Search

BackgroundLarge language models (LLMs) have demonstrated impressive performance on medical licensing and diagnosis-related exams. However, comparative evaluations to optimize LLM performance and ability in the domain of comprehensive medication management (CMM) are lacking. The purpose of this evaluation was to test various LLMs performance optimization strategies and performance on critical care pharmacotherapy questions used in the assessment of Doctor of Pharmacy students.MethodsIn a comparative analysis using 219 multiple-choice pharmacotherapy questions, five LLMs (GPT-3.5, GPT-4, Claude 2, Llama2-7b and 2-13b) were evaluated. Each LLM was queried five times to evaluate the primary outcome of accuracy (i.e., correctness). Secondary outcomes included variance, the impact of prompt engineering techniques (e.g., chain-of-thought, CoT) and training of a customized GPT on performance, and comparison to third year doctor of pharmacy students on knowledge recall vs. knowledge application questions. Accuracy and variance were compared with student’s t-test to compare performance under different model settings.ResultsChatGPT-4 exhibited the highest accuracy (71.6%), while Llama2-13b had the lowest variance (0.070). All LLMs performed more accurately on knowledge recall vs. knowledge application questions (e.g., ChatGPT-4: 87% vs. 67%). When applied to ChatGPT-4, few-shot CoT across five runs improved accuracy (77.4% vs. 71.5%) with no effect on variance. Self-consistency and the custom-trained GPT demonstrated similar accuracy to ChatGPT-4 with few-shot CoT. Overall pharmacy student accuracy was 81%, compared to an optimal overall LLM accuracy of 73%. Comparing question types, six of the LLMs demonstrated equivalent or higher accuracy than pharmacy students on knowledge recall questions (e.g., self-consistency vs. students: 93% vs. 84%), but pharmacy students achieved higher accuracy than all LLMs on knowledge application questions (e.g., self-consistency vs. students: 68% vs. 80%).ConclusionChatGPT-4 was the most accurate LLM on critical care pharmacy questions and few-shot CoT improved accuracy the most. Average student accuracy was similar to LLMs overall, and higher on knowledge application questions. These findings support the need for future assessment of customized training for the type of output needed. Reliance on LLMs is only supported with recall-based questions.

OBJECTIVES:. We aimed to summarize the most significant and impactful publications describing the pharmacotherapeutic care of critically ill patients in 2023. DATA SOURCES:. PubMed/MEDLINE and the Clinical Pharmacy and Pharmacology Pharmacotherapy Literature Update. STUDY SELECTION:. Randomized controlled trials and prospective studies of adult critically ill patients assessing a pharmacotherapeutic intervention and reporting clinical endpoints published between January 1, 2023, and December 31, 2023, were eligible for inclusion in this article. DATA EXTRACTION:. Articles from a systematic search and the Clinical Pharmacy and Pharmacology Pharmacotherapy Literature Update were included. An a priori defined three-round modified Delphi process was employed to achieve consensus on the most impactful publications based on the following considerations: 1) overall contribution to scientific knowledge and 2) novelty to the literature. DATA SYNTHESIS:. The systematic search and Clinical Pharmacy and Pharmacology Pharmacotherapy Literature Update returned a total of 1202 articles, of which 1164 were excluded. The remaining 38 articles underwent a three-round modified Delphi process. In each round, articles were independently scored based on overall contribution to scientific knowledge and novelty to the literature. Included articles are summarized and their impact discussed. Article topics included hydrocortisone for severe community-acquired pneumonia, inhaled amikacin for prevention of ventilator-associated pneumonia, methylene blue for septic shock, restrictive vs. liberal fluid management for sepsis-induced hypotension, andexanet alfa for major bleeding associated with factor Xa inhibitors, and early administration of four-factor prothrombin complex concentrate in patients with trauma at risk for massive transfusion. CONCLUSIONS:. This review provides a summary and perspective on the potential impact of the most relevant articles in 2023 describing advances in the pharmacotherapeutic care of critically ill patients.

