THE CLINICAL PHARMACEUTICAL SERVICE IT TEAM: ENHANCING MEDICATION WORKFLOWS AND PATIENT SAFETY IN EPIC
European Statement
Patient Safety and Quality Assurance
Author(s)
Christina Theil Schnor and Saranya Loganathan.
Why was it done?
In 2018, hospitals in Region Zealand (RZ), Denmark, transitioned to the electronic health record (EHR) system, EPIC. Following this, hospital pharmacists faced repeated medication order challenges causing adverse events such as inappropriate medication orders, dispensing and administration errors, and insufficient workflow coordination. These issues resulted in complex, time-consuming workflows impacting quality and patient safety. Additionally, collaboration between corporate IT and clinical staff was challenged by a lack of understanding of practical issues. To address this, pharmacists of RZ established the Clinical Pharmaceutical Service IT Team (CPS IT Team) to build specialized knowledge of the EHR medication module, aiming to assure quality, optimize workflows, strengthen interdisciplinary coordination, and support safer and more efficient clinical use.
What was done?
CPS IT Team standardized workflows, enhanced coordination of medication order tasks, and created a forum to effectively utilize professional knowledge and networks across areas.
How was it done?
To address diverse Clinical Pharmacy challenges, CPS IT Team became the bridge between internal organization (RZ Hospital Pharmacy and corporate IT) and external partners (EPIC and The Capital Region of Denmark (CRD)). For this reason, CPS IT Team was established with one team manager and two units: Internal and External unit. CPS IT Team continuously adapts to evolving Clinical Pharmacy needs.
What has been achieved?
The establishment of CPS IT Team has driven significant internal optimization and standardized workflows. Acting as a coordinating unit, it optimizes medication processes from ordering to dispensing and administration. Dialogue with IT has been strengthened, enabling more efficient, targeted communication across professional groups.
Collaboration with EPIC and CRD has enhanced quality assurance and optimized workflows. CPS IT Team efforts have helped prevent medication-related adverse events, improve workflows, and optimize medication processes. Interdisciplinary collaboration and professional consultation networks between regional clinics, hospital pharmacies, IT, and EPIC have been notably strengthened. These efforts have increased patient safety and fostered a safer, more coherent workflow in EPIC.
What next?
Fusion of RZ and CRD into Region Eastern Denmark will change CPS IT Team’s working conditions, opening new opportunities such as an expanded collegial network and broader range of tasks and needs. Systematic data use will support Hospital Pharmacy’s work, improving efficiency and quality in daily operations.
CYBER RISK ANTICIPATION: HOW TO GET YOUR HOSPITAL PHARMACY PREPARED IN (ALMOST) 2 HOURS?
European Statement
Introductory Statements and Governance
Author(s)
C. Delage1, P. Troude2, U. Gouot3, V. Bloch1, B. Isabey4, A. Desmons5.
1 Pharmacy Department, Lariboisière-Fernand Widal Hospital, GHU APHP.Nord, Assistance Publique – Hôpitaux de Paris, 2 rue Ambroise Paré, 75010 PARIS
2 Public Health Department, Lariboisière-Fernand Widal Hospital, GHU APHP.Nord, Assistance Publique – Hôpitaux de Paris, 200 rue du faubourg Saint-Denis, 75010 PARIS
3 Information Systems Department, Lariboisière-Fernand Widal Hospital, GHU APHP.Nord, Assistance Publique – Hôpitaux de Paris, 2 rue Ambroise Paré, 75010 PARIS
4 Hospital Executive Management, Lariboisière-Fernand Widal Hospital, GHU APHP.Nord, Assistance Publique – Hôpitaux de Paris, 2 rue Ambroise Paré, 75010 PARIS
5 General Affairs Department, Lariboisière-Fernand Widal Hospital, GHU APHP.Nord, Assistance Publique – Hôpitaux de Paris, 2 rue Ambroise Paré, 75010 PARIS
Why was it done?
With the 2024 Olympic Games approaching, French healthcare institutions faced an elevated cyberattack risk. Preparing healthcare services for such events is essential but often time-consuming, leading sometime to postpone or abandon this work. A rapid, pragmatic methodology was therefore needed to efficiently assess vulnerabilities and define concrete preparedness measures without overburdening teams.
