July 2010: Volume 36, Number 7
Radio Frequency Identification of Specimens
Can it Solve Labs’ Preanalytical Predicament?
By Bill Malone
Courtesy Tagent, Inc.
As clinical labs have become larger, more complex, and more automated, the need to produce timely test results for thousands or hundreds of thousands of samples on a daily basis has forced laboratorians to become experts in far more than biochemistry and medicine. Logistical and technological challenges on the preanalytical side of testing mean laboratorians have had to become systems engineers, not just analyzing specimens, but managing a complex matrix of inputs, outputs, and data, all enabled by technology that can be both powerful and bewildering. With the knowledge that each sample represents someone’s parent, sibling, child, or spouse, the frustration is enormous when samples are lost, misplaced, misidentified, never received, or even more consequential, when results for the wrong patient are returned.
The answer to many of the problems that plague the preanalytical phase of testing may soon be solved by a relatively simple and mature technology that is now being refined and miniaturized so that it can be ubiquitous, invisible, and inexpensive. First used by the British in World War II to identify friendly aircraft visible on radar, radio frequency identification (RFID) is quietly inching its way toward mainstream use in labs as a way to reliably identify and follow every patient sample throughout its journey. Today, RFID technology has advanced to the point that tracking lab samples is feasible but costly. But cutting-edge companies and the labs working with them say this will soon be a thing of the past.
“RFID is definitely going to hit the clinical lab and will be a very important technology for us; it’s just a question of price right now,” said Charles Hawker, PhD, adjunct professor of pathology at the University of Utah School of Medicine and scientific director for automation and special projects at ARUP Laboratories in Salt Lake City. “The good news is that companies are now developing technologies that should be able to break the current price barrier.”
Ready to Go
In the same way that computer processor chips have become simultaneously smaller and more powerful, RFID tags can now be manufactured as tiny as 2 mm by 2 mm and thin enough to be sandwiched into disposable, standard labels for blood and tissue specimens or even embedded within the blood tube. With these high tech labels attached to each individual specimen, labs can precisely identify and track patient samples as they move through the laboratory processing path, silently broadcasting their location and unique identity to electronic readers placed throughout the lab, hospital, or just about anywhere you can imagine.
Several companies are now focusing their efforts on developing RFID systems specifically for labs. Seattle-based Ambient ID launched its LabTrack system in 2009, and already has installed the application at four labs—including at least one large national reference lab—and 12 hospitals. Because Ambient ID is in the early phase of implementation, the company is unable to name its customers due to strict non-disclosure agreements. Other labs have internally developed RFID systems in place. An example is Mayo Clinic’s Rochester campus, where a functional RFID tracking system was developed and is used in the anatomic pathology lab and GI endoscopy areas (See Box, below).
Success Story at Mayo
RFID in Anatomic Pathology Provides Cost Savings, Efficiency
Since 2007, the division of anatomic pathology at Mayo Clinic (Rochester, Minn.) has been using RFID to track pathology slides and paraffin-embedded tissue blocks, finding huge gains in quality, workflow, and cost savings as a result.
“I think this is a huge patient safety issue,” said Schuyler Sanderson, MD, a pathologist and assistant professor of laboratory medicine at the Mayo Medical School College of Medicine who helped implement the RFID system. “The system maintains patient identity throughout the process of collecting the specimen and sending it to the laboratory. We are similar to many hospitals and laboratories in that we have had patient identities swapped during a procedure where suddenly patient A became patient B, simply because we applied the wrong label, misread the medical record number, or we had some other human error that we just can’t control. This technology protects us against that.” Prior to RFID Mayo used so-called hard stop policies to catch such errors. It worked very well in terms of fixing errors before there was any risk to patient safety, but was very time consuming, Sanderson explained.
In the Mayo system, RFID tags are attached to the sides of specimen bottles. A nurse in the procedure area places the bottle on a pad containing an RFID reader, then enters patient information and the instructions for that specimen into a database. The database then links the RFID tag with the identity of the specimen. The pathology lab is connected to the same database, so when the lab receives the specimen bottle, an RFID reader in the lab senses the tag and opens up the appropriate record in the database. This process has made it possible to leave behind paper requisition forms that lead to potentially dangerous transcription errors, Sanderson explained.
“Before this system, I had to ask a nurse to copy down very complicated data from the procedure area onto a piece of paper and then send that to a lab and have it re-transcribed in the laboratory information system, which put us at risk for another transcription error that could potentially be very significant,” he said. “But with the RFID technology, we avoid that completely, and we go from electronic system to electronic system here seamlessly.”
In a paper published last year, Sanderson and his colleagues found significant reductions in transcription errors, from very minor typographical changes with minimal potential for clinical impact, to major errors that could jeopardize patient safety (Am J Gastroenterol 2009;104:972–975). They classified these errors as class 1, 2, 3, with 3 being the most significant. Before implementing the RFID system, there were 646 class 1 errors, 112 class 2 errors, and seven class 3 errors out of 8,231 specimen bottles sent to the pathology lab. With RFID tracking in place, there were 35 class 1 errors, two class 2 errors, and two class 3 errors. However, both class 3 errors under the RFID system were recognized and corrected before the specimens were processed.
