Medical Device Link.

IN PERSON

Chris Christopher is vice president, global solutions, at Dade Behring Inc. (Deerfield, IL). He can be reached at chris_christopher@ dadebehring.com

IVD Technology: What have been the biggest technological advances in lab instrumentation and automation during the past few years?

Chris Christopher: I am absolutely confident that instrument connectivity has been a constant and a driving force, especially as customers are demanding open connectivity of automation systems.

We’ve seen progress related to the integration with informatics. Today’s advanced solutions—particularly in automation—take into consideration the work flows of the laboratory and incorporate process-control solutions that assist in generating predictable turnaround times. Scalability, or small footprints, along with tighter configurations and lower overall costs for the clinical lab, have also been significant trends. The ability to offer solutions for both small- and midsized labs, as well as very large ones, is important to the overall lab industry.

As far as the latest trends, from my perspective the primary one is customer-focused design. By that I mean the automation we’re seeing has open connectivity not only to the manufacturing company’s instruments but also to those from other vendors as well. I believe that companies offering automation solutions that connect only to their own instruments will find it very difficult to sell automation in the future.

The second trend I see relates to cost-effective solutions for specimen management. Our studies have shown that the number of add-on tests for many labs has increased significantly over the last five years, and that this is a result of medical-necessity testing. As more laboratories expand into the outreach market, lab personnel are faced with the challenges of effectively managing and storing samples for add-on test efficiency.

Would you say that the lab instrumentation and automation market is growing?

It’s been rapidly growing. Not only in the United States but also within the other major regions, such as Europe and Asia-Pacific.

There is a high level of interest in automation, primarily due to efforts to improve the safety of the operators. In addition, clinical labs want to improve productivity and efficiency while lowering their overall cost structures.

How will the trends that you’ve identified contribute to this growth?

I think this question really touches the heart of the matter.

Automation can provide process control of the various functions within clinical lab testing. For example, a laboratory has to deal with a wide number and variety of specimens that arrive at its door, and it has to perform a broad menu of tests on each of these specimens.

So, when you look at the preanalytical, the analytical, and the postanalytical processes involved, there are a lot of variations. In other words, a lab may receive a telephone call or experience some other interruption and might not be available at the right moment to move the specimens from the centrifuge to the analyzer. There are a lot of these so-called wait times in laboratories.

Automation can control all of these processes—transporting the samples to the centrifuge, removing and decapping them, then routing them to the appropriate instrument. All of these activities, when performed through automation, improve overall productivity and the number of billable tests per paid hour.

Designing for the Clinical Lab

What sorts of challenges do IVD manufacturers face when designing and developing laboratory systems?

Only a few of today’s laboratory instruments were initially conceived 5 to 10 years ago, when instruments were designed to connect to lab automation.

So, for the most part, connecting to automation systems has been an afterthought. In order to achieve the performance that a customer expects, a number of design characteristics must be taken into consideration.

First is the connection of the instrument to the automation line. Manufacturers have two choices: either to transfer the collection tube from the automation line to the instrument and then back, or to use point-in-space sampling. In the latter, automation enables the instrument to connect to or to reach over the top of the sample and to transfer that sample into the instrument for diagnosis. Collection tube transfer is the more expensive connectivity approach. Point-in-space sampling, which is the preferred method, typically requires a 510(k) submission.

Another design consideration relates to system integration. With the connection of instruments and automation track, a lot of information must flow from the instrument to the lab automation system, or LAS. As the demand to connect a large variety of instruments has increased, communication standards and protocols have become a challenge. Ideally, the industry would integrate plug-and-play technology, but I don’t see that for many years to come.

A third consideration is the overall scope and function of the LAS. It is important to find areas where we can effectively apply process control to preanalytical, analytical, and postanalytical production steps. The LAS needs to be able to support these activities. The goal is for the laboratory staff to touch the sample only once.

A fourth design consideration is overall flexibility. Today’s laboratory has many design and space constraints. No two labs have exactly the same floor plan. With fixed space due to either cost or facility design, flexibility becomes extremely important. Having an automation system that can be configured linearly, as well as with 90º shifts, becomes essential.

So, how do we overcome these complexities with lab automation? That’s a challenge, for sure. Successful IVD companies have dedicated their design and marketing teams to customer-focused design. They have also formed a number of alliances with experienced lab automation companies.

It seems that many of these challenges center around a primary concern with lab instruments and automation: ensuring testing efficiency and reliability.

When lab automation is installed in a core laboratory, the percentage of all tests that are processed on the automation line shifts from 60 to 80%. So, reliability is the number one concern. Second to this, of course, is how a company approaches the service and support of that automation line.

It’s a complex process because, to be truly effective and productive while meeting the customer’s overall goals, lab automation must not only connect to the instruments but also have a solid and robust connection to the laboratory information system, or LIS.

