1. Recommendations

Leadership and Advocacy

  • The Royal Pharmaceutical Society supports the responsible and effective use of AI across healthcare settings, including pharmacy services, to improve the safe and effective use of medicines in any setting, and achieve better patient outcomes.
  • The Royal Pharmaceutical Society will engage and collaborate with relevant stakeholders to identify opportunities for AI deployment to enhance pharmacy professional roles and advance the safe and effective use of medicines.
  • The Royal Pharmaceutical Society holds the view that AI technologies will enhance the value that pharmacy professional roles provide to people and is not a technology to replace roles.

Education and Training

  • Pharmacists must familiarise themselves with AI to ensure they have a level of awareness which allows them to contribute to the digital advancement of pharmacy practice
  • Pharmacy professionals should have education and training on the benefits and risks of AI systems in pharmacy practice, and these must be introduced into undergraduate and foundation training programmes (investment required)
  • The pharmacy workforce must be provided with the opportunities to develop the necessary skills and will require protected learning time to train and implement these new technologies
  • Knowledge and skills required by the pharmacy workforce are described within the RPS policy on Digital Capabilities

Development and Collaboration

  • Development of AI tools in pharmacy practice to improve the safe and effective use of medicines must be co-produced with pharmacists, data scientists, developers and patients
  • Investment is essential to ensure digital infrastructure supports the interoperability of systems, to optimise the benefits of AI tools.

Quality, Safety and Governance

  • Information quality, security, standards and governance must be adhered to in the development and deployment of AI technologies in pharmacy practice
  • AI technologies deployed within healthcare, including pharmacy services, for the safe and effective use of medicines must comply with regulatory and legislative requirements as these evolve with the evidence base over time
  • AI technologies, when deployed within pharmacy, must undergo rigorous validation testing and ongoing internal performance monitoring and evaluation to ensure intended benefits are realised and unintended consequences are mitigated
  • Regulation for healthcare, including pharmacy professionals and premises, related to AI, must not stifle innovation but achieve the right balance of quality and safety
  • Pharmacists should exercise caution in sharing patient or any personal information with third party AI tools
  • As the digital transformation of healthcare continues, and AI technologies expand across all sectors of practice, healthcare continuity planning and disaster recovery are vital considerations to support reliable service delivery.

Patient Engagement and Transparency

  • Data sets used to train AI tools must be diverse and representative of all patient populations to minimise bias and improve accuracy of AI predictions across different demographics
  • Patients should have transparency of how AI supports any decision-making and that they understand the limitations of AI driven recommendations (where applicable)
  • Patients must be fully informed when accessing services or receiving pharmaceutical care that involves AI technology.

 

2. Introduction

We have developed this policy to outline the opportunities, challenges and potential applications of artificial intelligence (AI) within pharmacy now and in the future.

The ongoing global digital transformation and significant pace of change in healthcare means that the RPS must optimise the opportunities that these advancing technologies can bring if we are to put pharmacy at the forefront of healthcare.

The content has been developed in consultation with RPS members, Expert Advisory Group members, Board members, multi-professional experts in digital technology and AI, and external stakeholders including the General Pharmaceutical Council (GPhC).

The pharmacy professional visions published in England, Scotland and Wales identified how the introduction of AI tools might enhance patient access to care, improve patient experience, support clinical decision-making and improve the safety and efficiency of the medicines supply chain1 2 3. AI capabilities are rapidly evolving within clinical and operational aspects of pharmacy, and over the last five years, there has been a significant increase in the number of academic research, cultural commentary and opinion pieces published about AI and its potential applications.

AI tools are already in daily use in our mobile phones, televisions and home smart devices. The successful application of AI tools in diagnostic imaging is well described in the literature and by the Royal College of Radiologists 4 However, the excitement surrounding AI is matched equally with concern about potential risks, and pharmacists and their teams need to have an awareness of AI in the round if they are to support the decision-making necessary around deployment of such tools in pharmacy practice.

What is artificial intelligence?

The Oxford English Dictionary defines AI as the capacity of computers or other machines to exhibit or simulate intelligent behaviour. It also describes software used to perform tasks or produce output previously thought to require human intelligence, especially by using machine learning to extrapolate from large collections of data5.

