July 2024—Instruments, assays, patient flow, and IT were some of what laboratory directors and IVD company representatives spoke of when CAP TODAY publisher Bob McGonnagle convened the group online on May 3. Here’s what they said about what instruments and assays are deployed where and the considerations that come into play in large health systems.
CAP TODAY’s guide to chemistry and immunoassay analyzers for mid- and high-volume laboratories begins here. The guide to analyzers for use at the point of care and in low-volume labs, in the June issue, is at https://bit.ly/CT_0624-PG.
During last year’s roundtable on chemistry and immunoassay systems, there was an emphasis on labs needing to be agile and adaptive and to know they might be a hub or a spoke in this environment of consolidation. Being nimble and flexible is important to being able to grow and scale, particularly being able to deploy instruments as conditions dictate.
Ryan Stephens, talk to us about the aging population, which has deep significance for the workforce and the patient population you’ll be testing, and consolidation. How do you see it coming?
Ryan Stephens, group marketing manager, automation and core lab digital solutions, Roche: One challenge with an aging population is access. How do you ensure access to laboratory medicine for this population, especially in rural areas where the draw station or closest lab might be a couple of hours away and they might not have reliable transportation. How do we work with new services that are coming out with at-home testing or sampling and integrate it into our testing practice? How do we provide access as staffing becomes more difficult? As the population ages, monitoring them with laboratory diagnostics will be increasingly important so you can get them the right therapies and treatments and help improve their health, which will ultimately reduce the overall cost of health care in the U.S.
In thinking about growth, flexibility, and the pressures our customers face, it’s important to think about how we support laboratories of different sizes, whether it’s a low-volume community hospital that needs a broad menu because it’s the only health system within a few counties, or a satellite site, a smaller hospital that’s part of a large integrated health network that sends most of its samples to a core laboratory. They still need stat capabilities and to operate their ED. We also have to support mega labs that are getting bigger by the day.
There are two big industry trends. One is health systems are getting bigger because of economies of scale, and cost and reimbursement pressures are causing them to have to join other systems. The second is large reference labs are buying up and partnering with health systems. We’re focused on offering a broad spectrum of solutions that meet each of those needs.
Allison Chambliss, UCLA has engaged in a lot of consolidation, acquisitions of physician groups and health care testing sites. Can you add to what Ryan said?
Allison B. Chambliss, PhD, associate clinical professor and director of laboratory stewardship, Department of Pathology and Laboratory Medicine, David Geffen School of Medicine at UCLA: We have three growing central laboratories across Los Angeles County, but we’re always looking to increase access for our patient population that expands the map of where our health system and medical centers are. We have expanded our draw locations over the past several years but not where our laboratories are, so we need to make sure we can stabilize specimens for testing and prepare them for potentially long transport times, or offer simple, onsite point-of-care-type testing when there’s a clinical need to catch the patient where they are and get a result right there.
We recently acquired a moderately sized hospital in our area. It will be a huge project for our health system and laboratory department and a challenge because we don’t have expanded resources to work through the acquisition. We’ll work with their laboratory to get them standardized to our approaches, whether it be analytical testing, reporting of results, or getting into our LIS because they’re not using the same LIS and EHR as we are. Our LIS, QA, and laboratory leadership staff are working through that and we don’t have extra people to do it.
Martu Richards, what are you hearing from laboratories as they look at instrumentation, and what are your thoughts on the need for consolidation and how to deal with the laboratory staff shortage?
Martu Richards, marketing director for Atellica instrument acquisition, diagnostics, Siemens Healthineers North America: We hear the same concerns from laboratories regarding the need for consolidation and standardization and how we can support their efforts in terms of their staff being able to multitask—if you’re in a network, can you work in a core lab and at one of the spokes? In our development process and with the products we’re bringing to market, we’re thinking about how we can address standardization. Can we redefine standardization? We’re familiar with total and task targeted automation. We know about total lab automation in terms of a larger track. How do we redefine automation in a way that these options are available not only to network and reference labs but also to smaller laboratories? We do that through our platforms and menus.
We’ve been talking about core labs for 15 to 20 years and the testing that belongs in a core lab for cost and other efficiencies, yet we’re constantly thinking about how we distribute testing in networks such as to smaller hospital labs, clinics, doctors’ offices, and EDs. It’s something you have to come to every day. David Alter, what are your thoughts about this?