What was done?
A fast-track business continuity plan (BCP) approach was developed by the management department of our university hospital and implemented in the pharmacy department to rapidly assess preparedness for cyber risks and IT disruptions. The structured process mapping allowed each sector to produce an operational continuity plan in only a few hours, evaluating process vulnerability, defining measures to maintain operations during a cyberattack, and identifying needs (new IT equipment, human resources, backup materials).
How was it done?
Dedicated meeting (1-4 hours) were held for each pharmacy sector (medicines, medical devices, sterilisation, radiopharmacy), involving the cyber risk pharmacist referent and sector head. The plan took the form of a table describing all sector processes (eg., procurement, storage, dispensing), their normal functioning and IT requirements (applications, devices, networked medical equipment…). Subsequent columns described degraded functioning and required resources (staff, equipment, premises) at three time horizons: 3 hours, 3 days, and 3 weeks after the start of a cyberattack.
What has been achieved?
Nine BCPs identified 64 processes. In just a few hours, the method provided a clear picture of which sectors/processes were resilient and which were unprepared. It revealed vulnerabilities such as reliance on IT systems and automation, lack of printed procedures, and absence of backup equipment. Simple measures were identified to improve preparedness: printing key documents (eg. hospital formulary), identifying critical data for regular backup, training staff, etc. This rapid diagnostic approach raised awareness among hospital management of resources needed to sustain pharmacy operations during cyber disruption.
What next?
Based on their BCP, each sector took measures to get prepared to IT shutdown. Using professionals’ deep process knowledge to design practical solution to ensure continuity of clinical activity makes preparedness for IT disruption more accessible. Given its simplicity, the methodology was generalised by hospital management to all departments.
CHATGPT-5 AS A POTENTIAL ALLY IN IDENTIFYING DRUG–DRUG INTERACTIONS?
European Statement
Patient Safety and Quality Assurance
Author(s)
Presenting author : H Decouvelaere
Co-author : C Lambert de Cursay
Why was it done?
Drug–drug interactions (DDIs) represent a major issue in clinical pharmacology, as they can lead to serious, sometimes fatal, adverse effects. The emergence of artificial intelligence models such as ChatGPT raises questions about their reliability in identifying DDIs. The literature reports that ChatGPT may generate nonexistent information (“hallucinations”) or provide inaccurate or incomplete data. However, evidence regarding its use in detecting DDIs remains limited.
What was done?
To evaluate the reliability of ChatGPT in detecting DDIs, particularly for recently marketed drugs or those under compassionate use authorization (CUA), for which conventional databases are sometimes incomplete.
How was it done?
Thirteen older drugs (marketed before 2010) and seven drugs marketed since June 2025 or under CUA were identified. All pairwise drug combinations were tested. A standardized script was used to query ChatGPT consistently, minimizing bias related to question phrasing. Results were checked using reliable sources (product characteristics summaries, scientific literature, DDI-Predictor, etc.). Searches were conducted between July and September 2025 using ChatGPT-5.
What has been achieved?
A total of 210 drug pairs were analyzed. ChatGPT’s responses were consistent with the literature for 72% of pairs (n=152). For 9% (n=18), the information was incomplete. Partial discrepancies were observed for 1% (n=2) and total discrepancies for 18% (n=38). Among these, 68% (n=26) corresponded to ChatGPT hallucinations and 32% (n=12) to undetected DDIs. Agreement with the literature was 65% (n=59) for older drug pairs, 95% (n=20) for pairs of two recent or CUA drugs, and 74% (n=73) for mixed pairs (old and recent drugs).
What next?
ChatGPT-5 can serve as a helpful tool for identifying DDIs, providing correct analysis in nearly 75% of cases. However, its performance remains limited due to the significant risk of hallucination or omission. ChatGPT’s responses were less reliable for older drugs, likely because of the vast and sometimes outdated documentation available. Conversely, newer or CUA drugs—although less documented—benefited from more recent and homogeneous sources, improving response quality. Therefore, ChatGPT-5 cannot replace human expertise or official databases. It should always be used as a complementary tool, with its outputs verified against trusted sources.
