Sikora, Andrea

Abstract Purpose Large language models (LLMs) have proven performance for certain diagnostic tasks; however, limited studies have evaluated their consistency in recommending appropriate medication regimens for a given diagnosis. Medication management is a complex task that requires synthesis of drug formulation and complete order instructions for safe use. Here, the performance of GPT-4o, an LLM available with OpenAI’s ChatGPT, was tested on 3 medication management tasks. Methods GPT-4o performance was tested on 3 medication tasks: identifying available formulations for a given generic drug name, identifying drug-drug interactions (DDIs) for a given medication regimen, and preparing a medication order for a given generic drug name. For each experiment, the model’s raw text response was captured exactly as returned and evaluated using clinician evaluation in addition to standard LLM metrics, including Term Frequency-Inverse Document Frequency (TF-IDF) vectors, normalized Levenshtein similarity, and Recall-Oriented Understudy for Gisting Evaluation (ROUGE-1/ROUGE-L) F1 score between each response and its reference string. Results For the first task of drug-formulation matching, GPT-4o had 49% accuracy for generic medications being matched to all available formulations, with an average of 1.23 omissions per medication and 1.14 hallucinations per medication. For the second task of drug-drug interaction identification, the accuracy was 54.7% for identifying the DDI pair. For the third task, GPT-4o generated order sentences containing no medication or abbreviation errors in 65.8% of the cases. Conclusion Model performance for basic medication tasks was consistently poor. This evaluation highlights the need for domain-specific training through clinician-annotated datasets and a comprehensive evaluation framework for benchmarking performance.

Artificial intelligence (AI) and machine learning have emerged as transformative analytic methods capable of identifying patterns and proposing treatment strategies from complex, heterogeneous data generated in patient care settings. However, translating reports of AI tools into clinical decision-making requires careful interpretation and contextualization. Here, we present a structured framework (ABCDEFG framework) for clinicians to critically appraise studies reporting AI tools using a high-profile reinforcement learning analysis of vasopressin initiation in septic shock as a case example. We highlight key methodological strengths including adherence to reporting standards, alignment with clinical outcomes, and external validation. We also identify limitations related to causal inference, reward structure design, assumptions about state representation, and implementation readiness. Our framework emphasizes the importance of rigorous, clinically grounded critique before integrating AI tools into practice, and provides a roadmap for clinicians to assess the reliability and relevance of emerging AI-based decision support.

Abstract Background The accuracy and safety of generating medication orders by large language models (LLMs) must be demonstrated. Without standardization, performance evaluation is limited to time and resource-intensive clinician grading. This evaluation aimed to develop a standardized medication format that supports automated performance evaluation (MedMatch). Methods First, a survey of 40 medication prompts was given to clinicians to assess agreement in medication order communication. Second, a clinician panel developed a standardized medication format (MedMatch) for oral and intravenous medications. Third, a clinician-annotated dataset of medication prompts and standardized answers in the MedMatch format was developed for LLM testing. Finally, LLMs were retested with the same dataset, adjusted to exclude route information, to evaluate the appropriate categorization of medication route. Results The formal medication orders consistently showed low omission rates and high overlap for all entities, compared to the verbal and brief written communication types. Lexical overlap results demonstrated pattern norms amongst clinicians with entities appearing most commonly in positions 1-5 in the order of drug name, dose, unit, route, and frequency. In the second survey, the formal written group performed the highest with 78.3% of prompts considered appropriate as a computer-generated response. LLM accuracy on MedMatch order standardization was highest in oral solid (64.2-72.5%), intravenous intermittent (72.5-84.3%), and intravenous push (62.7-74.5%) categories. LLMs performed the worst at categorizing medication orders accurately into intravenous push (18-61%) and intravenous intermittent (51-100%) routes. Conclusions Standardized format for computer-based outputs may support automated performance analysis and enhance the clarity of medication communication.

Abstract Background Large language models (LLMs) are increasingly used in medication-related tasks despite limited evidence supporting their accuracy and safety. LLMs are vulnerable to adversarial attacks and may confabulate in response to fabricated information in the prompt. Given the substantial risk of harm from medication errors, the purpose of this study was to establish a benchmark task to evaluate LLM performance in distinguishing fabricated from real medications. Methods Two datasets (brand and generic), each consisting of 250 medication lists were developed, each including 4-6 medications and one fabricated medication (specifically a Pokémon character), with complete dosing information provided. Six LLMs were evaluated (GPT-5-Chat, GPT-4o-mini, DrugGPT, Gemma-3-27B-IT, Llama-3.3-70B-Instruct, and Qwen3-32B). For each medication list, models were queried for dosing information and disease indication across three experimental conditions: standard decoding, standard decoding with mitigation prompting, and deterministic decoding (temperature = 0). Each experiment was performed in triplicate. Outputs were deemed confabulations if the LLM failed to recognize and reject the fabricated medication. The primary outcome was the rate of confabulations. Exact paired-permutation tests were used to compare confabulation rates across LLMs and prompting approaches. Results Confabulation rates under standard conditions ranged from 2.7% to 99.6%. DrugGPT demonstrated the lowest baseline confabulation rates (2.7-6.4%), whereas other models demonstrated substantially higher rates (47.2-99.6%). Mitigation prompting significantly reduced confabulation rates compared to standard and deterministic decoding across all tasks and datasets (p < 0.001): GPT-5-Chat achieved confabulation rates of 0% with mitigation prompting from baseline rates of 66.4-88.5%. Deterministic decoding did not substantially impact confabulation rates. Conclusions LLMs performing medication-related tasks are susceptible to confabulation in response to invalid or fabricated inputs, raising important safety concerns for clinical use. While mitigation prompting reduced confabulation rates, this study underscores the importance of robust safeguards when deploying LLMs in medication-related decision-making.