David Alter, MD, director of clinical chemistry, Department of Pathology, and associate professor of pathology and laboratory medicine, Emory University School of Medicine: The holy grail is a core lab that’s not on the perimeter of your system but is centrally located so your couriers can be equidistant from all your associated hospitals. Even at Emory we struggle with that because our core lab for outpatient testing is at the university hospital, which is around the corner from Emory University. To go to our next largest hospital, which is not far away, it still takes 15 to 20 minutes when you factor in traffic and parking.
Emory is fairly standardized across analyzers and instruments. Even though there is a core lab concept with some tests performed at the main hospital, there’s been a growth of testing focal to certain hospitals based on their patient populations or because they have more bandwidth. For instance, the immunology lab is at the main campus, but we’re on the pipeline to implement more immunology testing at our next biggest hospital because we don’t have an obstetrics population at the main hospital but do have one there. So we have to do more testing for syphilis, HIV, and other infectious diseases in the turnaround times they require. I’ve seen here something that sounds antithetical but breaks the paradigm in terms of moving tests to other locations that have the bandwidth to handle those tests and appreciating that you’re going to have to make concessions with regard to hospital B needing more timely HIV testing, for example. If our holy grail is a hub and a wheel, it looks like the wheel is getting bumpy.
My ask of in vitro diagnostic companies is to recognize that if labs or hospital systems adopt what I call a hubbed wheel, we need testing done on the periphery using smaller versions of the primary instruments so we don’t duplicate our resources.
It’s as if we’re in a constant triage—we bring tests into a large core lab and then find we need them concentrated in another site as well.
Ryan Stephens (Roche): With the hub-and-spoke model, we think about not only where the testing is but also where patient flow is. Are patients flowing the same way the tests and samples are, meaning are specialties being put into different hospitals? If a patient goes to a rural or satellite site and it’s discovered they have an advanced disease and need to go to a hospital with greater capability or a center of excellence, they oftentimes will be tested again. It’s critical to have standardized testing, reference ranges, and reagents across the health system. Roche is focused on having a standardized family of analyzers. The Cobas Pure, which is perfect for smaller sites, uses the same reagents and has the same reference ranges as the Cobas Pro, which offers flexibility for higher-volume laboratories. You can customize it and it’s modular. That has been a key investment for Roche and a philosophy. It’s driven by patient flow.
There’s patient flow and sample flow; those two can be distinct. The infrastructure needed to move things physically, and as a matter of data, comes to the fore. Martu Richards, do you agree?
Martu Richards (Siemens Healthineers): Absolutely. Siemens Healthineers is looking at it as a new way of working in terms of the instrumentation. Within the last year in the U.S., we added the Atellica CI analyzer to the Atellica Solution platform. The menu and software, reference ranges, et cetera, are standardized. We are redefining standardization in hub-and-spoke laboratories by using the same sample management on Atellica Solutions that are in larger laboratories that provide, for example, automated sorting and archiving on larger track systems. We’re moving those to the smaller instrumentation, the Atellica CI, without having to go to a larger track. It’s a new way of working for the laboratory, because the lab’s thought process is, “We need total lab automation.” And we’re saying we can provide those same functionalities.
As the industry continues to mature, the desire to manage data succinctly will get stronger. Many manufacturers in the marketplace, including Siemens Healthineers, offer data management packages, whether to manage the instrument or data across QC or patient moving averages. It’s important to have a conversation with the customer to figure out what they need and tie up the specific solutions with how their laboratory works, how they’re set up, because there’s a lot of flexibility in the data management tools.
Ryan Stephens, can you comment on how new assays such as for Alzheimer’s disease and traumatic brain injury are factored in?
Ryan Stephens (Roche): We need to ensure the new technology, the new assays that vendors are developing, especially around Alzheimer’s and traumatic brain injury, go to the right locations based on need. The same technology is in the Cobas Pro and Cobas Pure; it can allow a test, once approved, to go on the Pro for a high-volume laboratory, or if it’s needed in a satellite site, to be available on the Pure as well. As a lab’s patient mix changes and the disciplines change at each of the health systems to optimize their businesses and support their patients, we have the flexibility to support them with our systems.