ARTIFICIAL INTELLIGENCE AND AUTOMATION: TRANSFORMING HOW WE MANAGE DRUG RECALLS AND MEDICINE SUPPLY NOTIFICATIONS
European Statement
Patient Safety and Quality Assurance
Author(s)
Anna Lydon & Jonathan Day
Why was it done?
National drug recalls and Government issued Medicines Supply Notifications (MSNs) must be responded to promptly to minimise risk to patients and ensure stock safety. Within our Trust, comprising four hospital sites at different locations, an incident occurred where a recall email received over a bank holiday weekend which went unnoticed. The existing system relied solely on staff checking their emails, which posed a risk of delayed action—particularly during weekends, leave periods, and across multiple sites. This highlighted the need for a more robust and transparent process to ensure that all recalls and MSNs are received, actioned, and tracked in real time.
What was done?
A digital process was developed using Microsoft Power Automate and Artificial Intelligence (AI) to automate the handling of drug recalls and MSN emails. The flow triggers when a drug recall or MSN email is received. It extracts key information using AI, and automatically distributes the information across the relevant pharmacy and clinical teams across all Trust sites.
How was it done?
When a drug recall or MSN email is received, the attached PDF is analysed with an AI model trained with specific prompts to extract key fields including the drug name, MSN number, date, impact level and required actions. The extracted data automatically populates the MSN or drug recall Excel log – replacing what was a manual data entry process. Power Automate then initiates an approval process and posts a summarised Teams notification into a Teams channel for all relevant members. Each member receives the alert and one person from each site must acknowledge receipt. Following acknowledgement from a member of each site, Power Automate posts a confirmation in the Teams channel, enhancing transparency and providing assurance that the recall has been actioned. AI determines the appropriate clinical speciality for the drug and automatically directs a summary email to the corresponding pharmacists and clinicians working in that area.
What has been achieved?
The process ensures consistent and timely handling of recalls and MSNs, reducing reliance on individual inbox monitoring. It has improved visibility across all sites, eliminated missed notifications, and significantly reduced manual data entry time. Staff feedback has been positive, with greater confidence that all recalls and MSNs are captured and actioned promptly.
What next?
Future plans include integrating automatic escalation plans for unacknowledged alerts. The same model could also be expanded to other time-critical communications, such as National Patient Safety Alerts to further strengthen medicines governance across the organisation.
ARTIFICIAL INTELLIGENCE POWERED DOCUMENT MANAGEMENT FOR HOMECARE PRESCRIPTIONS
European Statement
Patient Safety and Quality Assurance
Author(s)
Anna Lydon & Jonathan Day
Why was it done?
Homecare prescriptions are physically sent to external homecare companies for dispensing, so the hospital must retain a copy for governance, audit and continuity of care. Previously, prescriptions were scanned in and manually moved to local folders with variable filenames, introducing risk of error, no audit trail and duplicated effort (pharmacy and clinical teams both scanning). This was a slow, manual process, with a typical delay of around 2-3 weeks before prescriptions were uploaded due to the high volume. We therefore needed a safer, faster and auditable process to retain copies of homecare prescriptions in the official patient record.
What was done?
We developed a Robotic Process Automation (RPA) which involved ustilising an Artificial Intelligence (AI) enabled workflow to capture homecare prescriptions, extract key fields, automatically file using standardised filenames, validate patient identifiers, and automatically upload documents into the patient record for all healthcare professionals (HCP) to view in real time.
How was it done?
We used Microsoft Power Automate and Azure AI to build an AI-driven RPA process to read scanned homecare prescriptions, extract key fields, standardise filenames, validate patient identifiers, and automatically upload documents into the official patient record. Within the automatic process, there are built-in validation checks to verify AI-extracted fields. If the AI confidence score is low or a check fails, the workflow prompts for human confirmation of the patient’s hospital number and then resumes processing.
What has been achieved?
To date, the solution has processed over 5,000 homecare prescriptions with 98% field-extraction accuracy. Average handling time has reduced from 3 minutes per prescription to 10 seconds per 50 prescriptions, equating to around 5 hours saved per 100 prescriptions. This novel process has released 7.5 hours of pharmacy staff time per week. Prescriptions are now available in the patient record in real time, improving information availability at the point of care and enabling staff to focus on higher value tasks. This has strengthened data quality and governance, and provides an audit trail whilst reducing duplication of work between pharmacy and multiple clinical teams.
