AI models can predict pharmacy foot traffic and peak service times, and even analyze patient histories to single out who picks up their prescriptions on time.
The health care industry has made significant strides in integrating artificial intelligence (AI) into its operations to date, but there are plenty more advancements in the wings. Within the pharmaceutical field, AI is solving problems that approach the limits of human capabilities, such as the rapid analysis of enormous datasets, paving the way for more efficient drug development and the rapid rollout of life-saving medications. What is emerging is a picture of the patient care journey in which AI plays a part every step of the way, from drug discovery to diagnostics to treatment.
The potential for AI to supersede existing guidelines for patient care—that is, offer more appropriate solutions than a physician could, based on available patient history—underscores the promise of the technology for population health. In some instances, this ability for AI to “intervene” in a physician or pharmacist’s recommended course of therapy is already here.
Whether developed to operate autonomously and continuously, or hand-in-hand with a medical practitioner, AI has the potential to change our lives for the better.
More efficient drug development, trials and rollout
The design of new drugs will be aided by AI Without getting technical, AI-powered predictive modeling can identify at a clip which part of the body a drug is going to target and how it’s going to help with a given disease. In this way, AI can fast-track the drug discovery and development process.
Moving on to the trial process, when new drugs are tested on people, AI can play a radical role in identifying health hazards. Trials are generally broken down into 4 phases, with each phase marked by drug administration to a larger trial population. With AI, we can not only fast-track phases 1 and 2 by predicting what smaller test groups of patients might experience, but apply it to subsequent phases, in which the technology can streamline patient scoring, virtual screening, and randomization, offering predictive insights into how the drug will be absorbed and metabolized, along with any indicators of toxicity.
From a patient perspective, this means cheaper drugs. Remember that medications are so expensive not because they cost a lot to make, but because of the costs that go into these preclinical and clinical trials.
The promise of AI in this branch of pharmaceuticals is enormous and is why, in 2023, we’re likely to see more partnerships between software companies with strength in AI competency and organizations in the health care space specializing in clinical administration, documentation of clinical trials, and hospital documentation.
Improved patient outcomes in clinical and retail pharmacy settings
In many ways the COVID-19 pandemic has fast-tracked utilization of AI in the health care space. At the retail pharmacy level, for example, we’ve seen a surge in robotics-assisted fulfillment as well as a rise in the popularity of standalone pharmacy kiosks for contactless dispensation.
Although regulations still mandate a human pharmacist be available for consultations with patients before they depart with their medications, these advancements in inventory management and workflow are having a big impact on patient care. AI models can predict pharmacy foot traffic and peak service times, and even analyze patient histories to single out who picks up their prescriptions on time.
The result is time saved (and less stress) for all: The patient is in and out of pharmacy faster, and the pharmacist can sync inventory more closely with pickups, minimizing inventory waste and the mundane task of restocking unclaimed meds.
AI can also prove helpful at another, crucial touchpoint of the retail pharmacy experience: Identifying which patients will benefit most from an in-person consultation with a pharmacist at the time of pickup.
As we look to the future of how human-AI interaction may play out in clinical settings, one need look no further than the drug utilization review edits already in practice at your local pharmacy. This is software that flags for pharmacists whether usage of a new prescription may be dangerous when combined with the patient’s existing drug therapy or medical condition(s), based on available patient data. When this warning is issued, pharmacists are required to address the interaction before proceeding with dispensation.
Applying this capability to a clinical interaction in the future, we get a real glimpse of the qualitative and quantitative powers of AI—assisting clinicians in deciding what’s best for a patient. For decades, physicians and medical workers have applied literal textbook guidelines to treat everything from vague complaints to serious conditions.
However, with an AI-powered data network configured between all points of a patient’s care journey —primary physician, specialist, hospital and pharmacist—it becomes possible for the technology to offer a more accurate prediction of the right treatment (medication) for a patient than, say, their physician, whose determination is evidence-based while reflecting those clinical guidelines.
Here's an example: If a patient is prescribed a hypertension medication in a physician’s office, the electronic input system the physician is using, having assessed the patient’s data holistically and instantly (everything from past prescriptions to emergency department visit summaries), may suggest that the patient would do better with an alternative medication. This scenario represents a great advancement for AI, but really it’s a leap for clinicians and pharmacists, whose occupational objectives center around getting the right medication to the right patient in a timely fashion.
Application of AI in the pharmaceutical space suggests a number of precedents when it comes to improved patient care, more efficacious drug therapies, and time savings every step of the way, from drug design and development to drug pickup at your local retail pharmacy. Across all health care-based settings, the theme of successful AI implementation translates to stronger matches of patient symptoms with accurate diagnoses and the right treatment.
News headlines tell us that AI is being implemented in every sector, modeling and automating workflows for any organization that receives data—and at an exponential, almost nimble, pace. Like the transition to any new system, there are sure to be hiccups and adjustment periods as databases expand to reflect the population as a whole.
But as we ask in medicine, do the benefits outweigh the risks? In the case of AI, signs indicate yes. That’s a win for patient and practitioner.
About the Author
Reema Hammoud, PharmD, BCPS, is assistant vice president of Clinical Pharmacy, Sedgwick.