Abstract Fluid overload, while common in the ICU and associated with serious sequelae, is hard to predict and may be influenced by ICU medication use. Machine learning (ML) approaches may offer advantages over traditional regression techniques to predict it. We compared the ability of traditional regression techniques and different ML-based modeling approaches to identify clinically meaningful fluid overload predictors. This was a retrospective, observational cohort study of adult patients admitted to an ICU ≥ 72 h between 10/1/2015 and 10/31/2020 with available fluid balance data. Models to predict fluid overload (a positive fluid balance ≥ 10% of the admission body weight) in the 48–72 h after ICU admission were created. Potential patient and medication fluid overload predictor variables (n = 28) were collected at either baseline or 24 h after ICU admission. The optimal traditional logistic regression model was created using backward selection. Supervised, classification-based ML models were trained and optimized, including a meta-modeling approach. Area under the receiver operating characteristic (AUROC), positive predictive value (PPV), and negative predictive value (NPV) were compared between the traditional and ML fluid prediction models. A total of 49 of the 391 (12.5%) patients developed fluid overload. Among the ML models, the XGBoost model had the highest performance (AUROC 0.78, PPV 0.27, NPV 0.94) for fluid overload prediction. The XGBoost model performed similarly to the final traditional logistic regression model (AUROC 0.70; PPV 0.20, NPV 0.94). Feature importance analysis revealed severity of illness scores and medication-related data were the most important predictors of fluid overload. In the context of our study, ML and traditional models appear to perform similarly to predict fluid overload in the ICU. Baseline severity of illness and ICU medication regimen complexity are important predictors of fluid overload.

Abstract Unsupervised clustering of intensive care unit (ICU) medications may identify unique medication clusters (i.e., pharmacophenotypes) in critically ill adults. We performed an unsupervised analysis with Restricted Boltzmann Machine of 991 medications profiles of patients managed in the ICU to explore pharmacophenotypes that correlated with ICU complications (e.g., mechanical ventilation) and patient-centered outcomes (e.g., length of stay, mortality). Six unique pharmacophenotypes were observed, with unique medication profiles and clinically relevant differences in ICU complications and patient-centered outcomes. While pharmacophenotypes 2 and 4 had no statistically significant difference in ICU length of stay, duration of mechanical ventilation, or duration of vasopressor use, their mortality differed significantly (9.0% vs. 21.9%, p

Abstract Prostacyclin therapy is a mainstay of the management of pulmonary arterial hypertension (PAH). Inhaled prostacyclins present safe and effective options for the management of PAH that limit systemic side effects. We describe the first reported case of life‐threatening bronchospasm and acute respiratory failure associated with inhaled prostacyclin administration.

Abstract While medication regimen complexity, as measured by a novel medication regimen complexity-intensive care unit (MRC-ICU) score, correlates with baseline severity of illness and mortality, whether the MRC-ICU improves hospital mortality prediction is not known. After characterizing the association between MRC-ICU, severity of illness and hospital mortality we sought to evaluate the incremental benefit of adding MRC-ICU to illness severity-based hospital mortality prediction models. This was a single-center, observational cohort study of adult intensive care units (ICUs). A random sample of 991 adults admitted ≥ 24 h to the ICU from 10/2015 to 10/2020 were included. The logistic regression models for the primary outcome of mortality were assessed via area under the receiver operating characteristic (AUROC). Medication regimen complexity was evaluated daily using the MRC-ICU. This previously validated index is a weighted summation of medications prescribed in the first 24 h of ICU stay [e.g., a patient prescribed insulin (1 point) and vancomycin (3 points) has a MRC-ICU = 4 points]. Baseline demographic features (e.g., age, sex, ICU type) were collected and severity of illness (based on worst values within the first 24 h of ICU admission) was characterized using both the Acute Physiology and Chronic Health Evaluation (APACHE II) and the Sequential Organ Failure Assessment (SOFA) score. Univariate analysis of 991 patients revealed every one-point increase in the average 24-h MRC-ICU score was associated with a 5% increase in hospital mortality [Odds Ratio (OR) 1.05, 95% confidence interval 1.02–1.08, p = 0.002]. The model including MRC-ICU, APACHE II and SOFA had a AUROC for mortality of 0.81 whereas the model including only APACHE-II and SOFA had a AUROC for mortality of 0.76. Medication regimen complexity is associated with increased hospital mortality. A prediction model including medication regimen complexity only modestly improves hospital mortality prediction.

