Accountable Care Organization

Episode 63: Reducing Misdiagnoses

September 2024

MaineHealth Associate Chief Quality Officer Mark Parker, MD, joins Heather and Mike to discuss a promising care variation reduction project that uses the delivery of timely lab results as a lever to improve diagnosis accuracy. 

Heather: This is BACON, brief ACO news from the MaineHealth Accountable Care Organization. A monthly podcast for healthcare providers. I'm Heather Ward.

Mike: I'm Mike Clark. Heather and I are practicing physicians and participants in the MaineHealth ACO. This month we talked to MaineHealth's Associate Chief Quality Officer about a new effort to reduce misdiagnoses.

Heather: This sounds intriguing. Let's get to it.

Mike: So back in May, the MaineHealth ACO hosted a value-based care symposium attended by a couple of hundred of the ACOs closest friends. It's a symposium you definitely want to attend in the future if you ever get a chance. And one of the more popular sessions at the symposium was a shark tank, where innovators introduced promising care variation reduction projects, and one of the most intriguing was presented by doctor Mark Parker, MaineHealth's Associate Chief Quality Officer.

Heather: To learn more about this project, we invited Mark on the show today. Welcome, Mark.

Mark: Hi. Thanks for having me.

Mike: We're glad you're here. And first off, tell us a bit about your role, the associate chief quality officer. That just sounds like a huge and kind of intimidating job and a big responsibility.

Mark: Yeah, not so intimidating. It's something I've grown into. I'm a nephrologist by training, and I've been in Portland and at MaineHealth for about 25 years. And about 7 or 8 years ago, I became the vice president for quality and safety in Portland. And then more recently, in the past two years, I've taken on some system responsibilities. And so the change in title, specifically in quality and safety, we have oversight for such functions as risk management, patient safety, performance improvement, quality reporting, accreditation, patient experience, and more recently, I have taken on specific roles at the system level for strategy around patient safety and continue to focus on all aspects of quality and safety in the southern region. So, the people I work with most closely are my boss, the Chief Quality Officer for MaineHealth, doctor Omar Hassan, and then a partner in crime in the southern region, is the regional VP for Quality and Safety, Helen Troy.

Heather: Fantastic. So, all of your work with risk management and patient safety sounds like it led you right to your project. Let's dive right into your care variation project from the symposium. As I understand it, the goal is to reduce misdiagnosis. What inspired you to tackle this problem and really take it on?

Mark: Yeah, well, it emanated from work we've been doing over the past year at MaineHealth around what is nowadays termed diagnostic excellence, which is a little bit of a flip from the years when it was referred to as diagnostic error. But it's talking about the same topic. And I've worked very closely in recent years with a colleague in Portland, doctor Bob Trowbridge. He's a hospitalist in Portland who also is a nationally respected expert in the area of preventing diagnostic error. We've had a lot of interest in this amongst clinicians and MaineHealth leadership, and it led us to form a committee in late 2023 to develop projects and efforts to reduce diagnostic error, and the group got together, utilized various resources such as the Safer Diagnosis Checklist, which is promoted by many healthcare safety organizations, and also the Leapfrog Safety Group recommendations around safety. And we came up with a few different areas that we wanted to focus on. And this project that we're talking about today was one of them.

Mike: This project within this whole big field of risk or misdiagnosis reduction and improving diagnostic excellence, this one touches on getting timely and accurate lab results. Is that right?

Mark: Yeah. That's correct.

Mike: Can you tell us why, among all the other endeavors, you chose this particular approach?

Mark: Yeah, sure. When you think about diagnostic excellence, there are a variety of ways you can approach it, but there's two broad categories of concerns in helping clinicians to come to the correct conclusion and make the correct diagnosis. One of them is the issue of inaccurate or delayed information, which leaves the clinician lacking the tools to come to the proper conclusion and to do it in a timely manner. And the other area of diagnostic excellence that we focus on is cognitive reasoning. In other words, the clinician has the information available, but for whatever reason comes to the incorrect conclusion. We have projects going in both of these areas, but specifically to the area of inaccurate or delayed information. We wanted to get into a project that would hopefully lead to relatively quick success, and there are a number of ways you can attack the problem of delayed or inaccurate results, and a number of areas you can look at, not just laboratory reporting, which is what we're talking about, but you can also look at it in terms of imaging, reporting and other types of test results. The reason we got into critical lab results reporting in part, was because there's already been some effort in this area, and we decided to kind of collaborate with colleagues who were already doing some work in this area and help them advance what they were trying to do and hopefully advance our overall desire to improve diagnostic excellence. We began partnering with some of these colleagues, such as Doctor Robert Carlson, who's the head of Nordx, and some of our leadership colleagues at MaineHealth Medical Group, who were already beginning to tackle this problem.

