Charna Albert
August 2025—For routine diagnostic testing, capillary blood is a reliable alternative to venous blood, say the authors of a study published in May (DiPasquale C, et al. J Appl Lab Med. Published online May 19, 2025. doi:10.1093/jalm/jfaf059).
The study was led by Babson Diagnostics, of Austin, Tex., an independent startup spun out of Siemens Healthineers. Babson says it’s the first to validate capillary blood as a viable option for routine testing.
Babson uses capillary blood in its blood collection service, BetterWay Blood Testing. BetterWay requires one-tenth the volume of conventional venous methods and does not require a phlebotomist for collection, though it does require a trained health worker. The study, which comes one year after the company began offering its testing, is a report of Babson’s methods for reducing sample volume without compromising assay performance and overcoming capillary sample quality problems.
BetterWay uses what the company calls an “ecosystem” of different devices in its testing: the MiniDraw (a handheld fingertip collection device developed in partnership with Becton, Dickinson) and a hand-warming device and sample preparation device, both developed by Babson. No more than two tubes—one for serum and one for whole blood—are collected, depending on the patient’s order. The collection and sample preparation are done at retail pharmacies and outpatient clinics. After preparation, the samples are transported to Babson’s laboratory for analysis on Siemens Atellica and other analyzers of major manufacturers.
“The data presented in this manuscript is just the tip of the iceberg, in terms of the validations that we have done for this entire ecosystem,” says Chris DiPasquale, study coauthor and vice president of assay development at Babson.
The JALM report is divided into two parts. The first verifies the performance of 20 assays that were optimized for lower sample volumes by comparing their key performance characteristics with those of their unmodified versions. The second evaluates specimen equivalence across 39 analytes (20 “miniaturized” and 19 full volume) by comparing samples collected with the BetterWay system to samples collected with conventional venous methods. “They are essentially equivalent, within the allowable error goals,” DiPasquale says. At the time of the study, the 39 analytes were the full BetterWay test menu; today they are the majority of the menu.
The Babson hand warmer and BD MiniDraw work in tandem to prevent hemolysis. “It’s one thing to collect a drop and stick it on a glucose meter. It’s another to fill a tube,” says Alan H. B. Wu, PhD, D(ABCC), study coauthor and co-director of the core laboratory at Zuckerberg San Francisco General Hospital. (Dr. Wu is on Babson’s scientific advisory board.) When more than a drop of blood is needed—and Babson collects six to 18 drops for its capillary testing—the finger typically must be massaged to facilitate blood flow, which introduces leakage of intracellular fluid and can dilute the sample.
The hand warmer heats the palm before and during collection, vasodilating the blood vessels, Dr. Wu says. A fitted finger sleeve on the BD MiniDraw has attached wings that facilitate blood flow when squeezed. The wings control squeeze location and pressure, and blinking lights indicate to the collection technician the correct squeezing rhythm.
“The hand warmer and MiniDraw help to collect high-quality samples in a reasonable amount of time,” DiPasquale says. “With the innovation in the BD MiniDraw—the use of a finger sleeve to apply even pressure—we have been able to achieve very low levels of hemolysis with these capillary samples that enable broader panels to be more reliable and accurate.”
Babson’s sample preparation device, which is located at each collection site, performs much of the post-collection, preanalytical work that a phlebotomist or other health care worker would do for a typical blood draw. A digital information system automatically matches patient orders to a unique visit code to ensure the correct tests are performed. Labeling and matching the sample to the patient ID is done through a factory-barcoded system. “Our sample preparation device scans that and marries that tube to the individual and their test orders and accounts in our system,” DiPasquale says.
The device also mixes the sample. “Each tube has a different number of inversions required by the manufacturer, and that can be variable when you’re letting people do that on their own,” DiPasquale says. “Our preparation device standardizes and makes sure those are inverted the right number of times.”
