R&D debuts at Vigilance Santé: an interview with the Director

Hello Jean-Félix Charest! Vigilance Santé recently created a “Research and Development” department. Could you tell us what motivated this decision and what the mission of this new team is?

Hello. This decision is a natural continuation of our approach. Vigilance Santé has always kept a close eye on emerging technologies to anticipate the transformations affecting clinical practice and clinical tools.

However, the arrival of artificial intelligence accelerated our thinking. We quickly realized this is a fundamental shift, one that must be understood, tested, and integrated rigorously. And since our clients and partners are also very interested, it felt important to move forward in step with our ecosystem and stay connected to real-world needs.

AI is also changing how information can be produced, structured, and used. Our value proposition is built on the reliability of the clinical information we provide to professionals. We therefore see an opportunity to go further and evolve our solutions into tools that are even more integrated into workflows and even more focused on decision support.

It is in this context that we created the Research and Development department. Its mandate is twofold: first, to strengthen our internal capabilities and the way we develop our products; second, to design and enhance AI-enabled solutions, always grounded in content validated by our pharmacists, to provide concrete support to health professionals. This direction aligns directly with our mission: facilitating excellence in medication therapy management for all health professionals.

We know that the term R&D can mean different things depending on the organization. Could you describe your team’s mission in a bit more detail?

Yes, with pleasure. In our case, the R&D team’s mission is structured around three major pillars related to product development.

• First, the research component involves maintaining active technology watch in digital health innovations, especially AI, to support product teams’ strategic thinking, including positioning, differentiation, and priorities as the market evolves.
• Next, the development component focuses on designing, experimenting, validating, and building technological innovations that improve both our solutions and our ways of working. In other words, we focus as much on optimizing products as on improving the efficiency and quality of our development processes.
• Finally, knowledge sharing acts as a catalyst: we structure innovation by turning ideas into learnings, then into prototypes, and ultimately into concrete solutions with demonstrated value for our users and for the organization.

Very interesting! I’d also like to talk about your role. Before leading R&D, you led the data team. How did that transition go? And how does your data experience influence your approach today?

I would say the transition happened quite gradually. At first, we set up an artificial intelligence committee by bringing together interested people and internal “champions.” This committee allowed us to launch a few initiatives and begin exploring AI’s potential.

Fairly quickly, we realized this approach had its limits. A few hours here and there were not enough to drive such a radical shift. So, we decided to go further by creating an incubator whose role was to explore opportunities concretely and validate our ability to deliver AI projects. The results were convincing, and that’s when we chose to evolve the incubator into a full R&D department that matches our ambitions. This new department is not limited to exploration, it also has a mandate to deliver products in collaboration with our development teams.

In my previous role as Director of Product, Data, I was already leading the AI committee and then the incubator. The transition to leading R&D therefore happened in a natural and logical way.

As for how my data background influences my R&D approach, it is central. Artificial intelligence tools learn from data, they rely on it to generate results, make predictions, and issue recommendations. A strong understanding of our internal data, combined with my understanding of the pharmacy context, helps me identify opportunities for new products, as well as levers to improve what we already do.

So, I see R&D as a natural convergence point between data, AI, and real-world needs in the field.

In this new function, what kinds of expertise will you be able to rely on? What does your team look like?

My team brings together colleagues with two complementary profiles: on one side, developers with several years of experience within the company who have trained in artificial intelligence; on the other, recent graduates in software development with a specialization in AI.

On my side, I bring expertise in pharmacy and data, which helps us keep projects grounded in real-world practice and in the quality of our inputs.

Finally, we work closely with the leaders of the company’s various product lines to ensure our initiatives align with priorities and translate into concrete improvements or new solutions.

Jean-Félix, when you look to the future, what do you see as the biggest challenges in developing AI-based clinical decision-support tools?

On the challenge side, we must learn to work with the limitations and risks of generative AI. The outputs are not always reliable, particularly because of hallucinations, which calls for validation mechanisms. The challenge is to put these tools in the hands of health professionals while ensuring that what is proposed remains robust, traceable, and trustworthy.

