Compass on ‘consumerizing health care’ and more

June 2022—What stood out among all that was seen and heard at the Executive War College? Compass Group members who were there answer CAP TODAY publisher Bob McGonnagle’s question in their early May virtual get-together, shortly after the War College took place. Here’s what they and other lab leaders said about retail lab testing, digital pathology and artificial intelligence, and their plans for the future.

The Compass Group is an organization of not-for-profit IDN system laboratory leaders who collaborate to identify and share best practices and strategies.

Many of you were at the Executive War College, as was I, and I want to share this perspective from the meeting. We have a lot of new lab participants and many of them were there to do deals or learn more about opportunities they could get or fit into. I saw an emphasis too on retail testing, point-of-care testing in pharmacies. Greg Sossaman, what struck you at the War College?
Gregory Sossaman, MD, system chairman and service line leader, pathology and laboratory medicine, Ochsner Health, New Orleans: There was a focus on the strategy for some of these new players. Truvian was one of the companies that presented its point-of-care device, which was interesting. I think in this post-Theranos world there will be a couple of players that come out with a device that works in a clinic-type setting.

There was also a focus on artificial intelligence as it pertains to digital pathology but also other areas, perhaps AI in conjunction with some of these new, smaller devices that could be put in a clinic. There was a newer focus on data and artificial intelligence or computational pathology than I have seen before. There were interesting discussions about anatomic pathology and as it evolves more toward automation and pairing of digital pathology and computational pathology. I’m going to see how those fit in our strategy.

Tony Bull, did anything at the meeting strike you?
Tony Bull, system administrative officer, Pathology and Laboratory Medicine Integrated Center of Clinical Excellence, Medical University of South Carolina: It looks like decentralized routine testing is starting to come to fruition, and I’m trying to understand what the impact will be and how it will affect our laboratories. I’ll be interested to see how it bears out and if the promises are met or if we find it has limitations.

Dwayne Breining?
Dwayne Breining, MD, executive director, Northwell Health Laboratories, New York: The era of digital pathology is upon us. We’re entering into it full steam ahead. COVID drove a lot of that with the demands for people to stay out of the office.

The other thing at the War College that struck me is the disruption and displacement by other players, and by that I mean chain drugstores like CVS, Walgreens, Rite Aid. There seems to be infrastructure going up, putting clinics next to retail pharmacies.

In New York we’ve been seeing a trend away from the classic primary care providers, your family doctor, and more into urgent care, episodic care, you don’t touch health care until you think you need it and you don’t have loyalty to anything. We have to keep a close eye on how that drives testing, and the big pharmacies will want their part of that, too.

Stan Schofield, do you think the momentum toward more point-of-care testing and testing distributed in a region is a pipe dream or is it becoming a reality where you operate in Maine?
Stan Schofield, president, NorDx, and senior VP, MaineHealth: It’s not a pipe dream. It’s happening in large, concentrated cities with high population densities and younger populations. Younger people tend to not have an affinity or an allegiance to a system, and they go when they want something. They want it now and they want it convenient.

Urgent care or walk-in clinics are here, but the drugstores are not taking over anything yet. Our population is still either government paid or you have insurance. There’s a lot of resistance to out-of-pocket payment for services. The younger generation probably doesn’t have that reluctance, but the older population and higher users of these services probably do. Drugstore testing is infrequent here, but I’ve seen it around the country and people I talk to say it’s growing and gaining traction. We’ll be one of the last ones. We have urgent care that took 10 years to get here—after everybody else had it. Drugstore testing will be five years after everybody has it, or at least in terms of it being used frequently and consistently.

The population is driving this because access to primary care is hard. You can’t get in to see a doctor quickly, and when people have something, they want to take care of it. To use urgent care centers with insurance, the deciding factor is the out-of-pocket expense, and that’s going to separate early adopters from later on in this cycle of evolution, at least in my region.

Pete Dysert, Dallas has a lot of socioeconomic diversity—is this a strategic consideration at Baylor Scott & White as you plan for the future?
Peter Dysert, MD, chief, Department of Pathology, Baylor Scott & White Health, Dallas: Our overarching strategy is defined around consumerizing health care. The success of our Epic app and people’s ability to communicate when everything was shut down has built a strong constituency who likes that platform. Anything our system decides is strategically related to supporting the consumer is something we will be asked to figure out.

