Home Stretch | Biostatistician on the bus

During the first wave of Covid-19, when hospitals were flooded with patients, biostatistician and bus driver Ruben Deneer retreated behind his computer screen. He developed a model for emergency physicians to quickly diagnose Covid-19 infections using standard blood values. His CoLab score, which is calculated by a smart algorithm, became a success.

He never imagined he would obtain a doctorate, did not consider himself the type of person to spend four years delving into studies and research. That was especially true when he interrupted his studies to pursue a childhood dream and was finally allowed to get his bus driving license at 21. But after eight years of driving around the East Brabant region as a full-time bus driver, the urge to study returned.

Ruben Deneer has been with the General Clinical Laboratory of the Catharina Hospital for several years now. There, he works as a biostatistician creating models to support doctors and nurses in making the right diagnoses; this Wednesday, he will defend his dissertation at the Department of Biomedical Engineering. “I used to think of it as a nice recruitment pitch: that as a biomedical engineer, you’re trained to bridge the gap between physicians and technology. The way I saw it was that in practice, you’d end up in one of these two worlds, but the great thing is: I’m truly right in the middle. This became even more apparent to me during the pandemic. We were able to apply our skills to relieve pressure on the emergency department (ED) wherever possible. That applicational aspect gives me immense satisfaction.”

Real-world data

His dissertation combines multiple studies, with the common denominator being the use of real-world data. “These data are collected for the purpose of allowing the physician to track what’s happening with the patient. It’s not like a study, where you can think about the setup beforehand and collect data systematically. Real-world data is very complex and has its pitfalls, but the advantage is that you can obtain a large and diverse data set in a short period of time.”

Pending the medical ethics committee’s approval of his own clinical study – Deneer wants to estimate the risk of re-admission or death in patients discharged from the hospital after admission for acute heart failure – he starts his doctoral period by joining an already ongoing study in collaboration with the Department of Cardiac Surgery. “After bypass surgery, some patients experience dysfunction of the new artery for various reasons. In order to identify this as soon as possible, we use a cardiac biomarker that can tell us something about the condition of the heart over time. It was difficult to make a model that could show abnormalities statistically; doctors ran into problems with the irregularly obtained data. Because in practice, not every patient is monitored at fixed times, and so we had to factor in a continuous time scale. The logistics within the care process have great variability.”

Golden data

But just as his own study is about to start, the Covid-19 pandemic strikes. It was a hectic time, says Deneer. “Soon, there was a whole raft of models being published in haste. At the beginning, I had my doubts whether we should join in on this, as most of these models turned out to be unusable. But – and this will sound very bad to those who had Covid-19 during that initial period – in Eindhoven, we were close to the epicenter in terms of infections. Our ED has an extensive laboratory package and combined with the many patients, this allowed us to obtain “golden data”. That was the deciding factor for me – we had an opportunity to actually contribute something practical.”

Thanks to a smart algorithm developed by Deneer in close collaboration with his co-supervisor Arjen-Kars Boer, it is possible to determine within an hour whether or not a patient has Covid-19 by means of a blood sample collected immediately upon entering the emergency department, as standard practice. At a time when the first PCR tests still take over a day to complete and the capacity of isolation rooms is strained, the CoLab score proves to be a very useful tool for doctors. It allows them to safely send patients with other conditions to the regular ward. Several Dutch hospitals start to adopt Deneer’s model to reduce pressure on care where possible during the initial peaks of Covid-19.

Infectiousness

The CoLab score has not left the ED, but it is being used in a different way now, Deneer explains. “Initially, the goal was to single out low-risk patients during the huge patient influx into the ED. Now that far fewer Covid-19 patients are being admitted, it is actually helping the doctor identify who has a high risk of Covid-19 infection. Thanks to a European grant, we were also able to incorporate the effect of the number of Covid-19 cases in the environment into our model, making the calculations even more accurate. But what we really want to know is not so much whether a patient has Covid-19, but whether they are infectious to other patients. After receiving my PhD, I will continue to work on this, in collaboration with the Zuyderland Medical Center and Maastricht UMC+. They have developed a new PCR test to measure infectious virus particles and we’re now going to examine whether a correlation exists between the development of the CoLab score and the degree of infectiousness.”

In addition to his scientific research, Deneer continues to drive the bus in the Eindhoven region. Still, it was in his conversations with patients that he realized that his job as a bus driver is more than just “a paid hobby to get all the numbers out of my head”, Deneer emphasizes. “By driving around on the bus, I can pierce my TU/e bubble and I feel closer to society. And you take that feeling back with you into the hospital.” But, he concludes with a big smile, “Above all, it’s just a lot of fun to drive people around on Friday or Saturday nights. This combination makes me a happy person.”

Share this article