Dataset Information

Prediction of acute appendicitis among patients with undifferentiated abdominal pain at emergency department.

ABSTRACT:

Background

Early screening and accurately identifying Acute Appendicitis (AA) among patients with undifferentiated symptoms associated with appendicitis during their emergency visit will improve patient safety and health care quality. The aim of the study was to compare models that predict AA among patients with undifferentiated symptoms at emergency visits using both structured data and free-text data from a national survey.

Methods

We performed a secondary data analysis on the 2005-2017 United States National Hospital Ambulatory Medical Care Survey (NHAMCS) data to estimate the association between emergency department (ED) patients with the diagnosis of AA, and the demographic and clinical factors present at ED visits during a patient's ED stay. We used binary logistic regression (LR) and random forest (RF) models incorporating natural language processing (NLP) to predict AA diagnosis among patients with undifferentiated symptoms.

Results

Among the 40,441 ED patients with assigned International Classification of Diseases (ICD) codes of AA and appendicitis-related symptoms between 2005 and 2017, 655 adults (2.3%) and 256 children (2.2%) had AA. For the LR model identifying AA diagnosis among adult ED patients, the c-statistic was 0.72 (95% CI: 0.69-0.75) for structured variables only, 0.72 (95% CI: 0.69-0.75) for unstructured variables only, and 0.78 (95% CI: 0.76-0.80) when including both structured and unstructured variables. For the LR model identifying AA diagnosis among pediatric ED patients, the c-statistic was 0.84 (95% CI: 0.79-0.89) for including structured variables only, 0.78 (95% CI: 0.72-0.84) for unstructured variables, and 0.87 (95% CI: 0.83-0.91) when including both structured and unstructured variables. The RF method showed similar c-statistic to the corresponding LR model.

Conclusions

We developed predictive models that can predict the AA diagnosis for adult and pediatric ED patients, and the predictive accuracy was improved with the inclusion of NLP elements and approaches.

SUBMITTER: Su D

PROVIDER: S-EPMC8759254 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:Abdominal pain represents greater than 20% of US Emergency Department (ED) visits due to a wide range of illnesses. There are currently no reliable blood biomarkers to predict serious outcomes in patients with abdominal pain. Our previous studies have identified three mRNA transcripts related to innate immune activation: alkaline phosphatase (ALPL), interleukin-8 receptor-β (IL8RB), and defensin-1 (DEFA1) as promising candidates to detect an intra-abdominal infection. The objective of this study was to evaluate the accuracy of these mRNA biomarkers to predict likely infection, hospitalization and surgery in Emergency Department patients with undifferentiated abdominal pain. We prospectively enrolled Emergency Department patients with undifferentiated abdominal pain who received an abdominal CT scan as part of their evaluation. Clinical outcomes were abstracted from the CT scan and medical records. mRNA biomarker levels were calculated independent of the clinical outcomes and their accuracy was assessed to predict infectious diagnoses, surgery and hospital admission. 89 patients were enrolled; 21 underwent surgery; 47 underwent hospital admission; and, no deaths were observed within 30 days. In identifying which cases were likely infectious, mRNA biomarkers' AUC values were: ALPL, 0.83; DEFA1 0.51; IL8RB, 0.74; and ALPL + IL8RB, 0.79. In predicting which Emergency Department patients would receive surgery, the AUC values were: ALPL, 0.75; DEFA1, 0.58; IL8RB, 0.75; and ALPL + IL8RB, 0.76. In predicting hospital admission, the AUC values were: ALPL, 0.78; DEFA1, 0.52; IL8RB, 0.74; and, ALPL + IL8RB, 0.77. For predicting surgery, ALPL + IL8RB's positive likelihood ratio (LR) was 3.97; negative LR (NLR) was 0.70. For predicting hospital admission, the same marker's positive LR was 2.80 with an NLR of 0.45. Where the primary cause for admission was a potentially infectious disorder, 33 of 34 cases (97%) had positive RNA scores. In a pragmatic, prospective diagnostic accuracy trial in Emergency Department patients with undifferentiated abdominal pain, mRNA biomarkers showed good accuracy to identify patients with potential infection, as well as those needing surgery or hospital admission.

Project description:STUDY OBJECTIVE:We compare the efficacy and safety of intravenous lidocaine with that of hydromorphone for the treatment of acute abdominal pain in the emergency department (ED). METHODS:This was a randomized, double-blind, clinical trial conducted in 2 EDs in the Bronx, NY. Adults weighing 60 to 120 kg were randomized to receive 120 mg of intravenous lidocaine or 1 mg of intravenous hydromorphone. Thirty minutes after administration of the first dose of the study drug, participants were asked whether they needed a second dose of the investigational medication to which they were randomized. Patients were also stratified according to clinical suspicion of nephrolithiasis. The primary outcome was improvement in pain scores of 0 to 10 between baseline and 90 minutes. An important secondary outcome was need for "off-protocol" parenteral analgesics, including opioids and nonsteroidal anti-inflammatory drugs. RESULTS:We enrolled 154 patients, of whom 77 received lidocaine and 77 received hydromorphone. By 90 minutes, patients randomized to lidocaine improved by a mean of 3.8 points on the 0-to-10 scale, whereas those randomized to hydromorphone improved by a mean of 5.0 points (mean difference 1.2; 95% confidence interval 0.3 to 2.2). Need for off-protocol "rescue" analgesics occurred for 39 of 77 lidocaine patients (51%) and 20 of 77 hydromorphone patients (26%) (difference 25%; 95% confidence interval 10% to 40%). Adverse events were comparable between groups. Among the subset of 22 patients with nephrolithiasis, lidocaine patients reported a mean improvement of 3.4 points on the pain scale, whereas hydromorphone patients reported a mean improvement of 6.4 points (mean difference 3.0; 95% confidence interval 0.5 to 5.5). CONCLUSION:Intravenous hydromorphone was superior to intravenous lidocaine both for general abdominal pain and a subset of patients with nephrolithiasis. A majority of patients randomly allocated to lidocaine required additional analgesics.