Heather: Fantastic. Building on that a little bit, I'm wondering if we can dive into some details. I'm wondering about really the scope of the project and where you are in the timeline.

Mark: We're still pretty early. We formed a steering committee just in the past few months. Earlier this summer, we identified what areas we want to work on, and as I pointed out, we had some areas where there's already been a head start. One area where there was a head start is in after-hours critical value results for practices that don't have on call providers. And actually, this was a revelation to me. I've been working in this medical community for 25 years, and I did not realize that there were some small medical practices within MaineHealth who don't have on call coverage in the overnight hours or sometimes on weekends. So you get an answering machine and it tells you, you know, go to the emergency room or do some such thing if you really need immediate assistance, but you don't have any way to get a call back from a provider or have someone acknowledge a test result. This problem was already recognized at MaineHealth Medical Group, and there have been some leaders at MaineHealth Medical Group who were beginning the work of identifying these practices and then, more importantly, developing some routing paths for critical lab results after hours, so that our colleagues at Nordx have someone to speak to and to handle the immediate need if there is a critical result out there after hours. That was one area that's part of the scope of the project.

Mark: Another area was brought up by my colleague Doctor Carlson specifically, and that is that there is in existence a closed loop reporting system related to the new Beaker laboratory module in Epic, which would improve some efficiencies in the timing and response to lab results that are critical. The current process, both of you are probably somewhat familiar with given your own work, is that after hours if a lab result is flagged as critical and requiring reporting, it will be called by a Nordx employee or staff member out to the person on call. And that leads to a lot of difficulties with some lab results. One of the examples I cited already is when there's no actual person on call for a practice, but even in the practices that have call coverage, sometimes the call coverage is listed incorrectly, sometimes the person who is receiving the call is managing multiple other problems and doesn't call back the page immediately, leading the Nordx employee to then, you know, continue to call back and take a lot of valuable time. And so this closed loop reporting system works within Epic to actually automate the contacting of providers with results through Haiku or Epic Chat. And it allows for acknowledgement of receipt of the results so that the Nordx staff members knows that the result was received. This would not only get the results out relatively quickly and accurately but would also cut down on the churn and the effort to get the results acknowledged.

Mark: This does require a high-fidelity call schedule where focusing on the ambulatory sphere initially although we hope to eventually work also in the inpatient sphere. We're going to start with 1 or 2 small practices to just pilot the work and work with our IT colleagues to get that piece moving. Those are probably the two biggest pieces of the initial scope of the project. We do have some other pieces that we're pulling in, including some further discussion of the discrimination of criticality. And what I mean by that is Nordx has some standards around critical lab results and what defines them, and what are the levels for critical results. There is some ongoing discussion about what's truly emergent and what's urgent, i.e., what needs to be discussed now and what could be discussed hours from now. So there may still be some opportunity in that realm. And then one last piece of the project that we're currently thinking about is Point of Care critical lab results. So ones that don't originate out of the Nordx laboratory but are maybe local lab results that are done through local testing, such as we have done in our some of our anticoagulation clinics with warfarin results, prothrombin and AHS and the like, and how are those critical results reported accurately and timely?

Mike: Wow. Wow. This is an incredible project, Mark. So many moving parts, right? So many things that you don't necessarily have control over. And a lot of stakeholders. That brings up for me the question of what potential pitfalls do you anticipate and how do you plan to avoid them or addressing them as you move forward with this project?