Abstract Background Identifying patterns within ICU medication regimens may help artificial intelligence algorithms to better predict patient outcomes; however, machine learning methods incorporating medications require further development, including standardized terminology. The Common Data Model for Intensive Care Unit (ICU) Medications (CDM-ICURx) may provide important infrastructure to clinicians and researchers to support artificial intelligence analysis of medication-related outcomes and healthcare costs. Using an unsupervised cluster analysis approach in combination with this common data model, the objective of this evaluation was to identify novel patterns of medication clusters (termed ‘pharmacophenotypes’) correlated with ICU adverse events (e.g., fluid overload) and patient-centered outcomes (e.g., mortality). Methods This was a retrospective, observational cohort study of 991 critically ill adults. To identify pharmacophenotypes, unsupervised machine learning analysis with automated feature learning using restricted Boltzmann machine and hierarchical clustering was performed on the medication administration records of each patient during the first 24 h of their ICU stay. Hierarchical agglomerative clustering was applied to identify unique patient clusters. Distributions of medications across pharmacophenotypes were described, and differences among patient clusters were compared using signed rank tests and Fisher's exact tests, as appropriate. Results A total of 30,550 medication orders for the 991 patients were analyzed; five unique patient clusters and six unique pharmacophenotypes were identified. For patient outcomes, compared to patients in Clusters 1 and 3, patients in Cluster 5 had a significantly shorter duration of mechanical ventilation and ICU length of stay (p

Abstract Tacrolimus (TAC) and mycophenolic acid (MPA) provide maintenance immunosuppression and is dosed empirically in elderly kidney transplant recipients (KTRs) resulting in health inequities. Limited immunosuppressive pharmacokinetics are available comparing adult ages. This secondary analysis compared TAC and MPA pharmacokinetics and adverse effects (AEs) among young, middle‐aged, and elderly Black and White KTRs. The 12‐h TAC and MPA pharmacokinetics with AE evaluation were conducted in 67 stable KTRs greater than or equal to 6 months post‐transplant. TAC regimens were adjusted to target troughs. MPA regimens were adjusted using clinical response. Participants were: young: less than or equal to 40 years; middle age: greater than 40 to 60 years, and elderly greater than 60 years. Noncompartmental pharmacokinetic analysis determined area under the concentration‐time curve 0–12 h (AUC0‐12h), clearance (CL), and CL/body mass index (BMI) with 0‐h troughs. MPA enterohepatic recirculation (EHR), MPA‐AUC6‐12h/MPA‐AUC0‐12h, and MPA glucuronide (MPAG)‐AUC0‐12h/MPA‐AUC0‐12h were determined. Univariate analysis of variance (ANOVA) was conducted using SAS version 9.4. No group differences were noted for estimated glomerular filtration rate, MPA, and TAC doses. EHR was reduced in elderly with decreased MPA‐AUC6‐12h/MPA‐AUC0‐12h (p = 0.049) and increased MPAG‐AUC0‐12h/MPA‐AUC0‐12h (p = 0.036). MPA troughs (p = 0.045) were reduced in the elderly. TAC CL/BMI (p = 0.043) was reduced in the elderly. For therapeutic MPA AUC0‐12h: 30–60 mg·h/L, 34.3% KTRs achieved this target with 55.2% greater than the therapeutic range. 77.6% KTR were in the TAC AUC0‐12h target: 100–190 ng·h/mL and 19.4% were below this range with no age relationship. In 44% young, 26% middle‐age and 7.8% elderly subjects achieved target AUC0‐12h for both medications (p = 0.036). Neurologic AEs were manifested in the elderly (p = 0.014). Immunosuppressive pharmacokinetics demonstrated age‐related differences with reduced TAC CL/BMI and MPA EHR and increased neurologic AE in the elderly. This immunosuppressive regimen may require age‐adjusted individualization to optimize allograft function.

