European Journal of Hospital Pharmacy

To investigate the performance and risk associated with the usage of Chat Generative Pre-trained Transformer (ChatGPT) to answer drug-related questions.A sample of 50 drug-related questions were consecutively collected and entered in the artificial intelligence software application ChatGPT. Answers were documented and rated in a standardised consensus process by six senior hospital pharmacists in the domains content (correct, incomplete, false), patient management (possible, insufficient, not possible) and risk (no risk, low risk, high risk). As reference, answers were researched in adherence to the German guideline of drug information and stratified in four categories according to the sources used. In addition, the reproducibility of ChatGPT's answers was analysed by entering three questions at different timepoints repeatedly (day 1, day 2, week 2, week 3).Overall, only 13 of 50 answers provided correct content and had enough information to initiate management with no risk of patient harm. The majority of answers were either false (38%, n=19) or had partly correct content (36%, n=18) and no references were provided. A high risk of patient harm was likely in 26% (n=13) of the cases and risk was judged low for 28% (n=14) of the cases. In all high-risk cases, actions could have been initiated based on the provided information. The answers of ChatGPT varied over time when entered repeatedly and only three out of 12 answers were identical, showing no reproducibility to low reproducibility.In a real-world sample of 50 drug-related questions, ChatGPT answered the majority of questions wrong or partly wrong. The use of artificial intelligence applications in drug information is not possible as long as barriers like wrong content, missing references and reproducibility remain.

Benedict Morath

Ute Chiriac

Elena Jaszkowski

Carolin Deiß

Hannah Nürnberg

Katrin Hörth

Torsten Hoppe-Tichy

Kim Green

Eur J Hosp Pharm

Performance and risks of ChatGPT used in drug information: an exploratory real-world analysis

ChatGPT在药物信息中的表现与风险：一项探索性真实世界分析

We read with interest the article on ‘Performance and risks of ChatGPT used in drug information: an exploratory realworld analysis’. Morath et al consider the performance and risks of using Chat Generative Pretrained Transformer (ChatGPT) to answer drugrelated questions. In a realworld sample of 50 drugrelated questions, Morath et al found that ChatGPT answered the majority of the questions wrongly or that answers were partially inaccurate. According to Morath et al, it is impossible to use artificial intelligence systems for drug knowledge as long as there are obstacles like inaccurate content, missing references, and repeatability. Artificial intelligence (AI) is a phrase used to describe computational software that draws conclusions from freely available data that may have come from reliable or suspect sources. AI cannot study, approve, or create confidential scientific or medical data. Without human supervision, AI should not be used to generate, update, or approve sensitive content. The reliability of data in AI is a significant and contentious issue. It is vital to think about how AI should be used responsibly and what responsible uses for AI should be. The ChatGPT may provide pertinent information right away even without user participation. Indeed, effective intake management techniques may help prevent misuse. Nonetheless in our opinion, it is crucial to keep in mind that whether or not AI is used ethically and practically is a decision to be made by people, not AI.

Amnuay Kleebayoon

Viroj Wiwanitkit

Performance and risks of ChatGPT used in drug information: comment

ChatGPT在药物信息中的表现与风险：评论

We report our experience with the use of ceftaroline in combination with daptomycin or vancomycin for methicillin resistant coagulase negative

Sunish Shah

Lloyd Clarke

Simi Padival

Ceftaroline combination therapy for methicillin resistant coagulase negative<i>Staphylococcus</i>bacteraemia and endocarditis

头孢洛林联合治疗耐甲氧西林凝固酶阴性<i>葡萄球菌</i>菌血症和心内膜炎

There have been cases of cardiotoxicity induced by osimertinib in patients with non-small-cell lung cancer (NSCLC). However, limited data exist for a comprehensive cardiotoxicity profile analysis for osimertinib use in NSCLC patients. The aim of this study was to report the entire profile of cardiotoxicities after the initiation of osimertinib in consecutive patients with estimated glomerular filtration rate (EGFR) mutation at a single health system.