AI can be categorised into narrow AI, which is designed to perform a narrow task such as facial recognition or language translation, and general AI which is largely theoretical (and not yet in use) and can perform any intellectual task that a human can do. AI technologies include machine learning, of which natural language processing is a subset. Machine learning algorithms allow computers to learn from and make predictions based on data. Natural language processing enables machines to understand and respond to human language. Large language models, such as those behind tools such as ChatGPT, are designed to understand, generate, and manipulate human language.

Although AI technologies are already a feature of many of our daily digital interactions, not all such systems actually involve AI. The term “algorithmic misperception” is used where people believe they are interacting with AI but are using simpler, rules-based algorithms. The Alan Turing Institute has published a useful glossary to support understanding of the language surrounding these technological developments6.

AI offers potential benefits in both clinical and operational aspects of pharmacy practice. These are discussed in more detail within this policy.

3. Scope for AI in pharmacy practice

AI technologies have the potential to transform the way that pharmacists and pharmacy teams work. The RPS supports the responsible and effective use of AI across healthcare settings, including pharmacy services, to improve the safe and effective use of medicines in any setting, to achieve better patient outcomes.

Concerns about job replacement are common at the advent of new technologies; however, there are real opportunities for AI to enhance job roles, rather than replace them. The automation of routine tasks could allow all staff to take on a more patient-centred focus and ultimately improve patient care6. While AI will radically alter how work gets done and who does it, the larger impact will be in complementing and augmenting human capabilities, not replacing them7.

A report by The Health Foundation showed that over half of the public (53%) thought AI will distance them from healthcare staff, while nearly two-thirds of NHS staff surveyed (65%) thought AI would make them feel more distant from patients. These results suggest that AI will need to be designed and deployed in ways that protect and enhance the human dimension of care8.

AI and antimicrobial stewardship

Researchers led by Dr Laura Shallcross, at University College London Hospitals NHS Foundation Trust, are developing and testing an antibiotic surveillance system called SamurAI. The system will utilise hospital electronic patient records, combining historical data for patients prescribed antibiotics with the findings from the specialist microbiologists and pharmacists as they review prescriptions.

Under specialist medical supervision, the computer system will learn what works best and when to start, stop or change the use of antibiotics. The team will compare the AI predictions with what the specialists would do.

The aim is for the system to identify and flag patients who require a review and provide support for decision making in antibiotic prescribing. The findings are expected to be published in 2025.

AI offers opportunities to transform the way that pharmacists and their teams work, but we are at an early stage of discovery, design, and deployment. Through the policy drafting process, several potential pharmacy use cases for the improvement of existing pathways and work processes, were identified.

Although not exhaustive, the list includes:

Pharmaceutical Industry and Medical Research

  1. Drug discovery by analysis of vast amounts of data
  2. Automating quality control processes in production
  3. Optimising clinical trial design and participant recruitment
  4. Analysis of large-scale data in post-marketing surveillance and pharmacovigilance
  5. Producing targeted marketing communications9

Education and Training

  1. Producing educational content in formats to suit learner needs
  2. Generation of assessment questions
  3. Analysis of training evaluations
  4. Patient specific information on their medication regimen

AI systems capable of reading and prioritising prescriptions offer significant advantages, particularly in the high-pressure environment of a busy pharmacy. By automatically categorising prescriptions by urgency, these tools ensure that critical or time-sensitive medications are processed without delay, safeguarding patient safety.

- Mohammed Sheikh, Director, Personal Homecare Pharmacy

Hospital Pharmacy

  1. Risk stratification of patients for pharmaceutical care interventions
  2. Decision support tools in electronic prescribing systems
  3. Automation of operational processes surrounding procurement, stock management and distribution10
  4. Streamlining dispensing processes
  5. Support for the production of aseptically prepared products

Primary Care

  1. Risk stratification of patients for pharmaceutical care interventions
  2. Decision support tools in electronic prescribing systems
  3. Automation of processes surrounding repeat prescription generation, records management
  4. Analysis of patient prescribing data to monitor cost-effective prescribing5
  5. Analysis of patient monitoring data to identify opportunities to improve outcomes in long term conditions5.

