October 2026—This conversation is part of a series of one-on-one virtual roundtables in which CAP TODAY publisher Bob McGonnagle speaks with representatives of a company to spotlight its laboratory solutions. He spoke on July 21 with Nick Fingland, PhD, PMP, senior director of R&D operations and science, BD, and Chris Farnsworth, PhD, D(ABCC), section head of clinical chemistry, WashU Medicine, and professor of pathology and immunology and of emergency medicine, Washington University. McGonnagle’s questions and their answers follow. The title of their talk: Where quality begins: the real impact of blood collection choices.
Dr. Farnsworth, you and Dr. Kimberly Huggler coauthored an article—“A Match Made in Heaven: Has Your Blood Collection Tube Been Appropriately Validated with Your Assays?”—that was published in the May issue of CAP TODAY and forms the nugget of today’s roundtable. How do you consider a tube verified when you get it from a reagent manufacturer?
Dr. Farnsworth (WashU): When talking about verifying a tube, we first have to discuss the terminology. The Clinical and Laboratory Standards Institute document GP34-A, “Validation and Verification of Tubes for Venous and Capillary Blood Specimen Collection,” differentiates validation and verification processes. The validation in this case is on the level of a manufacturer, unless I’m manufacturing my own tubes.
Generally, I’m going to verify a tube’s performance. I don’t have to demonstrate that a tube performs on a large study for precision and across all the analytes we can potentially use. In many cases it’s going to be a small study. At the crux of our conversation today is: When is it a verification or a validation, and when do I have to do nothing?
Dr. Fingland, what steps do you take before you hand off tubes to customers like Dr. Farnsworth for his verification in his laboratory?
Dr. Fingland (BD): It starts with understanding what the customer, i.e. a laboratory, needs our products to do. When we think about our validation process, we are looking at developing tubes at a high rate, at a global scale, and with precision across all the different lots, and we also look at the relevant considerations among and within different regions. We think about what assays and instruments our customers are employing and how they are employing them, because there are regional differences in how a product is used. We think about it not just from the moment of the test but also in transit prior to collection, during collection, post-collection, arrival at the lab, and even beyond that for it to complete the life cycle and use. We also consider the customer’s unique requirements and edge cases where the tube might break. We think about the challenges associated with operators who have different levels of education or experience in a lab or clinic and build that into our validation efforts.
With two of our recent FDA clearances, we tested over 50 analytes on each of the tubes. We subjected them to a battery of tests to make sure we’re covering those uses. Our company has a legacy of more than 75 years with many of our products. We’re ensuring that it is not only the new products that carry on the legacy but also the current products as they evolve, and we’re doing so with laboratories in mind.
A collection tube is involved in several steps within a laboratory workflow, starting with the tube itself, the sample that goes into it, which then meets a diagnostic reagent in an instrument, and we probably have to account for an automation line. Dr. Farnsworth, if a resident asks you how to do this with a new lot of tubes, what would you tell them?
Dr. Farnsworth (WashU): The question is more about changing tube types—going from a 4.5-mL lithium heparin tube to a 2.7-mL tube. How to demonstrate they are comparable really varies. There are scenarios in which you’ll want to do an n-of-20 comparison—you’ll go to the floor, draw across multiple cohorts, and try to span your measuring range with two different tubes. From an FDA perspective, these are multimillion-dollar studies for which Nick’s team must find labs that will help; doing it at a hospital level is difficult.
Oftentimes we’re getting healthy volunteers, which won’t yield results that span the measuring range but we’ll get an idea of whether the numbers will be approximately the same. That’s probably the best a clinical lab will be able to do in most circumstances, unless it has good access to patients and an institutional review board that’s willing to go along with that design.
Many labs do a study, but I don’t know that it’s always needed. In many circumstances, we can look to the literature. The CAP checklist requirement notes that the laboratory director can determine whether verification by the laboratory is needed based on a review of the clinical literature and by evaluating information from specimen collection container and instrument manufacturers. You don’t necessarily have to do a big study.
This has started to play out interestingly in a hospital system setting. Many of us live in 10-, 20-, 30-hospital systems. If we have a lab that has demonstrated equivalency between two products, I don’t think every lab in the system has to do the same. One lab having the data is probably sufficient, and we can disseminate that data throughout our labs, assuming they’re using the same testing manufacturer, et cetera. It’s case by case; it’s not as simple as a one-size-fits-all approach.
We’re in an era of enormous health care systems with many institutions. Dr. Fingland, your tubes are going into the mothership, in many cases, where there’s a large automated line for analysis, but many tubes remain local and go on equivalent analyzers at scale for the source they’re on. When BD says it has validated a tube in one place, much of the burden falls on BD to make sure that holds across the board, does it not?