When developing lab systems, do IVD manufacturers try to come up with completely new technologies? Or, through upgrades and integration, do they tend to build on current instrumentation and automation offerings?

In other words, are we removing steps from the process or are we adding more? For example, when we developed the connectivity for Dade Behring’s RxL chemistry system and StreamLab analytical work cell, we built it in such a way that all of the available instrument intelligence was being transferred to the lab automation control station or to the workstation. Doing so allowed the operators to know the status of the instrument at all times, freeing them from having to check whether the system needed any reagents or other attention. All of this intelligence is built into the instrument and the laboratory automation system through integration.

What challenges do software upgrades and advances present to connectivity? Is security a concern?

Absolutely. Connectivity to lab automation is key because without this, you’re transporting the sample from an input station to the workstation but aren’t able to take advantage of the transfer and processing of that sample. So, it is important that the connectivity is done well.

When Dade Behring connected to the Immulite 2000 immunoassay system by Diagnostic Products Corp. (DPC; Los Angeles), what we did was work out both the short-term and long-term needs in the event that connectivity was broken. Would such a failure be caused by the instrument? Would it occur from the StreamLab losing communication? Or would it be because information from the laboratory information system hadn’t transferred over? It’s very difficult for the user to determine the root of such a problem unless the different manufacturers have a good working relationship.

Protocols and Standards

In 2000, at the behest of the American Association for Clinical Chemistry, a group of IVD manufacturers and organizations formed the Connectivity Industry Consortium (CIC) to create connectivity standards for point-of-care instruments. Are the protocols that came out of CIC still in effect?

That’s a challenging question. Currently there are no far-reaching agreements on industry standards for connectivity. You could talk about Health Level 7 and other protocols that are widely used, but there is no laboratory standard on connectivity.

CIC reached agreements about what the ideal design would be and published them, but they are still voluntary. They are only guidelines, not rules that an IVD company has to adhere to.

What this has done is create a need for the automation supplier and the IVD companies that are going to connect those instruments to the automation line to work very closely together. As a result, software upgrades can be both tested and validated for the lab automation prior to a field service representative upgrading that instrument’s software.

Do you think that drafting required standards would be good for the long-term growth of the IVD industry?

I think that the industry is really calling out for a standardization of automation connectivity, not only from the instruments to the LAS, but from the LAS to the LIS.

The vast majority of information is transferred between the laboratory information system and the lab automation system. And again, there are no required standards for this. If we had communication protocols that everyone agreed to, it would help the industry achieve the goal of plug-and-play automation systems. In this scenario, an instrument could be rolled up to the automation and, in 30 minutes, could go on-line and be able to perform tests or verification protocols.

More-Efficient Testing

How has point-of-care testing affected the development of lab systems?

I believe that point-of-care testing is a complement to lab automation systems. It actually reduces the demand for staff within the clinical laboratory. This approach allows the LAS to do what it does best—complete the large work loads for routine and timed tests in a productive way.

Smart automation systems, such as the StreamLab, are designed to address many variables and to allow the instruments to complete their work load in the shortest amount of time. In the end, when point of care is used, the physicians are more pleased with the lab’s overall performance.

Both connectivity and point-of-care testing, as well as informatics and scalability, have had the effect of making testing faster and less expensive. What else has helped contribute to this trend?

In regard to emerging technologies, I believe that the next phase of automation will focus on informatics. This will help enhance process control as manufacturers continue to search for ways to further increase productivity and efficiency for the clinical laboratory.

As an example, I always ask myself what would happen if we could perform sample-quality measurements on a freestanding module before introducing them to the actual instrument. If we did this, could we improve our overall productivity by increasing instrument throughput while reducing the number of repeated tests?

I also believe that the challenge presented to automation engineering is how far automation can be taken. Many process-control enhancements could best be performed using a collective approach, with an LIS vendor working closely with an LAS supplier or company. As of today, I haven’t seen any evidence that this has happened. In the future, I expect the gap between what the LIS and the LAS can do to close. The reason for this has more to do with the design structure and the robustness of the LAS.

In large hospitals, a wide gap exists between the different types of LIS. But we are definitely seeing an interest in lab automation that expands beyond the large labs to the smaller labs. These small laboratories are looking at middleware—informatics—to solve their information and connectivity needs.

The Influence of Molecular Diagnostics

Molecular diagnostics has been garnering attention for a few years, but hasn’t reached the level of growth that many people have been hoping for. Even so, how have advancements in the area affected instrument development?

As you note, there continues to be a strong and growing worldwide interest in molecular diagnostics. Everywhere I travel, I see that people are investing in molecular testing. The cost to enter the field is very high. Doing so requires specialized instruments and highly skilled personnel. A number of manufacturers are working closely with traditional automation companies to determine which steps in the process can be automated.