Our lab-initiated strategy is to be an observer as the forces in the community work these issues out and as the role of the primary care network versus an app-based approach to accessing care works its way through our system.

If you were to learn that a Quest or a Labcorp or a Sonic were planning to put a large network in pharmacies or groceries in the Dallas area, do you think it would be a precipitating event for Baylor to be more active to counteract that competition?
Dr. Dysert (Baylor Scott & White): Yes, we’d take notice of that. The pressure would arise from our primary care doctors as their patients start to demand the convenience of accessing those services over having to come to the doctor’s office. We recently offered in-home immunizations, as one example, through the app, to save people time and play on the convenience of our services.

Dhobie Wong, how is Sutter viewing this?
Dhobie Wong, MBA, MLS(ASCP), CLS, VP of laboratory services, Sutter Health, Sacramento, Calif.: We have full-service labs within our foundation setting, and we are exploring that model. There is heightened demand for more access to test results, receiving results more rapidly, especially in oncology. We’re determining the cost-benefit analysis. Does it make sense to decentralize our testing? What testing is absolutely needed in the outpatient physician setting?

Have you thought about adding rapid testing capability in your patient draw centers?
Dhobie Wong (Sutter Health): We have some testing available in our outpatient setting, including full-service lab tests. Several are co-adjacent with urgent cares, so we can provide some level of rapid testing for our patients and physicians. We are evaluating that model—the cost, decentralization, equipment, redundancy.

Wally Henricks, talk about digital pathology in AI or this question of should we do more distributed testing, and perhaps we need to because we can’t afford to lose patients from the system.
Walter Henricks, MD, vice chair, Pathology and Laboratory Medicine Institute, and laboratory director, Cleveland Clinic: We’ve implemented self-swabbing capability and are looking to expand it. We’re pursuing more rapid testing at acute care centers so it can be done in one place. We expanded and enhanced point of care at those outlying centers. In the future we’ll look at more frequent courier service and improvements in logistics.

Regarding AI, there’s justified excitement and at the same time there’s much hype. Hooman Rashidi, MD, recently joined us and is directing and developing a new Center for Artificial Intelligence and Data Science in Pathology and Laboratory Medicine. Dr. Rashidi had done a lot of innovative AI work at UC Davis and developed an AI platform. He’s going to champion our program regarding applications of machine learning and artificial intelligence in pathology. We think there are opportunities to drive outcomes.

There are a lot of unknowns for laboratories regarding AI and machine learning. What regulations or guidance will apply to labs using these tools? The FDA is actively looking at these. Laboratories might contribute to the rigor of validation of AI/ML tools. We are used to doing that and are tuned in to how to analyze test validation, test performance, and other data. Pathologists and laboratories are well suited to ask good questions about the foundations on which these algorithms are built. Another question is the extent of lab and pathologist responsibility. If the result or interpretation ends up in a CLIA-type test report, then that’s one level of responsibility. When lab values already reported by a lab are used in some other AI algorithm that incorporates clinical or other data that are outside the purview of the lab to generate an output, it’s a different level of oversight responsibility.

We see great opportunity. But what the value is for the investment needs to be looked at rigorously, just like digital pathology in its early days.

There’s discussion about the use of AI in infectious disease and some therapeutic drug monitoring. It’s not restricted to surgical pathology.
Dr. Henricks (Cleveland Clinic): We have active efforts in both anatomic pathology and laboratory medicine. AI is not only about digital imaging, and it will be applied to non-image data sets familiar in clinical pathology.

Sterling Bennett, one of the conversations at the War College in a session on AI in surgical pathology was how many institutions are being advised to credential their pathologists in many states or are doing it on their own, almost as if they were a national AP laboratory. Is that happening at Intermountain?
Sterling Bennett, MD, MS, senior medical director, pathology and laboratory medicine, Intermountain Healthcare, Salt Lake City: As the Intermountain footprint grows in seven states and looks at a fully integrated pathology service, it’s important for our pathologists to be credentialed or licensed in many states. One of the advantages in digital pathology is we can more readily use the subspecialty expertise among our pathologists. To use it fully requires they be credentialed in multiple institutions and licensed in multiple states, and that will open the door to more use of AI.

