IN PERSON
 |
Joseph Meyer is vice president of
marketing for the centralized diagnostics business unit at Roche Diagnostics (Indianapolis). He is responsible for setting and executing the company’s U.S. marketing strategy and communicating product and market requirements. He can be reached at joseph-f.meyer @roche.com. |
To learn about the challenges facing both instrument manufacturers and their laboratory customers, IVD Technology editor Richard Park spoke with Joseph Meyer, vice president of marketing for the Centralized Diagnostics business unit of Roche Diagnostics (Indianapolis).
In this interview, Meyer discusses the changes overtaking central clinical labs and how new products and technologies enable them to cope. He talks about how the job of the automation salesperson sometimes involves demonstrating the value to the wider healthcare system of automating the lab.
IVD Technology: What have been the biggest technological advances in lab instrumentation and automation during the past few years? And what are the latest trends?
Joseph Meyer: The drivers for the development of both instrumentation and automation have consistently been the amount of skilled labor available to laboratories today, and the cost of that labor. The skilled clinical laboratory workforce is aging. Instruments have to be simpler in order to be operated by new, less-skilled, laboratorians. And the need for automation will continue to increase in order to reduce the amount of labor required per test for samples that come to the laboratory.
A couple of technological advances come to mind.
One is to bring clinical chemistry and immunochemistry testing together on the same platform so that essentially one sample can be tested for everything ordered for that sample or for that patient. That can effectively reduce the number of samples handled by 30%, plus or minus, depending on the testing split.
On the automation side, there haven’t been significant recent advances, conceptually, in what automation has provided. Taking a sample in a tube and—without a technician doing anything to the sample—spinning it down, centrifuging it, and delivering a sample volume of serum to an instrument for testing—none of that has changed. The advance we’ve seen is to be able to provide that automation—that front-end piece—in a more flexible way so that the lab doesn’t have to reconfigure itself around automation. Rather, the automation solution can be configured to fit the lab. This allows the cost of automation to come down so not just the largest laboratories can afford it.
Another big trend in automation has been for systems to be what is called open, built so that you can configure the automation so as to have more than one vendor’s instruments connected to it.
You mentioned trying to make instruments simpler to accommodate the shortage of skilled labor and the aging workforce. How have IVD manufacturers such as Roche Diagnostics addressed that issue?
We’ve done a couple of things. One is to produce instruments that operate in the way I suggested—like our Cobas 6000–series analyzer—instruments for laboratories that have not previously been able to get chemistry and immunochemistry together on the same platform. Labs using these instruments can load one sample and have both of these panels of tests provided. So, Roche has responded to the labor issue by designing an instrument from the ground up that delivers integration of the basic existing technologies.
Our automation R&D has much the same driver—the sense of creating a solution that can be configured to fit the lab and not require the lab to be configured around automation. Our customers are rarely building a new laboratory. Often, we have to make automation solutions fit into old structures. Despite the increase in construction for hospitals, there are a lot of aging hospitals in the market today—buildings that are 30 and 40 years old. Our response was to build a very configurable system.
Delivering Added Value
What are the primary challenges that IVD manufacturers encounter in designing and developing lab instrumentation and automation? And how do they overcome them?
The size and performance trade-off for a given instrument or a piece of automation is one. Another challenge we take seriously is to build instrumentation and automation that is easy to install and maintain, and to use. Laboratories are 24/7 operations. Given the shift in the skilled labor pool, along with the heavy activity, the automation systems in them have to be fault tolerant and robust.
They have to perform well, and with very high reliability. And when something does require operator involvement, it has to be designed in a way that’s simple enough for existing laboratorians to effectively maintain and use these systems.
Another challenge comes from workforce mobility. Within an integrated health network, employees might work in different laboratories. Also, it’s common nowadays for people to have a second job elsewhere than in their home hospital. Our ability to provide software that’s consistent across a portfolio is quite important, so that these people don’t have to learn a new system for each lab.
When we take some of these automation solutions to lower-volume labs, we very much have a case to prove on the economic side. The capital outlay for these systems is not insignificant.
So, one challenge for manufacturers is to present this as a solution for the many interested smaller labs by appealing beyond the clinical laboratory. We have to demonstrate to a hospital or a health network a value exceeding what the laboratory alone might see. For example, if it takes six or seven fewer laboratorians to staff a lab that’s automated, fine. But if automation can help reduce adverse patient outcomes through improved quality and safety, and improve emergency room turnaround times, then the manufacturer has created an economic and clinical benefit for the hospital.
Therefore, a challenge we have in selling in the laboratory itself is to connect those dots. It’s not always easy to do, because the lab has at times an independent profit-and-loss view of its role in the hospital.