Community Pharmacy

  1. Enhancing public health services offered through pharmacies by providing patients with tailored ongoing support and signposting e.g. smoking cessation, sexual health.
  2. Decision support tools in electronic prescribing systems and patient pathways e.g. Pharmacy First
  3. Virtual consultations for minor ailments6
  4. Supporting medication adherence with reminders, notifications and educational resources11

Operational Activities

  1. Optimising administrative tasks such as diary and email management
  2. Automating analysis of business data and generation of reports
  3. Analysis of workforce and budgetary data to identify areas of focus
  4. Writing job descriptions.

AI at Sciensus

Sciensus logoAt Sciensus, we receive prescriptions from NHS hospitals through various channels, such as:

  • Paper prescriptions delivered by Royal Mail, which are then scanned into the Sciensus system
  • Emailed prescriptions, which are uploaded into the Sciensus system
  • Prescriptions directly uploaded by NHS hospitals via the Sciensus Connect NHS portal.

In cases where copy prescriptions are received, the original follows via Royal Mail a few days later for archiving purposes.

Once the prescriptions are received and scanned, our AI technology plays a vital role. The AI system 'reads' the scanned copy, matches the prescription to the Sciensus Patient Record, then compares the medication details with previous prescriptions. After this, the prescription is validated by the pharmacy team before dispensing. The introduction of AI has streamlined our processes, eliminating the need for manual transcription and one of the manual approval steps. This efficiency is welcomed by our dispensary staff, allowing them to focus on their core clinical skills.

Proactive data validation is essential for good governance and risk management. Since implementing this AI technology, we’ve observed a significant improvement in accuracy, as evidenced by our internal near-miss reporting.  We have also delivered significant advantages in our ability to prioritise urgent and dose change prescriptions, further enhancing patient safety.

Looking ahead, we are exploring how different AI models can support various aspects of our processes, continually seeking ways to enhance efficiency and patient care.

- Alex Clarke, Commercial Products Operations Manager, Sciensus

4. Considerations for AI in pharmacy practice

Development and Collaboration

Development of AI tools in pharmacy practice to improve the safe and effective use of medicines must be co-produced with pharmacists, data scientists, developers, and patients.

By seeking professional connections with data scientists, teams will be able to identify the problems that require a digital solution and consider whether an AI tool is the appropriate route, delivering improved outcomes and value for money. AI systems should be built following clinical guidelines and peer-reviewed research. If developing a new model, co-producing the prototype and supporting the testing phases with real data is an important factor in ensuring the tool is fit for purpose and can be manually validated.

Depending on the processes involved, the pharmacy and AI development team may be working within a wider multidisciplinary team to ensure unintended impacts of the technology are identified and mitigated and that the application meets real-world needs.

Information quality, security, standards, and governance must be adhered to in the development and deployment of AI technologies in pharmacy practice.

To be able to validate outputs, it is important that we can critically appraise AI tools, with a focus on explainability, to reduce the risk of the “black box” phenomenon where we are unable to ascertain how the tool reached the output from the original data input.

A risk of adopting AI models in work processes is the incorrect or misleading results that AI models generate, called AI hallucinations. One way is to consider how the tool is trained, i.e., on which dataset and ensuring that there is a valid evidence base for the tool under consideration. The House of Commons Science, Innovation and Technology Committee published an interim report12 into the governance of AI which describes 12 challenges, including access to data, bias, liability, and privacy.

Deployment of AI must be used to reduce health inequalities and not to widen them. Mitigation against bias within the data sets used to train AI tools is crucial. Most large language models are trained on vast, unstructured data from the internet; efforts should be made to balance data sets to better represent underrepresented populations. The Professional Records Standards Body (PRSB), in its position statement on AI and Health Information Standards13, emphasises the importance of rigorous health and social care information standards as well as upholds the need for standards that respect confidentiality and consent, promote transparency, and provide clarity on the accountability for AI outputs and clinical decisions based on these outputs.

Regulatory risks

Generative AI tools may inadvertently collect and store sensitive data without legal authorisation. This poses a risk, as such data could be used to train future AI models, thereby introducing unprecedented challenges to data security and privacy, including intellectual property. Privacy and data governance should ensure government and best practice data security principles are not breached by the adoption of AI. Utilising free off-the-shelf AI tools can expose organisations to regulatory risks related to data collection and usage that may not apply to proprietary tools, where more stringent data policies are enforced through subscription or licensing agreements. 