This study presents a method in which historical AIS data are used to predict the future trajectory of a selected vessel. This is facilitated via a system intelligence-based approach that can be subsequently utilized to provide enhanced situation awareness to navigators and future autonomous ships, aiding proactive collision avoidance. By evaluating the historical ship behavior in a given geographical region, the method applies machine learning techniques to extrapolate commonalities in relevant trajectory segments. These commonalities represent historical behavior modes that correspond to the possible future behavior of the selected vessel. Subsequently, the selected vessel is classified to a behavior mode, and a trajectory with respect to this mode is predicted. This is achieved via an initial clustering technique and subsequent trajectory extraction. The extracted trajectories are then compressed using the Karhunen–Loéve transform, and clustered using a Gaussian Mixture Model. The approach in this study differs from others in that trajectories are not clustered for an entire region, but rather for relevant trajectory segments. As such, the extracted trajectories provide a much better basis for clustering relevant historical ship behavior modes. A selected vessel is then classified to one of these modes using its observed behavior. Trajectory predictions are facilitated using an enhanced subset of data that likely correspond to the future behavior of the selected vessel. The method yields promising results, with high classification accuracy and low prediction error. However, vessels with abnormal behavior degrade the results in some situations, and have also been discussed in this study.

The use of electrical precompensation for high-speed operation of lumped MZMs is demonstrated for 50 Gb/s PAM4 modulation through experiment and simulation. An accurate equivalent circuit model is fitted to S11 measurements and then used to analyse design trade-offs in terms of system performance, energy consumption and drive voltage requirements. If precompensation is not used, careful selection of source resistance with respect to electrode and packaging inductance is shown to be crucial for high-speed modulation. Increasing the electrode inductance is shown to be beneficial as it introduces inductive peaking, which allows optimum performance with a higher source impedance. Furthermore, combining precompensation with a slight inductive peaking and optimised source resistance is demonstrated to reduce drive voltage requirements and further reduce the energy dissipated in the equivalent circuit.

Aim: The cost–effectiveness of treatment options (anticholinergics, β3-adrenoceptor agonists, onabotulinumtoxinA, sacral nerve stimulation and percutaneous tibial stimulation [the latter two including new rechargeable neurostimulators]) for the management of overactive bladder (OAB) were compared with best supportive care (BSC) using a previously published Markov model. Materials & methods: Cost–effectiveness was evaluated over a 15-year time horizon, and sensitivity analyses were performed using 2- and 5-year horizons. Discontinuation rates, resource utilization, and costs were derived from published sources. Results: UsingMedicare and commercial costs over a 15-year time period, onabotulinumtoxinA 100U had incremental cost–effectiveness ratios (ICERs) gained of $39,591/qualityadjusted life-year (QALY) and $42,255/QALY, respectively, versus BSC, which were the lowest ICERs of all assessed treatments. The sensitivity analyses at 2- and 5-year horizons also showed onabotulinumtoxinA to be the most cost-effective of all assessed treatments versus BSC. Conclusion: OnabotulinumtoxinA 100U is currently the most cost-effective treatment for OAB.

Reverse triggering is an underdiagnosed form of patient-ventilator asynchrony in which a passive ventilator-delivered breath triggers a neural response resulting in involuntary patient effort and diaphragmatic contraction. Reverse triggering may significantly impact patient outcomes, and the unique physiology underscores critical potential implications for drug-device-patient interactions. The purpose of this review is to summarize what is known of reverse triggering and its pharmacotherapeutic consequences, with a particular focus on describing reported cases, physiology, historical context, epidemiology, and management. The PubMed database was searched for publications that reported patients presenting with reverse triggering. The current body of evidence suggests that deep sedation may predispose patients to episodes of reverse triggering; as such, providers may consider decreasing sedation or modifying ventilator settings in patients exhibiting ventilator asynchrony as an initial measure. Increased clinician awareness and research focus are necessary to understand appropriate management of reverse triggering and its association with patient outcomes.

In-system clinical laboratories have proven themselves to be a fundamentally important resource to their institutions during the COVID-19 pandemic of the past year. The ability to provide SARS-CoV-2 molecular testing to our hospital system allowed us to offer the best possible care to our patients, and to support neighboring hospitals and nursing homes. In-house testing led to significant revenue enhancement to the laboratory and institution, and attracted new patients to the system. Timely testing of inpatients allowed the majority who did not have COVID-19 infection to be removed from respiratory and contact isolation, conserving valuable personal protective equipment and staff resources at a time that both were in short supply. As 2020 evolved and our institution restarted delivery of routine care, the availability of in-system laboratory testing to deliver both accurate and timely results was absolutely critical. In this article, we attempt to demonstrate the value and impact of an in-system laboratory during the COVID-19 pandemic. A strong in-house laboratory service was absolutely critical to institutional operational and financial success during 2020, and will ensure resiliency in the future as well.