I think the general challenge for IVD manufacturers is to constantly think about the state of the business. We have to ask ourselves, if smaller laboratories are interested in automation, how do we build flexible automation that delivers the real value those laboratories are after? Roche has done that. In R&D, we prioritize those market needs.
Have clinical labs been demanding a demonstration of added value in instrumentation and automation? Could manufacturers address such requirements by increasing the technological capabilities of automated systems?
The laboratory is under a lot of pressure, especially for STAT testing for cases like the chest pain patient who arrives at the emergency department. Roche and other companies have been able to demonstrate the tremendous value of automation as it relates to turnaround times for emergency cardiac testing. That value is self-evident. Being able to get a troponin T turned around predictably in 40 minutes or less in a central lab with test quality and control much higher than possible with a point-of-care device represents tremendous value—to the emergency department and generally.
Roche as the supplier helps labs get from the point of accepting the concept of automating to actually deploying an automation system. There’s a lot between those two points. The value-add from Roche is to provide the service teams, the project management teams, the consulting teams that help that laboratory go from where they are today with an unautomated, typically kind-of-chaotic environment to one that is smoothly running once automation is installed. That’s a service we see the marketplace demanding of us.
Product Development
In developing lab instruments and automation, do manufacturers aim for something completely new very often, or do they tend to build on current offerings through upgrades, modular additions, and the like?
In general, the industry builds on what it has in the marketplace today. Completely new instruments or automation systems are introduced to the market only over long cycles. Historically, the product cycle is shorter for instrumentation than it is for automation.
Short-term developments are focused more on the available menu for a particular instrument. Once the laboratory has its chemistry and immunochemistry testing on one platform, and even connected to automation, its goal then becomes how can it get as much as possible of its routine testing done on that platform.
With good planning, maybe some good luck, we do have something uniquely new to talk about every year.
But there’s a balance, in that you can be excited about an instrument you’ve had for a while that is a fantastic product, that delivers great value, the same as being excited about something that is new.
I’ve been really fortunate because we’ve had the Cobas 6000 instrument to talk about for the last couple of years. We launched it last year, and we could talk about it a bit before that. But a tremendous amount of customer insight and research and development is wrapped up in that instrument. That’s given us something that is genuinely new, and also unique, to talk about.
This year we’re bringing to market a couple of pieces connected to automation that are a bit different. One is prepackaged integration with Stago connecting to our automation line. We’re doing similar work with Sysmex so as to allow connecting their instruments to our automation line out of the box, if you will.
The other piece we have to announce this year, which is more on the analytical side, is a refresh of our software. There are probably 30 or 40 really important new features. This is quite significant for us. We’ll make our Modular Analytics line look and feel basically like a new entity.
Connectivity
You just introduced the idea of interconnectivity. What role does connectivity play now with respect to lab instrumentation and automation?
It’s significant. As I mentioned, for laboratories to get the full benefit of the limited personnel they have, they need to get as much testing as possible onto one automation line, or as few automation lines as possible. Connectivity is quite important for having as much testing as possible on one line.
What’s interesting is that it takes a fair amount of lab-specific research and analysis to determine what the optimal mix is. It turns out that it’s not optimal to have all the testing on one line, because of volumes and throughputs and so forth. But the general rule is, more is better—more testing and more of a mix of tests on one line.
Connectivity has been central to labs being able to fully capitalize on the benefits of automation.
Is instrument connectivity with the laboratory or hospital information system, including through the Internet, a natural part of the conversation you have with customers these days?
Absolutely. Information is the backbone of all this testing, and keeping track of it used to be a very manual process. Even the process by which Roche communicated operating information to customers was manual.
But today, with the Cobas 6000 series for example, all of that is available on the Internet. A big chunk of that information comes down to the instrument automatically so that the laboratorian doesn’t have to do anything except acknowledge changes—no more manually keying changes in at the instrument from a paper document.
We even have the ability to look at the performance of the instrument remotely. An individual lab can begin to have some benchmark comparisons that show how its instrument is running with respect to the performance of all instruments in the world like it. A much better ability to fine-tune performance through remote troubleshooting and diagnostics is available. The whole service experience is improved.
So, both of those elements are dependent on these systems being connected across the network. There’s huge value for the labs in having systems connected across the network and in having the kind of remote connectivity the Internet provides. The Internet is critical, absolutely fundamental, to the design of the new generation of systems.
Point-of-Care Technology
How has point-of-care testing affected the development of lab instrumentation and automation technologies?
In some cases, such as cardiac testing in the emergency department, it has increased the pressure on the central laboratory for high service and fast turnaround time.
