-Photoroom%201.png)
.svg)
Cardiff Met University: Scaling Remote Runner Research Outside the Lab With Always-On Wearable Data Capture
Key Results
2.5x
more participants
per study
through real-world, out-of-lab testing
~2,000
real-world runs
automatically captured
via connected wearables
12-week
baseline behavior context captured pre-enrollment
to measure true change over time
Case Study Summary
Cardiff Metropolitan University partnered with DashLX to run a remote 4-week gait retraining intervention study in runners with patellofemoral pain. By enabling participants to connect their own wearable devices, including Garmin, Coros, Fitbit, Polar, and Suunto, the research team automatically captured objective running load data before and during the intervention.
This approach helped Cardiff Met scale real-world runner research outside the lab by expanding both the number of participants supported and the depth of contextual data captured, while reducing reliance on self-reporting and minimizing manual work required to monitor adherence across the study. It also demonstrates a shift toward always-on research access, where researchers can reach specific populations on demand and continuously learn from real-world behavior over time.
Study Overview
Institution
Cardiff Metropolitan University
Study Type
Runner research study
Study Duration
4 weeks
Survey Tool
Qualtrics
Wearable Integrations
Coros, Fitbit, Garmin, Polar, Suunto
DashLX Role
Device connection + automated capture of running load and run-level details
“DashLX has opened a new door for injury rehabilitation in runners. We can now perform gait retraining in the real-world due to DashLX’s seamless integration with wearable devices, which provide crucial information regarding changes to running gait and distance run, allowing us to easily quantify adherence to our gait retraining protocol.”
Dr. Izzy Moore,
Associate Professor in Human Movement and Sports Medicine
Cardiff Metropolitan University
The Challenges: Scaling Runner Research in the Real World
1. Self-reporting limited data depth and increased internal effort
With traditional approaches, researchers typically relied on self-report measures of a small number of running load variables, such as total distance per week and number of runs per week. This constrained the depth and accuracy of insights while increasing manual effort for research and product teams to collect, clean, and interpret the data.
2. Lab-based research models constrained scale and real-world validity
Historically, gait retraining studies often required participants to attend in-person lab sessions. This created barriers to accessibility and scalability, limited participant diversity, and reduced the ability to observe real-world running behavior in natural training environments.
3. Limited contextual data made adherence and outcomes harder to measure
Relying on self-reporting restricted the volume and granularity of contextual running data available. This made it difficult for teams to accurately assess participant adherence, link training behaviors to outcomes, and understand how runners engaged with the program over time, while also increasing participant burden.
Overview of the Project: Remote Gait Retraining With Automated Real-World Data Capture
Cardiff Metropolitan University launched a remote 4-week intervention study in runners with patellofemoral pain involving three groups. The study measures pain, function, psychosocial measures, and monitors running load throughout participation, with follow-ups at 1, 2, and 6 months.
A key research focus is understanding how alterations in running load during an intervention impact pain and function. At the same time, the exact mechanisms behind pain reductions remain unclear, making it important to capture more complete, objective running behavior data without placing additional burden on participants.
Working with DashLX allowed the research team to use participants’ existing wearable devices (Garmin, Coros, Fitbit, Polar, and Suunto) to gather more detailed running load data than self-report alone.
Rather than relying on fixed timepoints, this approach supports continuous learning in real-world environments, making it possible to access specific populations when needed and treat each participant as an ongoing source of real-world insight.
Complete a series of surveys during the study about your knee pain
You will be allocated to 1 of 3 intervention groups
Follow a 4-week running intervention
Use your own device to share your running metrics with us
Why This Matters Beyond Research Teams
DashLX enables a world where research isn’t limited to discrete studies or narrow test groups. Brands can move toward always-on insight—accessing key personas and segments when needed, and turning every customer into an opportunity to learn through real-world behavior.
The Highlights
Initiative
Remote 4-week gait retraining intervention in runners with patellofemoral pain, designed to move injury research out of the lab and into real-world conditions.
Lived Experience Data
Captured objective running behavior at scale, including 12 weeks of baseline history and ~2,000 runs during the study. Run-level data included time/date, distance, duration, pace, step rate, elevation, walk/run duration, and average heart rate.
Automation
Participants connected their own wearables (Garmin, Coros, Fitbit, Polar, Suunto) for passive data capture, with AI-triggered surveys sent post-activity, reducing self-reporting and manual adherence monitoring.
Integrated Insights
Combined wearable-based LX data with post-activity survey responses to link running behavior, gait changes, pain, and functional outcomes, enabling real-time adherence verification and deeper insight into injury rehabilitation in real-world conditions.
Impact: Results at Scale in a Real-World Setting
1. Scaled the Study Beyond the Lab With a Larger Real-World Testing Pool
The study successfully supported 53 eligible participants in a remote intervention setting, helping expand participation beyond in-lab constraints and enabling runners to take part in real-world conditions.
Proof point: 2.5x more participants enrolled.
2. Built a More Complete Picture of Running Load Over Time (Baseline + Intervention)
Instead of limited self-report variables at a few timepoints, the study captured more complete running load context, including a 12-week objective baseline snapshot prior to enrollment and detailed run-level tracking throughout participation. This creates a model for ongoing insight access, enabling researchers to understand specific groups on demand and learn continuously from real-world behavior over time.
Proof point: 12 weeks of running history captured at baseline + ~2,000 runs recorded during the study.
Run-level fields captured included:
- date/time completed
- distance
- duration
- step rate
- pace
- elevation
- duration walk/ran
- average heart rate
3. Verified Adherence at Scale Using Real-World Run Behavior Signals
Having access to all runs across participants allowed the research team to check adherence to intervention guidance throughout the study. Preliminary data indicates participants are adhering to the interventions.
For example, in one of the three groups, participants were provided with guidance on increasing step rate and initially decreasing running load before gradually rebuilding based on knee pain ratings. Capturing all runs made it possible to confirm adherence to this approach.
Example Adherence Snapshot (Step Rate + Load Reduction Group)
Mean total weekly run distance and step rate of participants from the step rate + load reduction group who completed the 4-week intervention:
