Artificial Intelligence (AI): How It’s Impacting Hospital Pharmacy
Ever since platforms like ChatGPT hit the media, AI probably has you wondering: is it going to help your job—or take it?
Artificial Intelligence (AI0

Artificial intelligence (AI) has been around since the 1950s, and pharmacy has always found a way to expand its scope of practice every time a technological advancement improves efficiency. So, let’s examine some ways AI could enhance hospital pharmacies.

Fewer Shortages Through Improved Manufacturing Processes

Over the last few years, drug shortages have been a major concern for just about everyone in healthcare. Not only have supply chains failed, but major recalls may have left your hospital scrambling to take care of your patients. With AI, drug manufacturing and supply chain management should become streamlined.

AI can predict when a machine may fail or require maintenance by learning how it normally operates and performs. AI monitors equipment in real time and detects any deviation from normal activity. Earlier detection decreases equipment downtime and increases overall manufacturing productivity.

Similarly, AI can improve batch quality and consistency through more accurate quality control measures. It uses data from current quality test results to detect defects in real time. Fewer batch failures mean fewer recalls.

Lastly, certain AI algorithms work to predict drug demand. So, it can help optimize inventory, production schedules, and distribution, improving the supply chain.

Lower Drug Prices by Reducing Drug Development Costs

You know that the customer ultimately pays for the billions of dollars and more than 10 years it takes drug companies to successfully get a new drug to market. Now, AI may help reduce those costs.

Using large datasets of chemical structures and their related activities, AI may help predict how new drug candidates might behave in the human body. Along the same lines, AI can find patterns in biological data and disease progression to identify potential drug targets. So, drug companies may save on resources in two ways: finding what to target and selecting the optimal drug candidate for that target.

This same concept is also being used for drug repurposing. AI can analyze the structure of an already approved drug and find a potential new indication, which is significantly cheaper than getting a new drug candidate to market.

Beyond drug discovery and repurposing, AI can help optimize clinical trials from recruiting to analyzing data. Finding patients and evaluating protocols and trial designs can happen faster. Getting real-time trends on data means trials can more easily adapt and pivot if needed.

All of this means drug companies have less overhead and expenditure, which could lead to cost savings for hospitals.

Improved Patient Outcomes Through Better PKPD Data

Few things are more frustrating than to research a drug-related question only to find it hasn’t been researched in your patient population. Can this drug be given to someone with kidney disease? Or during pregnancy? Or an infant? AI may be the solution.

Historically, animal studies have been used to evaluate the pharmacokinetic and pharmacodynamic (PKPD) activity of a drug. These studies are time-consuming and with small sample sizes, and they may not accurately predict what will happen in humans.

But now, PKPD activity may be predicted using AI. Some models can analyze how humans would react to the drug. This means an accurate safety, efficacy, and toxicity profile can exist before the drug is even given to an animal or human.

Other models focus on individualized care by analyzing patient-specific data. This can be used to predict how an individual may respond to a particular treatment, including disease management, side effects, and potential toxicity.

The Downside of AI

Like everything, AI has some downsides that you should keep in mind:

  • Unintended biases can happen, especially with limited inputted data. Examples of this include rare diseases and under-represented populations (race or gender) in clinical trials.
  • Validating and regulating an AI model can be difficult. The industry and FDA provide limited guidance on this.
  • Ethical considerations, like patient privacy and rights, need to remain a top concern as AI expands.

Like every technology, AI has its pros and cons. But one thing is certain—it’s here to stay. While most of the benefits hospitals may see are byproducts of other companies utilizing AI, there is no doubt that AI will eventually integrate into hospitals, optimizing your workflows and allowing your pharmacy to grow even more.

References:

Vora LK, Gholap AD, Jetha K, et al. Artificial Intelligence in Pharmaceutical Technology and Drug Delivery Design. Pharmaceutics. 2023 Jul 10;15(7):1916. doi: 10.3390/pharmaceutics15071916. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10385763/

Sultana, A., Maseera, R., Rahamanulla, A. et al. Emerging of artificial intelligence and technology in pharmaceuticals: review. Futur J Pharm Sci. 9;65 (2023). https://doi.org/10.1186/s43094-023-00517-w

U.S. Food and Drug Administration. Center for Drug Evaluation and Research. Artificial Intelligence in Drug Manufacturing. 2023. https://www.fda.gov/media/165743/download.