How Artificial Intelligence Can Combat The Opioid Abuse Epidemic [transcript] [audio]

Guest: Brad Bostic

Presenter: Neal Howard

Guest Bio: Brad Bostic was inspired to improve healthcare service levels when his mother was battling cancer and her journey exposed that most healthcare professionals did not understand her unique needs. Brad founded to create a new approach to healthcare relationship management that analyzes relationships and behaviors with advanced cloud technology to drive positive change. Brad serves on the Techpoint board, Indiana University Dean’s Advisory Council, and is a member of the Indiana Chamber of Commerce. Brad is also a member of Young Presidents Organization (YPO) and has been recognized on the Indianapolis Business Journal 40 Under 40 list as one of the top business leaders in the state. Brad has a degree in business with an Information Systems concentration from the Indiana University Kelley School of Business.

 Segment Overview: Brad Bostic, CEO of, an award-winning healthcare CRM, discusses how artificial intelligence can mine clinical data lakes and be useful in dealing with opioid abuse and other public health issues.


Neal Howard: Hello and welcome to Health Professional Radio. Public health issues like opioid addiction and diabetes are burdening our healthcare system as well as our economy as a whole. Our guest today is Mr. Brad Bostic, CEO of It’s award-winning healthcare CRM and he’s joining us today on the program to talk about how artificial intelligence can mine clinical data links and be very useful in dealing with opioid abuse and other public health issues. Welcome to the program Brad Bostic.

Brad Bostic: Thanks a lot for having me on.

N: Great. Give our listeners just a bit of background about yourself and let’s jump right into artificial intelligence and how opioid abuse can be addressed.

B: Sure. So I’m the CEO of and we’re really focused on solving very difficult challenges in healthcare and what we found is that one of the biggest challenges in healthcare is around how people can be engaged in a more personalized way. So you describe that HC1 is really a healthcare CRM or what we call “Healthcare Relationship Management Company”. And if you look at these major public health issues like opioid addiction and diabetes really, they’re issues that we need to deal with by becoming much more proactive because way too much what we measure now are all in the rearview mirror in terms of being lagging indicators. And the good news is while these are really problematic issues with the emergence of these cloud technologies like HC1 provides, we actually now have an opportunity that’s pretty unprecedented to bring together live information in a very rich way to start becoming more predictive about things, so that we aren’t just looking at with the opioid epidemic specifically things like death rates as an indicator of how bad this is going, but instead we’re looking at in a live manner, what is happening with the actual usage of drugs and how do we proactively institute programs to have the right kind of action, so that we can prevent the type of misuse and abuse that’s been going on. So machine learning is really a powerful way to do that, which gives you the power of an unlimited number of people trying to organize and find these issues that historically with people, you couldn’t do and with machine learning and artificial intelligence now, you can identify these trends and patterns and become very predictive. Much like trying they create a weather forecast with things like opioid abuse, we need to create a much better view into what’s likely to happen in the future so we can institute the right programs to cut down on that.

N: Can the AI or learning machine differentiate between clinical data and how it’s important and how it relates and claims data and how it’s important and it relates when it comes to opioid use based on prescription drugs and then once those drugs are introduced into the illegal or black-market and physicians are dealing with them as they come into their practices, how does AI differentiate or combine the two types of data in order to make some meaningful decisions?

B: Well that’s an insightful area to focus because it calls out a part of this challenge that just humans alone without really powerful technology, it would take years and years to even collate the data. So if you take a data set like healthcare claims as you pointed out which is going to give you a part of the picture, and then you take another set of data that’s about positivity rate of actual drug usage based on drug screens, and you look at that across an entire population including not just healthcare but workplace regulated industries and criminal justice. Now you’ve got a body of information that when correlated can start allowing you to truly unearth very powerful patterns that not only speak to where and why is abuse occurring but also can tie back to what is the cause of that abuse. So both in terms of claim telling you what was prescribed for example but also downstream when there are claims that go against the kinds of treatment activities that are required for the abuse and misuse of drugs. So what it really comes down to is things like to understand even across the United States, what’s going on with positivity rate when you’re bringing in toxicology drug orders, drug test orders and results, you need to be able to map all of that to a common lexicon with what we apply it to as Siloed. And Siloed is a way that creates a standard view into, this is the same test and result type as another test and result type which would be coded differently just because of the variances and how different systems actually output the data. So we use machine learning to normalize all of that, up to that common standard and then also can map that to align with claims information to correlate that either by provider or geography or any number of different attributes so you can start bringing together that full story of, where is prescribing happening and where our drugs getting appropriately administered. And so what kind of quality level is going on in the way that for example in pain management, drugs are being prescribed and monitored and you can look in the data automatically and identify what we call drug consistency which tells you, “Hey, this is a sign that if I see in the claim that there are prescriptions written and then I see in the actual drug scanning results that’s ongoing, there’s the right consistent level of that drug. Now I’m seeing a case where this is appropriate usage of that drug and that has one course of action and then on the other side if I see it where, “Hey, this is an irregular pattern” and it also appears as though these groups of individuals have other types of substances in their blood, now that’s another indicator. But the simple answer is this machine learning and artificial intelligence horsepower that we have now is what allows four billions of these data points to be correlated and then insights to be gleaned from that.

N: As far as you’re concerned in the future of AI, can law enforcement and healthcare data be seamlessly, I guess analyzed to help out both of those aspects as we try to spearhead some type of solution to the opioid epidemic and some of the other healthcare issues that we’ve talked about?

B: I’d say absolutely. There’s application across the board and the way that we really view and are approaching, it’s really like a disease state, abuse and misuse of these drugs. Many people get exposed to opioids through some kind of a medical encounter where they have pain and that can result in addiction and it’s something that then needs to be managed to get all the chemical balance back in a way where you can move forward with your life and not be addicted to those drugs. And so there’s certainly a desire to help people overcome these issues and also from a law enforcement perspective, make sure that you have the right type of training and the right type of treatments available that can come by way of law enforcement. Naloxone for example is sort of the miracle reversal of an overdose that can save lives. And if you have really rich insight into what the patterns are of drug use and abuse across a geography and you can actually predict where that’s going to hit hardest, you can say “Hey, we need to make sure that law enforcement is a quick deal with that”. So yes, both across healthcare as well as law enforcement and really public health in general, this is truly the future of how we get away from the reactive approach to healthcare and public health and get to a proactive approach. It’s similar to, you would never sit back and just wait blindly for a hurricane to hit you during hurricane season. You would be looking at the forecast and preparing for that. We’re putting the same type of capability in the hands of public health officials powered by this sort of correlated healthcare data then we’re just so excited about the positive impact that we can make on public health.

N: Now where can we go online and learn more about HC1.

B: So you can visit, that’s HC, the number And there’s also specifically for the opioid dashboard which is an offering from HC1, it’s just Those are both great resources that you can get to to learn more about what we’re building and really how we’re working to unlock answers to solve this major problem of addressing the public health crisis related to opioid abuse.

N: Great. Brad Bostic, CEO of

B: Please invite me anytime and I’ll be here.

N: You’ve been listening to Health Professional Radio. I’m your host Neal Howard. Transcripts and audio of the program are available at and

Liked it? Take a second to support healthprofessionalradio on Patreon!
Become a patron at Patreon!


Leave a Reply

You must be logged in to post a comment.