Prediction of good neurological recovery after out-of-hospital cardiac arrest: A machine learning analysis
پیش بینی بهبود عصبی خوب بعد از ایست قلبی خارج از بیمارستان: تجزیه و تحلیل یادگیری ماشین-2019
Background: This study aimed to train, validate and compare predictive models that use machine learning analysis for good neurological recovery in OHCA patients. Methods: Adult OHCA patients who had a presumed cardiac etiology and a sustained return of spontaneous circulation between 2013 and 2016 were analyzed; 80% of the individuals were analyzed for training and 20% were analyzed for validation. We developed using six machine learning algorithms: logistic regression (LR), extreme gradient boosting (XGB), support vector machine, random forest, elastic net (EN), and neural network. Variables that could be obtained within 24 hours of the emergency department visit were used. The area under the receiver operation curve (AUROC) was calculated to assess the discrimination. Calibration was assessed by the Hosmer–Lemeshow test. Reclassification was assessed by using the continuous net reclassification index (NRI). Results: A total of 19,860 OHCA patients were included in the analysis. Of the 15,888 patients in the training group, 2228 (14.0%) had a good neurological recovery; of the 3972 patients in the validation group, 577 (14.5%) had a good neurological recovery. The LR, XGB, and EN models showed the highest discrimination powers (AUROC (95% CI)) of 0.949 (0.941–0.957) for all), and all three models were well calibrated (Hosmer–Lemeshow test: p >0.05). The XGB model reclassified patients according to their true risk better than the LR model (NRI: 0.110), but the EN model reclassified patients worse than the LR model (NRI: 1.239). Conclusion: The best performing machine learning algorithm was the XGB and LR algorithm .
Keywords: Out-of-hospital cardiac arrest | Outcome | Machine learning analysis
Setting up standards: A methodological proposal for pediatric Triage machine learning model construction based on clinical outcomes
تنظیم استانداردها: یک پیشنهاد روش شناختی برای ساخت مدل یادگیری ماشین تراشی کودکان براساس نتایج بالینی-2019
Triage is a critical process in hospital emergency departments (ED). Specifically, we consider how to achieve fast and accurate patient Triage in the ED of a pediatric hospital. The goal of this paper is to establish methodological best practices for the application of machine learning (ML) to Triage in pediatric ED, providing a comprehensive comparison of the performance of ML techniques over a large dataset. Our work is among the first attempts in this direction. Following very recent works in the literature, we use the clinical outcome of a case as its label for supervised ML model training, instead of the more uncertain labels provided by experts. The experimental dataset contains the records along 3 years of operation of the hospital ED. It consists of 189,718 patients visits to the hospital. The clinical outcome of 9271 cases (4.98%) wa hospital admission, therefore our dataset is highly class imbalanced. Our reported performance comparison results focus on four ML models: Deep Learning (DL), Random Forest (RF), Naive Bayes (NB) and Support Vector Machines (SVM). Data preprocessing includes class imbalance correction, and case re-labeling. We use different well known metrics to evaluate performance of ML models in three different experimental settings: (a) classification of each case into the standard five Triage urgency levels, (b) discrimination of high versus low case severity according to its clinical outcome, and (c) comparison of the number of patients assigned to each standard Triage urgency level against the Triage rule based expert system currently in use at the hospital. RF achieved greater AUC, accuracy, PPV and specificity than the other models in the dychotomic classification experiments. On the implementation side, our study shows that ML predictive models trained according to clinical outcomes, provide better Triage performance than the current rule based expert system in operation at the hospital.
Keywords: Machine learning | Emergency department | Triage | Data science | Clinical decision support systems
PALLIATIVE CARE SYMPTOM MANAGEMENT IN THE EMERGENCY DEPARTMENT: THE ABC’S OF SYMPTOM MANAGEMENT FOR THE EMERGENCY PHYSICIAN
مدیریت علائم مراقبت از بیمار در بخش اورژانس: مدیریت علائم ABC برای بیمارستانی-2018
Background: Palliative care is a rapidly evolving area of emergency medicine. With an estimated 5,000 to 10,000 baby boomers per day reaching retirement age, emergency departments (EDs) are treating more patients with chronic and serious disease. Palliative care offers comprehensive care for patients with advanced medical illness, aims to alleviate suffering and improve quality of life, and plays an important role in caring for these patients in the ED. Objectives: We sought to increase the emergency physician’s knowledge of and comfort with symptom control in palliative and hospice patients. Discussion: Having the skills to deliver efficient and appropriate palliative and hospice care is imperative for emergency physicians. Palliative care should be considered in any patient suffering from symptoms of a life-limiting illness, whereas hospice care should be considered in the patient with likely <6 months left to live. Palliative care is appropriate earlier in the course of disease, and is appropriate when the practitioner would not be surprised if the patient died in the next 2 years (‘‘The Surprise Question’’). This article discusses management in the ED of pain, nausea, dyspnea, agitation, and oral secretions in patients appropriate for hospice and palliative care. Conclusion: The need for palliative and hospice care in the ED is increasing, requiring that emergency physicians be familiar with palliative and hospice care and competent in the delivery of rapid symptom management in patients with severe and life-limiting disease. 2017 Elsevier Inc. All rights reserved.
