How to ... use AI to manage patient medications

From accelerating the introduction of important new medicines and improving ordering and inventory processes, to helping older adults better manage multiple prescriptions, artificial intelligence’s reach is seemingly ubiquitous. As one pharmacist observes, “more than 50% of the vendors at pharmacy shows are now touting some sort of AI. It’s everywhere.”

1. Like any new technology, AI calls for patience and a period of adjustment.

“In this ever-evolving healthcare landscape, the first step for pharmacists and nurses to adapt to working with AI is through education,” says Allison Rainey, head of nursing and clinical informatics at MatrixCare. Rainey strongly urges both to familiarize themselves with AI tools’ capabilities and limitations.

LTC professionals must learn how to best integrate AI into their work, which will increasingly focus on critical thinking rather than information gathering, says Ignacio Rossani-Ortega, director of business intelligence and analytics for Guardian Pharmacy Services.

“Adopting a ‘trust but verify’ approach to AI-generated [results] is critical, and this includes getting patients to buy into the value of AI,” he says.

The transition to working alongside AI in medication management requires a proactive approach, says Joe Kramer, vice president of sales for Geri-Care Pharmaceuticals Corp.

“Training programs can be implemented to familiarize healthcare professionals with AI-powered systems, emphasizing their role as collaborators rather than replacements,” Kramer adds.

2. The benefits of managing medications with AI are growing.

“AI can quickly identify high-risk medications and potential adverse drug reactions and can help pharmacists quickly scan medication profiles and triage assessments that improve residents’ quality of life,” says Erin Donatelli, PharmD, BCGP, FASCP, vice president of clinical and consulting services for Remedi Senior Care.

Risk management alone is a significant upgrade.

“AI can dramatically transform our approach to medication safety and the early detection of health issues among patients, especially in older adults,” adds Rossani-Ortega. “With AI, we see a future in which predictive technology advises pharmacists and nurses of potential risks, such as falls or a hospitalization from adverse drug events.”

Rainey says AI also can “sift” through thousands of records, aggregating care-related data points for clinicians. “The medication management decision-making AI tools alone contribute to job satisfaction, reduce burnout and turnover risks, and retain highly skilled clinicians within facilities for the long term,” she says.

3. Nursing home residents stand to benefit.

“AI can readily identify medication duplication, contraindications, potential drug interactions, predictable adverse reactions, and high-cost medications with suitable alternatives,” Rainey says. “This information empowers clinicians to act appropriately in near real-time — prior to the patient receiving their first dose, being discharged, or experiencing a negative outcome.”

“AI facilitates communication between patients, caregivers and healthcare providers, fostering a collaborative approach to medication management and promoting autonomy and empowerment among residents,” adds Kramer.

4. Despite being 25 years old, AI still has potential drawbacks.

Its machine-learning algorithms are based on medicine and evidence for an average patient population — not on the aging and high-risk residents in US nursing homes, says Shelly Spiro, executive director of the Pharmacy HIT Collaborative. For now, many pharmacists and nursing home managers believe everyone should proceed with caution.

“Pharmacy AI technologies we’ve explored tend to be very expensive — for smaller facilities, the ROI may be hard to realize,” notes Stephen Creasy, PharmD, director, clinical services for PharMerica. “We are trying to understand how we can use it effectively and in a manner that doesn’t introduce potential liability.”

Donatelli says it’s crucial for providers to remember that AI “should never replace human expertise and clinical judgment.”

Overreliance on any technology poses problems, echoes Chad Worz, PharmD, chief executive at the American Society of Consultant Pharmacists.

“These are all tools that aid clinicians in decision-making and insights,” he notes. “So long as nursing homes don’t hand off decisions to robots, I don’t see problems.”