The RFID system has also brought significant cost savings, Sanderson noted. “Back in 2006, we took a quick measurement of how much time we spent just trouble-shooting labeling errors, and just GI specimens alone, that was about $125,000. We’ve done some more recent calculations, and we look at our future version of this system as offering an internal rate of return of about 30 percent.” The pathology labs using RFID have now gone to only single-digit errors per month since publishing the paper and those errors are captured and corrected immediately.
Even with these significant savings, Sanderson still believes the biggest impact has been on patient safety. “It would be easy to say that cost is the most significant benefit, but that’s not completely true, because we really have experienced such a tremendous quality improvement,” he said. “We were willing to spend a little extra money up front to get a high quality product and truly benefit patient safety and patient care overall.”
When asked about the benefits of RFID tracking of lab specimens, Carlton Burgess, director, Healthcare/Diagnostic Laboratory Solutions for Ambient ID, enthusiastically describes the technology. “RFID adds visibility to your laboratory processes by showing you where every sample is along the way to help improve your turnaround times. We light up the path that samples take, all the way from the collection site, through to the testing facility, and then internally within the testing facility, through storage and then final disposal.”
RFID tags enable a lab to identify and alleviate bottlenecks in real-time while samples are going through the system and also improve upon current barcoding systems that can be highly tedious, Burgess explained. For example, to identify a blood tube that’s barcoded on a rack holding 72 tubes, a person has to touch every single tube to get it in front of the barcode reader to scan. With RFID, a reader device can scan the entire rack and capture the exact identity of every single sample without manipulating any of the tubes.
Tracking samples with RFID tags is especially helpful when samples frequently move among various locations, whether it be at a reference lab, regional lab, or core lab of a hospital group. “There are very few truly independent hospitals, so there is a lot of movement of samples between facilities, which is why it’s so valuable to track them, even if you’re just moving samples internally within one location,” Burgess said.
Mountain View, Calif.-based Tagent is also developing RFID tags for use in labs and has agreements with several labs to pilot the company’s technology in the coming months. Technological breakthroughs by Tagent and other companies now mean that tags can be made so small that they don’t interfere with other factors, such as the size of a blood tube or label that labs already use.
“The advantages of the technology that we are developing is that the tag is a very small, single, self-contained chip, and it’s small enough that it fits easily under the label of a blood tube, on a pathology slide, or within a pathology cassette,” said Geoff Zawolkow, vice president of marketing and business development for Tagent. “Unlike some other small tags, we can get 10 meters’ worth of range out of it, which is critical for the laboratory, and so we’re also able to locate those tags within a meter of the fixed system and down to the individual tube with a specialized handheld seeker. These new tags are more precise and more robust, in order to minimize the kinds of changes that a lab would have to make in order to implement the technology.”
The size and robustness of new tags makes for a seamless introduction of an RFID system, Zawolkow explained. For example, in a blood draw center, as soon as the label is printed and put on the tube, an RFID reader can sense the embedded tag and link all the information from the tube with the number that’s on the tag. From that point on, the lab can confirm that the specimen exists and where it is.
Aside from the tags themselves, an RFID system also requires readers at various spots within the lab itself or other structures through which samples pass. Like miniature radio towers, readers pick up the tiny signal sent out from each tag and upload information into a database. Tagent’s tags don’t require a battery because they collect energy from radio waves, then transmit back the unique identification of the tags using that energy. Tagent’s system is unique in using a third piece of hardware, called a power node, to send out the radio signal that gives the tags the energy needed to broadcast their location. Other kinds of tags don’t require power nodes, but may need to be larger to accommodate a separate antenna or battery.
Just like the tags themselves, the readers or power nodes can also be made virtually invisible to a lab operation, and require about the same amount of work as setting up a local area network for computers, Zawolkow explained. “It depends on the situation in the lab. In general, it can be done in a way that’s not disruptive,” he said. “RFID readers can be placed in doorway entrances and into ceiling tiles. Since most labs have drop ceilings with tiles, we would just replace a tile with a reader. You have to string connection wires to it, similar to what you’d do with your computer network.”
A Question of Cost
While the basic benefits of RFID tracking are pretty easy to grasp, any new technology that laboratorians evaluate for their lab has to make sense economically by either fitting an already tight budget or paying for itself with new cost savings. The companies developing RFID systems for labs say that the latest technology can pass the test, with both a modest initial investment and big payoffs in productivity. However, most labs will have to wait a few years before enough customers are buying RFID tags in bulk to bring overall costs down to just pennies a tag.
“The tags that we’re developing are meant to be tens of cents, though we haven’t settled on the exact price point yet,” said Zawolkow. “I think the main question is, how much is the lab going to save by implementing this, and is it enough to pay for the technology? Of course there are other benefits in quality besides just the savings that they would like to have, but we understand that a lab must have those savings as well in order to implement anything new.” Tagent’s research indicates that labs would need tags to come in at something under 30 cents, according to Zawolkow.
