Unscripted – The AMCP Podcast: AI in Managed Care Pharmacy: Ethics, Efficiency & Patient Empowerment

Fred Goldstein – 00:01
Welcome to Unscripted, the AMCP Podcast, a look inside Managed Care Pharmacy.
Listen in as experts explore the challenges, innovations, and opportunities shaping healthcare for millions of patients.
Today, we'll be discussing AI and its use and potential in managed care pharmacy.
My guest on today's show is Joe Honcz, Senior Vice President of Market Access with Petauri.
I'm an AMCP board member. Welcome, Joe.

Joe Honcz – 00:26
Hey, good morning. How are you?

Fred Goldstein – 00:28
I'm doing really well—glad to get you on. So tell us a little bit about your background, particularly as it relates to AI.

Joe Honcz – 00:34
Yeah, sure. You know, as a growing capability—let's just use it as that for now—AI seems to permeate everyone's conversation, but the use of it is still pretty nascent. It's pretty basic, if you will.
I started thinking about the use of AI and artificial intelligence back when I was in Aetna's Innovation Lab. I spent a couple of years there. It was an amazing place to work. And our charge really was to look at different ways we can use technology—or innovation, if you will—to help improve the member’s health.
Siri was starting to come around, and Alexa for Amazon. Those are really AI agents, if you will. They were very rudimentary back in the day, but my thought process was—and sort of the hypothesis we tried to prove—can that individual become your health concierge? Your person that helps track where you're going, helps you figure out where to go next.
So that’s where it started. And the imagination is unlimited here in what we can do with AI in healthcare.

Fred Goldstein – 01:38
Fantastic. So before we dive into the depths of this area, what are some of the fun ways maybe you're using AI in your own work or play?

Joe Honcz – 01:46
Yeah, so I'm a pharmacist by background. I'm a basic writer of things—I have to type all the time, create slides, and really try to sound smart, right?
So every time I write something, I ask my AI assistant, if you will, to make it sound better. And Fred, I have to tell you, each and every time, it has made my prose sound a little bit better, a little more refined.
I'll be honest—it's very basic, but it has upped my game from a literary perspective pretty meaningfully.
Also, my family likes to break stuff around the house, and I’m constantly trying to fix things. Searching out different solutions or fixes—AI has been tremendously more powerful, even more so than Google.

Fred Goldstein – 02:27
Fantastic. So what do you see as the most impactful uses for AI in healthcare?

Joe Honcz – 02:32
First and foremost, I think the patient should feel way more empowered. The provider and everyone else around them should recognize that these patients are starting to see really in-depth information—stuff that can be tailored to their life.
They can ask the AI bot to make it an eighth-grade level so they can understand it—or say, “Give me the PhD level. I want to understand the depths.”
That’s what AI does right now—it levels the playing field and gives patients an opportunity, if they’re using it.
For health systems, providers, and even managed care, we're seeing incremental efficiencies. For example, bots are now following the healthcare provider and helping refine the notes taken during a visit—refining them down for billing purposes. AI bots can do that much more efficiently.
So I think we’re going to see people be able to do their jobs better, faster—and hopefully spend more time with each other, talking. Getting back to being a bit more human.

Fred Goldstein – 03:31
It sounds like that example is really around efficiency. Are there other areas or use cases where AI is improving efficiency?

Joe Honcz – 03:37
Yeah, there are specific examples. One I really like is from Sean Wang, the previous head of AI at Elevance—formerly Anthem.
He was on a panel with me about a year ago, and he shared how Elevance is working to connect patients with the right physician.
If you’ve ever tried to find a specialist, it can be extremely challenging. AI can do that homework—figure out the best physician match for what you're dealing with.
That’s not only efficient, but it’s better medicine. Instead of bouncing between doctors, let’s get you the right one the first time.

Fred Goldstein – 04:12
That’s a great example. So as organizations think about using AI, are there resources they should go to?

Joe Honcz – 04:18
Believe it or not, AI itself is a tremendous resource.
AMCP did a pre-conference recently on AI. There was a little selection bias—most people raised their hands when asked if they had used AI. But in a general session, I think less than half would have.
So just use it—Copilot, ChatGPT—ask it to teach you about AI. There are like 12 different types of AI that exist. They all kind of interconnect.
That’s a rare case where the tool itself can teach you how to use it.
Other options include Coursera. They have a great course on AI in healthcare that you could finish in a week using the free trial.

Fred Goldstein – 04:55
Really fantastic. One of the big media stories around AI has been its use in prior authorization. How do we ensure that doesn’t go wrong?

Joe Honcz – 05:04
If you think about AI as a person doing a job—evaluating clinical criteria—people can make bad decisions. So can AI.
It requires training, just like we do with ethics and bias training.
It’s about making quality decisions and rooting out biases.
This is about patients getting access, so we really need to up our training game—and fast. I think we’ll see exponential improvement.

Fred Goldstein – 05:27
There’s also concern about bias in the training data. What should organizations consider from a governance perspective?

Joe Honcz – 05:33
Absolutely. Each organization should have an AI governance council—or at least a set of guiding principles.
That includes ethics, morality, bias reduction, and ensuring high-quality decisions.
These should shape every step of the process to ensure your AI “employee” is functioning well.
We’ve seen some legislative and regulatory momentum. There’s a group in DC called CHI working to make sure AI functions properly.
Everything is moving fast right now and not yet fully integrated. But I think in a year, we’ll have a more refined view.

Fred Goldstein – 06:06
What about drug discovery? Do you see AI playing a role there?

Joe Honcz – 06:09
Yes, absolutely. I work with manufacturers, and AI is moving rapidly in that space.
The use of AI to explore compound viability is taking off like wildfire.