There is also the issue of confidentiality, which is non-negotiable. This is something we approach very proactively and in a structured way. Concretely, our approach is to put in place infrastructure and guardrails that allow us to maintain control of data: ensuring it is not used to train commercial models, that it does not leave Canada, that it is protected according to best practices, and that all legal obligations regarding information protection are fully respected. In other words, innovation will not come at the expense of security, this is a baseline condition of our approach.

Thank you. I’d like to come back to the notion of trust. Your tools have been used in real clinical settings for more than 30 years. How do you reassure your clients that, even in an R&D context, performance and trusted data will still be delivered?

What makes us strong is the high level of trust that health professionals place in our data and content. These resources have always been produced, and above all, validated, by pharmacists, and that is a principle we want to preserve. Our goal is not to replace human clinical expertise with automation, but to support it.

Concretely, AI becomes a lever to accelerate certain steps in content production and integration, while maintaining systematic validation by the pharmacy team. It helps us make information available more quickly in our tools and easier to access on a day-to-day basis.

And beyond content, AI also opens the door to new functionalities that were not possible before, always with the same priority: supporting health professionals with reliable, contextualized, and useful information.

Could you give us a concrete example of a feature or use case that AI makes possible today, and that would have been much more difficult, if not impossible, to achieve before?

Of course. We can take the classic example of a conversational agent. Today, everyone knows tools like ChatGPT, Gemini, or Claude, which allow you to ask questions in natural language and receive an answer. The challenge with these general-purpose tools is that they can make mistakes, and they are not necessarily adapted to our specialized clinical context.

On our side, we are currently developing a conversational agent that answers questions by relying only on our data and content. It is also able to indicate the source of the information used to formulate its response. Concretely, a person no longer needs to navigate through multiple pages to find what they want, they can simply ask their question and get the relevant information faster.

And that is only a first step. By combining this capability with information from the patient record and the results provided by our analysis engine, we open the door to more personalized and contextualized recommendations. Before generative AI, this kind of experience, conversational, fast, and anchored in a controlled knowledge base, was much harder to offer.

That will certainly be a highly appreciated feature. To stay on this topic, I’d like to know how, in concrete terms, your R&D projects will improve the solutions already used by healthcare teams.

Our objective is to enrich our solutions through AI in ways that are very tangible in the day-to-day work of care teams. The conversational agent is a good example: it simplifies access to information and reduces the time spent searching.

Another important lever lies in improving our analysis engine. Take patient records: they often contain a large amount of information in free-text notes. Until now, these notes were not used by our analysis engine simply because they were difficult to interpret automatically. With AI, we will be able to better understand and structure this type of information and integrate it into our analyses.

The result: recommendations will be more relevant, more complete, and above all, better contextualized for each clinical situation, because we will be able to go further into the patient record data.

Will your team focus more on improving existing products or on creating new solutions?

We will work on both fronts: improvement and development.

We have many ideas for using AI to enrich our existing solutions, for example, a conversational agent in RxVigilance, more advanced analysis in our engine, or adding a transcription tool in RxConsultAction.

At the same time, we also have ideas for new products, such as tools to read a prescription and automate entry into a patient record, detect changes between the prescription and the patient record, or detect issues on a prescription and propose pre-filled interventions, among others.

That balance is important. You regularly collaborate with health professionals and partners. Concretely, how do these people contribute to your R&D projects? Do you have examples where their feedback can truly guide, or even reshape, a project?

Yes, absolutely. First, we want to establish an Innovation Advisory Committee composed of pharmacists, physicians, nurses, and other stakeholders. The objective is to stay grounded in real-world needs: to confirm we are working on the right problems, and then to ensure what we develop truly addresses the issues raised.

Then there is the whole aspect of pilots and testing. In the past, we could do a lot of validation internally. With AI, what we can do internally is less representative, because performance and value really emerge in real usage contexts. We therefore need external feedback, especially from pharmacists, physicians, and nurses, to confirm that our solutions are relevant, reliable, and well integrated into day-to-day practice.