What next?
We plan to extend the workflow so that all AI-extracted fields are written back to the Homecare Patient SQL database, creating a single, queryable source of truth and enabling automated checks, dashboards and reporting. We also plan to roll the solution out for other prescription types (outpatient and discharge), replacing off-site paper storage with searchable patient records, reducing storage costs, improving retrieval times, and strengthening governance.
INTEGRATING ARTIFICIAL INTELLIGENCE: STREAMLINING MEDICATION HISTORY DOCUMENTATION WITH CHATGPT AT ODENSE UNIVERSITY HOSPITAL
Pdf
European Statement
Clinical Pharmacy Services
Author(s)
Fjóla Høg Nielsen, Gine Cecilie Stobberup
Why was it done?
The initiative was driven by the need to support a growing number of patients who require medication history. To ensure that pharmacy technicians could complete as many medication histories as possible, have enough time to thoroughly investigate potential issues, and maintain the desired quality of medication history, we initiated the use of ChatGPT to handle the journal note-writing aspect. This initiative was launched in January 2024 and implemented in August 2024.
What was done?
At Odense University Hospital, pharmacists and pharmacy technicians conduct daily reviews of hospitalized patients’ regular medication based on data from the Danish Medicines Agency’s system “The Shared Medication Record”, prescription deliveries, and patient statements. The patient’s usual medications are reviewed to determine what they are taking at home and to identify potential issues, such as compliance problems. A note is written in the medical journal for the attending physician, providing an overview of the patient’s regular medications and any concerns. To improve efficiency and consistency in this process, we implemented the use of ChatGPT to write these notes after the pharmacy technician has completed the medication history, ensuring standardized documentation, increased safety, and for saving time.
How was it done?
ChatGPT was programmed to document the medication history following the standard note format previously used. One of the key challenges was ensuring that ChatGPT could meet the specific documentation needs and minimizing errors in the generated notes. After the initial programming, pharmacy technicians were asked to use ChatGPT and keep track of how many medication histories were written with its assistance, as well as to identify any recurring errors. Based on their feedback, ChatGPT was adjusted to reduce the occurrence of similar errors in future notes.
What has been achieved?
Pharmacy technicians have reported that the time required to complete a medication history has decreased, particularly for patients with long medication lists. The system also ensures that the notes are always written in a consistent manner, reducing the likelihood of missing important information.
What next?
Moving forward, we will continue refining the system to further eliminate errors and improve accuracy. This initiative showcases the successful integration of advanced technology into healthcare, with potential applications across other healthcare settings.
COMPUTER DEVELOPMENT OF THE MAGISTRAL FORMULATION MODULE FOR UNIFICATION OF THE PRODUCTION AREA
Pdf
European Statement
Production and Compounding
Author(s)
A. Henares-López, V. Collados-Arroyo, C. Mayo-López, R. Fernández-Caballero, L. Carrasco-Piernavieja
Why was it done?
A new magistral formulation and mixes management program was developed in the Farmatools® software to optimize the procedures and records of this area.
What was done?
The correct management of compounded formulation is important for the labor of a pharmacy service, ensuring safe preparation and adequate traceability. In our pharmacy service we only had a simple formula registration program that only allowed formulas to be added to the stock but the rest of the work (protocols, batches, preparation staff, validation) was recorded on paper.
How was it done?
Taking advantage of the implementation of the Farmatools® software in our pharmacy service that already had a simple compounding module, it was necessary to carry out a development to add improvements such as registration of raw materials, quarantine states, expiration notices, creation of manufacturing protocols for each preparation, formula programming calendars, improvement of the traceability system (new batch records and other necessary data in the case of biological material), quality controls such as mass uniformity control for capsules, validation section of the formula prepared for the pharmacist, registration of manufacturing staff, improvement of data present on the manufacturing label.
This work was possible thanks to the help of the farmatools® IT staff with whom we held regular meetings.
What has been achieved?