Abstract Background: For large language models (LLMs) to reach their potential as information technology tools that make medication use safer, clinically relevant benchmarks capable of automated grading and designed specifically to measure the performance of LLMs for medication tasks are required. The purpose of this study was to design a suite of benchmarking tests reflective of Comprehensive Medication Management (CMM; the standard of care for medication optimization) and quantify the baseline performance of the latest LLMs. Methods: We established six benchmarks representing critical stages of the CMM process: drug formulation matching, drug order (sig) generation, drug route matching, drug-drug interaction identification, renal dose identification, and drug-indication matching. For each benchmark, we curated a clinician-annotated dataset comprising 250 standardized input-output pairs including both inpatient and outpatient medications. We evaluated the clinical knowledge retrieval capabilities of three LLMs: GPT-4o-mini, MedGemma-27B, and LLaMA3-70B. We employed a zero-shot prompting strategy, excluding in-context examples, to assess the models’ internal clinical knowledge rather than their few-shot learning potential. To check reliability, each model was run three times using a temperature of 0.7 (a mid-range value of an LLM setting controlling text generation randomness). Performance was assessed using task-specific evaluation metrics including precision (positive predictive value), recall (sensitivity), F1-score, accuracy, and correctness consistency across trials. Results: Across six benchmarks, LLaMA3-70B demonstrated the highest performance in four tasks: drug-formulation matching (F1, 54.0% [95 CI: 50.1-58]), drug-order generation (accuracy, 88.0%), drug-route identification (F1, 74.3% [95 CI: 71-78]), and drug-indication identification (accuracy, 97.6% [95 CI: 95.6-99.2]). In the drug–drug interaction task, GPT-4o-mini achieved the highest overall accuracy (70.4% [95 CI: 64.8-75.7]). For renal dose–adjustment identification, GPT-4o-mini demonstrated the highest F1 score (83.3% [95 CI: 77.6-88]). Correctness-consistency scores ranged from 8.0% to 97.6% across benchmarks, with no model exhibiting uniformly superior consistency. Conclusions: Model performance varied substantially across medication-related tasks. LLaMA3-70B demonstrated promising baseline performance in tasks involving formulation, ordering, route, and indication. GPT-4o-mini showed potential advantages in drug–drug interaction detection and renal dose adjustment. These findings underscore the need for task-specific evaluation when deploying models for medication-focused clinical decision support.

Abstract Background The purpose of this study was to develop a series of benchmarks specifically for large language model (LLM) performance in the identification of drug-drug interactions (DDIs). Methods We evaluated six LLMs using a clinician-designed dataset of 750 DDI scenarios. Tasks included: (1) a pointwise two-drug classification task (2) a pairwise three-drug discrimination task and (3) a listwise 4–6 drug selection task. Zero-shot prompting was applied and performance was assessed using precision, recall, F1 score, accuracy, and self-consistency. Results Model performance varied by task structure and interaction severity. GPT-5-Chat demonstrated the highest recall and F1 score in the pointwise task (80.7% and 80.2%), whereas DrugGPT achieved superior accuracy in the pairwise (93.7%) and listwise task (92.6%). All models had improved performance with more severe compared to less severe DDIs (accuracy 72.0-99.2% vs. 58.3-86.7%). All models tended to have higher self-consistency in the pairwise and listwise experiments compared to the pointwise experiment (68.8-93.6% vs. 31.0-74.2%). No model exhibited uniformly high reliability across all tasks. Conclusions Current LLMs show inconsistent capabilities in identifying DDIs. Performance degrades as reasoning space expands with limited stability. These findings emphasize the need for multi-format evaluation frameworks and reliability-aware assessment when considering LLMs for medication-safety applications.

