From 3M Health Information Systems
Podcast Episode Transcript: Confronting the opioid crisis
Dr. Gordon Moore: Welcome to 3M’s Inside Angle Podcast. This is your host, Dr. Gordon Moore, and with me today is Dr. Eugene Christian. He’s the Chief Medical Affairs Officer at Bon Secours Mercy Health System. Welcome, Dr. Christian.
Dr. Eugene Christian: Thank you.
Gordon: Last time we spoke, you were in Virginia, and now you’re in Cincinnati. Tell me a little bit about that move.
Eugene: Interestingly enough, Gordon, last time we spoke, I had started with Bon Secours Health Care System as the CMO at St. Mary’s Hospital in Richmond, having moved from Charlotte that year. I was with Atrium Health, which was then Carolinas Health Care System. Started that in August of 2017, and then within six months, we had an announcement that we were merging with Mercy Health of Ohio, which is another Catholic health care system. And that merger became completed September 1 of 2018.
And as the reorganization took place in November of that year, I was offered the position of Chief Medical Affairs Officer for the new health care system ministry, and that job happened to be in Cincinnati. So, after having made a move to Richmond and enjoying working with the Bon Secours system, then made another move to Cincinnati to work within the larger, now newly emerged ministry as the Chief Medical Affairs Officer—which has been very exciting and, so far, a lot of fun.
Gordon: Give me a sketch of the Bon Secours Mercy Health Care System. What is that?
Eugene: We are a combination of what was the legacy Bon Secours Health Care System, which was a Catholic health care system based mainly in Virginia, Maryland, and South Carolina. We also have a few facilities in New York and also in Tampa, Florida. So we’re really stretched along the Atlantic coast.
Mercy Health of Ohio, another Catholic health care system primarily based in Ohio—actually the largest health care system in Ohio—also has a few facilities in Paducah, Kentucky. Bon Secours also had a facility in Ashland, Kentucky. So it was a really natural merger of our two health care systems; very, very similar with our mission and values, very alike in size between our two facilities as far as hospitals and overall patients and employees between the two systems. So, in a combination of the two, we now have employees of around 56,000, and we have 43 hospital facilities.
And just recently, we have acquired five hospitals in Ireland, which were legacy Bon Secours facilities in Ireland and will become a part of our health care system officially on the first of July. So Bon Secours Mercy Health is now an international health care company.
Gordon: Fantastic. Maybe we can do a face-to-face podcast recording on site in Ireland.
Eugene: That would be a lot of fun. Bring your golf clubs.
Gordon: I’ll do that. Well, the reason I wanted to speak with you today is that I heard you give a presentation a bit ago on some work that Mercy has been doing around managing opioids. And it’s a hot topic and very interesting, and what you talked about was really compelling in terms of the approach, the use of data, figuring out how to get behavior change, and the results. So, I’d like you to talk some about that. Why don’t you start with how did this whole thing start? Why did your health system get into this?
Eugene: I think both of our health systems had started focusing on this as the opioid crisis really became a national conversation. I know Mercy Health had been doing some work with the state of Ohio specifically on this over the past couple of years, but really began some of their work back in 2015 as the board at Mercy and the Sisters who were looking at the crisis as it existed—and the number of overdose deaths that they were seeing in Ohio and Kentucky was really astounding and really frightening to the ministry. So it became a priority with the health care system—really coming from the board level—that we needed to address this the best way that we could. So from the Mercy perspective, I think it was driven there.
On the Bon Secours side, that conversation, I think, began a little later. But in a similar manner—with our health care system leaders and our board leaders and our Sisters of Bon Secours really looking at this as a national crisis that, as a health care system, we really needed to do something about to help the patients and the populations that we serve. So it really came as a priority from the highest levels of both organizations. And then of course with the merger, it remained a high priority with the merged system.
Gordon: So how did you get started with this work?
Eugene: How did I get started with the work?
Gordon: How did the health system get started? So you had the boards saying that this is important—we want to get going. Where did they start?
Eugene: I think we had to look at where did the problem exist. And we know the problem exists in a couple of different areas. One is on the prescribing side. So it was the problem of looking at how providers treated pain. And as a background to that, one of the emphases that occurred back in the late 90s and early 2000s—and I think many providers will remember this—that it was really a Joint Commission and CMS prerogative that pain was a fifth vital sign. And that, as health care providers, we really needed to address pain better and really help our patients get through that. And as part of that, opioids became more and more frequently prescribed.
So I think the problems began, somewhat, back then. And I think the healthcare pharmacy companies were more than happy to help contribute to that with the different types of opioids—and we went from the short-acting to the long-acting opioids—which really started to contribute to the addiction problems that started to sprout from that.