The device is also a centrifuge. “We have some unique IP in that as well,” he says. “We use reverse centrifugation.” The serum tubes, he explains, are centrifuged upside down. The red cells, white cells, platelets, and gel consolidate in the cap when the tube is turned upright, yielding only serum in the tube.
With other microtainers, he says, when serum or plasma is separated by gel, the analyzer probe may get contaminated or clogged if it goes below a certain level in the tube. Consequently, an unusable amount of sample, known as “dead volume,” may be left at the bottom of the tube. Thanks to the reverse centrifugation process, the analyzer probe can contact the bottom or sides of the MiniDraw tube without issue, he says. “The gel and cells are in the cap and [get] discarded.”
“When the MiniDraw sample gets to the lab, you have an automation-ready sample, meaning you can de-cap the sample and you’re left with only serum that can go directly onto these automated analyzers,” DiPasquale says. BD manufactures and distributes adapters for the MiniDraw tubes, he notes, which allow them to be run on such analyzers.
Samples are kept in controlled, refrigerated storage before courier pickup. While in transit to the laboratory, a tracking device monitors temperature and disturbances like bumps in the road. “We own the sample from the moment it’s collected,” he says. “We monitor the quality of the transportation and make sure it’s delivered to the laboratory as pristinely as possible.”
Miniaturizing assays was a critical component of Babson’s methods for enabling the use of capillary specimens.
Companies that have attempted in the past to use capillary samples for broad-panel blood testing relied on predilution of samples, DiPasquale says. BetterWay, in contrast, “runs neat, undiluted samples directly on the analyzers.”
“By diluting the sample,” Dr. Wu says, “you are reducing the sensitivity of the test.”
“We’re not diluting unilaterally and then presenting that to the analyzers. This is targeted to particular parts of the analysis,” he says, “and therefore the preanalytic part is adjusted.”
Adds DiPasquale, “Assay optimization avoids the quality and accuracy issues caused by sample dilution.” BetterWay reduces sample volume through efficiency, he says, noting that the reverse centrifugation process reduces sample waste by 93 percent, while assay miniaturization reduces sample consumption by 55 percent. “With that reduction in sample waste and consumption, it is easy for us to run broad panels of tests with less than 10 percent of the typical blood volume.”
DiPasquale and his coauthors reduced the required sample volume to approximately 40 μL across 20 chemistry and immunoassay methods. (These included total cholesterol, LDL cholesterol, lipoprotein(a), HDL cholesterol, triglycerides, glucose, BUN, creatinine, ALT, AST, ALP, total bilirubin, direct bilirubin, total protein, albumin, total CO2, calcium, phosphate, TSH, and PSA.) They adapted FDA-cleared assays on the Siemens Atellica to use the reduced volumes and used unmodified reagent kits from Siemens to develop the miniaturized assays.
Says DiPasquale, “We’re leveraging the same reagents or formulations that Siemens currently offers to all clinical laboratories and shrinking down those reactions, while making sure the standardization, calibration, interferences, and accuracy are not compromised.” To ensure the performance characteristics met the design requirements, they used primarily venous serum for analytical validation. They found that the limits of quantitation of the full-volume assays were maintained and, in some cases, superior.
One innovation DiPasquale notes is Babson’s approach to the miniaturized triglycerides assay, which they modified to address interference from hand sanitizers and lotions. “We’ve implemented a reagent system, in addition to the Siemens formulation, that essentially wipes out exogenous glycerol that could be falsely counted as triglycerides,” DiPasquale says. “Conventional capillary collection, with more traditional triglycerides tests, is susceptible to that type of interference. We’re essentially enzymatically degrading the interferent before we even start quantifying triglycerides.”
All evaluations of miniaturized assay accuracy versus predicate assays resulted in correlation coefficients ≥0.995 when testing venous samples between both methods, DiPasquale and his coauthors write. Results below the lower limit of reportable results were excluded from analysis, meaning some analytes had fewer evaluable pairs. Lipoprotein(a) had the lowest number of evaluable pairs, at 73. Calcium, in contrast, had the highest, at 199.