An operationally implementable predictive model has been developed to forecast the number of COVID-19 infections in the patient population, hospital floor and ICU censuses, ventilator and related supply chain demand. The model is intended for clinical, operational, financial and supply chain leaders and executives of a comprehensive healthcare system responsible for making decisions that depend on epidemiological contingencies. This paper describes the model that was implemented at NorthShore University HealthSystem and is applicable to any communicable disease whose risk of reinfection for the duration of the pandemic is negligible.

Previously we identified a sodium caseinate (NaCas) hydrolysate, LFC25, which significantly increased secretion of satiety hormone glucagon-like peptide-1 (GLP-1) in vitro and reduced food intake in mice when administered intraperitoneally. This study investigates whether LFC25, administered orally, promotes GLP-1 secretion and/or reduces food intake in vivo. Over an 8-hour period, mice received LFC25 by oral gavage had similar food intake to mice received NaCas. Postprandial blood glucose, plasma active GLP-1, amino acids, insulin and food consumed at the next meal were not significantly different in pigs that consumed LFC25 compared to NaCas in a dairy beverage, at a dosage relevant for human consumption. Simulated in vitro gastro-duodenal digestion of LFC25 revealed a significant reduction in bioactivity. In contrast, the harsh conditions of the upper gut appear to functionalize intact NaCas as a GLP-1 secretagogue. In conclusion, LFC25 will need enteric protection to be used as a food ingredient for satiety.

Objective: Inhalation injury is an independent risk factor for mortality in burn patients. The purpose of this study was to observe the effect of inhaled iloprost on gas exchange in patients with inhalation injury and acute respiratory failure as measured by an improved PaO2/FiO2 ratio. Methods: Patients admitted to the burn intensive care unit from 2013 to 2014 meeting Berlin criteria for acute respiratory distress syndrome (ARDS) with a diagnosis of inhalation injury and who received inhaled iloprost were included. Medical records were reviewed to collect patient demographics, characterize iloprost prescribing practices, and observe changes in oxygenation and hemodynamic status after iloprost administration. Differences were evaluated using a t-test with cluster corrected standard errors. Results: A total of eight patients were included with 157 different PaO2/FiO2 ratios calculated. All patients had moderate or severe ARDS with a baseline PaO2/FiO2 ratio of 131.9 mmHg (IQR 119.3–197.3). Median duration of iloprost therapy was 5 days (IQR ± 7). A statistically significant increase in PaO2/FiO2 ratio was observed after iloprost administration with a mean increase of 9.7 mmHg (95% CI 1.8–17.7, p = 0.023). Inhaled iloprost had no effect on hemodynamic parameters. Conclusions: Inhaled iloprost for inhalation injury and ARDS after burn injury was associated with a small but statistically significant improvement in oxygenation. Keywords: Iloprost, Inhaled prostacyclin, Inhalation injury

This article examines the intricacies of environmental credit generation from concentrated animal feeding operation (CAFO) farm systems. Livestock production generates large amounts of manure (solid and liquid waste) and consumes a high volume of water that producers must manage to control odors and reduce pollution. To mitigate environmental impacts such as nutrient releases and greenhouse gas emissions, anaerobic digesters (ADs) can be used in CAFOs to avoid certain nutrient releases and capture the methane (CH4) produced when manure is broken down anaero­bically. Policy incentives have increased the use of ADs to reduce waste volumes and produce biogas for energy or bioelectricity, but the overall digester adoption rate is still very low (U.S. Environmental Protection Agency [U.S. EPA], 2010; 2014). To cover the higher cost of AD adoption relative to other forms of waste management, farmers may supplement the revenues they generate from the conventional outputs of a livestock operation (e.g., milk or meat) by selling credits into environmental markets. One question that arises is whether a single operation can sell into multiple credit mar­kets by “stacking” credits—that is, receiving multiple separate environmental payments to finance the conversion to AD technology. The issue is that the use of stacked credits introduces the possibility that some of the stacked credits might be for benefits that are “non-additional” in that they do not produce incremental pollution reductions and thus are suspect for use in off­setting a buyer’s pollution. This article describes the stacking problem and explores possible solu­tions, such as temporal constraints on credit issuance and discounting credits to account for additionality problems.