Project description:BACKGROUND:The HEART Pathway (history, ECG, age, risk factors, and initial troponin) is an accelerated diagnostic protocol designed to identify low-risk emergency department patients with chest pain for early discharge without stress testing or angiography. The objective of this study was to determine whether implementation of the HEART Pathway is safe (30-day death and myocardial infarction rate <1% in low-risk patients) and effective (reduces 30-day hospitalizations) in emergency department patients with possible acute coronary syndrome. METHODS:A prospective pre-post study was conducted at 3 US sites among 8474 adult emergency department patients with possible acute coronary syndrome. Patients included were ≥21 years old, investigated for possible acute coronary syndrome, and had no evidence of ST-segment-elevation myocardial infarction on ECG. Accrual occurred for 12 months before and after HEART Pathway implementation from November 2013 to January 2016. The HEART Pathway accelerated diagnostic protocol was integrated into the electronic health record at each site as an interactive clinical decision support tool. After accelerated diagnostic protocol integration, ED providers prospectively used the HEART Pathway to identify patients with possible acute coronary syndrome as low risk (appropriate for early discharge without stress testing or angiography) or non-low risk (appropriate for further in-hospital evaluation). The primary safety and effectiveness outcomes, death, and myocardial infarction (MI) and hospitalization rates at 30 days were determined from health records, insurance claims, and death index data. RESULTS:Preimplementation and postimplementation cohorts included 3713 and 4761 patients, respectively. The HEART Pathway identified 30.7% as low risk; 0.4% of these patients experienced death or MI within 30 days. Hospitalization at 30 days was reduced by 6% in the postimplementation versus preimplementation cohort (55.6% versus 61.6%; adjusted odds ratio, 0.79; 95% CI, 0.71-0.87). During the index visit, more MIs were detected in the postimplementation cohort (6.6% versus 5.7%; adjusted odds ratio, 1.36; 95% CI, 1.12-1.65). Rates of death or MI during follow-up were similar (1.1% versus 1.3%; adjusted odds ratio, 0.88; 95% CI, 0.58-1.33). CONCLUSIONS:HEART Pathway implementation was associated with decreased hospitalizations, increased identification of index visit MIs, and a very low death and MI rate among low-risk patients. These findings support use of the HEART Pathway to identify low-risk patients who can be safely discharged without stress testing or angiography. CLINICAL TRIAL REGISTRATION:URL: http://www.clinicaltrials.gov . Unique identifier: NCT02056964.

Project description:BackgroundChest pain is amongst the most common reason for presentation to the emergency department (ED). There are many causes of chest pain, and it is important for the emergency physician to quickly and accurately diagnose life threatening causes such as acute myocardial infarction (AMI). Multiple clinical decision tools have been developed to assist clinicians in risk stratifying patients with chest. There is growing recognition that machine learning (ML) will have a significant impact on the practice of medicine in the near future and may assist with diagnosis and risk stratification. This systematic review aims to evaluate how ML has been applied to adults presenting to the ED with undifferentiated chest pain and assess if ML models show improved performance when compared to physicians or current risk stratification techniques.Methods and findingsWe conducted a systematic review of journal articles that applied a ML technique to an adult patient presenting to an emergency department with undifferentiated chest pain. Multiple databases were searched from inception through to November 2020. In total, 3361 articles were screened, and 23 articles were included. We did not conduct a metanalysis due to a high level of heterogeneity between studies in both their methods, and reporting. The most common primary outcomes assessed were diagnosis of acute myocardial infarction (AMI) (12 studies), and prognosis of major adverse cardiovascular event (MACE) (6 studies). There were 14 retrospective studies and 5 prospective studies. Four studies reported the development of a machine learning model retrospectively then tested it prospectively. The most common machine learning methods used were artificial neural networks (14 studies), random forest (6 studies), support vector machine (5 studies), and gradient boosting (2 studies). Multiple studies achieved high accuracy in both the diagnosis of AMI in the ED setting, and in predicting mortality and composite outcomes over various timeframes. ML outperformed existing risk stratification scores in all cases, and physicians in three out of four cases. The majority of studies were single centre, retrospective, and without prospective or external validation. There were only 3 studies that were considered low risk of bias and had low applicability concerns. Two studies reported integrating the ML model into clinical practice.ConclusionsResearch on applications of ML for undifferentiated chest pain in the ED has been ongoing for decades. ML has been reported to outperform emergency physicians and current risk stratification tools to diagnose AMI and predict MACE but has rarely been integrated into practice. Many studies assessing the use of ML in undifferentiated chest pain in the ED have a high risk of bias. It is important that future studies make use of recently developed standardised ML reporting guidelines, register their protocols, and share their datasets and code. Future work is required to assess the impact of ML model implementation on clinical decision making, patient orientated outcomes, and patient and physician acceptability.Trial registrationInternational Prospective Register of Systematic Reviews registration number: CRD42020184977.