A brief introduction?
My name is Ghislain Leduc, I have been an epidemiologist for ten years, and I have been working for three years with the Sécurité Naissance network, which is the perinatal care network of the Pays de la Loire region. It is an association to which various professionals and stakeholders working around perinatal care are affiliated, whether before birth (such as obstetricians and midwives), around the time of birth, as well as pediatric services.
Within the Sécurité Naissance network, there is also the Grandir Ensemble network, which is the regional follow-up network for vulnerable children, particularly very premature infants. It aims to monitor these children from birth until the age of seven, in order to detect developmental disorders.
Could you tell us a bit more about your project?
The Grandir Ensemble network is mainly centered around referring physicians who agree to take part and who conduct medical consultations for children from birth until they are seven years old. These physicians are really at the heart of the network. Around them, there is a small coordination team of four or five people, whose role is to provide operational support in terms of tools for carrying out these consultations, as well as coordinating and animating the network.
Within this framework, physicians are required to refer families and children to a wide range of regional resources, whether professionals or organizations. And it was in this context that we decided to work with you.
There was always the difficulty, or very often the difficulty, of knowing which professional to refer families to and how to easily find those professionals. Two years ago, I developed a small Shiny application, really intended for internal use by the coordination team of four or five people, to help find, for a given family, the closest available resources from a geographical point of view.
This worked quite well for about one or two years for the coordination team. But we quickly realized that if we wanted to expand the user base and move towards something more industrialized, we would need support. With my own skills and the time I had available, I would not have been able to take the application further in terms of industrialization.
Why did you choose R Shiny rather than another technology?
I chose R Shiny for a very simple reason. My background is in epidemiology. Since I started working, now ten years ago, I gradually started using R, and then much more intensively, I had mainly been trained in SAS during my studies. Even though I tried to learn a bit of Python or other technologies in parallel, it remained quite anecdotal. R is really the tool I master the most today.
So when we identified a need and I started looking for solutions to address it with limited resources, using R and Shiny ultimately became the obvious choice for me, because it was the technology I was able to use and deploy independently.
What led you to contact Data Champ’? Why did you choose Data Champ’ over another agency?
We had this observation that the application was working well for the internal needs of the coordination team. We then had the intention to extend it, or at least to consider expanding it to more users, and ultimately to the network’s referring physicians. And the conclusion we reached was that this application was clearly handcrafted, and that there were some weaknesses that I had identified and that would clearly be blocking if we wanted to go further and expand the user base.
In my view, these weaknesses fell into three different categories: there was securing the access and the application itself, there was the need for more reliable user management including authentication and login methods, and then there were the application’s performance (obviously linked to hosting, as we were using shinyapps.io up to that point) as well as its long-term reliability.
There were several aspects where I felt blocked, and so we decided to call on a service provider to address these different issues.
I contacted four or five potential providers to request methodological and budgetary proposals for this project, based on specifications. Among these four or five potential providers, we identified two that clearly stood out. And upon reflection, Data Champ’ seemed to be the one whose proposal was, at least methodologically, the most detailed and the most aligned with our needs. And of course, in terms of budget, it also fit within our constraints.
Another important aspect was the quality of communication and exchanges, their simplicity. That was really what gave us the impression of the best overall balance in terms of fluid communication, a very human-scale approach, and a high-quality response.
How did the collaboration go? What aspect did you appreciate the most?
The collaboration went in an ideal way from my point of view. For several reasons.
The first reason is that the outcome is clearly there. The objectives were achieved. The application works, it has not yet been deployed on a larger scale, but that will happen in the coming weeks and months, so in terms of performance I cannot yet assess everything, but overall the result and all the levers that we had identified have been activated.
The quality of the exchanges and the simplicity of communication, which we had clearly identified as a key selection criterion, were fully confirmed throughout the project. The exchanges were smooth and responsive. Whenever we had questions, we received answers very quickly. You also showed enough flexibility to adapt to needs that may not have been clearly identified or were underestimated at the start.
On a more personal note, what I particularly appreciated (and which, to be honest, I did not necessarily expect) was the pedagogical approach you demonstrated. Particularly the audit of the existing application that you carried out, which, beyond laying the foundations for what you were going to implement, was also a major driver for improvement for me, very useful going forward.
So overall, the service, based on what we had identified, was very well delivered. And there was this additional bonus, which was truly very much appreciated.
What do you think of Data Champ’s working methodology?
We were impressed. We are a small organization, and we try to work with a structured methodology in terms of project management. Nevertheless, we were impressed by the project management and the tools you put in place. There was a real, clear visibility into the project’s progress, the resources that were planned, and those that were actually mobilized for each aspect.
The project management approach was clearly in place, and it was very valuable to have a very clear and detailed view of how you organize and operate in practice.
Are you satisfied with what was delivered as part of this collaboration? Did it meet your expectations?
The final delivery fully met our expectations. One of the points that had been specifically targeted as an area for improvement: the application was functional, it met the team’s needs. However, there were also usability issues that needed to be improved, as well as interface design and structural layout — which are closely linked to both the visual aspects and the application’s code.
The application as it was delivered fully meets our functional needs, as well as expectations in terms of usability improvements and user experience. And that was also an area where our expectations were less clearly defined, because it requires a certain level of expertise to imagine what is feasible and in which direction to go. Even from that perspective, our expectations were met.
Is there any aspect we could improve to make our service even better suited to your needs?
On a personal level, and I believe this was also a shared feeling within the team, this project management approach is, in my view, perfectly relevant for many projects (and certainly for projects much larger than ours). We were not entirely sure that it was absolutely necessary for a project of our size and structure. But in reality, those are considerations that mainly concern you and impact you internally. For us, everything we needed, we received.
And even if, hypothetically, this could be seen as a negative point, we absolutely do not consider this project management method to be inappropriate. We simply questioned whether the resources allocated to it were fully aligned with the project’s scope. Other than that, I have no additional negative points.
What does the future look like for the Grandir Ensemble network’s application?
There will be several future phases in the development of this directory. The first phase will be opening access to referring physicians very soon. There will be an initial phase where physicians will take ownership of the tool, we hope, will use it, and will most likely provide feedback on their experience and how they use it (and whether or not they find the information and features they need). We do not necessarily plan to define this precisely in time, but we will collect their needs at this level.
We are fully aware that the application is not meant to remain static from today onward. The idea is for it to evolve in order to best respond to the needs that will be identified.
And then the future developments, that is precisely the value of having chosen a Shiny application: there are a number of developments that I will probably be able to implement myself, over time, depending on available resources. And if there are developments that seem significant enough and beyond my capabilities, I will most likely call on you again to implement them.
The idea is always to adapt to physicians’ needs, and above all to provide them with something that is as useful and as relevant as possible. At this stage, it is difficult to anticipate what will emerge from all this. We will present the application to them, hope they will adopt it and see real value in it. And then we will see which direction to take it next.
Comments