Building and improving decision support tools based on real world practices: Meet new professor William Van Woensel

William Van Woensel was hired in summer 2022 as an assistant professor in health informatics and information systems at the Telfer School of Management. He received a PhD in applied computer science from the Vrije Universiteit Brussel in Belgium. Prior to his arrival at Telfer, he worked as a research associate at Dalhousie University. We interviewed him to learn more about his research interests.

Why did you choose to study health informatics and information systems? Any personal motivation?

My background lies in computer science, specifically, knowledge representation and reasoning, as well as mobile computing. I found health informatics and information systems to be an interesting and fitting, yet also challenging, application area. My motivation in my research is the clear need for decision support tools that are based on business and/or clinical knowledge, as well as the need for mobile apps for end-users to help with monitoring and health decision-making.

How does your PhD training inform your current research program?

During PhD research projects, clinicians would explain how things are done in practice, which was often different from the “official” clinical guidelines. Real-world experiences would often inform care for atypical cases, such as multiple co-occurring illnesses. While my current research includes the capturing of decision logic from guidelines, it thus also extends into the discovery of real-world processes found in event logs. At the same time, it has become clear that the rapid prototyping of evidence-based mobile apps, which help patients with health decision-making, could be highly beneficial as well.

Do you have any upcoming publications to share? 

I submitted a paper to the Journal of Biomedical Informatics (JBI) on clinical decision support centred on multimorbidity, a highly complex problem. The paper introduces and applies a community-based evaluation: instead of reading other teams’ papers, this type of evaluation directly includes those teams in setting up and performing the evaluation, leading to more consistent and objective results.
I further believe it is imperative that recommendations provided by decision support systems are explainable, in terms of the applied logic and domain knowledge. When providing decision support to clinicians, such explanations engender trust. When helping patients with health decision-making, explanations can help with education and behaviour change.

How can your research influence business communities in Canada?

On the one hand, automated decision support has the potential to improve compliance with best-practice guidelines. On the other hand, discovering real-world processes and outlining deviations from official guidelines has the potential to have a significant impact on organizations. Instead of simply enforcing compliance, organizations can use this research to gain insights into, and ultimately learn from, real-world processes. Moreover, providing health decisional aids to patients themselves, tailored to their context and needs, can empower patients in their self-care.