Dr. Fingland (BD): It does. And I agree with Chris’ comment on the comprehensiveness of the studies we have to design. What might be a similar workflow within a hospital network could be very different within the U.S. versus other regions of the world. We have to understand that variability and test to that variability.
It can be hard to cover the analytical measurement range for your own laboratory studies, particularly if you’re looking at analytes used in only certain populations. We design tests to cover relevant analytical measurements.
Dr. Farnsworth, how is tube oversight affected when there’s an FDA approval for a new test for an analyte we haven’t been testing before with an FDA-approved test, or there’s a new instrument that has launched and the instruments you have from company A are being replaced by instruments from company B?
Dr. Farnsworth (WashU): When we change an assay reagent—for example, from a Jaffe to a creatinine-based method, two totally different methods—I’m going to be particular about covering the measuring range, demonstrating we won’t have significant changes in our reference interval, and demonstrating precision. Let’s say when that original creatinine Jaffe assay comes to market, BD does the validation on its side, demonstrating its tube works the same as the predicate tube. But if Roche or Abbott updates that assay, I’m not sure Nick will have the chance to update the BD tube and say BD has tested it on this one as well.
There’s a lack of knowledge for most labs to comprehend that we’ve changed from a previous iteration of an assay to a new iteration—we know the tube works on this assay, but we haven’t shown it works on the new one. A package insert from one of our reagent manufacturers, as an example of how they do it, reads: “The sample types listed were tested with a selection of sample collection tubes that were commercially available at the time of testing. Not all available tubes or manufacturers were tested. They were tested from various manufacturers, contained different materials, which could affect the test results.”
So they acknowledge they can’t test them all. We as a lab have no idea which ones they tested, how much they tested them with, how many samples they were in, and if they tested the whole range. We’re taking it at face value. From the reagent side, that’s the gap we see that we’re hoping someone is filling on the back end, but we’re not always sure.
Dr. Fingland, I envision IVD reagent manufacturers have you on speed dial so they can tell you when they’re bringing out a new assay or new instrument, or the sample size for their new assay is much smaller than it has been before, and they anticipate needing a different-size collection tube or dilution, et cetera. Tell us from your perspective what starts the ball rolling at BD when that news arrives.
Dr. Fingland (BD): We maintain strong connections with instrument companies through a liaison group. It allows us to have a finger on the pulse of what’s going on such that as companies release these assays or instruments, our tubes are in the mix.
As we develop new products, we’re seeking to generate evidence throughout our product development process and post-launch so we’re able to publish articles that show the evidence of our products working with different analytes. That can help give labs the basis for lightening their verification process. If that’s of interest to your lab, we would like to hear from you and partner in those efforts.
The tubes BD manufactures are used on a variety of instruments with a variety of assays. Do you have a universality of providing solutions?
Dr. Fingland (BD): Yes. As I mentioned earlier about how we go about our validation and the two recent clearances in the U.S., we had over 50 analytes on each of the tubes because we know the wide range of use cases and panels that are done on them. As we think about multimillion-dollar clinical studies that we would endeavor, we consider potential uses for each region, use cases, and customer communities to make sure we’re covering those as comprehensively as possible. Needs change and evolve over time—for example, high-sensitivity troponin versus troponin I—so it’s important we target what’s relevant.
We’re trying to conserve as much blood as we can on patients. Dr. Farnsworth, tell us about that and how it might affect your relationship with a tube manufacturer.
Dr. Farnsworth (WashU): There still is a desire among our providers to get frequent draws in both adult and pediatric patients, with a recognition of trying to miniaturize what we can draw and how much blood we need. Many of our assays require a small amount of blood, but much of the automation in that space and with those manufacturers requires quite a bit of blood. This is an important component we don’t usually think about until we’ve tested our automation and realized it doesn’t work with the smaller tubes we’ve selected. It’s often an afterthought.
As an aside, I wish reagent manufacturers and automation vendors would get together to figure out the right amount of blood that is needed.
Many tubes have important additives, gels, EDTA, and so forth. Talk about the importance of that for sample stability, knowing how many samples from hub-and-spoke health systems are on the road or in airplanes. Dr. Farnsworth, do you ask your tube vendors along with the reagent manufacturers?
Dr. Farnsworth (WashU): We haven’t asked what their stability is in various scenarios. Many laboratorians tend to look first at the reagent manufacturer instructions for use. We don’t default to the tube manufacturers. It doesn’t commonly state in their IFU what the stability is on cells, what the stability process is, what the stability of frozen is. We need to know that information so we can ensure the sample is treated accordingly and we aren’t providing bad results. We don’t always have access to that information or look for it in the right spots, so we often have to do an internal validation study to demonstrate the same.