Right now, how much a lab can save using RFID is highly variable. The cost/benefit analysis has to take into account how each vendor plans to make the technology available, and what systems a lab has in place already. For example, Ambient ID’s system sells on a per-sample basis, and its customers have demonstrated significant savings that exceed the costs, according to Peter Allison, Ambient ID’s president. “Depending on the scenario, the savings have been anywhere from 150 percent of the cost, all the way up to five or six times the cost,” he said. “These savings come from faster turnaround time, processing, and a lot of employee hours.”
At the same time, for very large core or reference labs, even 10 cents per sample quickly adds up if every sample gets a tag, noted Hawker. Some reference labs print hundreds of thousands of labels each day, which translates into tens of thousands of dollars a day—and millions per year—to add an RFID tag to each label.
On the other hand, specimen processing centers for large labs have become cumbersome and expensive operations. “If a client sends us 100 specimens a day, and a processor can only handle 30 specimens an hour, that’s almost three-and-a-half hours of labor,” Hawker said. “So in this particular instance, if I could cut that down to a half-hour using an RFID system, that’s enough savings in labor to pay for the cost of the tags, even if they’re 10-to-20 cents apiece.”
The Needle in the Haystack
The unique ability of RFID systems to capture the identity and location of samples provides a new level of insight that can illuminate stubborn problems in sample handling and automation, emphasized Burgess. Software can document each RFID-tagged sample’s path, so the lab can establish appropriate dwell-time for samples between various handoff points. Then, if any of those dwell times are exceeded, an automated software alert lets the lab know how many samples are delayed so it can be proactive in real time. “This lets you find the needle in the haystack, pinpointing exactly which samples you have an issue with,” Burgess said.
Knowing the identity and location of samples beyond the reach of barcode scanners or the human eye not only makes misplacing a sample much less likely but also enables the lab to say for sure that they received it in the first place. “At ARUP, we have half a dozen people whose full time job is just looking for specimens for which the client sent an electronic order and we never got the specimen,” said Hawker. “Usually, the specimens arrive in a later shipment, but we see great potential for RFID helping us with quality, control, tracking, and accuracy issues.”
Although barcode systems have come a long way, RFID brings unique enhancements that barcodes can’t compete with. “The difference, and the advantage RFID has, is that barcodes are read at close range and one at a time, so you can’t bulk read what’s coming in and going out of any place in a lab with a barcode, it’s got to be handled one at a time and read one at a time,” said Zawolkow. “That limits when you know where the samples are, and how quickly you can either ship them out or intake them in a given location. With RFID, you can read them in bulk and locate them in bulk so you can know where individual tags are at any given time.”
The precise and comprehensive data RFID tracking makes available to a lab also enhances other quality and process improvement initiatives, said Allison. “Our RFID system helps improve the productivity of everybody that’s involved in the process,” he said. “When you put in a system like this, it’s almost like having a Six Sigma event. We typically do pilot projects with new clients to demonstrate how our system works, and there are a whole lot of ‘aha’ moments during that pilot project, and that’s why people buy the system.”
As companies continue to refine RFID and push prices lower and lower, there is also consensus that the technology itself will continue to evolve and add even more benefits, experts say. Labs’ dependency on such tracking systems will grow as they become more consolidated and volumes increase, making the logistics of sample processing ever more burdensome.
“Our goal is to know at all times the current and past locations of all samples from the time they are obtained until they are discarded,” said William Neeley, MD, medical director of the Detroit Medical Center University Laboratories. Neeley has followed the progress in RFID technology closely with an eye on fixing what now seem to be intractable problems.
Neeley sees the future in a new breed of cutting-edge RFID tags that boast read-write capabilities, where the patient information, requisition, and history of a sample gets programmed into the internal memory of the chip itself. “Read-write RFID tags have the potential to allow us to conveniently track all samples at all times,” he said. “These records will dramatically improve quality and provide accountability.”
Cost, size, and other factors have kept read-write tags out of the lab market for now, but Neeley believes they could be widespread within a decade. “It should be faster than that, but change in medicine is like turning a battleship. However, once people have this, they’ll wonder how they ever did without it,” he said.
“Sometimes labs don’t look at the big picture and see the last frontier, the preanalytical, which is our land of opportunity,” Neeley continued. “We do a pretty good job of getting data back to people, we do a better job every year with automation and do things that we only used to dream about, but the preanalytical is the orphan. That’s the one that’s lacking and where most of our money goes. I have more people in processing than I do in the chemistry lab. It’s in processing where most of our error comes, that’s where we have the greatest absenteeism, and the least stability.”
Following in the footsteps of early adopters like Mayo Clinic, the pattern so far has been that reference labs and others initially will test RFID on the anatomic pathology side, where high-value and irreplaceable samples make it easier to justify investment in new technology. “Ultimately, everything will be tagged, but initially most of the people that we talk to look toward the high-value tissue samples,” said Ambient ID’s Allison. “But with the mandates to cut costs, what we see a lot more of now is the reaching out to other industries like ours to find technologies and methodologies for healthcare systems to operate more economically. And it’s tools like our RFID LabTrack system that are going to catch on big time as healthcare looks to eliminate paperwork, improve patient safety, and boost efficiency.”