Artificial intelligence (AI) is reshaping health care, including how medications are used. On November 13, 2025, PQA hosted PQA Convenes: Artificial Intelligence in Medication Use Quality to bring together PQA members, health technology leaders, and other stakeholders for a discussion on how AI and machine learning are being used to identify, understand, and engage patients in their medication use.
This is the first blog in a four-part series on the event, and it covers the opening session, “AI in Practice Today: Lessons Learned and Emerging Practices.” The panel was moderated by Jeffrey Rochon of the Pharmacy HIT Collaborative, who was joined by Sarah Blendermann of Pfizer, Anita Patel of Walgreens and Kempton Presley of AdhereHealth.
Given the wide-ranging interactive panel discussion, no quotes or views are attributed to any panelists or organizations. The perspectives shared via this blog are intended to support continued dialogue about the role of AI in medication use quality.
Understanding Three Eras of AI
AI is not a single technology. It has evolved through three major paradigms that shape how we use it today.
- Predictive AI analyzes trends and patterns. It helps health systems anticipate factors such as medication adherence risks, hospitalization likelihood and remission predictions.
- Generative AI synthesizes information and creates new content. These tools summarize literature, produce clinical drafts and support communication workflows.
- Agentic AI goes a step further by taking action. It identifies a need, drafts a plan and initiates the next step, such as scheduling follow-ups or prompting interventions.
Health care organizations benefit most when these three capabilities work together within clinical workflows, not as standalone tools.
Lessons Learned: Challenges in AI Implementation
AI brings enormous potential, but the road to successful implementation is not simple. The panel identified several recurring challenges, including:
- Integrating data sources. Claims data, clinical data and patient-generated information don’t always align. Without complete, unified data, AI can’t drive meaningful next-best actions.
- Bias and trust. Communities differ, patients differ and algorithms can misinterpret those nuances. Ensuring that AI recommendations match real-world patient needs requires ongoing evaluation.
- Regulatory uncertainty. The rules for AI in health care are still emerging, and the “wild west” environment often forces teams to balance innovation with compliance.
- Cost and consumption. AI tools can be expensive. Leaders must weigh whether the value of automation is high enough to justify the ongoing computing costs.
- Behavioral change inside organizations. This may be the biggest barrier of all. Employees often say they “need AI” before clearly understanding the problem they’re trying to solve. Even when the right tool is available, building the habit of using it consistently can be difficult.
Lessons Learned: How AI Is Improving Patient Experience
Patients may not understand AI, but panelists agreed they will feel its benefits through smoother, faster, more personalized care.
- Faster access to treatment. AI is used to analyze wastewater data and predict COVID-19 spikes, enabling supply chains to deliver antivirals to the right places before outbreaks escalate.
- Better-targeted outreach. Pharmacies and health plans use AI to identify the right patients to contact and the right time to reach out, reducing unnecessary interruptions.
- More meaningful clinician interactions. By offloading administrative tasks, AI allows clinicians to focus more on the patient in front of them instead of a computer screen.
Emerging Practice: Leveraging AI for Meaningful Impact
The panel shared examples of how organizations across health care are leveraging AI to meaningfully impact how they engage and care for patients. Organizations are:
- Leveraging predictive models to improve patient engagement. AI now helps forecast patient behavior, optimize medication management and streamline returns programs.
- Transitioning from manually creating hundreds of thousands of content pieces each year to automatically generating diverse content variations.
- Analyzing provider interaction data to uncover insights that traditional methods missed. By pairing data science with behavioral science, clinical workflows are being redesigned to improve engagement.
These examples demonstrate that the key isn’t the size of the project. It’s how well teams understand the problem, test solutions and evaluate impact.
Emerging Practice: Thriving With AI Instead of Competing Against It
The question everyone was waiting for: Will AI replace clinicians, pharmacists, analysts, or other health care professionals?
The answer was clear. AI will not replace humans. Humans who use AI will replace humans who don’t. The difference lies in curiosity, critical thinking and judgment, all of which are skills AI cannot replicate.
Health care professionals bring empathy, nuance and lived experience to decisions. AI can accelerate tasks, generate drafts, or provide insights, but it cannot independently navigate the ethical, emotional and interpersonal parts of care. In fact, properly implemented AI should free up clinicians to spend more time connecting with patients, not less.
The Future of Health Care and AI
The panel closed with a powerful message. AI is not a technology issue but a people issue.
The organizations that will thrive are the ones that invest in AI fluency, integrate tools directly into workflows and center human connection at every step.
Artificial intelligence will amplify our impact, not replace it. The health care leaders who embrace this transformation with curiosity and courage will define the next chapter of patient care.
PQA Convenes: Artificial Intelligence in Medication Use Quality was made possible by the generous support of Arine, Merck, Pfizer and PQS by Innovaccer. PQA does not endorse, recommend or favor any organization, or its products or services. PQA general funds also supported this event.