I heard a definition of the AI abbreviation that I like: assisted intelligence. Fundamentally every laboratorian wants to get it right, and if there are tools to help do that in a better way, whether it’s better optics on a microscope, digital imaging, or assisted intelligence, I think the majority of pathologists will be open to it once they get past the fear of being replaced by a computer.

Pete Dysert, does the shortage of pathologists lead you to think more about digital pathology and AI?
Dr. Dysert (Baylor Scott & White): It does. If you can leverage your investment in that platform across your practice, then potentially you could augment the skill sets of the community-based pathologists. You still have to cover frozens and the outside areas and connect them to our experts. We hope it will be an offset to improve the efficiency and productivity of our existing staff. We’ve struggled with getting everything put in place to allow our histology labs to be able to optimize our pathologists’ productivity, because I think we will face a potential shortage, at least in the short term. And then you have the issue of access to the specialty physicians, subspecialty experts.

I met someone at the War College who is representing a group whose objective is to acquire pathology groups and outfit them with a national digital pathology network and bid for all kinds of work, a little like a NightHawk Radiology. Steve Carroll, what do you think about those ideas?
Steven Carroll, MD, PhD, chair, Department of Pathology and Laboratory Medicine, Medical University of South Carolina: It’s something we’ll be seeing more frequently. The day of the small private pathology practices is coming to an end. A lot of it is driven by surgeons, oncologists, and so forth wanting a subspecialist to look at their cases. Small groups can’t support subspecialists, so you will have to go to a larger group. Therein lies the rub because if you are, say, a neuropathologist, those cases are few and far between, so you have to have a larger catchment area to justify your existence and generate enough revenue for things to move forward. As a result we’re going to see more consolidation of pathologists into larger groups, and the natural evolution will be national and international capture of cases so you can take maximal advantage of that structure.

John Waugh, what do you make of the discussion about greater disparities of care? I’m hearing more about how the local community practice is letting patients down because they don’t have the subspecialty expertise of all the people involved in cancer care.
John Waugh, MS, MT(ASCP), system VP, pathology and laboratory medicine, Henry Ford Health System, Detroit: That dovetails with the digital pathology conversation, and we use a model that has subspecialty pathologists and they cross-cover two or three different subspecialties. It helps with covering meetings, vacation times, tumor boards, those kinds of responsibilities.

Community practice at our community hospitals has more of a generalist need, but often there are areas that seasoned pathologists can come into on that. They can read different cases with great confidence. We have daily conferences in which difficult or interesting cases are compared on digital screens across sites. We can have a multi-group sign-out on a given case. Nobody is out there on their own.

What is your volume of send-outs for consults or second opinions?
John Waugh (Henry Ford): Our send-outs are almost nonexistent. We keep a lot of cases in-house. There are consensus sign-outs that give people a good opportunity to be teachers of each other, to share opinions, to reach consensus if there are challenging cases. The only types of cases we send out are those that need a referee. There are a handful of those over the course of a year.

Julie Hess, tell me about the staff or pathologist shortages at AdventHealth. Are you still involved in trying to improve subspecialty coverage?
Julie Hess, VP, laboratory services, Advent-Health, Orlando, Fla.: Our pathology groups have independent contracts. They are recruiting different specialties based on some of the service-line growth in our hospital system, so as the needs arise they focus on that and work closely with our clinicians. We cover a large network, so we can have more specialized cases funnel into one or two people. We have an advantage because our pathology is centralized to our Orlando campus, so those specialists can sit at that location and it’s efficient.

As we recruit some of our newer pathologists or people new to the field, with our growth at the number of facilities and the need to have medical directors over different campuses, we’re challenged with finding people who want to cover the clinical pathology side and be a medical director, because they were expecting to be more specialized or stick strictly to the anatomic side. It’s something we’re having to balance.

Milt Datta, talk about Allina’s perspective on these topics.
Milton Datta, MD, chair of pathology, Abbott Northwestern Hospital, Allina Health, Minneapolis: We look at our regional hospital lab directors for CP, and we’ve been lucky to find a couple of people who are willing to drive to rural or suburban places. When we do cover operative cases there, it’s usually senior pathologists who are willing to go there and work. We’re expanding our telepathology to those sites for remote reads, advising, everything from the gross examination to the frozen sections.