When Vigilance Santé invests in R&D initiatives, how do you plan to measure their real impact, whether on medication safety, clinical outcomes, or workflow efficiency?

We certainly want to implement metrics, as I mentioned earlier, to monitor application performance in real time. But beyond technical indicators, we also need on-the-ground feedback: understanding what care teams truly gain from our tools and how they make a difference.

In short, we need data and measurement to ensure the tool works well, remains stable, and stays reliable over time, and we can’t do that without relying on real usage data.

Then there is the clinical and operational impact. We want to ensure our solutions genuinely help care teams, whether by reducing time spent on certain tasks, improving analysis quality, or supporting decision-making.

The exact measurement approach will depend on the projects, but the objective is clear: to develop tools whose value can be seen both in the indicators and in real-world experience.

Since the R&D department is a relatively new service, how do you define your development priorities? And more concretely, what healthcare challenges, especially in pharmacy, are you aiming to solve in the short and medium term?

When it comes to concrete projects, our focus is on simplifying and enabling excellence in medication therapy management for health professionals.

That said, trying to project too far ahead would be unrealistic in our context. Emerging technologies evolve quickly, as do market needs and practice environments.

So, our priority is as much about direction as destination: staying agile, continuously adjusting our development strategy, and delivering concrete improvements in short cycles. To move faster, our approach is to explore different avenues, prototype quickly, and validate in the field as early as possible. That is exactly what we have started doing since establishing the R&D department.

Take pharmacy as an example. In recent years, we have seen a significant expansion of scope of practice, but at the same time there is a staffing shortage and growing operational pressure. We therefore want to develop solutions that make it easier, according to the highest standards, to carry out the full range of clinical acts that these professionals are now authorized to perform.

At the same time, we also want to optimize operational efficiency in pharmacies.

Here is a concrete project. We recently started designing a prototype aimed at automating prescription entry and validation in pharmacies. These tasks are repetitive, high-volume, and often low value-added, and they are also a source of errors when manual transcription is involved. For us, this is not only an opportunity to improve productivity, it can also improve the reliability of information.

The prescription is often the starting point of clinical activity. If it can be analyzed even before it is entered into the software, we can detect issues earlier, perform a more relevant analysis, automatically flag certain medication changes, and even simplify documentation and billing for clinical services.

Ultimately, we want pharmacists to spend less time on administrative work and more time with the person under their care. This is only one project among others, but it illustrates our development approach well: test quickly, learn fast, and focus our efforts where the impact is real in the daily practice of health professionals.

Jean-Félix, if we were doing this interview again in five years, what results would make you say that creating an R&D department at Vigilance Santé has been a true success and has created tangible value for the healthcare system?

In five years, if we have helped concretely transform the way pharmacy work is done by reducing operational burden and strengthening the quality of clinical decisions, we can consider that we achieved our objective.

What I mean by that is: we will have enabled pharmacy teams to carry out their full range of acts smoothly, especially with the expansion of their scope of practice, without feeling overloaded.

To get there, it will involve new products and new solutions, but also adding key features to our existing solutions to lighten daily work, improve workflows, and support clinical practice.

Thank you very much. That’s inspiring. To close, do you have one last message for our readers, Mr. Charest?

I think I would say that health professionals will not have a choice but to embrace AI and new technologies. But I would like to reassure them and tell them that the solutions Vigilance Santé develops will be there to support them in a responsible, practical, and useful way in their day-to-day work.

In addition, we are always looking for pharmacists who would like to test new solutions and feel comfortable sharing their feedback, whether through pilots, tests, or other formats.

I’ll end on a human resources note. Since our team is very young, we are looking for AI-specialized developers; we are always open to meeting people who share the same passion as we do.

In all cases, you can always reach out to us with questions or to learn more about our projects.

Perfect! Thank you, Jean-Félix Charest. This was very interesting.

It was my pleasure! I hope we’ll have the opportunity soon to discuss projects we will have successfully delivered.