Background: Alien limb phenomenon occurs in 50–60% of patients with corticobasal syndrome (CBS) and usually presents with an “alien hand” phenomenon. The “alien foot” presentation is rarer and may be misdiagnosed, as foot involvement can lead to erroneous localization of the clinical problem to the knee, hip, or back. Subsequently misdiagnoses such as myelopathy, radiculopathy, functional disorder, stiff leg syndrome, neuromyotonia, and painful leg moving toes syndrome may occur.Case report: We describe two patients with alien foot symptoms that resulted in multiple opinions from different specialists, multiple diagnostic and therapeutic procedures, and delayed diagnosis. Eventually a diagnosis of CBS was made in both. Alien foot symptoms may be more common than initially thought and can result in a delayed diagnosis of CBS.Conclusion: The inclusion of this clinical finding in recently proposed diagnostic criteria highlights the need for increased clinical awareness.

A novel diagnosis technology combining the benefits of spectral kurtosis and wavelet transform is proposed and validated for early defect diagnosis of rolling element bearings. A systematic procedure for feature calculation is proposed and rules for selection of technology parameters are explained. Experimental validation of the proposed method carried out for early detection of the inner race defect. A comparison between frequency band selection through wavelets and spectral kurtosis is also presented. It has been observed that the frequency band selected using spectral kurtosis provide better separation between healthy and defective bearings compared to the frequency band selection using wavelet. In terms of Fisher criterion the use of spectral kurtosis has a gain of 2.75 times compared to the wavelet.

Many of the world’s poorest people live in areas with high malaria rates and suffer the associated physical, economic, and social hardships. These same areas are often undergoing extensive forest conversion and degradation. While causality has generally not been established, the scientific literature makes it abundantly clear that the juxtaposition of deprivation, deforestation, and disease is not pure coincidence. We chart a course for using transdisciplinary research to develop more effective policies to control malaria, protect forests, and alleviate poverty. First describing the malaria problem, including its etiologic roots, and its social toll, the paper then examines some shortcomings of contemporary societal responses. We discuss why understanding the role of deforestation in linking malaria to poverty is important and present the mixed empirical evidence on the malaria-deforestation-poverty link from macro- and micro-economic studies. The paper concludes with a proposal for strategically linking research and policy at the malaria-deforestation-poverty nexus in a comprehensive decision-analysis framework that channels research to the most pressing policy needs, informs policy with the most conclusive research, and ensures stakeholders are effectively informed about their options.

Bearing defect diagnosis is traditionally done using the demodulation/enveloping technology. Diagnosis is mostly based on the spectrum of the squared envelope signal. In literature, the use of the higher order spectra (HOS) has shown to have a tremendous potential for vibration based diagnostics. In this paper we implemented and experimentally validated the higher order spectra based on the envelope analysis for the diagnosis of ball bearing defects. The implemented technology employs the spectral kurtosis to obtain a frequency band for the demodulation and the third order normalized spectra, i.e. the bicoherence for diagnosis of bearing fault. The high effectiveness of the diagnostics of the implemented technology has been experimentally revealed and compared with that of well-known demodulation/enveloping technology.

The benefits of laparoscopic surgery over conventional abdominal surgery have been well documented. Reducing postoperative pain, decreasing postoperative morbidity, hospital stay duration, and postoperative recovery time have all been demonstrated in recent peer-review literature. Robotic laparoscopy provides the added dimension of increased fine mobility and surgical control. With new single port surgical techniques, we have the added benefit of minimally invasive surgery and greater patient aesthetic satisfaction, as well as all the other benefits laparoscopic surgery offers. In this paper, we report a successful single port robotic hysterectomy and the simple process by which this technique is performed.

OBJECTIVE: To describe the prevalence and severity of periodic limb movements during sleep in amyotrophic lateral sclerosis patients and to explore this fact as a predictor of severity of the condition with respect to mortality. METHODS: In this case-control study, questionnaire and polysomnographic data were analyzed from 35 amyotrophic lateral sclerosis patients. Controls were matched by age, genre, and body mass index. A Kaplan-Meier curve was used to compare the survival time of patients with periodic limb movements of sleep index below or above 5. RESULTS: The number of amyotrophic lateral sclerosis patients with an index greater than five was higher than controls (19 (53%) versus 4 (11%); p