“Not every male we enrolled and tested in these populations had detectable PSA,” DiPasquale says. “Both the venous sample and the capillary sample were below the limit of detection, meaning that person did not have any real circulating levels of PSA that we could measure.” Many also came in below the limit of detection for Lp(a), even with the full-volume assay. “Or they have such high circulating levels of these assays that they would be either too low or too high.”
Study participants were recruited across the ambulatory range, including targeted enrollment of participants with various chronic health conditions. “The BetterWay service is meant for ambulatory patients, at first. But that does not mean these tests do not have the analytical acuity to distinguish between abnormal and normal patients.” To validate the far ends of the measuring interval, additional samples were spiked to simulate extreme values, if those values were not present in the enrolled population.
In the second portion of the study, DiPasquale and his coauthors evaluated the equivalence of capillary samples to venous comparators for broad, routine diagnostic panels.
“We tested venous samples on unmodified tests and compared that to capillary samples tested with a combination of both unmodified and miniaturized tests,” DiPasquale says.
Participants were recruited from the general Austin population. Capillary and venous blood samples were collected from each participant using BD Vacutainer tubes, the BD Microtainer serum separator tube, and BD MiniDraw Capillary Blood Collection System devices, and they were processed to serum or whole blood for testing. For glucose, the comparators were conventional venous collection methods and conventional capillary methods; for total CO2, the comparator was conventional capillary collection only.
DiPasquale and his coauthors performed regression for bias analysis to demonstrate analytical equivalence between the different methods. For each analyte, bias estimates were compared to predefined acceptance criteria derived from various sources, including an equation that derived bias limits from allowable total error limits and maximum allowable standard deviation/coefficient of variation.
For each participant, two phlebotomists independently collected a BD Vacutainer venous sample and BD Microtainer SST capillary sample, and two health care workers independently collected a BD MiniDraw capillary sample. Where two venipuncture and two capillary samples were successfully collected, 46 to 114 evaluable pairs were included in regression equations for agreement and estimation of bias.
“With the [conventional] capillary collection, we had a lot of collection failures,” DiPasquale says. “Either not enough blood was in the tube for us to qualify that as a successfully collected sample, or hemolysis was too high in the conventional capillary, and therefore it was excluded.”
Test results below the lower limit of reportable results were excluded from the analysis, which affected evaluable pairs for lipoprotein(a), direct bilirubin, and PSA. Regression analysis demonstrated acceptable correlation and agreement, with correlation coefficients (r) ranging from 0.65 to 1.00, slopes between 0.95 and 1.11, and percent bias estimations ≤± 11.0 percent.
“The BetterWay launch menu includes 39 directly measured parameters from both whole blood and serum,” DiPasquale says. “We found that for 37 of the 39, we were within the error limits or allowable bias limits that would allow us to say that the capillary sample from MiniDraw with our methods is equivalent to venipuncture, or conventional testing.” Glucose collected with the MiniDraw was equivalent to conventional venous collection methods (105 evaluable pairs; correlation coefficient 0.986) and conventional capillary collection methods (77 evaluable pairs; correlation coefficient 0.989). Total CO2 was equivalent to only conventional capillary collection methods (79 evaluable pairs; correlation coefficient 0.648).
“There is a well-established physiological difference between total CO2 levels in capillary, arterial, and venous blood,” DiPasquale says. For total CO2, “We didn’t even attempt to demonstrate equivalence or lack of equivalence between the MiniDraw and the Vacutainer.”
Generations of conventional analyzers and assays have been designed for blood tubes of two to 10 milliliters, DiPasquale and his coauthors write.
In DiPasquale’s view, manufacturers often place a premium on making tests faster or more affordable, and some of those business decisions preclude innovations that would allow for the use of capillary samples on automated analyzers. “The assay developer is not necessarily concerned with sample volume limitations. However, there are strategies and methods to reduce sample volume while still making high-performing assays,” he says.