A correct stock of the amount of real raw materials existing in the pharmacy was achieved, complete computerization of the registry of preparations of magistral formulation, including batches, expiration dates, quarantines and necessary quality control records, creation of an electronic recipe book, approval or rejection of elaborate formulas and eliminating any paper records.
What next?
There are still new developments to be implemented, such as the creation of a calendar of scheduled work and a list that allows viewing the preparations pending validation by the pharmacist. Finally, it would be necessary to implement tablets that contain work protocols to avoid the use of paper.
With this program it has been possible to unify the work of the production area in a single program, which results in an improvement in traceability and fluidity in the work.
Design, implementation and impact on patient safety of a mobile application for patients on antiretroviral therapy
Pdf
European Statement
Patient Safety and Quality Assurance
Author(s)
Esther Chamorro de Vega, Carmen Rodriguez Gonzalez, Belen Mayo Canaleja, María Pilar Montero Anton, Sergio Herrera Bermejo, Rosa Romero Jimenez, Vicente Escudero Vilaplana, Roberto Collado Borrell, Ana Herranz Alonso, María Sanjurjo Sáez
Why was it done?
The profile of patients receiving antiretroviral therapy (ART) has changed. Patients are more active, autonomous and have greater information and communication needs.
Mobile health, especially apps, can help to healthcare professionals to optimize pharmacotherapeutic follow-up and provide patients greater autonomy facilitating communication with healthcare professionals and contributing to the humanization of their care.
Through this application we seek to achieve the following objectives:
– Allow patients to participate in their illness and/or treatment, promoting their compliance and safety and communication with the healthcare professional.
– Strengthen pharmacotherapeutic monitoring of patients receiving ART, avoiding visits to the emergency room and hospital consultations.
– Generate pioneering evidence on the impact of mobile applications on the pharmacotherapeutic monitoring of patients on ART.
What was done?
FarMCuida® is a mobile application designed for a closer and better pharmacotherapeutic monitorization of patients on antiretroviral therapy (ART). Through this app the patient can actively participate in his or her disease and treatment.
How was it done?
A multidisciplinary group was created to design and develop an app called FarMCuida® for patients on ART in a tertiary hospital. Once the app was designed and developed, we offered the app to patients starting or undergoing ART. Identification and inclusion of candidate patients was performed in the Pharmaceutical Care Consultation of the Pharmacy Service. The pharmacists responsible for the Infectious Diseases area performed real-time pharmacotherapeutic follow-up of the patients.
Records of adverse effect (AE) and Safety-related messages received (Adverse effects, interactions, drug administration) can be analyzed. All data can be exported from FarmCuida®. Informed consent is requested from all the patients included.
What has been achieved?
FarMCuida® was designed as a digital health mobile application that allows our patients to participate in their disease and/or treatment, promoting treatment safety, as well as making available a source of information and bidirectional communication, in real-time, between patients and healthcare professionals.
During a period of 13 months, a total of 152 patients were included in FarMCuida®. Seventy-two percent of the patients were diagnosed with HIV infection while 27.6% of them were included in the HIV pre-exposure prophylaxis programme.
In AE module, 58 adverse events (AEs) were registered. Twenty-two percent of the patients registered at least one AE.
Thirty-two percent of patients used the Messages module to communicate with healthcare professionals.
One hundred and twenty-six telematic consultations were carried out. Two hundred and five messages were exchanged between the patient and the pharmacist.
Forty-three percent of the consultations corresponded to doubts about interactions: 66.7% interactions with other drugs, 28.5% with herbal products and 4.8% with vitamin supplements. Three type D interactions and one type X interaction were identified.
In addition, 11% of the consultations corresponded to doubts about the administration of ART while 5.2% were related to adverse effects experienced by the patient.
At least three visits to the emergency department were avoided, resulting in health care and economic savings for the health system.
What next?
Based on the limited studies on the impact of mobile applications on the pharmacotherapeutic follow-up of patients, the results obtained will allow us to generate evidence in this regard. It is worth highlighting the flexibility of FarMCuida® design, being easily adaptable to different pathologies and hospital centres.
