Drug information center: challenges of the research process to answer enquiries in hospital pharmaceutical practices

Flôres, Damiana da Rocha Vianna; Sartori, Alexandre Augusto de Toni; Antunes, Julia Borges; Pinto, Alessandra Nunes; Pletsch, Julia Tauana; Pizzol, Tatiane da Silva Dal

doi:10.1136/ejhpharm-2017-001417

Cited by 7 publications

(7 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Human pharmacists often rely on tertiary references, such as Lexicomp, which are curated sources providing concise and easily accessible information. 22 , 23 In contrast, GPT-4 is trained on an extensive dataset from the internet encompassing primary references, such as original research articles, clinical trials, and case reports, as well as secondary references like review articles and guidelines. While primary references offer the most current and detailed information, they may be more difficult to interpret and apply in real-world situations due to their complexity.…”

Section: Discussionmentioning

confidence: 99%

Safety and quality of AI chatbots for drug-related inquiries: A real-world comparison with licensed pharmacists

Albogami,

Alfakhri,

Alaqil

et al. 2024

DIGITAL HEALTH

View full text Add to dashboard Cite

Introduction Pharmacists play a pivotal role in ensuring patients are administered safe and effective medications; however, they encounter obstacles such as elevated workloads and a scarcity of qualified professionals. Despite the prospective utility of large language models (LLMs), such as Generative Pre-trained Transformers (GPTs), in addressing pharmaceutical inquiries, their applicability in real-world cases remains unexplored. Objective To evaluate GPT-based chatbots’ accuracy in real-world drug-related inquiries, comparing their performance to licensed pharmacists. Methods In this cross-sectional study, authors analyzed real-world drug inquiries from a Drug Information Inquiry Database. Two independent pharmacists evaluated the performance of GPT-based chatbots (GPT-3, GPT-3.5, GPT-4) against human pharmacists using accuracy, detail, and risk of harm criteria. Descriptive statistics described inquiry characteristics. Absolute proportion comparative analyses assessed accuracy, detail, and risk of harm. Stratified analyses were performed for different inquiry types. Results Seventy inquiries were included. Most inquiries were received from physicians (41%) and pharmacists (44%). Inquiries type included dosage/administration (34.2%), drug interaction (12.8%) and pregnancy/lactation (15.7%). Majority of inquires included adults (83%) and female patients (54.3%). GPT-4 had 64.3% completely accurate responses, comparable to human pharmacists. GPT-4 and human pharmacists provided sufficiently detailed responses, with GPT-4 offering additional relevant details. Both GPT-4 and human pharmacists delivered 95% safe responses; however, GPT-4 provided proactive risk mitigation information in 70% of the instances, whereas similar information was included in 25.7% of human pharmacists’ responses. Conclusion Our study showcased GPT-4's potential in addressing drug-related inquiries accurately and safely, comparable to human pharmacists. Current GPT-4-based chatbots could support healthcare professionals and foster global health improvements.

show abstract

Section: Discussionmentioning

confidence: 99%

Safety and quality of AI chatbots for drug-related inquiries: A real-world comparison with licensed pharmacists

Albogami,

Alfakhri,

Alaqil

et al. 2024

DIGITAL HEALTH

View full text Add to dashboard Cite

show abstract

“…The overall median (IQR) time to resolve a DI request was 5 (3)(4)(5)(6)(7)(8)(9)(10)(11)(12)(13)(14)(15) mins. Requests from specialty care practices took significantly more time to address compared to primary care (P ¼ 0.0001).…”

Section: Description Of Drug Information Requestsmentioning

confidence: 99%

“…Additional examples can be found in Appendix A. Median (IQR) time to resolve DI requests was 5 (2-10) mins for training, 5 (3)(4)(5)(6)(7)(8)(9)(10)(11) for active information, and 10 (4-15) for passive information. Pharmacists spent significantly more time with passive information than with requests addressable with training (P < 0.0001) or active information (P ¼ 0.0031).…”

Section: Training Active and Passive Informationmentioning

confidence: 99%

“…5,8,9 The ease of accessing medical literature has caused some to disregard DI services as a specialized skill. 8 Because most healthcare professionals lack the time and training to effectively answer complex DI questions, 8,10 pharmacists, who are trained extensively in pharmacotherapy and on how to evaluate medical literature, have traditionally assumed that responsibility. 11 In 2014, 97.2% of hospitals in the United States reported pharmacists as the most frequently cited source of DI for prescribers.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Identifying Training Needs and Active Information Opportunities in Primary Care Through the Analysis of Drug Information Requests

