Artificial intelligence in pharmacy: The challenges and opportunities

The internet is exploding with opinions from professionals who think it will replace them

We have a tool that is revolutionising healthcare. It can be a great tool to maximise skillset and create workflow efficiencies if used wisely. But rely on it as the ultimate solution, and it will probably cost you your license.

Let me introduce you to the new helper of mankind.

Me: “Hey, ChatGPT.”

ChatGPT: “Hi, how can I help you?”

Me: “Do you think you will replace pharmacists?”

ChatGPT: “As an AI language model, I do not have the ability to replace pharmacists. While I have access to a vast amount of information about medications and their uses, pharmacists perform a range of important duties that cannot be replicated by an AI, such as providing personalised care and advice to patients, ensuring medication safety and effectiveness, and managing drug interactions and side effects. Additionally, pharmacists play a critical role in healthcare teams, working closely with doctors, nurses and other healthcare professionals to ensure that patients receive the best possible care. While technology can certainly support and enhance the work of pharmacists, it cannot replace the human touch and expertise that pharmacists bring to their practice.”

Me: “Well, that is reassuring.”

ChatGPT: “I’m glad to hear that. The role of pharmacists is evolving as healthcare technology advances, but their unique skills and knowledge will continue to be an essential component of patient care.”

Undoubtedly, you will have seen headlines about OpenAI, with the promise that it will revolutionise (or maybe take over the world) and replace some jobs. But will it replace pharmacists?

What role does it have in healthcare, if any? Can it help pharmacists? And what happens if patients start using it for healthcare purposes?

ChatGPT is a large artificial intelligence (AI) language model designed to generate text from user input. It can answer general questions by crawling the required information from the web. The ChatGPT technology represents a breakthrough in AI, with implications for its use in a wide range of applications.

ChatGPT’s capacity to produce insights from patient data to assist in more focused, personalised and successful therapies is one of its most significant advantages. It can analyse large amounts of data to uncover patterns and correlations that human providers may not notice immediately. This application of ChatGPT technology may help to improve patient outcomes, support healthcare equity efforts through increased access to care, lower the risk of adverse events or complications and lower healthcare costs by ensuring patients receive the most appropriate treatments for their bodies and conditions.

The internet is exploding with opinions from professionals who think it will replace them. I do not believe that AI will replace pharmacists, contrary to the opinions expressed on many social media platforms. It will, along with robotics, replace dispensing, inventory management, and all the mundane tasks you could think of that can be (and should be) automated. But here is what it cannot do:

‘A pharmacist’s knowledge may be replaceable, but a pharmacist’s expertise is irreplaceable and unparalleled’

It can answer queries, do the maths and provide evidence (albeit not always accurate). But it cannot provide empathy; it cannot put information in the context of the patient’s lifestyle and values, and provide an in-depth understanding of the patient’s concerns that may be affecting medication adherence. It cannot form an overall judgment or opinion. It is simply what it is: artificial.

Pharmacists need to recognise and understand this because the most common use you will encounter is patients using it to answer medication queries. Dr ChatGPT will take over Dr Google.

They will then contact you to verify the authenticity and accuracy of the information provided. Your job will be to discern the accuracy of the information and fill in the patient’s knowledge gaps.

Negating its existence and asking patients not to use it is futile; they will not listen because, in reality, every person is using it, and to some extent, it is accurate.

Fighting AI may make you appear less trustworthy; after all, everyone else sees its value and ability to transform work processes and answer almost any question under the sun. Your position is to help them filter the information and discern what is accurate and reliable from what is not.

Legal implications: Can you use it as a pharmacist? Is it against regulation?

There are currently no well-defined regulations to address the legal and ethical issues that may arise due to the use of AI in a health setting. Like any other information resource, AI platforms can answer your questions and provide evidence; however, they are dishonest in the accuracy of the evidence they produce. So be warned. It is not peer-reviewed for accuracy like NZF or any other resource you may be using. In other words, if patient harm occurs as a result of a false answer you provide or if questioned in court, you will not stand a chance by defending your position with ChatGPT as a resource for your answer.

Despite its potential in surfing the internet and regurgitating an answer in split seconds, I would not document it as a definitive resource for queries, as its lack of accuracy means it is not a legitimate resource to build opinions/recommendations on.

AI uses in healthcare

ChatGPT has been hailed for having a myriad of uses in healthcare. It can be implemented in clinical documentation to create accurate and detailed clinical notes, save time and increase the efficiency of documentation. It could take over tedious tasks such as appointment scheduling and answering general queries, freeing up time for tasks requiring a human interface.

