Search

Introduction: Glucagon-like peptide-1 agonists have gained attention in recent years due to their efficacy in managing type II diabetes mellitus and their emerging role in weight management. The purpose of this study was to characterize glucagon-like peptide-1 agonist exposures reported to a single United States regional poison center over nine years, including causes of exposure, associated clinical effects, and potential areas for improving patient education and safety., Methods: This retrospective cohort study analyzed all poison center calls involving glucagon-like peptide-1 agonists submitted to a single United States regional poison center from 14 January 2014 to 1 May 2023. Data were abstracted from the electronic medical record of the poison center, including demographics, call volume, drug involved, type of exposure, frequency of hypoglycemia, and other side effects., Results: Two hundred and thirty-seven cases involving glucagon-like peptide-1 agonists were reported to the poison center. The annual number of cases increased sharply over this period. Most patients ( n  = 166, 70.0%) were females. Most calls ( n  = 164, 69.2%) were due to unintentional therapeutic errors. Semaglutide was the most frequently involved medication ( n  = 72, 36.0%). Hypoglycemia was identified in eight patients (3.4%). The lowest mean (±SD) blood glucose concentration in these hypoglycemic patients was 49.6 ± 23.7 mg/dL (2.76 ± 1.3 mmol/L)., Discussion: Unintentional therapeutic errors were involved in 164 (69.2%) cases. Despite the generally mild clinical effects observed in this study, the occurrence of hypoglycemia in a subset of patients, often requiring hospitalization, is of concern. With reports of the acquisition of these medications through online platforms and poorly regulated compounding sources, this trend may pose public health risks., Conclusions: This study demonstrates the increasing incidence of glucagon-like peptide-1 agonist exposures reported to a United States regional poison center, predominantly due to unintentional overdoses, which highlights the need for ongoing patient education.

Methadone is an opioid receptor agonist with a high potential for abuse. The current study aimed to compare different machine learning models to predict the outcomes following methadone poisoning. This six-year retrospective longitudinal study utilizes National Poison Data System (NPDS) data. The severity of outcomes was derived from the NPDS Coding Manual. Our database was divided into training (70%) and test (30%) sets. We used a light gradient boosting machine (LGBM), extreme gradient boosting (XGBoost), random forest (RF), and logistic regression (LR) to predict the outcomes of methadone poisoning. A total of 3847 patients with methadone exposures were included. Our results demonstrated that machine learning models conferred high accuracy and reliability in determining the outcomes of methadone poisoning cases. The performance evaluation showed all models had high accuracy, precision, specificity, recall, and F 1-score values. All models could reach high specificity (more than 96%) and high precision (80% or more) for predicting major outcomes. The models could also achieve a high sensitivity to predict minor outcomes. Finally, the accuracy of all models was about 75%. However, XGBoost and LGBM models achieved the best performance among all models. This study showcased the accuracy and reliability of machine learning models in the outcome prediction of methadone poisoning.

Introduction: Efficient and complete medical charting is essential for patient care and research purposes. In this study, we sought to determine if Chat Generative Pre-Trained Transformer could generate cogent, suitable charts from recorded, real-world poison center calls and abstract and tabulate data., Methods: De-identified transcripts of real-world hospital-initiated poison center consults were summarized by Chat Generative Pre-Trained Transformer 4.0. Additionally, Chat Generative Pre-Trained Transformer organized tables for data points, including vital signs, test results, therapies, and recommendations. Seven trained reviewers, including certified specialists in poison information and board-certified medical toxicologists, graded summaries using a 1 to 5 scale to determine appropriateness for entry into the medical record. Intra-rater reliability was calculated. Tabulated data was quantitatively evaluated for accuracy. Finally, reviewers selected preferred documentation: original or Chat Generative Pre-Trained Transformer organized., Results: Eighty percent of summaries had a median score high enough to be deemed appropriate for entry into the medical record. In three duplicate cases, reviewers did change scores, leading to moderate intra-rater reliability (kappa = 0.6). Among all cases, 91 percent of data points were correctly abstracted into table format., Discussion: By utilizing a large language model with a unified prompt, charts can be generated directly from conversations in seconds without the need for additional training. Charts generated by Chat Generative Pre-Trained Transformer were preferred over extant charts, even when they were deemed unacceptable for entry into the medical record prior to the correction of errors. However, there were several limitations to our study, including poor intra-rater-reliability and a limited number of cases examined., Conclusions: In this study, we demonstrate that large language models can generate coherent summaries of real-world poison center calls that are often acceptable for entry to the medical record as is. When errors were present, these were often fixed with the addition or deletion of a word or phrase, presenting an enormous opportunity for efficiency gains. Our future work will focus on implementing this process in a prospective fashion.

Acetaminophen is one of the most commonly used analgesic drugs in the United States. However, the outcomes of acute acetaminophen overdose might be very serious in some cases. Therefore, prediction of the outcomes of acute acetaminophen exposure is crucial. This study is a 6-year retrospective cohort study using National Poison Data System (NPDS) data. A decision tree algorithm was used to determine the risk predictors of acetaminophen exposure. The decision tree model had an accuracy of 0.839, an accuracy of 0.836, a recall of 0.72, a specificity of 0.86 and an F1_score of 0.76 for the test group and an accuracy of 0.848, a recall of 0.85, a recall of 0.74, a specificity of 0.87 and an F1_score of 0.78 for the training group. Our results showed that elevated serum levels of liver enzymes, other liver function test abnormality, anorexia, acidosis, electrolyte abnormality, increased bilirubin, coagulopathy, abdominal pain, coma, increased anion gap, tachycardia and hypotension were the most important factors in determining the outcome of acute acetaminophen exposure. Therefore, the decision tree model is a reliable approach in determining the prognosis of acetaminophen exposure cases and can be used in an emergency room or during hospitalization., (© 2021 Nordic Association for the Publication of BCPT (former Nordic Pharmacological Society). Published by John Wiley & Sons Ltd.)