Third party AI suppliers should be investigated and monitored to ensure standards, ethics and governance align to your own. Users should not share personal or sensitive data into any AI tool unless there is consent from patients (whose data is being shared with the third party) and the necessary corporate agreements are in place.

In remote and rural areas, our challenge is finding more time from finite professionals and keeping care close to people.

It’s necessary for us to practice with great breadth and versatility, whilst keeping in mind the challenges in the long term with increasing volumes of work, especially complex work. Prioritising our professional, high skill-high value time towards person-centred care to maintain or reach better outcomes for people, isn’t within our reach without embracing new ways of working with digital tools.

We’re actively working on our digital foundations here to support us to embed digital competence and digital tools to support us with improving the health of our communities.

- Anthony McDavitt, Director Of Pharmacy, NHS Shetland, Interim Depute Director Shetland Community Health And Social Care Partnership

Education and Training

Pharmacists must familiarise themselves with AI to ensure they have a level of awareness which allows them to contribute to the digital advancement of pharmacy practice.

The pharmacy workforce will require education and training to optimise the impact of digital technologies as they continue to transform the way in which we work. Education on AI is required within initial education and training, evolving to relevant levels of digital skills as the workforce proceeds through their career and post-registration curricula. Training content will vary depending on the interaction with AI systems and must be tailored to suit an individual’s learning needs. Digital skills training can be a burden if the training is not relevant to the individual’s scope of practice.

The NHS Digital Academy has published an Artificial Intelligence (AI) and Digital Healthcare Technologies Capability framework14 to ensure health and care professionals can work in a digitally enhanced environment. To address the skills barrier to adopting AI in the workforce, the Department for Science, Innovation and Technology collaborated with the Bridge AI programme to publish an AI Skills for Business Competency Framework15.

The pharmacy workforce must be provided with the opportunities to develop the necessary skills and will require protected learning time to train, in an interdisciplinary environment, and implement these new technologies. People must understand how to work alongside new technology and have the competencies to use it effectively and safely16. The policy statements and requirements for the Digital Capabilities of the pharmacy workforce are covered in the RPS Policy published here.

AI in undergraduate education

The integration of broader digital information governance concepts is essential, particularly regarding the provision of patient data to AI-enabled systems and the subsequent clinical insights they generate.

This theoretical groundwork can be reinforced through targeted activities and assessments where students analyse AI-generated materials relevant to various clinical scenarios, providing direct experience in identifying advantages and constraints, in addition to reflection on the suitability of use of such platforms, allowing students to gain their own professional outlook which is cognisant of various professional and regulatory inputs.

Student pharmacists should also receive guidance on maximising the broader educational benefits these technologies offer, including their application as study companions, organisational assistants, and analytical and critical friends.

Training should incorporate instruction on leveraging common consumer-focused AI platforms to enhance independent learning beyond traditional classroom settings. For instance, students can engage AI systems to simulate patient interactions for developing diagnostic skills, condition management expertise, and counselling proficiency.

Recent technological advances, particularly in voice interaction and multimodal functionality, enable engaging and authentic practice environments that, when appropriately implemented, allow students to cultivate essential clinical competencies outside formal teaching sessions, allowing limited physical simulation resources within schools of pharmacy to be more effectively and efficiently used, without detriment to educational outcomes.

For educators, AI technologies present opportunities to streamline various aspects of teaching delivery, enabling greater focus on meaningful student engagement. These applications span the efficient creation of both formative and summative assessments, implementation of streamlined feedback processes, and the development of advanced digital learning tools that expand student support capabilities beyond traditional resource constraints.

However, educators, like their students, must ensure these approaches prioritise accuracy and align with established ethical and professional standards in pharmacy education.

- Dr Dan Corbett, Senior Lecturer (Digital Education) | School of Pharmacy, Queens University Belfast
- Dr Suzanne Cutler, Senior Lecturer in Pharmacy Education, University of Liverpool

Governance

The Chartered Institute of Ergonomics and Human Factors (CIEHF) Human Factors and Ergonomics of Healthcare AI white paper17 stresses the need for AI systems to be developed with the end-user in mind, ensuring they are intuitive and enhance user experience.

