therefore, several strategies have been put in place to increase patient value while trying to reduce spending. Several indicators have arisen to evaluate the performance of healthcare processes. The main one is Length of Stay (LOS). In this study, classification algorithms were used to predict the LOS of patients undergoing lower extremity surgery, an increasingly common condition given the progressive aging of the population. The context is the Evangelical Hospital "Betania" in Naples (Italy) in 2019-2020, which augments a multicenter study conducted by the same research team on several hospitals in southern Italy. All selected algorithms show an Accuracy above 90% but among them, the best is Logistic Regression with a value reaching 94%.

Modeling Cholecystectomy Hospital Stay Through a Linear Approach.

Cholecystectomy is among the most frequent procedures in general surgery. In the healthcare facility organization, it is important to evaluate all interventions and procedures that have a great impact on health management and that have a clear effect on the Length of Stay (LOS). The LOS represents, in fact, an indicator of performance and measure the goodness of a health process. This study was conducted with the aim of providing LOS for all patients undergoing cholecystectomy at the "A.O.R.N. A. Cardarelli" of Naples. Data were collected in the two years 2019 and 2020 and included 650 patients. A MLR model is created in the work to predict the value of LOS as a function of the following variables: gender, age, pre-operative LOS, presence of comorbidities and complication during surgery. The results obtained are as follows: R=0.941 and R2=0.885.

Prediction of Hospital Length Stay for Patients Undergoing Mastectomy.

The prolonged length of stay is an important aspect to be considered for the healthcare management since this affect both the health-related expenditure of the hospital and the quality of the offered service. In the light of these consideration is important for hospitals to be able to predict the LOS of patients and to work on the principal aspect affecting it in order to reduce LOS as much as possible. In this work we focus on patients undergoing mastectomy. The data were collected form 989 patients who underwent mastectomy surgery in the Surgery Department of the AORN "A. Cardarelli" of Naples. Different models have been tested and characterized and the one with the best performance was identified.

COVID-19 in Eye Surgery: The Case of a University Hospital.

Coronavirus epidemic has quickly become a global health threat. The ophthalmology department, like all other departments, have adopted resource management and personnel adjustment maneuvers. The aim of this work was to describe the impact of covid on the Ophthalmology Department of University Hospital "Federico II" of Naples. In the study logistical regression was used for a comparison between the pandemic and the previous period, analyzing patient features. The analysis showed a decrease in the number of accesses; reduction of the length of stay; and the statistically dependent variables are as follows: LOS, discharge procedures and admission procedure.

Predicting In-Hospital Mortality During the COVID-19 Pandemic in Patients with Heart Failure: A Single-Center Exploratory Study.

The aim of this study was to investigate whether exposure to the pandemic was associated with increased in-hospital mortality for health failure. We collected data from patients hospitalized between 2019 and 2020 and we assessed the likelihood of in-hospital death. Although the positive association of exposure to the COVID period with an increased in-hospital mortality is not statistically significant, this may underscore other factors that may influence mortality. Our study was designed to contribute to a better understanding of the impact of the pandemic on in-hospital mortality and to identify potential areas for intervention in patient care.

Evaluation of different machine learning algorithms for predicting the length of stay in the emergency departments: a single-centre study.

Background: Recently, crowding in emergency departments (EDs) has become a recognised critical factor impacting global public healthcare, resulting from both the rising supply/demand mismatch in medical services and the paucity of hospital beds available in inpatients units and EDs. The length of stay in the ED (ED-LOS) has been found to be a significant indicator of ED bottlenecks. The time a patient spends in the ED is quantified by measuring the ED-LOS, which can be influenced by inefficient care processes and results in increased mortality and health expenditure. Therefore, it is critical to understand the major factors influencing the ED-LOS through forecasting tools enabling early improvements. Methods: The purpose of this work is to use a limited set of features impacting ED-LOS, both related to patient characteristics and to ED workflow, to predict it. Different factors were chosen (age, gender, triage level, time of admission, arrival mode) and analysed. Then, machine learning (ML) algorithms were employed to foresee ED-LOS. ML procedures were implemented taking into consideration a dataset of patients obtained from the ED database of the "San Giovanni di Dio e Ruggi d'Aragona" University Hospital (Salerno, Italy) from the period 2014-2019. Results: For the years considered, 496,172 admissions were evaluated and 143,641 of them (28.9%) revealed a prolonged ED-LOS. Considering the complete data (48.1% female vs. 51.9% male), 51.7% patients with prolonged ED-LOS were male and 47.3% were female. Regarding the age groups, the patients that were most affected by prolonged ED-LOS were over 64 years. The evaluation metrics of Random Forest algorithm proved to be the best; indeed, it achieved the highest accuracy (74.8%), precision (72.8%), and recall (74.8%) in predicting ED-LOS. Conclusions: Different variables, referring to patients' personal and clinical attributes and to the ED process, have a direct impact on the value of ED-LOS. The suggested prediction model has encouraging results; thus, it may be applied to anticipate and manage ED-LOS, preventing crowding and optimising effectiveness and efficiency of the ED.