So initially it was, “Hey, we want to control pain.” But at what expense? And I think that’s where the opioid crisis really started to get bad—in the honest approach to trying to control pain, but not looking at it more broadly as what were the alternatives. And I think now that we’ve seen what those alternatives for pain control are, we’re doing a much better job with it. So I think it started back with how do we as prescribers and providers do a better job in controlling that.
And I think the other piece to this is recognizing that the addiction piece of this is very real, and how do we help people when they enter our system who screen positive for heavy opioid use and have addiction issues that we can help shepherd them in the right direction to get help? So, in our emergency departments and in screenings through our primary care offices, we’re trying to identify those patients who are struggling with this and try to provide help for them.
I think those are really the two places we’re trying to work on: How we do a better job with helping our patients avoid the opioids, first and foremost; and then if they needed them and had trouble with them, how do we help them move off of opioids, taper away from them. Or, if they are chronic pain patients or cancer patients who needed it, how do we appropriately prescribe it and keep them safe with it?
Gordon: I remember as you were presenting this earlier, you talked about the example of total joints and managing that with opioids—and the complications that flowed from that use and how that shifted.
Eugene: Absolutely. I think orthopedic joint replacements are a great example of trying to do a better job with overall pain management. So we know that when we were giving opioids to patients early on, it was harder for them to mobilize, it took longer for them to get into physical therapy, their lengths of stay were three or four days.
And over time, what we’ve discovered is that actually eliminating opioids, using local anesthetic and long-acting local anesthetic, pain blocks and joint blocks for control of pain—and using multimodal analgesia such as using Tylenol judiciously and using Celebrex and some of the other products that were out there—when we use these in combinations, they worked actually better than the opioids, particularly with the elderly patients. We didn’t have the CNS effects from it. They didn’t have the balance issues. We actually saw them rehabilitate faster—and it’s really helped push the metric to getting joint replacement as an outpatient procedure.
I think that’s a great example of how looking at pain differently and how we control it really helped out patients, particularly in that scenario with joint replacement.
Gordon: And the pain management was equally effective with the non-narcotic approach?
Eugene: Oh, yes. As a matter of fact, when we talk to patients now—and if you look at their pain scores and you look at objective VAS scores and so forth—the patients who are getting multimodal analgesia now oftentimes report lower pain scores than the patients who are getting opioids. It’s remarkable.
Gordon: Wow. And of course, therefore, by not using the opioids—as you mentioned before—they’re less likely to be lying in bed out of it on opioids. And also when they do get up—more stable. So that’s a clear double win for them as you were mentioning.
So, one of the things that’s implicit in what you’ve been saying so far is the ability to track metrics. What kinds of things are you guys tracking? And what’s important to you in terms of observing data flow over time?
Eugene: When we looked at the best way to track improvement here, it’s looking at total opioid burden. How much opioid was being prescribed across the system? And then looking at what best practices would be for morphine daily dose equivalents. One of the things that also helped drive this is the state of Ohio, the state of Virginia—some of the states in the last two years have actually legislated a limit on daily morphine equivalent. So you can take all the opioids that are out there and you can calculate, through an algorithm, a morphine equivalent dosage. So by doing that, we’re able to measure equally across all opioid prescribing morphine equivalents, so we can speak the same language and be able to measure them all the same equally.
So, for instance, 30 daily morphine equivalents is a limit that the state of Ohio set for the use of acute pain management for short-acting opioids. Of course, we chose that as one our metrics, as a target for saying that’s the limit that we would set for the daily prescription dose that we would send. The states also limited the number of days you can have on a prescription. Some are three, some are five to seven. Every state was a little different. So we had to work a little bit on how we helped our providers prescribe this in our electronic health record, and it had to be state-specific to meet those requirements.
But that, plus being able to calculate an overall opioid burden in the system— how many prescriptions were we writing, how big were those prescriptions, and what types of dosages were attached to them—by looking at both those metrics, we’re able to track the results of the things we were putting in place, like how we used our electronic health record to help our prescribers actually prescribe the proper dosages and make it easy for them to do it without the prescribers having to actually calculate what those were. So that really helped us be able to track our progress, just being able to do that across the system and make it really equal as far as how we measured everything.
Gordon: Reducing work burden is something that’s really interesting to me, because I see that as one of the things that really good technology can do. I just don’t see a lot of really good technology. I hear mostly from our colleagues that they’re frustrated by, “It’s not making life easier for me.” But it sounds like you were using that—“make it easy to do the right thing”—when you were implementing policies and procedures around this?
Eugene: Absolutely. We were thinking about what these limits were on the prescribing—how do we help our physicians and providers prescribe properly and know what the doses were that they were going to be doling out and the number of days. What we built into our electronic health record was an opioid pain prescription module. So, instead of having ad-hoc abilities to just order opioids by clicking on a specific drug, we actually put all of the opioids into a specific pain set. So if you wanted to prescribe an opioid, you had to open that pain set. It was separate from any of the other non-opioid pain-management solutions that we had.