Reduced-volume assays that enable testing of a broader set of specimens can bring “more timely blood testing to a broader audience at affordable cost,” he says. Babson has shown that routine diagnostic panels can be run with such microsamples, he says, and others are working on similar efforts. “As this body of evidence grows, it could influence decisions that development scientists make in the coming years,” he says.
Prior studies have shown evidence of “drop-to-drop” variation in blood volumes less than 100 μL (Cao J, et al. Pract Lab Med. 2017;9:24–27). “The design of the MiniDraw collects a minimum of 225 microliters, mitigating this issue,” DiPasquale says. “We’ve demonstrated equivalence across all those parameters and do not see drop-to-drop variability with this collection system.”
In terms of interstitial fluid contamination, he says, “even across all our serum analytes, you don’t see evidence of consistent negative bias that would be indicative of ubiquitous dilution across the panels we test. I can’t say that if we were collecting 50 microliters or a drop of blood per collection that we wouldn’t have that. But the lavender and gold tubes collect 225 and 435 microliters, respectively, and those volumes successfully mitigate this effect.”
Since trained health care professionals performed all capillary collections, the conclusions from the study do not apply to self-collection, DiPasquale and his coauthors note. “At-home or self-collection is something on the road map that Babson is interested in and will be explored. However, the current clearance requires a trained health care worker to collect those samples,” DiPasquale says.
Self- or at-home collection is an industry objective, says Dr. Wu of San Francisco General Hospital. But the BetterWay system provides more than just a blood draw, he says. “It has the whole processing step, which you’re not going to get with self-collection.”
Babson is working now to expand its testing menu, DiPasquale says. For BetterWay’s initial launch, Babson focused on the laboratory tests typically ordered with a physical exam. In a future, phase two rollout, the testing menu will include less frequently ordered tests. Babson also hopes to expand its capabilities beyond its onsite laboratory. “Determining the methods and specifications for putting these samples in other laboratories is something we’re acutely focused on now,” he says.
Another goal is demonstrating that BetterWay could be a good fit for a more traditional health care ecosystem. It already has been used to good effect in community settings, says Elan Shoulders, MPH, who worked with Babson to bring BetterWay testing to a community-based men’s health event in New Jersey.
Shoulders is manager of community outreach and engagement at Hackensack Meridian Health’s John Theurer Cancer Center. At the event, which the cancer center cosponsored, Babson provided screenings for PSA, lipids, and HbA1c.
“Typically, when we are in the community we provide traditional laboratory testing,” Shoulders says. But venipuncture can be a deterrent for some who might otherwise get screened, she says, particularly when health or past addiction issues come into play. “We don’t want to create barriers to getting people screened for cancer, so whatever we can do to meet the community where they are, we try to adapt to that.”
She learned about BetterWay after BD, a Hackensack Meridian Health Foundation partner, offered to provide finger-prick health screenings at cancer center community health events. “When we met with BD, they didn’t screen for prostate cancer, which is a priority screening for our team,” Shoulders says. BD referred them to BetterWay, whose testing menu includes PSA.
Babson brought health care workers to the event to conduct the screenings and handled all other logistics, including sample processing and shipping the samples back to the company’s laboratory in Texas, Shoulders says. Test results were available the next day and participants were alerted via text message that results were available in their secure web portal.
At the event, Babson collected about 60 samples, which may have been a challenge for the cancer center’s one per diem phlebotomist, had they gone the route of traditional laboratory testing. “I don’t know that she would have been able to handle that large of a crowd by herself, being the only phlebotomist,” Shoulders says.
As with traditional laboratory testing, there may be a few patients who are unable to be drawn, she says. But at the event, “more than the majority of people were able to get their labs done.”
“In terms of community [testing], it works well,” she says. “It expands your level of reach and your level of engagement.”
Charna Albert is CAP TODAY associate contributing editor.