The other side of that discussion is, the quality and the control that are possible in a central laboratory can be different than with a point-of-care device. The very interesting conversation in the marketplace is one of speed versus assured quality. In some cases, point-of-care testing—especially when implemented in the right way—can bring speed and high quality. But that’s not true in every hospital. Overall, point-of-care testing seems not to have stolen business from the central lab. Basically, the volumes of testing in both places have increased.
I think then the question becomes one of how to have a laboratory structure that can support both point-of-care testing and central lab testing in a consistent way so that the physicians have confidence in any results they get back, and they don’t have to wonder about what platform the test was performed on or question its quality.
Is making tests simpler even more important when it comes to point-of-care testing, since these tests are often run by people lacking background in lab medicine and lab testing techniques?
Yes, that becomes a very important decision for the group that wants to deploy point-of-care testing. These devices have to be designed to be simple enough for nonlaboratorians to use. But when a hospital or a physician decides to have point-of-care testing, questions about training the people who will run the handheld instrument and calibrate it arise. It is really important that they’ve got the right support in place.
Roche has a very robust point-of-care business. We know that when those things are in place, this testing can be quite effective.
Interestingly, the speed and perceived convenience of point-of-care testing has put pressure on the central laboratory to be much more efficient, to turn around their tests very predictably and quickly. And that in turn applies pressure on the labs to deploy integrated and automated solutions to meet those turnaround times.
Is there much discussion about or demand for wireless connectivity for IVD devices?
We do have discussions in those areas. I was part of our new business development team before I came into my current role, and we tested a piece of software that would act as a central place to record handheld patient test results and then allow their distribution to physicians wirelessly or through another communication channel of the user’s choice. Roche is doing quite extensive exploration in the area, especially with our blood glucose monitoring technology.
Molecular Diagnostics
How have developments in molecular diagnostics affected instrumentation development and automation?
I think the clinical laboratories provided more of a roadmap for molecular diagnostics than the other way around, because we’re a little further down the road with having tests be automated and having somewhat-less-skilled personnel work on the instrumentation. With respect to things like ease of use and automation, molecular diagnostics has leveraged the experience of clinical diagnostics.
In terms of how molecular diagnostic developments have affected the lab, I think there is a synergy between molecular and clinical chemistry. For example, after patients have been identified as having hepatitis C, they can be monitored with both clinical chemistry and molecular tests for different targets.
So, the chemistry testing side of our business is trying to lay the foundation for what the molecular test designers will eventually go through. The molecular testing business operates a little bit differently from clinical chemistry and immunochemistry testing. On a spectrum of high-skill to low-skill labor requirements, molecular diagnostics is at the higher-skilled end and is more labor-intensive.
Considerable work has been and is being done to simplify the way molecular tests are run to make them more conducive to automation.
Do you have anything that holds promise for automating multiplex testing, where the DNA in a small sample is amplified so multiple tests can be performed on it?
We already offer multiplexing with our chlamydia and gonorrhea tests. We’re very interested in multiplexing, and we have the technology in our blood screening pipeline.
Ensuring Customer Satisfaction
How do IVD manufacturers find out what their customers—laboratories, laboratorians, physicians, and other end-users—need most to get from their lab instruments and automation?
When we’re working on a new test or instrument, we conduct a series of in-depth discussions, including focus groups, and take input from customers who get hands-on experience with the actual prototype system or reagent that’s under development.
Beyond that, our sales and marketing teams have extensive day-to-day dialogue with customers, which creates the context within which we evaluate all the survey input. So, the body of primary research that Roche invests in every year to understand what our customers need is quite extensive.
And when what you hear from customers is about problems they’ve encountered or concerns they have, do you take those comments to your research and development team and say, “What can we do about it?”
Yes. That dialogue happens in a number of ways.
We have very-well-structured quarterly meetings where we look at our portfolio and the investments we’re making. We have routine, day-to-day dialogue with the product managers in our development centers in Europe and the United States. We also look at any customer phone call or field action we’ve had that makes us think there’s a way to improve our products. We’ve got various systems, databases and so forth, in which we capture all of that gathered information. And then our global colleagues work through those issues in a structured way.
What challenges do you expect to emerge in the area of lab instrumentation and automation?
Laboratories have been under the pressure we talked about for quite some time, and I don’t see it subsiding. I think the laboratory—just like the U.S. healthcare system generally—is going to continue to be asked to do more with less. We’re going to continue to respond to the pressures to make the economics of new systems work and to have the most configurable and flexible systems possible in order to meet a broader market requirement. I think we’re just going to see more of the same over the next three to five years.