Mark: Right now, it's a really good question, Mike, and it kind of gets to how we how do you approach quality improvement or performance improvement projects in general? And I think one of the things you do right at the beginning is you make sure that you A., Have the right stakeholders at the table, and B., That you truly have people truly engaged and feeling like this is an important problem that they truly want to solve. Because you'll get down a lot of rabbit holes if you don't have those two pieces aligned. I think we've established already that we've got a fair amount of engagement. There's a lot of enthusiasm around this. We've got participants from all our local health organizations across MaineHealth. We've got a lot of investment from the MaineHealth Medical Group leadership. Nordx and their leadership have been very interested and involved in this obviously. And we've got some initial partnership with it. So, I think that's a great place. You know, we're in a great starting place as far as that goes. The second thing is, you know, sticking to the scope of the project and starting small and then scaling it up because this is, as you can imagine and understand, this is a big problem, and it's not limited to the areas that I've already talked about. Critical lab results reporting spans the entire spectrum of medical care, both ambulatory and acute care. It spreads into transitions of care. And there's a lot we need to do ultimately, but as I said earlier, right now we're kind of biting off what we can chew initially. We're looking for the quicker wins. And we're going to pilot in a small way and then see what we can scale up and spread. I think our pitfalls would be if we don't do those things, if we try to sort of boil the ocean right off the bat and solve the entire problem in one fell swoop. And clearly, it's not the type of problem we can do that for.

Heather: That's fantastic. So you had already you've already mentioned like you're looking for the quicker wins. I'm wondering what you think you need to make this project successful and really, what is going to be your measure of success?

Mark: That's a good question, Heather. I think as with any project, we want to identify the 1 or 2 singular aims that we're trying to get to and then understand how we're going to measure it. And if you can measure something, then you have a good chance of identifying if you've been successful. There's a few different ways we can measure this. One of this, one of the ways is through adverse event reporting. Currently we use the RL data system for adverse event reporting, and we get quite a few reports of critical lab results. Admittedly, one of the difficulties with adverse event reporting systems is they are voluntary. So if you don't report it, then you don't know it happened. But we have a fairly sizable trend regarding critical lab results reporting issues. And we'll follow that trend and we'll follow the nature of the reports over time. And this will give us a sense of whether we've truly made improvements. We can also get another sense of this through provider satisfaction, through provider engagement. It's a little hard to tease out the impact of this particular project on provider engagement relative to other things that impact provider satisfaction, but this is a this can be a big dissatisfier for providers when they're not getting their lab results in a timely and accurate fashion.

Mark: Following the results of engagement surveys and some of the comments that are made in engagement surveys will help us understand our results as well. Patient satisfaction would be an ideal way to measure this, but again, it's very hard to separate the impact of one small project like this from all the other things that influence patient satisfaction responses. While I had aspirations of also looking at patient satisfaction scores and survey results connected to this project, I also temper it in my mind with the fact that it's going to really be hard to make a clear association or a cause and effect when we do things here and how it looks with the patient experience scores. As far as what we need to make the project successful, I think it's the engagement with the right team members. So the Nordx team, the IT team, a lot of our after-hours providers and how we work with the individual practices to have this set up. I think those are the things that will help make us successful.

Mike: Wow Mark, this is a fabulous project and so important in the way it touches so many domains, from the provider experience to the ability to deliver excellence and really achieve what we all aspire to as providers. And in the end, to really take good care of our patients. We are thrilled to have you explain this and bring this forward and are so grateful for the work you and your team are doing. And I do have one last question as we conclude is to our listeners, to people like me and Heather, do you have any advice for others who might be thinking about taking on an improvement project like this?

Mark: Yeah, I think it goes back to some of the things I said in one of your earlier questions. I think it's having the right stakeholders, the right people at the table, making sure that the people that you involve in the project are truly engaged and invested in the outcome. Sticking to the scope of the project. Asking a lot of questions. Start small when it's feasible and those things that are successful, scale them up and spread them and use the PDSA cycle, the "plan do study act" cycle meaning look at things in an iterative fashion, look for rapid cycle change. We're not doing the original approach to intensive research here. We're looking for relatively quick wins. And so use those PDSA cycles to help define what's going on, where the improvements need to occur, how to continue the improvements and then how to scale and spread the results.

Heather: Wow. Well, this has been so interesting and really inspiring. Thank you so much. Mark, for joining us today.

Mark: Oh thank you I really enjoyed it.

Heather: Thanks for listening to BACON this month. You can find all our episodes on your podcast app and at our web page, MaineHealthACO.org/BACON. And if you have questions, comments or suggestions, we'd love to hear from you. Please email us at bacon@mainehealth.org. That's Bacon@mainehealth.org.

Mike: BACON is produced by the MaineHealth Accountable Care Organization. Thanks for joining us. See you next time.

Heather: See you next time.