Too good to go – sharing medication near expiry with a digital solution
Pdf
European Statement
Clinical Pharmacy Services
Author(s)
Charlotte Sørensen, Karin Aagot Møller Jørgensen, Anne Grethe Nørgaard Kyndi, Charlotte Lund Madsen, Line Scriver Poulsen, Charlotte Hjorth, Gitte Slyngborg Morbjerg, Janni Østergaard Jensen, Mette Juul-Gregersen, Rune Dalsenni Rask
Why was it done?
A sustainability project in 2020 highlighted manual and time-consuming workflows when checking expiry dates in hospital medicine rooms. For example, manual reading, interpretation and notation of expiry dates in paper forms every half year. Therefore, a smartphone application was developed and implemented in a large university hospital in 2022. The application was a huge success and colleagues from other Clinical Pharmacy departments in other hospitals within the same region wanted to be part of the success of the initiative.
What was done?
A smartphone application for managing expiry dates in medicine rooms was applied in one hospital in 2022 and in four other hospitals in 2023. The data-matrix of medicine packages is scanned with a smartphone camera and the application gives a sorted overview of medication near expiry. In the application you can register a medicine package as either used, discarded or released. By releasing medicine packages, the medicine is made available to colleagues in other medicine rooms.
How was it done?
The smartphone application was implemented in Clinical Pharmacy departments in four other hospitals, one by one from February 2023. After implementation the application is used in 373 medicine rooms.
What has been achieved?
Clinical Pharmacy staff are happy with the application. It takes significantly less time to check and scan expiry dates with the digital solution.
From February to September 2023 (five hospitals)
• 47491 packages were scanned
• 5251 packages were released (11%)
• 2664 packages were moved between medicine rooms to avoid medicine wastage (51%; 147000 €). Most activity (78%) was seen at the hospital that had been using the application since 2022. Most packages were moved between medicine rooms at the same hospital; however, in 215 cases (8%), medicines were exchanged between hospitals as well (20.000 €)
• 10918 packages were discarded due to expiry (743168 €; 0.5% of bought medicine).
What next?
The full potential of the digital solution remains to be achieved.
Pharmaconomists have to release medicine in the application a few months earlier than they do now. To go shopping in the application before ordering from the pharmacy, will probably also result in discarding less medicine. Sharing medication across hospitals to a greater extent will also be relevant to consider.
Design, build and implementation of electronic prescribing and medicines administration in neonatal units across four UK hospital trusts
Pdf
European Statement
Introductory Statements and Governance
Author(s)
Holly Jones, Nicola Vosser
Why was it done?
Neonatal units were out of scope during the initial EPMA go-lives due to the complexities in prescribing and the specialist resource required for implementation. This project therefore brought neonatal units in line with all other areas, transitioning from paper drug charts to EPMA. In addition, it supported an ICS wide goal to align practice between trusts, standardising prescribing and facilitating effective use of digital systems in healthcare.
What was done?
A neonatal formulary was developed on an existing Electronic Prescribing and Medicines Administration (EPMA) system shared by four hospital trusts that make up a regional Integrated Care System (ICS). This involved interdisciplinary collaboration and shared decision making to align practice and prescribing guidelines. Complex protocols included continuous variable rate drug infusions, intravenous fluids and parenteral nutrition.
How was it done?
Key guidelines were shared with a collaborative approach to identify and harmonise differences in local practice, including syringe volumes and dose calculation ‘factors’. The EPMA team (comprising specialist pharmacists and medicines management technicians) completed a thorough gap analysis of the existing adult and paediatric drug catalogue and managed all new build work for neonates. Clinical decision making and validation were multidisciplinary tasks with input from specialist clinicians, pharmacists and nurses. Training, go-live planning and implementation were led by the EPMA team.
What has been achieved?
A bespoke neonatal medication build has been successfully implemented, supporting safe and efficient prescribing in neonates across the ICS. The development provides the ability to prescribe medications and document administration clearly and safely, including adjustment of infusion rates. Clinical pharmacists have full visibility of this information and are able to accurately complete pharmacy reviews, interventions and verification. In addition, alignment of practice and standardisation of care has been reached across the ICS, with benefits to both staff and patients.
What next?
The neonatal build is updated and optimised based on feedback from users to maintain the usability and safety of the system. Development of electronic fluid balance charts for neonates, including drug infusion volumes, is also in progress. Details of the neonatal project are being shared with other UK hospital sites using the same EPMA system.