And we built into that—and it took some extensive work to do it, but it really worked out beautifully—all the different narcotic pain medications were listed, and if you chose the proper narcotic at the proper dose, if you clicked on that, it gave you the exact dose, the number of days, and your prescription then printed out. And you didn’t have to go to another step. If you wanted to, you could go beyond that. If you wanted to order more and you wanted to order more days, you could do that. But then what it would take you to is another screen that would warn you: “Hey, you are outside the parameters that have been set for this. Do you really want to do this? If you really want to do this, then you have to take another step.”
So essentially what we did is we made it really easy for the providers to do the right thing—and made it really hard for them to do the wrong thing. And we initially had concerns from the providers that this might slow them down and it would interfere with their work flow. But the way we constructed it, and with the education we put out, it was quickly adopted and the providers actually loved it and found that: “Wow, I can stay compliant. If I have to go outside that, I can, and here’s how I have to do it.” And there are times when they just had to do it for a particular patient—which is fine.
But making it easy to do the right thing and hard to do the wrong thing—and I don’t necessarily use the word “wrong”—but hard to work outside what we felt were the proper parameters really seemed to be workable for our physicians, and they were happy to embrace that.
Gordon: And you, therefore—I’m guessing—did not experience torches and pitchforks at your door?
Eugene: Surprisingly, we did not. I think the communication about this was pretty widespread, and I think our providers, for the most part, understood what we were trying to do—and what we were trying to do to help their patients, really, and keep everybody within safe ranges.
Also, for patients who had chronic pain and were in pain clinics—it was set up—and again, all of this was driven by the states: If you’re a chronic pain patient, you had to have a contract with only one provider. You could only order from one pharmacy. So we also set that up for the chronic pain patients who were in pain management to make that easy for our providers to do. And along with that, knowing that these were higher doses and more dangerous doses, allowed for naloxone prescriptions to be added in there so that patients could take that with them as a safety net in case, for whatever reason, they might run into trouble. Their family members are taught how to use it. And we have some examples of how that rescued some patients from an overdose problem.
Gordon: Was it complicated coming up with these standards? Or was it pretty much you take what the state legislated and said, “We’re just going to do that?”
Eugene: We started that. That was our base. “Here’s what the state says we’re going to do. We’re going to comply with that.” But when you looked at best practice as far as morphine equivalents and treatment for acute pain with short-acting opioids for the opioid-naïve patient—these really were best practices from a clinical standard. So no one really had a hard time complying with the state regulations on this because it really did line up with what was felt to be best practice and safe practice for opioid prescribing.
Gordon: And I presume you had a typical structure—you get a bunch of medical staff together and multidisciplinary teams, that kind of thing?
Eugene: Yes. We actually created an Opioid Task Force, and this task force was multidisciplinary. So we had surgical specialists, medical specialists, pharmacy supply chain, nursing—everyone in those particular areas was part of this system task force that met and looked at how to design this. And once the decision was made how we wanted to construct it, then the clinical informatics teams then went into Epic, and our builders then built these parameters into our prescribing algorithms. So it was really a multidisciplinary approach to this. We wanted to make sure that every stakeholder that would have anything to do with the opioid world was involved.
And of course, very heavy on ED involvement, because ED docs are the first people—they are going to contact many of these patients when they come in—that might need opioids. Also, the ED had a sub-task force to look at: how can we make our emergency departments as opioid-free as possible? So the concept of the opioid-free emergency department.
For the most part, in many of our markets, we were able to drive that metric fairly well. And that was a decreased amount of opioid prescribing by ED docs for patients that are coming in for emergency care that might need something to control pain. And they would, of course, be discharged—they would not necessarily be admitted to the hospital for their problem. But they may have a simple fracture that was taken care of, splinted—and trying to use alternative methods for pain while we were giving patients close follow-up and getting them into their specialists as quickly as possible so the patients weren’t out waiting any great length of time to get seen for their definitive care.
So access we worked on. And then, trying to use alternative prescribing practices for using non-opioids was a metric that the EDs collectively decided they wanted to measure amongst themselves and hold themselves to. And there was no significant number that they were trying to reach. It was just as far down to zero, actually, as possible—since zero was the goal, knowing that it was probably not doable in every situation. But driving it out of the ED was a big part of the work that we did.
Gordon: When you were describing this earlier, I remember you showed a slide showing a really dramatic drop in ED visits, particularly around that cohort of people who might have been seeking drugs—maybe because of abuse risk—and that that drop in ED visits could be alarming to administrators. Is that something that came up with you guys?
