In the realm of ambitious scientific endeavours, the Mobilise-D project stands as strong evidence of hard work to balance research aspirations and the often unpredictable nature of real-world challenges. In a recent interview with Brian Caulfield, the leader of Work Package 3 (WP3) within the Mobilise-D consortium, and Arne Kuederle, David Singleton and Hugo Hiden, a vivid picture emerged of the complexities faced by the team. As they reflect on the journey, it becomes clear that the work undertaken in WP3 was not merely a technical checklist but a dynamic and evolving process, akin to coaxing the project through its journey one step at a time.
A broad mandate, a shifting landscape
At the project’s inception, WP3 faced an extensive mandate: the team was tasked with decoding and implementing the data requirements for two significant studies—the technical and the clinical validation studies. The roadmap seemed clear: understand data needs, define data pathways, establish a central platform, and orchestrate the intricate connection of data from ingestion to statistical analysis. However, as the WP3 team tells it, the reality was far from the neatly outlined plan.
The regulatory landscape, initially assumed to provide a clear guide, turned out to be somewhat blurry. Expectations of predefined standards were shattered, with the team realizing that they were venturing into uncharted territories. The absence of a roadmap, and the Mobilise-D consortium handed the task to develop one, meant WP3 had to adapt and innovate as they encountered each challenge.
One of the central predicaments faced by WP3 was the volatile nature of data requirements. Unlike traditional clinical trial scenarios where the data team come in once the protocol has been locked down, Mobilise-D found itself in the unique position of dealing with evolving data needs from the trials and the various cohorts (Parkinson’s Disease, Multiple Sclerosis, Chronic Obstructive Pulmonary Disease and hip fracture recovery). The team discovered that the customary approach—where protocols provide a rigid structure for data collection—was not in the books for Mobilise-D.
As Brian recalls, the initial data gathering process was anything but straightforward. In the first two year of the project, clinical experts, researchers in movement science, data analysts, and pharma industry involved in the project were still refining the study protocols and thus unable to provide concrete data specifications. But WP3 could not wait for that to be ready, so the team found themselves chasing a moving target, making decisions based on the limited information available, and accommodating unforeseen complexities such as multilingual variations and diverse data sources.
Adapting to the unforeseen: working as a team
As the team struggled with these challenges, the unanticipated arrival of the COVID-19 pandemic further amplified the complexity. Clinical sites, inundated with pandemic-related priorities, found it challenging to prioritize Mobilise-D’s recruitment of patients and the global lockdown created a scenario where the clinicians’ minds, hands and working time had to be prioritized for the population health. Hence, the process for fine-tuning data specifications for the project was slowed.
While we were moving forward from this and clinical protocol was almost in place, a new challenge appeared: sensor unavailability. The initial plan to use the McRoberts Dynaport sensor had to be adjusted in real time due to a worldwide chip shortage that limited the supply chain. The solution, which was the addition of the Axivity AX6 sensor, brought new technical challenges but the team worked with colleagues in the technical WP2 and McRoberts to modify the pipeline and distribution of sensors throughout the network of clinical sites, showcasing the project’s adaptability.
“From a pure computational perspective, once we adapted the data, the rest just worked”, said David Singleton.
David expands on the evolving requirements, emphasizing the collaboration with Clario (Mobilise-D CRO consortium partner) that provided us the platform to collect the data from patients. But here as well, there they were unforeseen challenges, like new parameters or forms that had to be developed in eScience and integrated, multiple languages, etc. Brian sheds light on how the study’s scale, with up to 500 data points being captured from participants at each visit, impacted Clario’s platform, revealing the unprecedented magnitude of the project: “I’ve certainly come around to the view that the scale and the complexity of what we were trying to achieve here really hadn’t been done before.”
Arne highlights the impact on the workforce, as introducing new tools and platforms required researchers to step out of their familiar tooling. The need for cross-disciplinary skills emerged, emphasizing the importance of a team capable of understanding both the domain-specific and technical aspects of the project: “We suddenly needed to teach the people, e.g. clinicians, recruiters, and students that needed to interact with the platform, the new skills to do that.”
Despite all the challenges experienced, every member of the WP3 team expresses their pride at being part of such an impactful programme and were enormously thankful for the collaborative spirit and learning opportunities that they experienced in Mobilise-D. Though they might say that they would do things differently if they could go back to 2019, they would definitely say yes if asked to do it all over again!
The long-term vision: looking ahead.
As we explore into the complexities faced by WP3, it becomes clear that the work undertaken was not a straightforward technical implementation but a dynamic process of collaboration, adaptation, and problem-solving.
As the project advances, the consortium’s next focus is on extending the study for specific cohorts. The team has emphasized the ongoing efforts to create an infrastructure and foundation for the long-term viability of Mobilise-D’s outcomes: this includes preparing algorithms and data for potential external sharing, as well as exploitation of data by the consortium members and the research and pharma community. They are also acknowledging the complexity involved in such endeavours and the time and budget needed to continue.
While the interview provided a comprehensive overview of the challenges faced and solutions implemented, there is a clear anticipation for future blog posts. Topics may include the impact of the project, its outcomes, and a deeper exploration of the data-sharing infrastructure being developed. As the Mobilise-D journey continues, the team remains committed to overcoming challenges and contributing valuable insights to the scientific community.
Stay tuned for more updates on Mobilise-D’s groundbreaking research!