As we have seen in the core lab, instrument manufacturers must rethink their development approach in order to connect to an automated line. This is also true with molecular diagnostics. From my perspective, though, it will take at least five years or more for molecular diagnostics to get to where we are today with automation for the core laboratory.

How can an IVD manufacturer justify investing in something that is so expensive to develop when current returns aren’t very high?

You’re absolutely right. The challenge is that for many molecular diagnostic tests, the cost for the laboratory is higher than the billing price. And until there is enough value associated with molecular diagnostics, whereby the payers—whether they’re the insurance providers or the federal government—see satisfactory returns on their investments, there will always be very slow growth in the area in the foreseeable future.

One way to lower costs is to automate those highly manual processes and to reduce expensive labor. This is very much an issue related to volume. The greater the testing volume, the lower the overall cost structure and the more a company can justify its investments.

For automation in molecular diagnostics to catch up with automation in the clinical laboratory, manufacturers will have to make further investments in connectivity, sample distribution, and the results-generation process. Once these technological advances have been realized, I believe that the cost of molecular diagnostics will decrease and the field will begin to take off.

Focusing on the End-User

How do IVD manufacturers find out what laboratories, laboratorians, physicians, and other end-users need most from lab systems? In particular, how do manufacturers go about accommodating the needs of the different sizes of markets you mentioned?

The approaches vary widely, but the one that seems the most effective is through education, beginning with quality-improvement initiatives such as Six Sigma and lean principles. These help eliminate most of the waste in the laboratory process and enable greater efficiency. Then, it is much easier to determine what the overall impact of laboratory automation will be.

In most cases, the productivity of the individual medical technologists will improve significantly, as will the overall safety of the medical technologies. Labs will produce fewer errors and will be able to remove the variability of the production process.

I am firm in my belief that not all labs need full automation, but they do need some elements of automation. Scalability is essential. You’ve got to be able to take advantage of today’s technologies and present solutions that make sense and are affordable to each of the different market segments.

For more than eight years, we’ve had a healthcare solutions team that has been working with all sizes of laboratories. We have been into literally hundreds of laboratories around the world. The information that we’ve collected from these labs has helped guide our company’s focus and direction in lab automation.

The Laboratory of the Future

What trends can we expect to see this year and in the future? What challenges do you see arising?

For certain, the trends toward intelligent integration will continue. In order to achieve this next level of integration, the instrument design requirements will extend beyond traditional connectivity to the LIS. Designers will need to consider the new LIS communication protocols that focus on improving overall lab productivity.

On a second front, I believe that lab automation system informatics will continue to evolve and take over many of the responsibilities that the traditional laboratory information system has performed. Examples include the automated receipt of samples, verification of results, and alert messaging of the users.

Looking toward the future, the two greatest challenges facing lab automation and instrumentation will be integration and small sample sizes.

With more than 500 different laboratory information systems worldwide, maximizing the system integration and productivity at an affordable price can seem a huge, highly complex, and sometimes seemingly impossible task. While development and connectivity fees in Europe average 2500 to 5000 euros, these same fees in the United States are frequently 10 to 20 times higher. This creates a difficult environment for both the laboratory and the lab automation system supplier.

In addition, manufacturers face the challenge of automating pediatric small collection tubes. It’s very difficult to get a bar code on these tubes and then to be able to transfer them to the instrument and identify the bar code. At the present, small collection tubes cannot be placed on any of the automation lines. The reasons for this include the lab automation system itself, the connected instrument, and the lack of information from the laboratory information system noting the size of the container introduced. In many hospital laboratories, pediatric small collection tubes can represent up to 30% of the work load. As a result, the lab automation system has less of an overall impact in these environments.

As there is a growing interest in being able to use a smaller amount of sample taken from the patient, we will have to be able to address this problem to maximize lab productivity. I do believe that this challenge will be solved in the future with automation, but it’s going to take a lot of engineering and cooperation with our customers to be able to overcome it.

You mentioned the necessity of new communication protocols for intelligent integration. What types of standards do you see emerging?

The protocols that I’m talking about are being able to pass more information and more data from the instrument to the lab automation system. This, in turn, will provide the LIS with the ability to look at, analyze, and then process that information without the laboratorian having to touch the sample once it has been placed on the automation line.

An example would be the sample going through the first process and being stored in a refrigerated storage-and-retrieval system. Then, when an add-on test is received, this new request would pass automatically from the LIS to the LAS. The LAS would then retrieve the specimen from the storage unit, place it back onto the track, process the additional test, and restore it in a high-quality and high-safety manner.

This is an exciting and growing field. Worldwide interest in it is at an all-time high. And I think that as we gain more experience in lab automation and the intelligent integration of systems, the productivity, efficiency, and costs of the laboratory five years from now will be far different from those of the laboratory today.