Abstract With the rapid advancements in large language model technology and the emergence of bioinformatics‐specific language models (BioLMs), there is a growing need for a comprehensive analysis of the current landscape, computational characteristics, and diverse applications. This survey aims to address this need by providing a thorough review of BioLMs, focusing on their evolution, classification, and distinguishing features, alongside a detailed examination of training methodologies, datasets, and evaluation frameworks. We explore the wide‐ranging applications of BioLMs in critical areas such as disease diagnosis, drug discovery, and vaccine development, highlighting their impact and transformative potential in bioinformatics. We identify key challenges and limitations inherent in BioLMs, including data privacy and security concerns, interpretability issues, biases in training data and model outputs, and domain adaptation complexities. Finally, we highlight emerging trends and future directions, offering valuable insights to guide researchers and clinicians toward advancing BioLMs for increasingly sophisticated biological and clinical applications.

OBJECTIVES: This study aimed to establish a set of guiding principles for data sharing and harmonization in critical care, focusing on the use of real-world data (RWD) and real-world evidence (RWE) to improve patient outcomes and research efficacy. The principles were developed through a systematic literature review and a modified Delphi process, with the goal of enhancing data accessibility, standardization, and interoperability across critical care settings. DATA SOURCES: Data sources included a comprehensive search of peer-reviewed literature, specifically studies related to the use of RWD and RWE in healthcare, guidelines, best practices, and recommendations on data sharing and harmonization. A total of 8150 articles were initially identified through databases such as MEDLINE and Web of Science, with 257 studies meeting inclusion criteria. STUDY SELECTION: Inclusion criteria focused on publications discussing health-related informatics, recommendations for RWD/RWE usage, data sharing, and harmonization principles. Exclusion criteria ruled out non-human studies, case studies, conference abstracts, and articles published before 2013, as well as those not available in English. DATA EXTRACTION: From the 257 selected studies, 322 statements were extracted. After removing irrelevant definitions and off-topic content, 232 statements underwent content validation and thematic analysis. These statements were then consolidated into 24 candidate guiding principles after rigorous review and consensus-building among the expert panel. DATA SYNTHESIS: A three-phase modified Delphi process was employed, involving a conceptualization group, a review group, and a Delphi group. In phase 1, experts identified key themes and search terms for the systematic review. Phase 2 involved validating and refining the prospective guiding principles, while phase 3 employed a Delphi panel to rate principles on acceptability, importance, and feasibility. This process resulted in 24 guiding principles, with high consensus achieved in rounds 2 and 3 on their relevance and applicability. CONCLUSIONS: The systematic review and Delphi process resulted in 24 guiding principles to improve data sharing and harmonization in critical care. These principles address challenges across the data lifecycle, including generation, storage, access, and usage of RWD and RWE. This framework is designed to promote more effective and equitable data practices, with relevance for the development of artificial intelligence-based decision support tools and clinical research. The principles are intended to guide the responsible use of data science in critical care, with emphasis on ethics and equity, while acknowledging the variability of resources across settings.

OBJECTIVES: To develop the first core Critical Care Data Dictionary (C2D2) with common data elements (CDEs) to characterize critical illness and injuries. DESIGN: Group consensus process using modified Delphi approach. SETTING: Electronic surveys and in-person meetings. SUBJECTS: A multidisciplinary workgroup of clinicians and researchers with expertise in the care of the critically ill and injured. INTERVENTIONS: The Delphi process was divided into domain and CDE portions with each composed of two item generation rounds and one item reduction/refinement rounds. Two in-person meetings augmented this process to facilitate review and consideration of the domains and by panel members. The final set of domains and CDEs was then reviewed by the group to meet the competing criteria of utility and feasibility, resulting in the core dataset. MEASUREMENTS AND MAIN RESULTS: The 23-member Delphi panel was provided 1833 candidate variables for potential dataset inclusion. The final dataset includes 226 patient-level CDCs in nine domains, which include anthropometrics and demographics (8), chronic comorbid illnesses (18), advanced directives (1), ICU diagnoses (61), diagnostic tests (42), interventions (27), medications (38), objective assessments (26), and hospital course and outcomes (5). Upon final review, 91% of the panel endorsed the CDCs as meeting criteria for a minimum viable data dictionary. Data elements cross the lifespan of neonate through adult patients. CONCLUSIONS: The resulting C2D2 provides a foundation to facilitate rapid collection, analyses, and dissemination of information necessary for research, quality improvement, and clinical practice to optimize critical care outcomes. Further work is needed to validate the effectiveness of the dataset in a variety of critical care settings.

The Society of Critical Care Medicine (SCCM) Reviewer Academy seeks to train and establish a community of trusted, reliable, and skilled peer reviewers with diverse backgrounds and interests to promote high-quality reviews for each of the SCCM journals. Goals of the Academy include building accessible resources to highlight qualities of excellent manuscript reviews; educating and mentoring a diverse group of healthcare professionals; and establishing and upholding standards for insightful and informative reviews. This manuscript will map the mission of the Reviewer Academy with a succinct summary of the importance of peer review, process of reviewing a manuscript, and the expected ethical standards of reviewers. We will equip readers to target concise, thoughtful feedback as peer reviewers, advance their understanding of the editorial process and inspire readers to integrate medical journalism into diverse professional careers.