IN PERSON
 |
Steve Bailey is director of new product development in worldwide R&D at Ortho-Clinical Diagnostics (Rochester, NY). Since joining in 2002, he has led the development and delivery of new platforms for clinical chemistry and immunodiagnostics markets. He has over 25 years of experience in on-time systems delivery in defense, medical imaging, telecommunications, and medical diagnostic industries. He can be reached at sbailey@ocdus.jnj.com. |
To learn more about this effort to appeal to small, medium, and large lab markets within the limits of resource allocation and clinical construction, IVD Technology editor Richard Park spoke with Steve Bailey, director of new product development in worldwide R&D at Ortho-Clinical Diagnostics (Rochester, NY). In this interview, Bailey shares his views on where he thinks the IVD industry is headed and how future development lies in connectivity and asking the right questions. He also discusses the initiative to create core technology, expand test menus, and gather crucial customer feedback before, during, and after the design process is complete.
IVD Technology: What have been the biggest technological advances in instrumentation development during the past few years?
Steve Bailey: Lab automation has taken on a new level of importance in the ability to actually get instruments hooked up to various lab automation systems (LAS). We’re being challenged to provide a solution for both medium and high lab automations, as well as some of the low- and medium-sized customer solutions. Also, access to high-speed telecommunications has enabled us to provide remote diagnostics to virtually any customer around the world. Now, we can program instruments to provide large amounts of data on a daily basis, which in turn allows us to look inside the instruments and help service customers more quickly.
What are the main factors that drive technological advances in instrumentation development?
There is a continuing struggle to find qualified lab technicians to run all the equipment in addition to the cost-cutting factor in each of the labs. So the overall focus on integration within laboratory instrumentation development, whether it’s on a single box where labs are conducting multiple tests or whether they’re trying to bring two boxes together, is to provide the ability to limit the number of lab or hospital personnel who have access to the samples during the testing process.
Expanding the Menu
What are some of the latest trends in instrumentation development?
Most of the trends in instrumentation development are focused on integration. Sometimes we consider that to be lab automation. But other times, it’s simply adapting more tests to share a single platform or a single piece of equipment, whether it’s two disparate technologies combined in a work cell or a fully integrated system built from the ground up.
In the past, we’ve recognized a need to bring a variety of tests together on a single box or with a single customer experience, so we could ensure more-efficient labs and technician expertise on the boxes. Looking back at the instrument offerings from Ortho-Clinical Diagnostics and our competitors five or six years ago, we were all running 20, 30, 40, 50 tests maximum on a given box. Now we’re incorporating 80, 90, and even 100 or more tests for a single box. So we’re definitely increasing the menu capacity on one instrument.
What are the primary obstacles that IVD manufacturers encounter when developing their instruments?
Customers continue to want more menu capacity, more tests on a single instrument. Certainly, there are limitations for how many detection technologies or how much software and hardware we can incorporate on one box that will provide the breadth and depth of testing that a customer wants on a single instrument. But the technologies can be made only so small, and only so many reagents, samples, and hardware can be included on one box.
In order to meet these challenges, we have to provide the connectivity of our instruments to other vendors’ instruments. This could be done through an open system environment, or by building systems that allow other testing designs to plug into our system environment, where customers can get all of their answers without having a collection of disparate individual systems in their labs. It’s a challenge because there’s a finite amount of R&D capital to go around, and our budget is obviously limited like everybody else’s.
How do you address issues related to quality control in developing instruments?
We are committed to making sure that human intervention and data verification are not excluded from the instruments’ operation. For example, our sample indices provide the ability to look at a particular sample and assess the hemolysis or turbidity. We don’t assume that we know better than a qualified technician who would be analyzing the sample. Instead, we flag it and say that, based on our testing and analysis, this may be a suspect result. This way the customer can take the data, look at them closely, and decide whether to rerun that patient’s test.
I think the same thing is true with autoverification. A lot of vendors have autoverification features built into their architecture. However, I think we want to guarantee that we’re asking a qualified technician or a follow-up system to make that decision. We’re not going to make that call for them. We’re always focused on trying to guarantee the quality of results, and not trying to second-guess what a qualified technician may have to determine.
How do IVD companies overcome these challenges?
IVD companies have to develop a road map, perhaps change their overall development strategy and process, and look at what core technologies or detection sensitometry they want to use and make them scalable. One thing we’re embarking on is a product platform strategy. We’re looking ahead at the next 5–10 years, not just at the next platform, but a series of three, four, five platforms and deciding how we’re going to make them both scalable and introduce them to the market quickly.