Dr. Fingland, talk about your studies and ideas around sample stability.
Dr. Fingland (BD): We start our validation process by thinking about it from the lab’s perspective. We think about all the inputs for preanalytical error. A key factor is how the product is handled. Are the number of inversions followed? Is it stored upright or on its side? Is it refrigerated? What is the product going through before it gets in the lab’s doors?
We pride ourselves on understanding different stability conditions at different temperatures for our products, but not everything is tested. When labs are working through a particularly unique workflow, our company is interested in helping make that successful. We have programs that help navigate through preanalytical errors and answer those questions.
On blood conservation, I agree with Dr. Farnsworth in that there’s a paradox—tubes draw a large amount of blood but the tests continue to miniaturize. One potential solution is how we can make the tube more performant so you can draw a smaller amount of blood with an additive ratio that’s conducive to a smaller blood draw, providing more operational flexibility with the same result.
With the emergence of new biomarkers that labs are starting to look at, as well as the proliferation of AI and connecting dots we previously didn’t consider, another question is whether we can do more with the blood than the test we’re initially using it for. Is there another use for the blood that can put a different lens on patient health? Is there something we could do that we’re not considering today?
Dr. Farnsworth, what should laboratories prioritize when they’re evaluating blood collection solutions?
Dr. Farnsworth (WashU): Know what has been assessed in that tube. Look at your reagent and tube manufacturer IFUs and know what the claims are and what has been tested. You may have to ask what tubes were used for a particular validation or what conditions they tested their tubes in for a particular study. There’s opportunity for us to collaborate more with reagent and tube manufacturers and ask that they provide the data. Also, look for it in the literature. See if it exists. Those are great first steps before we invest in laborious, difficult studies that are probably underpowered to demonstrate what we want them to do.
The onus is still on us to do wet studies to demonstrate verification, if there’s not good literature showing something works in the way it should. There are probably gaps in which a tube has not been addressed appropriately for a particular assay or vice-versa, and we can step in and do larger studies and publish them so other people can work off of them.
Start by looking for what the gaps are in what a tube has been tested for and push on that; then try to do your own work to demonstrate the tube is doing what it should.
Dr. Fingland, same question. If you were speaking to one of your laboratories, what should they prioritize as they evaluate blood collection solutions? And what are the biggest opportunities for improvement in current practice?
Dr. Fingland (BD): Operator variability in the sample collection process presents many challenges in terms of the uniformity of collection and getting the sample ready for the lab. There’s also a bit of a communication divide between the lab and the clinical space in which it’s hard for labs to understand what truly went wrong with the sample. Education is key. We think a lot about developing products that take the variability out of the collection process.
When labs are selecting a tube vendor, the testing that’s been done on the tubes should go into the equation. Also, understand a vendor’s ability to deliver at scale, as your needs grow and evolve, and as the world changes. For example, many of our products are affected by supply chains in the Middle East, so having a provider that can be resilient to changes in the market and still be able to provide quality with the results you expect and rely on for patients is an important consideration.
Dr. Farnsworth (WashU): I second that.
Dr. Farnsworth, WashU is a formidable teaching institution. You have residents and fellows in multiple specialties, many of whom will need to use the clinical laboratory. They will need to order IVD blood tests as part of what they know and do every day. What education around these issues do you provide when you give a lecture to internists or oncology residents?
Dr. Farnsworth (WashU): I gave a talk this past year to a medical school in Hong Kong and was shocked to see how much clinical pathology they get in their medical school. It’s largely absent from the programs I’ve experienced in the U.S. In our residency program, I teach about our blood tubes and collection draw order, and for some, it’s the first time they’ve heard about the tube types, what is in the tubes, and why we draw in a particular order, which is shocking. There are many opportunities at the medical level to train medical students more on what a tube is, why it’s important, and what’s in it.
With residents, we drill down on preanalytic error. The importance of how we get the right tube, matrix, and draw order are crucial for anyone going on to be a lab director, but it expands into all providers.
Dr. Fingland, a final word from you on this question of the education of the coming physicians and other providers who will need to collect blood and understand the tubes they’re using and their role in taking care of patients.
Dr. Fingland (BD): The majority of error is in the preanalytical pathway, so there is a role for us to remove the sources of error through creative design. Can we remove the guesswork associated with draw order in case an accident happens? Can we give labs flexibility around draw volume for different operational parameters? How can we do that together and partner with you? In your daily life, you’re putting together many puzzle pieces so you can confidently deliver the right diagnosis. If we can work together to remove guesswork, we’re doing the right thing.
CAP TODAY does not endorse named products in commercial roundtables, and the publisher’s comments are his own and not necessarily reflective of CAP TODAY or College of American Pathologists policy.