Salgado

Barker

Frerichs

et al. 2021

Journal of Pharmacy Practice

View full text Add to dashboard Cite

Introduction: Drug information (DI) services should work toward efficiency by identifying knowledge gaps and actively creating resources to address those needs. The aim was to identify training needs and active information opportunities in primary care by analyzing DI requests and to calculate labor cost associated with DI requests addressable with training or active information. Methods: DI requests received in 2016 and 2017 by ambulatory care pharmacists were independently classified by 2 authors into: training (i.e., delivery of content meant to be retained as knowledge and used when needed); active information (i.e., resources created preemptively and consulted when needed); or passive information (i.e., not addressable with training or active information). Inter-rater reliability was calculated using Cohen’s Kappa. Median time spent by category and across practice settings/professional types was compared using bivariate analysis. Thematic analysis categorized specific training and active DI requests and labor costs were calculated. Results: Of 2,041 DI requests, 330 (16.2%) were classified as training, 454 (22.2%) active information, and 1257 (61.6%) passive information (kappa = 0.769). Median (IQR) time to resolve requests was 5 (2-10) mins for training, 5 (3-11) active information, and 10 (4-15) passive information. Pharmacists spent 132.1 hrs = $8,956.98 answering questions addressable with training or active information. Areas warranting training or active information included: controlled substances, immunizations, patient assistance programs, policy/regulations, medication preparation/administration, storage/stability, disposal, availability/ordering medications, and patient-related resources. Conclusion: Several opportunities for training and active information were identified. Despite the single-institution nature, the method described can serve as an example for other institutions.

show abstract

“…One such role is the medicines information (MI) pharmacist, a specialist pharmacist role providing evidence-based answers to complex medication-related queries asked by clinicians [8][9][10]. MI centres [2] in Australia are primarily located within hospitals and are staffed by pharmacists experienced in information retrieval and critical review of evidence-based medicine [8,11,12].…”

Section: Introductionmentioning

confidence: 99%

A Retrospective Review of COVID-19 Medicines Information Queries in a Quaternary Hospital with Unique COVID-19 Border Controls

Misko,

Rawlins

2023

Pharmacoepidemiology

View full text Add to dashboard Cite

Background: Medicines information (MI) is a specialist area of pharmacy that provides evidence-based answers to often complex medication queries, utilising resources such as textbooks and databases. With the advent of the COVID-19 pandemic, there was a need to change the way COVID-19-related queries were answered due to the rapid evolution of information on vaccination, treatment and prevention. Methods: Medicines information queries were retrospectively reviewed utilising the centre’s medicines information database from January 2020 through December 2022 using the COVID-19 keyword to retrieve relevant queries. Information was collected on the enquirer’s role, query category, time taken to complete the query, relevant keywords and references accessed. Keywords and references were analysed further to determine the types of queries asked and which references were helpful. Results: The centre received 214 COVID-19-related queries, predominantly in 2022. Most queries were from pharmacy staff (95.8%) and related to vaccination (n = 95, 44.4%) or treatment (n = 87, 40.7%). Government and specialist organisation websites were used most commonly as reference sources (24.6% and 16.5%, respectively) for their currency with COVID-19-specific resources (such as national guidelines, COVID-19 treatment interaction checkers) and textbooks/databases used less commonly. Conclusions: MI pharmacists have demonstrated their ability to obtain reliable COVID-19-related information, utilising and interpreting information from less traditional sources.

show abstract

Drug information center: challenges of the research process to answer enquiries in hospital pharmaceutical practices

Cited by 7 publications

References 15 publications

Safety and quality of AI chatbots for drug-related inquiries: A real-world comparison with licensed pharmacists

Safety and quality of AI chatbots for drug-related inquiries: A real-world comparison with licensed pharmacists

Identifying Training Needs and Active Information Opportunities in Primary Care Through the Analysis of Drug Information Requests

A Retrospective Review of COVID-19 Medicines Information Queries in a Quaternary Hospital with Unique COVID-19 Border Controls

Contact Info

Product

Resources

About