Regulation for healthcare, including pharmacy professionals and premises, related to AI, must not stifle innovation but achieve the right balance of quality and safety.

In the UK, many AI products designed for use in healthcare are regulated as medical devices by the Medicines & Healthcare Products Regulatory Authority (MHRA). Rapid advancements in AI and other technologies can outpace the regulatory framework, making it challenging for the MHRA. They have worked with the FDA and Health Canada to develop 10 guiding principles for Good Machine Learning Practice18.

The MHRA has developed a comprehensive framework for regulating software and AI as medical devices. This framework ensures the product is fit for purpose, safe and effective. The software must be transparent and explainable and undergoes rigorous testing to ensure compliance, with post-marketing surveillance in place19. NHS England Transformation Directorate's AI Lab has also commissioned the AI and Digital Regulations Service to support adopters and developers in navigating the regulatory and access pathway20.

The General Pharmaceutical Council (GPhC), regulator for pharmacy professionals and pharmacy premises, will incorporate advancements in artificial intelligence (AI) deployment into future updates of professional standards. Pharmacists remain accountable for professional decisions in the pharmacy, whether they are made by professionals alone or supported by AI-powered decision support tools. Ensuring any pharmacy workflows supported by AI have a human “in the loop” will contribute to the controls in place to mitigate some of the risks described.

Adoption of AI should enhance trust between healthcare professionals and patients. Patients should have transparency of how AI supports any decision-making and that they understand the limitations of AI driven recommendations (where applicable).  Where AI tools are deployed in pharmacy practice, pharmacists should provide opportunities for patients to ask questions to fully understand how their data may be used within AI systems.

As practice evolves to integrate AI tools across healthcare, the accountability framework, including regulatory frameworks and professional standards (encompassing pharmacy professionals and premises), must adapt swiftly. This ensures that regulation fosters innovation while maintaining the necessary controls to support the high-quality, safe, and effective use of medicines and the provision of pharmacy services in the future. AI, when deployed within pharmacy, must undergo rigorous validation testing and ongoing internal performance monitoring and evaluation to ensure intended benefits are realised and unintended consequences are mitigated.

The Royal Pharmaceutical Society has cautiously adopted Machine Learning and AI into internal processes. Machine learning and AI has been used by our Data Science department to improve our own internal systems and processes.

We have adopted Natural Language Processing (NLP) for improved technical capabilities, such as vectorised search functionality, sentiment analysis, entity recognition, topic modelling and text classification. This has allowed us to extract meaningful insights from large volumes of text data, whilst optimising our internal workflows.

The Royal Pharmaceutical Society does not use Large Language Models (LLMs) to make clinical decisions/recommendations or replace the intellectual thinking of our expert employees. AI is only used for tooling to improve visibility and interpretability of data, enabling our team to make enhanced decisions.

- Ben McCarthy, Head of Data Science, RPS

Environmental impact

The environmental impact of AI must not be overlooked. This relates to the significant energy consumption required for training and running large AI models, which contributes to carbon emissions e.g. electricity power and water for cooling the servers.

Efforts are being made to develop more energy-efficient algorithms and use renewable energy sources to mitigate this impact21.

5. Conclusion

The Royal Pharmaceutical Society aims to position pharmacy at the forefront of global digital transformation.

With AI tools already integrated into everyday devices and some clinical practices, we must emphasise the importance of awareness and informed decision-making among pharmacists to navigate the benefits and risks of AI deployment in pharmacy practice.

