A recent study of the performance of an autonomous robotic phlebotomy device (ARPD) in a clinical setting revealed some surprising results. Not only did the first-stick success rate rival that of manual phlebotomists, but patients were highly enthusiastic about using the device in the future. Eighty-two percent indicated they would either prefer to use the ARPD again, or had no preference.
To learn more about the study, the patients' response, and the implications for phlebotomy, we spoke with lead researcher, Thijs C. van Holten, Clinical Chemist at St. Antonius Hospital in Utrecht and Nieuwegein, the Netherlands.
But first, we caught up with one of the ARPD developers, Luuk Giesen, CMO, Vitestro, to hear what the study results mean for the manufacturer.
What, for you, was most striking about the ARPD clinical study results?
I was quite surprised by the patient experience results. Ninety percent of patients said they found the blood draw by machine less painful, far less painful, or similar to a manual blood draw. Quite a number of patients also said that next, time, they would prefer using the machine for their blood draw.
When we started on this venture, several years ago, patient acceptance of an ARPD was still an open question. But the study results suggest patients are very receptive to the idea.
I also heard, anecdotally, from the clinics, that patients were really quite willing to participate in the study and try out the ARPD
How might the ARPD evolve in light of the findings?
We have a large engineering team, of between 60 and 70 people, who are continuously working on iterating and refining the technology. It remains a challenging project, and we'll never be done refining it.
How do you picture the future of phlebotomy in the light of these study results?
I think that going for a blood draw will become a little like going to the supermarket – patients will be offered the choice of whether to be served by a human or a machine.
And I think that choice will extend to community settings eventually. While our current focus is on automation for high-throughput, centralized settings – namely major hospitals – we are aware of the trend towards decentralization of care. I predict that one day we will see ARPDs in community settings such as GP practices, and maybe even pharmacies.
What’s next for the ARPD product development?
Alongside launching the product in Europe, we're also working on obtaining FDA approval in order to gain access to the US market.
There might be further publications to come out of the current study, and we have another trial still ongoing in Europe, testing new iterations of the technology. Eventually, we plan to conduct clinical trials in the US.
Ultimately, our vision is that the ARPD becomes the standard of care for blood draws in phlebotomy.
Read more about Vitestro's ARPD development here.
What challenges led you to explore an automated approach to phlebotomy?
One of the main drivers was a staffing issue. We were dealing with a high level of sick leave among employees, which created a significant shortfall in our workforce. As a result, waiting times in the blood draw department became very long, and that was something we needed to address.
Were there also technical or workflow challenges you were hoping to improve?
Yes, alongside the staffing pressures, we were facing the usual day-to-day challenges associated with phlebotomy. Things like selecting the correct tubes, ensuring the right fill volumes, proper mixing – all of those routine but critical steps.
These are issues every blood collection service encounters, and our department is no exception. The idea with automation is not only to help with capacity, but also to improve consistency and reduce the likelihood of these kinds of errors in the process.
How did you first become involved with the development of robotic phlebotomy?
Back in 2016, while I was training in Delft as a resident in clinical chemistry, I was approached by one of the entrepreneurs with a simple question: “Do you think it’s a good idea to automate blood draw?”
To be honest, my initial reaction was a bit hesitant. But at the same time, I found the idea interesting, so we stayed in touch and had a number of discussions in those early stages.
What led to the collaboration developing further?
Things really progressed when I moved to St. Antonius Hospital. It’s a non-academic hospital, but there’s a strong focus on innovation and research, and a real effort to make those activities possible in practice.
At that point, I suggested we should explore working together more formally, because the environment was well suited to it. Around the same time, the ARPD manufacturer also relocated to Utrecht, so we were physically very close, just a short distance from the hospital.
That made collaboration much easier, and it’s really how we moved from early conversations into active development and research together.
How did staff respond to the introduction of robotic phlebotomy?
Overall, the response from phlebotomists was quite positive. Of course, there were some initial reservations, which is understandable. Whenever a new technology is introduced, especially one that could affect daily work, people naturally ask what it means for their role and their future.
How did you address concerns about job security?
From an early stage, we were clear that we don’t see robotic blood draw as a replacement for manual phlebotomy. There are clear limitations to the system. For example, it’s designed for adult patients and isn’t suitable for young children. There are other patient groups where manual blood draw will remain the better option.
The goal is really to support the service rather than replace it. It can help improve turnaround times and increase availability, especially during busy periods or when staffing is stretched.
Communicating that clearly helped reassure staff. Once it became clear that the technology is there to assist rather than replace them, people were much more comfortable and, in many cases, genuinely interested in seeing how it could fit into their workflow.
How did patients respond to the idea of robotic blood draw?
There was actually a high level of enthusiasm from the outset. In the early stages, we conducted interviews to gauge willingness, and around 90 percent of people indicated they would be open to having their blood drawn by a robot.
One interesting observation was the demographic spread in how people responded. We found that younger patients were, somewhat surprisingly, a bit more reserved about having their blood drawn by a machine. In contrast, middle-aged and older patients were often more open, even quite enthusiastic.
What did you observe once patients experienced the system in practice?
Around 90 percent reported that they were satisfied with the robotic blood draw, in terms of comfort. Of the small minority who reported dissatisfaction, it was generally not related to the procedure itself, but more to broader concerns about automation and the idea of robots replacing human roles. From a procedural standpoint, acceptance was very high.
Was there anything the patients didn’t like about the ARPD approach?
As with manual phlebotomy, you do still see occasional bruising. That’s not something you can completely prevent, and it’s comparable to what you would expect in routine practice.
Are there specific areas you’re looking to improve?
Yes, there are a few practical aspects we’re working on refining. In particular, things like applying the bandage and minimizing any blood spillage after the draw. These are relatively small parts of the overall procedure, but they do matter for the patient experience.
Where do you see further opportunities for automation in the workflow?
One area we’re actively exploring is integration with pneumatic tube systems, so that once a sample is collected, it can be sent directly to the laboratory.
At the moment, tubes are collected in a container and still need to be handled and transported by staff. That’s a step that could be automated, and it would streamline the overall process.
Automating other aspects of the workflow, such as tube selection and patient identification could improve both efficiency and safety. Those steps are still manual, and they’re also where errors are most likely to occur.
Do you see specific clinical settings where this technology could have additional impact?
One area that stands out is the emergency department. We see relatively high rates of hemolysis there, and I would be very interested in evaluating whether a robotic system could help reduce that.
Of course, it wouldn’t be suitable for all patients. For example, it’s not designed for patients who are unconscious or unable to cooperate. But for ambulatory patients coming into the emergency department, it could be a useful addition, allowing blood to be drawn quickly and consistently by the system.
How do you picture the future of phlebotomy in the light of these study results?
The way we envision it is a blended workflow, with manual phlebotomy and robotic phlebotomy in parallel. Patients will be invited to choose between the two options for their blood draw.
For the phlebotomy staff, this will introduce a hybrid role. On one day, a staff member might be overseeing and coordinating the robotic system, making sure everything runs smoothly. On another day, they would be carrying out manual blood draws as usual.
So rather than replacing existing roles, it creates a more varied way of working. It combines traditional skills with oversight of automated systems, which many of our staff find both engaging and a natural next step in how the field is evolving.
Newsletters
Receive the latest pathologist news, personalities, education, and career development – weekly to your inbox.