Allison Chambliss, the flexibility and new assays are great, but these decisions aren’t made alone. You must spend a lot of time thinking about where to put what. Should we offer the new assay in seven locations or three? Take us through the story of your day as you look at a new assay for the system.
Dr. Chambliss (UCLA): I like the idea of following the patient rather than just the sample. Sometimes we have to do that in a reactive, rather than proactive, way. They don’t always consult the lab when they think about introducing a new practice, changing a practice location, or opening a clinic with a specialty focus. We might have a little lead time or none at all, so we have to be flexible and nimble. Having standard modular platforms of different sizes can help us, whether it’s to introduce new tests or shift tests between locations. Having new tests available on standard platforms rather than having a standalone, boutique, one-test-only type of instrument is what we’re looking for.
Consolidation can be good for certain tests, such as manual tests, tests that require expertise, like protein electrophoresis, whether it be the instrumentation used or the interpretation required by a pathologist. But redundancy can also be good. When it comes to the tests ordered in multiple locations, we have some redundancy between our lab locations. We’ve consolidated a couple of tests to our outreach reference lab or main hospital location. We also have tests that we run at all three locations, and when there’s an unexpected downtime or we run out of reagent, we can pivot and send samples across town to the other laboratory to have them run on the same platform with the same reference range and analytical accuracy and performance. This model of completely consolidating or only performing tests at one location is no longer the way we want to go.
There has to be a lot of dialogue not only among laboratory leadership but also clinical staff to make sure there’s going to be satisfaction with the way the lab distributes this testing. I hear more and more about people spending time in meetings with chief medical officers and service heads so that lab needs are closely attuned to clinical demands. David Alter, what’s your experience with that at Emory?
Dr. Alter (Emory): I’m a little separated from those discussions but I have seen some. For instance, we are a primary Beckman site and have BNP throughout. One of our sites has Roche and has brought on, through a cumulative agreement, NT-proBNP for patients who are on Entresto. It’s an example of how we’ve been nimble in addressing something like that. The esoteric markers, such as for Alzheimer’s, for now will be the provenance of the main hospital where the critical mass of expertise is among the technical staff.
The real winners over time will be independent companies like BioPorto Diagnostics, which developed an NGAL assay and worked closely with IVD companies so the assay could be adapted to Roche, Olympus, Beckman, Siemens, and so forth. People are uncomfortable being tied down to a platform because there are a few assays they can’t find anywhere else. I was impressed that Emory brought the NT-proBNP onboard to address the concerns of the cardiologists because we were lucky enough to have another site that had the ability for the assay.
Martu Richards (Siemens Healthineers): On the issue of having to use NT-proBNP and BNP in the same network and being able to give your cardiologists that choice, one of the ways industry can support that is with the approach Siemens Healthineers has taken by making those two assays available on the Atellica Solution and the Atellica CI.
Ryan Stephens, your thoughts on this?
Ryan Stephens (Roche): If you look at the menus across the major IVD vendors, there may be a 15, 20 percent difference. Each vendor specializes. We’re focused on cardiometabolic and neurology, for example, but we also need to be flexible when we partner with other companies, not only with the assays but also automation. It is hard for each vendor to offer every test and so we are seeing an increase in demand to connect third-party vendors to both our automation and digital products. With our digital solutions, Cobas Infinity, which is getting rebranded to Navify Lab Operations, we want to use that platform for laboratories to be able to connect with other analyzers, other companies’ products. Our Navify products can help labs with operational simplicity and efficiency.
We’re talking about analyzers and traditional analytics, but really it’s about the total solution and how you approach your lab business and lab medicine. It’s not just analyzers and reagents; it’s automation, it’s digital, it’s how we integrate with LISs and make sure it’s seamless. We can lighten our customers’ burden by having a robust solution that enables better interoperability and fewer workstations.
Much of the burden for optimal data handling resides in the middleware, and the instrumentation company, which didn’t necessarily sign on to do this job, is the principal agent of this data handling and integration into the system. Martu Richards, do you agree?
Martu Richards (Siemens Healthineers): I do. It’s another opportunity for manufacturers to help laboratories in this new way of working. For example, how many of those functions can we put on the instrument itself without needing middleware? Because middleware is an additional cost for the laboratory and an additional responsibility to manage. Some of the things we’re doing uniquely on our user interface or in on-instrument software involves the lab evaluation studies that require middleware or off-the-shelf software to manage lot-to-lot comparison data. We’re bringing that into our software so laboratories don’t need a second or third product. How much can we provide on the instrument without the lab having to add an additional instrument or additional software? You will see more of that from Siemens Healthineers as we move forward.