We’re making a fairly large transformation internally to decide what components we want to put in place that we can then plug and play into small, medium, and large-scale instruments more quickly based on the industry’s needs.
How do you determine what will be the clinical labs’ instrumentation needs in five to 10 years?
We are looking at other industries and researching how big manufacturing companies develop four or five major pieces of a copier or printer and repackage it or scale it, then continue to feed the market with new products on a regular basis. What we’re trying to do is build the core technology. We’re going to take our hardware and software and make it scalable so that we can walk over to a shelf, pick the different parts we need, and start assembling them so that everything from a small, desktop unit to a large unit is compatible.
Regarding the 5–10 year forecast, that’s difficult to say because we don’t necessarily know where the industry’s going, but there are some things we do know. We have to be able to scale. We have to go all the way from a point-of-care device to a mainframe something that handles 1 million or 2 million tests a year. We have to resist reinventing detection technologies every time we do that. We have to always guarantee that a particular assay can work on everything including small, medium, large, or extra-large systems.
When developing instruments, what model do IVD companies tend to follow: coming up with completely new instruments or building on current instrumentation offerings?
What we see for most IVD companies is the model of building on current instrumentation. Upgrades or modular additions in the case of moving toward integration are finding their way into work cells and software that will let two boxes continue to work the same way they do independently, but aggregate the results. So the customers feel like they’re getting an integrated solution, but we call that a first-generation integration.
Unfortunately, it can limit how IVD companies are addressing the customer’s needs if they’re not integrating from the bottom up. What we’re trying to do is come up with completely new instruments as a starting point, but that doesn’t address the high cost of development because it’s very expensive to put together new platforms from the ground up. When companies apply the product platform approach, that first instrument might cost a lot. But then the derivatives or products that spin off from that platform get a lot of bang for the buck. That’s when companies can reduce overall instrument development costs and start going after the high diagnostic test market.
Making Connections
What role has connectivity played in instrumentation development?
Telecommunications technology dramatically increasing data throughput made us go back and think about what data we want to collect in hopes of increasing an instrument’s serviceability and monitoring the quality of data that customers find the most important.
With our previous platform, to look across the entire hardware and software platform and ask the following questions: What data do we have? What do we want to store? What could be useful to both customers and our service organization? That translates to new onboard storage capabilities, new ways to group data together, and new ways to anticipate problems before they occur. It’s similar to a predictive maintenance approach that always tries to find ways to help maintain the quality of the instrument and reduce service, which increases uptime for the customers.
The other piece of it is providing new opportunities to help customers further reduce their workload, such as autoreplenishment of reagents and consumables. Many companies are looking at opportunities to monitor the usage of the instrument, number of tests, consumables, and reagents, and provide a service to our customers to let them know when they’re about to run out of a particular consumable and help place the order or reorder automatically.
How has point-of-care testing affected instrumentation development?
The biggest challenge is to find ways to provide the exact same quality and result from the point-of-care all the way up through a scalable mainframe instrument. We look at developing a new marker or diagnostic test and begin analyzing how we’re going to get that test to run, how it will run on current platforms, and how could we run it on a point-of-care platform.
It’s difficult to maintain the same quality of results while also driving improvements to detection technology. Sometimes it is simply not feasible when the solution needs to sit in the palm of the hand versus something that resides in a lab.
Connectivity and point-of-care testing fall under the realm of making things faster, smaller, and cheaper. What other such similar issues have emerged and will continue to emerge that will drive the development of instruments?
Having every instrument be able to handle all the tests the same way and provide all the capabilities and functionality that meet the customers’ needs becomes a big challenge, especially if IVD companies try to address both desktop and small-market segments of the industry as well as large core labs.
The key for us is to continue to focus on our new medical content strategy. We want to spend a lot of time going after the new medically relevant diagnostic tests and figure out how we can run them on our entire suite of instruments against anything from point-of-care to mainframe.
What role has microfluidics played in instrumentation development? How does it influence instrumentation development efforts?
We include very small volumes of microliters inside the instrument per our reagents and samples. So technologically, we’ve been there for many years with small volumes of fluid. But if we want to keep making our instruments smaller and move toward the desktop or point-of-care solution, the ability to take hardware and scale it down to a point where we can continue to deliver microfluidic levels of samples and reagents, and perhaps even smaller, will be very important.
How have developments in molecular diagnostics affected instrumentation development? How will it play a role in the future?
When I look at where molecular diagnostics has gone during the past several years, clinical diagnostics seems to follow a very similar path where the customers want to continue to get increased work flow and more automation on a series of tests with a platform. So it’s doing more with fewer people and having it integrated onto single platforms or through automation.