It can be used as a translation tool, and that is one exceptional use that pharmacists can maximise. Type your English sentence and ask ChatGPT to translate it for you to whatever language you use. (I must say this is the best use I’ve found for it so far.)

Having passed the Medical Knowledge Self-Assessment Program (MKSAP) with a score of 75 per cent, it is foreseen that it could become a medical chatbot that can assist doctors in speeding up diagnosis. However, caution will need to be exercised as it can exacerbate and introduce its own bias.

It can also help provide information tailored to a patient’s context and needs, but nothing should be disseminated without a clinician reviewing and editing it.

The opportunities of AI applications in pharmacy

AI has the potential to revolutionise the pharmacy industry. Here are some examples of its myriad uses:

• Medication management – one of the most critical roles of pharmacists is to ensure that patients receive the right medication at the right dose and at the right time. AI can automate such a task by matching a patient’s medication list against their medical history, allergies, and current medications, reducing the likelihood of medication errors. It can be used to identify potential drug interactions and alert pharmacists and healthcare providers of any potential issues. It can also be used to provide patients with personalised medication reminders and instructions, improving medication adherence. By streamlining the medication management process, AI can help pharmacists reduce errors, improve patient safety, and save time.
• Workflow optimisation – AI can be used to optimise pharmacy workflows and improve efficiency. For example, it can be used to monitor medication inventory levels, predict demand and automatically reorder medications when supplies run low. This can help reduce waste and ensure that medications are in stock. AI can also be used to automate medication dispensing, reducing the time pharmacists spend on manual dispensing tasks and allowing them to focus on more complex patient care tasks.
• Personalised medicine – the one-size-fitsall approach to medicine may not always be effective, as patients can respond differently to the same medication. AI can help address this issue by analysing patient data. Its algorithms can identify genetic and other factors that influence a patient’s response to medication. This allows healthcare providers to tailor treatment plans to each patient’s unique needs, potentially improving treatment outcomes. This is an area of precision medicine known as pharmacogenomics.
• Drug development – discovering novel therapies is a long and expensive process. AI can significantly speed up the drug discovery process by analysing large volumes of data and identifying patterns that human researchers may miss. AI algorithms can be trained to recognise disease-specific molecular structures and predict treatment efficacy. This enables researchers to concentrate their efforts on the most promising ideas, potentially shortening the time and expenses associated with drug discovery
• Predictive modelling – AI can also be used to analyse patient data and predict future health outcomes. For example, AI algorithms can be trained to predict the likelihood of developing certain conditions, such as diabetes or heart disease. This allows healthcare providers to take proactive measures to prevent these conditions before they occur. By identifying at-risk patients early, healthcare providers can take steps to intervene early in the disease process.

The challenges

However, this will not come without challenges that will need to be overcome for successful adoption and implementation, such as:

• Data quality and availability – in many cases, the data may be incomplete, inaccurate or biased, which can lead to incorrect conclusions and flawed decision-making. Ensuring data quality and availability is, therefore, a crucial challenge to successful AI adoption.
• Ethics and bias – AI algorithms can inadvertently perpetuate existing biases in data or decision-making. For example, if an AI algorithm is trained on data that disproportionately represents a certain group or ignores certain factors, it can produce biased outcomes that perpetuate inequality. Hence, ensuring ethical and unbiased AI is a critical challenge that requires careful consideration and attention.
• Regulatory and legal considerations – as with any new technology, AI raises questions about regulatory and legal frameworks, including liability and accountability issues. There is a need to establish clear legal and regulatory frameworks that address issues such as data privacy, security and ethical considerations.
• Human acceptance and adoption – AI can be perceived as a threat to jobs, so ensuring human acceptance and adoption of AI is a critical challenge, requiring education about the benefits and limitations of AI.
• Interoperability and integration – many healthcare organisations already use various software and technology platforms, and integrating AI into existing systems can be challenging. Ensuring that AI systems can be easily integrated and interoperable with existing systems is a crucial challenge for successful AI adoption.

In conclusion, while AI has the potential to transform the pharmacy sector by enhancing efficiency, accuracy and decision-making while lowering costs and saving time, it is critical to recognise and address the issues that come with it to assure ethical and unbiased outcomes. AI technologies offer a wide range of applications, from medication management and drug discovery to personalised medicine, and can lead to better patient outcomes and a more efficient healthcare system. As AI technologies advance, we can anticipate even more prospects for AI integration in the pharmacy business.

AI must be used in conjunction with human expertise and judgement rather than as a replacement. By doing so, we can harness the full potential of AI while ensuring that patient care remains the top priority.

Disclaimer: The content and opinions expressed in this article are my own and do not represent any organisation I’m affiliated with.