One of the concerns you hear at the lab informatics meetings is how to get LIS results into the EHR in a way that’s useful for the ordering physician. Many of those, as you see in the display, are disappointing and even misleading sometimes.
We have wonderful machines and we’re getting great results that we’re putting into sophisticated, expensive LISs. But a physician and others will look at it on an EHR, not an LIS. Allison Chambliss, do you see that dilemma at UCLA?
Dr. Chambliss (UCLA): We do. One of my major focuses and interests at UCLA is laboratory stewardship, and to practice laboratory stewardship we rely on data from many sources. It’s a struggle to pull those pieces together from different sources and ensure we are getting an accurate picture of the data and that we’re interpreting it correctly across the different systems in order to identify areas of priority and need, whether it be over- or underutilization or to support the laboratory industry in demonstrating laboratory medicine’s value, which is a term that’s used often but with little concreteness to it. We don’t know what it means or how to do it, and a big part of that is we don’t have the data in one place. I would love to see a consolidation of the data. Can we have information about the platform that the lab result came from? That’s a big piece of it. As the number goes into the EHR and people outside of the laboratory use that data, they’re not thinking about what instrument it came from or what method it used. If I pull the same data from two years ago, was it the same system? Can I compare those numbers? I would like to be able to see it all in one place and compare and show quantitatively the value of laboratory medicine.
Dr. Alter (Emory): Complicating that is the IT department, which is often short-staffed with many more projects on its list than can be accomplished. Since they have to prioritize the projects, they might push back when we ask them to build a separate result for the same test for a different manufacturer. I’ve made such requests over time and their response is that it’s going to take too much bandwidth, too many resources, none of which they have. It would be helpful if Epic came to us and said not only have we worked with hospitals that have Roche analyzers but we have an expert in our company who is able to navigate Roche instruments, or our expert is connected with the IVD companies so is comfortable working with Roche, Beckman, Siemens, and so forth. Instead of: “We have experience at other hospitals with this analyzer line. Let’s hook you up with them and see what they did.”
It’s an excellent point, but in fairness to Epic and to Cerner, they have a lot of staff devoted to supporting many clinical departments and in particular laboratory. There’s a laboratory working group in clinical and anatomic pathology that convenes fairly regularly at Epic’s headquarters, and Cerner has a similar entity.
Years ago, labs had more internal IT operatives than they do today. They have been taken away from the lab and put into centralized IT, and I think that is a rub that David is well aware of and it’s good to call out. Martu Richards, do you agree?
Martu Richards (Siemens Healthineers): I agree completely. Data management programs, like the middleware that comes from various manufacturers, have blossomed in the past few years because manufacturers are taking on more because hospitals don’t have the resources. Dr. Alter, my recommendation is to talk to the manufacturers you’re working with because we all have deep data management and informatic solutions and are so customized to your needs in the laboratory that we can bring in expertise to help you in specific areas.
Dr. Alter (Emory): I don’t want to say that Emory is getting a short shrift and we’re not asking the right questions. We are. I just want to emphasize that it can only get better.
Ryan Stephens, some people might believe that some of these interfaces, which traditionally were difficult, are now more simplified with application programming interfaces and advances in IT. No one would contemplate a large engagement with a major IVD company without having the IT solutions front and center as you discuss these engagements, correct?
Ryan Stephens (Roche): Yes. With the challenges that hospitals are having around IT, it’s a best practice to bring them in early. There should be work with the vendors on how to approach the IT group to make sure we have the right information, because it is complex. The centralized resources are spread thin because they’re focused on other priorities throughout the hospital.
Allison Chambliss, can you give us a final comment on this topic?
Dr. Chambliss (UCLA): When there’s a new instrument that requires a new interface, we try to get IT onboard early. Everyone knows they need to be, but there are many competing priorities outside of the lab. The earlier we can plan and set a timeline and have resources the better. We want to know if another laboratory that is using the same instrument and same EHR has already gone live with this. What can we learn from them to make this as streamlined as possible?