6. References

  1. The Kings Fund + Royal Pharmaceutical Society. A Vision for Pharmacy professional Practice in England. https://eu-health-news.info/england/vision-for-pharmacy-practice-in-england%3C/a%3E [Accessed 20 August 2024]
  2.   Royal Pharmaceutical Society. Pharmacy: Delivering a Healthier Wales. https://eu-health-news.info/wales/pharmacy-delivering-a-healthier-wales%3C/a%3E [Accessed 20 August 2024]
  3.   Royal Pharmaceutical Society + National Pharmacy Technician Group Scotland. Pharmacy 2030: A professional vision. https://eu-health-news.info/pharmacy2030%3C/a%3E [Accessed 20 August 2024]
  4.   Royal College of Radiologists. AI Registry. AI Registry Listing | The Royal College of Radiologists (rcr.ac.uk) [Accessed 1 October 2024]
  5.   Oxford English Dictionary. (2023) Oxford: Oxford University. [Online] https://www.oed.com/search/dictionary/?scope=Entries&q=artificial+intelligence. [Accessed 20 Aug 2024]
  6.   The Alan Turing Institute. Defining data science and AI. https://www.turing.ac.uk/news/data-science-and-ai-glossary [Accessed 1st October 2024]
  7.   Wilson HJ & Daugherty PR. Collaborative Intelligence: Humans and AI Are Joining Forces. Harvard Business Review. July–August 2018 issue (pp.114–123) [Online] [Accessed 20 August 2024]
  8.   The Health Foundation. AI in health care: what do the public and NHS staff think? 31 July 2024 [Online] https://www.health.org.uk/publications/long-reads/ai-in-health-care-what-do-the-public-and-nhs-staff-think
  9.   Goundrey-Smith. S. Artificial Intelligence in Pharmacy and the Pharmaceutical Industry – Opportunities and Ethics. PM Healthcare Journal. Issue 08. Spring 2024. Pp 11-18.
  10.   González-Pérez, y. et al. Introducing artificial intelligence to hospital pharmacy departments. Farmacia Hospitalaria 48 (2024) TS35–TS44 https://doi.org/10.1016/j.farma.2024.02.007
  11.   Crilly P. Opportunities and threats for community pharmacy in the era of enhanced technology and artificial intelligence.  International Journal of Pharmacy Practice, 2023, 31, 447–448 https://doi.org/10.1093/ijpp/riad065
  12.   UK Parliament Science Innovation and Technology Committee. The Governance of artificial intelligence:interim report - Report Summary. HC 1769. 31 August 2023. [online] https://publications.parliament.uk/pa/cm5803/cmselect/cmsctech/1769/summary.html  [Accessed 21 August 2024}
  13.   The Professional Record Standards Body. Position Statement on AI and Health Information Standards. 28 Mar 2024. [Online] https://theprsb.org/ [Accessed 21 August 2024]
  14.   Health Education England, & The University of Manchester. (2023). Artificial Intelligence (AI) and Digital Healthcare Technologies Capability Framework.[Online] Framework structure | Digital Transformation.[Accessed 24th October 2024]
  15.   The Alan Turing Institute, Innovate UK, Department for Science, Innovation and Technology (DSIT), Digital Catapult, Science and Technology Facilities Council, British Standards Institution, & Alliance for Data Science Professionals. (2024). AI Skills for Business Competency Framework (2.0.0). Zenodo. https://doi.org/10.5281/zenodo.11092677
  16.   Expert opinion. Dr Kate Preston. Co-Chair, Chartered Institute of Ergonomics and Human Factors Digital Health & AI Special Interest Group.
  17.   Human Factors and Ergonomics in Healthcare AI. CIEHF. 2021. [online] www.ergonomics.org.uk [Accessed 6 July 2024]
  18.   MHRA. Good Machine Learning Practice for Medical Device Development: Guiding principles. 27 Oct 2001 [Online] Good Machine Learning Practice for Medical Device Development: Guiding Principles - GOV.UK (www.gov.uk) [Accessed 26 August 2024]
  19.   MHRA. Software and artificial intelligence (AI) as a medical device. 13 June 2024. [online] https://www.gov.uk/government/publications/software-and-artificial-intelligence-ai-as-a-medical-device/software-and-artificial-intelligence-ai-as-a-medical-device [Accessed 26 August 2024]
  20.   NHS England. The AI and digital regulations service. https://transform.england.nhs.uk/ai-lab/ai-lab-programmes/regulating-the-ai-ecosystem/the-ai-and-digital-regulations-service/ [Accessed 1 October 2024]
  21.   Strubell E, Ganesh A, McCallum A. Energy and policy considerations for modern deep learning research. In Proceedings of the AAAI conference on artificial intelligence 2020 Apr 3 (Vol. 34, No. 09, pp. 13693-13696).