Keywords: emergency medicine; end of life; palliative care; symptom management
Crisis in the Emergency Department The Evaluation and Management of Acute Agitation in Children and Adolescents
بحران در بخش اورژانس ارزیابی و مدیریت هیجان حاد در کودکان و نوجوانان-2018
KEYWORDS : Agitation ، Aggression ، Restraint/seclusion ، Emergency department ، Delirium
Recurrent use of inpatient withdrawal management services: Characteristics, service use, and cost among Medicaid clients
استفاده مجدد از خدمات مدیریت بسته های بستری: مشخصات، استفاده از خدمات و هزینه در میان مشتریان خدمات درمانی-2018
Reducing repeat use of costly inpatient services, such as inpatient withdrawal management, among Medicaid members is a target of healthcare reform. However, characteristics of frequent users of inpatient withdrawal management are understudied. We described the characteristics, service utilization, and costs of New York Medicaid clients who use withdrawal management services by analyzing data from Medicaid records from 2008. We examined follow-up care for individuals with different levels of repeat withdrawal management. We found 32,196 Medicaid withdrawal management patients with a total of 67,073 episodes and we divided patients into low (1 episode, n = 19,602), medium (2–3 episodes, n = 8619) and high (≥4 episodes, n = 3978) use cate gories. High users had almost 8 times the withdrawal management cost of low users. Similarly, they had 5 times more emergency department visits than low users. High users had high levels of homelessness (75%), 20% had HIV/AIDS, and 40% had Hepatitis. High withdrawal management users were less likely than low users to receive any follow-up treatment services. Medicaid clients with high utilization of inpatient withdrawal management are a small but costly population with poor follow-up rates to subsequent treatment services. They are a socially disenfranchised group that may benefit from targeted services to address their complex clinical needs.
Keywords: Withdrawal management ، High utilizers ، Medicaid ، Service utilization ، Cost
Sickle Cell Disease in the Emergency Department: Complications and Management
بیماری سلولی سقط در بخش اورژانس: عوارض و مدیریت-2018
Sickle cell disease is the most com mon blood disorder in the United States, affecting 100 000 people. A genetic mutation creates hemoglobin S. In the deoxygenated state, hemo globin S polymerizes, creating sickled hemoglobin. Sickled hemoglobin causes a cascade of complex patho physiologic events that lead to hemo lysis, chronic anemia and endothelial damage. This results in clinical com plications, end organ dysfunction and a shortened life expectancy. The acute nature of many sickle cell complica tions makes the emergency depart ment a common setting where sickle cell patients present. Common com plications (vaso-occlusive episode, fe ver, acute chest syndrome, stroke) and less common complications (splenic sequestration, priapism, aplastic cri sis, ocular emergencies) will be dis cussed. Public health implications will be discussed briefly.
Keywords: sickle cell disease; anemia; compli cations; vaso occlusive crisis; vaso occlusive episode; acute chest syn drome; stroke; splenic sequestra tion; priapism; aplastic crisis
Preventing Emergency Department Violence through Design
جلوگیری از خشونت گروهی اورژانس از طریق طراحی-2017
T rends and news reports highlight a growing concern about random violence in public venues. Health care settings traditionally have been considered sacred ground for vulnerable ill or injured patients, as well as care providers, who are considered part of the public safety net. However, not all hospital or health-system leaders fully appreciate the dynamic situations that arise when fear, pain, drug use, or mental-health behaviors put patients, staff, and visitors in harm’s way. It is crucial that staff partner with administrators, facility leaders, and safety officers to design emergency departments with evidence based concepts to minimize or eliminate risks to safety and security. This article provides a comprehensive review of best design practices to help guide clinical user groups in meetings with hospital leaders, architects, and engineers.