We’re planning our five-year strategy for the health care system, and the lab is trying to step up the analytics. Mike Laposata, MD, PhD [professor and chair of the Department of Pathology at the University of Texas Medical Branch at Galveston], gave a talk describing the Laboratory 2.0 movement. We invited senior executives to listen to the opportunities in data analytics for population health metrics. One of the arguments we’re putting forth is if we do a good job shifting from reactive treatments of critical lab values to proactive result trends identification, we will take some of the pressure off our frontline medical teams and reduce burnout, and that’s one of the ways we’re selling it as a benefit.

How is the recruitment and supply of pathologists in Minneapolis?
Dr. Datta (Allina): We have recruited extremely well. We’ve recruited several people mid-career from academic centers who started looking at their volume of responsibilities. We’re a private practice group yet we offer subspecialty work and other opportunities for publications and so on that are a little less pressured, and I can’t help but think it’s that.

Wendy Kleckler, what’s your reaction to the discussion of point-of-care testing in grocery stores and pharmacies?
Wendy Kleckler, VP of Business Development, ACL Laboratories, West Allis, Wis.: We do a fair amount. We’re selective on where point-of-care instrument assays are located. Our medical groups are employed by Advocate Aurora Health, so we have a little control over that. We have point-of-care testing in some of those clinics. The lab oversees what we purchase and what tests are done, but there’s an Advocate Aurora team that determines who runs the point-of-care in those systems. The lab’s not completely responsible. We have clinics within Walgreens but no lab testing. Those clinics are staffed with nurse practitioners and some medical assistants. They do collections and our couriers pick up samples and bring them to the laboratory.

One focus for us is how we reach the consumer. Our young people don’t want to go to the doctor. They would rather walk into a pharmacy and pick up a kit from the shelf. How can we look at direct-to-consumer from a population standpoint? It would allow us to go outside our region in Illinois and Wisconsin and might give us a bigger bandwidth. I’m on a strategic planning team to identify how we can do that. Can we build it? Do we buy it? What does it look like for the future so if a kit was sent home for a lab test, it would come back to ACL and we could perform the testing and get results out via our LiveWell app?

Are you in the initial stages of that effort?
Wendy Kleckler (ACL Laboratories): Yes. One of the key markets is our Medicare Advantage population—how we can get, for example, colorectal fecal immunochemical tests [FIT] to market so they don’t come into the office prior. We’re starting with a pilot population of FIT tests in that patient population.

There’s a notion that we need to glue the consumer to our health care system because that’s where the big bucks come from. The belief used to be that we got inpatients through the emergency room or through the affiliated doctors. Wally, what are you doing in Cleveland to keep patients coming to the clinic and from not going to places like Houston or San Francisco?
Dr. Henricks (Cleveland Clinic): We have significant marketing outside of northeast Ohio. We have a spot on the Super Bowl each year and ads on huge walls at the airport. We have different ways to get the name in front of people in unexpected areas.

There is a lot of community outreach, more than there’s ever been in my time here. We’re also partnering with larger employers for specialty care, especially in cardiac medicine, providing opportunities for large self-insured employers to have Cleveland Clinic as a primary or secondary consultation that’s covered by their plan. Our electronic consultation program and virtual visits support these efforts.

Sterling, what are the practice models of pathology within Intermountain?
Dr. Bennett (Intermountain): We have one group that has about half of our pathologists, but we have multiple affiliated groups. We have one small group employed by Intermountain. It’s both a generalist model and a mixed generalist-subspecialist model.

Is there a system push to have one kind of model within the system?
Dr. Bennett (Intermountain): We don’t have a clear, consistent read from the organization about one model. But we think as we enter the digital pathology age, it would be beneficial to have a single model.

Dwayne, I came away from the War College in part thinking there are a lot of forces urging us toward one model for pathologists serving in a network. Did you come away with that same impression, and if so, what do you think will happen at Northwell?
Dr. Breining (Northwell): All roads are leading to a subspecialized central service. In Michigan they put forward a model of the pathology hospitalist who can cover frozen sections and AP/CP demands, but then you have a mothership of subspecialists that can be readily consulted. [See CAP TODAY, April 2022: “Pathology hospitalists in place at UMich.”] Digital pathology will facilitate that as it gets established.

It will be helpful because maintaining an equal high standard of care and equal access to subspecialists when you need it will become more important as different cancer services get regionalized and need to be provided at a high level, like some places have done with cardiac programs. It’s already starting to happen for cancer, and that’s the way we’ll integrate.