Gathering Feedback
How do IVD manufacturers find out what laboratories, physicians, and other end-users need the most from their IVD instruments?
We have a number of different techniques that we use with our current platforms or instruments in the field. Whenever we launch a new instrument, and even up to two, three, or four years after the market release, we conduct surveys with customers. Ongoing input helps us analyze how the instrument is performing against their needs. From that, we get to see new trends, features, and capabilities that we need to add to future platform or instrument development.
We also do surveys with customers when they come in for training. As they’re getting trained on new instruments, we ask them: What kinds of things do you see missing? What kinds of things are hitting the mark within your particular laboratory environment?
When we do a new platform development, it’s substantially more involved. As we enter the concept phase, we take a very broad overview within the industry to make sure that we’re on the right track. When we begin development, we spend a lot of time conducting targeted market research and putting together very detailed questions. This helps us understand individual feature trade-offs, which then influence exactly how the instrument is going to be designed.
The next step, which is what we call the subsystem or detailed design phase, is to bring customers in from around the world and have them give very specific hands-on feedback about features before we lock them down and build them into the instrument. This tells us what kinds of major trade-offs customers need and want, and how features should look and feel to provide the most value.
Certainly all the trade shows, such as IFCC, AACC, and many of the smaller shows around the world, give us constant feedback that we’re either on the right track or some corrections need to be made in a new platform development. In the last four or five years, the voice of customer exercises has become increasingly important to make sure we’re more in tune with what their needs are.
Smaller labs are going to have very different automation and instrumentation needs compared with large reference or hospital labs. How are those factors taken into consideration? What further challenges does that present to IVD companies in their instrumentation development efforts?
When IVD companies are developing a single instrument, they certainly can’t service every market at once. It’s simply not feasible. It’ll either be too fast, too slow, too big, too small, etc. Initially, they must define what market segment they’re targeting. We understand what features are the most important attributes of the instrument within those given segments. For example, let’s say we’re going after a medium segment industry customer base, and they may administer 200,000 to 1 million tests a year. That’s a pretty wide range.
Within that range, we have to understand the work flow of those customers. We ask the following questions: How does it vary? How is it the same? How many tests do they run? What types of tests do they run? How fast are they expecting to get results? We went after those answers in very exhaustive detail with our new instrument platforms.
In the old days, when IVD companies were only servicing a clinical chemistry world, the instrument was more static and similarly used from customer to customer. When companies look at an integrated world, the variations that it brings are substantial. Everything from a batch mode to peak-hour use of the day to the types of tests that are run during different times of the day to the way the customer walks up and uses the instrument is taken into consideration. It was really an eye-opener to see what we had to accomplish to satisfy a wide, and in some cases unpredictable, range of customer uses all within one integrated environment.
One thing we focused on right off the bat was throughput. When we looked at all the attributes of customers’ needs, the throughput of the instrument came back every single time as being probably—aside from the quality of the result—the most important thing for customers. So from day one, we developed a throughput model that reflects how our instruments run 100%. With that model, we’re able to take live customers’ LAS output, put it through the model, and tell them exactly what their throughput is going to be, as well as the walkaway time and their consumption.
Being able to help the customers understand how our instruments are going to perform before they buy is a huge differentiation for us. That’s a big change in the industry as we progress toward an integrated environment.
What new trends can we expect to see this year and in the future in instrumentation development? What future challenges will emerge in instrumentation development?
I think the industry will see more of a product platform approach. We want to be able to react more quickly to the ever-changing market landscape and provide products more quickly from the small-, medium-, and large-scale segments. So that’s a key feature we’re going to see in development.
The second thing is tailored solutions. Today, it’s one-size-fits-all, but in emerging markets, they have different needs. They have different turnaround-time requirements, different tests, and different protocols. We’re becoming aware that one size doesn’t fit all, and a product platform approach will allow us to begin creating tailored solutions for these emerging markets, developing countries, and places that haven’t had the luxury or the ability to put IVD instrumentation in place.
As for future challenges, there’s always the ongoing discussion of multiplexing technologies. If we go with the multiplexing technology approach, does that allow more tests onto a single chip or a single slide? Will it allow tests to get faster and smaller? A smaller footprint is going to continue to be important, although there is certainly a limit to that.
Finally, we’ll also see a move toward greater sensitivity. As we go after these new medical content and diagnostic tests to try to find more ways to increase the quality of life for the patient, we need to come up with more highly sensitive detection technologies that can provide the assays for the results customers need.