Samuel Akaakole Mensah

Syed Ahmad

Waleed Alruwaili

Rutu Raval

Karthik Gonuguntla

Brijesh Patel

Cardiovascular events in estimated glomerular filtration rate-mutation non-small-cell lung cancer patients on osimertinib

估计肾小球滤过率突变非小细胞肺癌患者使用奥希替尼的心血管事件

<h3>Objectives</h3> Automated dispensing cabinets (ADCs) offer improved medication safety, greater efficiency and return on investment. However, integrating ADCs into medication dispensing processes can be challenging in complex hospital environments. This study aimed to draft suggestions to help hospitals adopt ADCs. <h3>Methods</h3> Two-day visits were organised in seven European hospitals operating ADCs. Investigators used an observational grid, a questionnaire and interviews, each divided into the themes of medication processes before and after the introduction of ADCs, the major steps followed and the resources involved, ergonomics and staff perceptions. <h3>Results</h3> ADCs were integrated into four global hospital medication dispensing systems (packs of drugs are distributed from the central pharmacy to wards for dispensing) and three nominative systems—that is, patient-specific ones (drug doses prescribed for individuals are distributed from the central pharmacy to wards with ADC as supplementary stock). A general ADC project implementation timeline was shaped: main drivers of automation to initiate the project, visit of other sites, pilot test (with IT integration and staff training), and evaluation phase (satisfaction, safety, efficiency) to justify a possible expansion. Users (7 pharmacists, 21 nurses, 7 data engineers) identified facilitators (such as a dedicated project manager, a pilot phase, an intuitive device), barriers and any improvements needed (training for incoming staff, reorganisation of ward workflow, dynamic inventories). <h3>Conclusions</h3> Despite their diverse pharmacy organisations, each hospital raised similar challenges and reported analogous major steps in project implementation. Although integration processes are complex, ADCs rapidly provide users with benefits. By following the practical advice and recommendations from these hospitals, new adopters might reduce the risks of failed ADC projects and accelerate their integration.

Olivia Francois

Edith Hufschmid Thurnherr

Cedric Blatrie

Etienne Cousein

Ana Herranz

Farshid Sadeghipour

Pascal Bonnabry

Integrating automated dispensing cabinets into the medication dispensing process

短名	Eur J Hosp Pharm
Journal Impact	1.41
国际分区	PHARMACOLOGY & PHARMACY(Q3)
期刊索引	SCI Q4中科院 4 区

ISSN	2047-9956, 2047-9964
h-index	24
国内分区	医学(4区)医学药学(4区)

涉及主题	医学生物内科学护理部药理学外科病理家庭医学心理学政治学法学精神科遗传学化学药店生物化学数学免疫学经济急诊医学药方统计
出版信息	出版商: European Association of Hospital Pharmacists (EAHP)，出版周期: ，期刊类型: journal，开源期刊: 否
基本数据	创刊年份: 2012，原创研究文献占比: 97.22%，自引率:N.A.， Gold OA占比: 0.00%
平均审稿周期	网友分享经验：较慢,6-12周
平均录用比例	网友分享经验：容易

European Journal of Hospital Pharmacy

期刊引文格式

有 1 位以上作者的期刊

有 2 位作者的期刊

有 3 位作者的期刊

有 5 位以上作者的期刊

书籍引用格式

学位论文引用格式

网页引用格式

专利引用格式

手工熬夜修改参考文献？研飞自动匹配期刊，一键轻松成稿，支持 Word/WPS

投稿经验分享

Performance and risks of ChatGPT used in drug information: an exploratory real-world analysis

Performance and risks of ChatGPT used in drug information: comment

Ceftaroline combination therapy for methicillin resistant coagulase negativeStaphylococcusbacteraemia and endocarditis

Cardiovascular events in estimated glomerular filtration rate-mutation non-small-cell lung cancer patients on osimertinib

Integrating automated dispensing cabinets into the medication dispensing process

Ceftaroline combination therapy for methicillin resistant coagulase negativeStaphylococcusbacteraemia and endocarditis

Cardiovascular events in estimated glomerular filtration rate-mutation non-small-cell lung cancer patients on osimertinib

Integrating automated dispensing cabinets into the medication dispensing process

Neutropenia possibly caused by cefoperazone/sulbactam

Importance of ambient temperature on the efficacy of elastomeric infusion pumps