Presentations for hypoglycemia associated with diabetes mellitus to emergency departments in a Canadian province: A database and epidemiological analysis
ارائه برای هیپوگلیسمی مرتبط با دیابت به بخش اورژانس در استان کانادا: پایگاه داده و تجزیه و تحلیل اپیدمیولوژیک-2017
Aims: The prevalence of diabetes mellitus was reportedly 9% in 2014, making it one of the most common global chronic conditions. Hypoglycemia is an important complication of diabetes treatment. The objective of this study was to quantify and characterize hypo glycemia presentations associated with type 1 or 2 diabetes made to emergency depart ments (EDs) by adults in a Canadian province. Methods: A retrospective cohort study was conducted using reliable administrative data from Alberta for a five-year period (2010/11–2014/15). Records of interest were those with an ICD-10-CA diagnosis of diabetes-associated hypoglycemia (e.g., E10.63). A descriptive analysis was conducted. Results: Data extraction yielded 7835 presentations by 5884 patients. The majority (56.2%) of presentations were made by males, median patient age was 62, and 60.5% had type 2 dia betes. These episodes constituted 0.08% of presentations to Alberta EDs. The annual rate of presentations decreased by 11.8% during the five-year period. Most presentations (63.4%) involved transportation to ED via ambulance. Median length-of-stay was four hours. For 27.5% of presentations, an X-ray was obtained. Most hypoglycemic episodes (65.2%) were considered to be moderate, while 34.3% were considered to be severe. Conclusions: Diabetes-associated hypoglycemia presentations to Alberta EDs are more com monly made by patients with type 2 diabetes, who are more likely to be transported via ambulance and also admitted. Each year, approximately one percent of Albertans with diabetes presented with a hypoglycemia episode; however, knowledge of the variation across regions can guide a strategy for improved care.
Keywords: Diabetes mellitus | Hypoglycemia | Emergency department | Epidemiology | Administrative database
Asthma Home Management in the Inner-City: What can the Children Teach us?
مديريت آسم در خانه و در داخل شهر: کودکان ما چگونه مي توانند ما را آموزش دهند؟-2017
Objective: Knowledge of asthma home management from the perspective of poor, minority children with asthma is limited. Method: Convenience sampling methods were used to re cruit families of low-income children who are frequently in the emergency department for uncontrolled asthma. Thir teen youths participated in focus groups designed to elicit re flections on asthma home management. Data were analyzed using grounded theory coding techniques. Results: Participants (Mean age = 9.2 years) were African American (100%), enrolled in Medicaid (92.3%), averaged 1.4 (standard deviation = 0.7) emergency department visits over the prior 3 months, and resided in homes with at least 1 smoker (61.5%). Two themes reflecting multifaceted chal lenges to the development proper of self-management emerged in the analysis. Discussion: Findings reinforce the need to provide a multi pronged approach to improve asthma control in this high-risk population including ongoing child and family education and self-management support, environmental control and housing resources, linkages to smoking cessa tion programs, and psychosocial support. J Pediatr Health Care. (2017) 31, 362-371.
KEY WORDS : Asthma | children | focus groups | poverty | self-management
Effect of Certificate of Need Law on the intensity of competition: The market for emergency care
تأثیر قانون صدور گواهینامه بر کثرت رقابت: بازار خدمات اورژانسی-2017
Purpose: This article aims to contribute to the academic literature in better understanding the impact of Certificate of Need (CON) Law on Emergency Department (ED) care. Impact of CON Law on ED compe tition remains an unanswered empirical question. Methods: We examine the impact of CON Law and its stringency on the intensity of competition (rivalry among competitors) between EDs measured by the Herfindahl-Hirschman Index (HHI). We then esti mate the effects of CON Law on HHI by treating CON as an exogenous (endogenous) variable. Findings: On average the CON legislation enhances ED competition. A possible reason is that the law hinders predatory behavior, and therefore acts as an effective anti-trust tool. Other findings indicate that competition is found to be positively related to a states population size and median income and negatively related with the prevalence of employer provided insurance and magnitude of illegal immigration in a state. Practical implications: This article sheds some light on how political regulations could affect healthcare market and hence may provide public policy makers some insights on reducing healthcare cost. Originality: Our analysis of the impact of CON regulation on ED competition significantly contributes to the healthcare and strategy literature. The law potentially serves as an anti-trust tool in the hands of the government. We extend the empirical literature by treating CON Law and its stringency as exogenous (endogenous). Our comprehensive analysis considers a host of control variables such as population demographics, their health status and access to health care, healthcare facilities, political environment, in addition to the CON features.
Keywords: CON Law | Health policy | Competition | Herfindahl-Hirschman Index | Emergency department