Eugene: It did. One of the things that we found in some of our ED volume tracking is we saw that some of the volumes in the emergency departments did drop. Once we instituted what we call the opioid-free emergency department, word gets out in the community fairly quickly. And once people found that, “Hey, I can’t go to that hospital’s emergency department and get a narcotic because they’re not going to give it to me”—those patients didn’t come. And we did see some drop off in volume. But it was the kind of volume that, honestly, doesn’t need to be in the emergency department.
We want people to come to the emergency department that need care. And we see everybody that comes in and want to care for their acute problem. But for those people that didn’t have an acute problem and were just really drug seeking, we saw a drop off in that. And it’s amazing how many there are, because we saw some overall volume shifts that occurred because of that. And initially, there’s always the alarm from the administration side—“Oh my gosh, our volume in the ED’s dropping.” Well, that’s because some of that volume didn’t really need to be in the ED to begin with. And we really have to make sure that we’re giving the right care at the right time at the right place. And so I think that that’s all okay.
And when we look at it from the standpoint of we’re providing the right care for our patients, and we’re doing the right thing for the population—what happens over time is that as your emergency department gets a little less congested from those people who are waiting to get that, wait times actually get better, and people get seen in a more timely manner. And over time, some of the EDs bounced back because the word got out: “Hey, you don’t have to wait as long to go to that ED. Let’s go there.” Because, as you know, in many communities, ED wait times are posted on the billboards now. So it did make a difference.
Gordon: I can imagine that a person with an opioid-use disorder doesn’t need to come to an emergency department and get more opioids. They would more likely benefit from substance-abuse treatment. So that’s a good thing. It’s tough, though, sometimes, when I think about pushback around revenue when that can be pretty lean in hospitals. And it sounds like you guys have weathered that pretty well.
Eugene: We think so.
Gordon: So I hear, then: You have an evidence base. You have making it easy to do the right thing, getting everybody on board, using a multidisciplinary team approach. Also, informing leadership: “Hey, we’re likely to see some changes in data. That might be a little bit concerning, but here’s why it’s important.” Because that’s connected to the mission that works pretty well. I remember you also mentioned that there was a difference when you looked at EDs across state lines in Kentucky versus Ohio. Do I recall that correctly?
Eugene: Yes. What we saw with Kentucky versus Ohio, for instance—we’ve got a big market in Cincinnati, and the Ohio River is right there and you can walk across the bridge and you’re in Kentucky. Kentucky did not have the same parameters that their state had put forth as far as amounts of opioid you could dispense and a limit on the days, for instance, that you could do it. So, because they had not embraced that part, it took longer to get our facilities in Kentucky to see some improvement because there wasn’t that push at the state level that mandated doing that. So it took a little longer to really socialize the work we were doing across the system there to start to see results.
But once we started instituting it and providers began to see the benefit in it—despite the fact that they weren’t necessarily required to do it—we began to see some improvement. I think a big part of it was just between Ohio and Kentucky, we saw that—just because they were different in what they were requiring.
Gordon: I’m often thinking about the impact of policy on care delivery. And most of the time, when I think about that and talk with folks about those impacts, it tends to be kind of negative. But this seems to be one of those cases where the policy in Ohio around appropriate use seems to be reasonably appropriate and translate into better care delivery. So that sounds like a win.
Gordon: That’s a pretty good lesson. I want to thank you for your time today and ask you if there are any last thoughts or wrap-ups you have for folks who are listening.
Eugene: I would say that, really for us and for any health care system that’s trying to institute change, having the support from the top-down for something like this really was helpful. Everyone knew what the right thing was, it was supported from the board level on down, and the board made sure that the resources were available to get the work done. So there were plenty of resources allocated to get the builds done in our EHR.
There were plenty of resources allocated to work on—we didn’t really talk very much about the access to care and recovery piece—so we also had an EsPeR screening process where we were screening and looking at who screened positive and who we could get into recovery and treatment. And in a lot of our areas, we were able to get local community help to partner with us, particularly in some of the treatment centers, to quickly get people into treatment once we saw them and were able to screen for them. We were actually able to get Suboxone treatment for three days while we were waiting for people to get into treatment and get access to that treatment.
So there’s that whole other piece of it that we started in some of our markets. And depending on what the resources were in the markets, we were able to expand, more in some places than in others. But we continue to do that work. We continue to partner with our communities on helping our addicted patients with that treatment piece of this to help them get off of it and away from it. Because even though we may not prescribe it, people can still go buy it on the street. And if they can’t get that, they’ll try to get something different. But if we can help them with their overall problem and get them into treatment, we can save some lives. And that’s really what we’re driving towards.
The addiction problem is terrible, and the death from it is terrible. And if we can save some lives with the work we’re doing, then we’ve done what we’ve come to do to take care of our patient population.
Gordon: Dr. Eugene Christian, thank you so much.
Eugene: Thank you so much, Gordon.