Assessing real-world gait with digital technology? Validation, insights and recommendations from the Mobilise-D consortium.

The sample was a convenience sample of 108 participants, and nine participants were excluded from analysis because data were unavailable. The cohorts were small, including 12 participants with congestive heart failure, so results may not apply to all patients. Monitoring lasted 2.5 hours, so it may not capture day-to-day fatigue or longer-term mobility changes. The study did not compare wearable-derived outcomes with a laboratory-based reference system, because it focused on real-world gait. The ranking method was simplified and did not fully assess algorithm performance separately for slow, medium, and fast walkers. Real-world factors such as turns, slopes, stairs, obstacles, crowdedness, and visibility were not accounted for and may have affected results. The effect of walking aids on algorithm performance was not assessed, although walking aids were used by participants in several cohorts. Stride length estimates had large errors, so they are not ready to guide clinical decisions about walking distance or gait change.

Funding came from the European Union's Horizon 2020 research and innovation programme and the EFPIA via the Innovative Medicines Initiative 2 Joint Undertaking Mobilise-D project, NIHR facilities, the Spanish Ministry of Science and Innovation, the Generalitat de Catalunya, and the IMI2 JU. The supplied text does not include a separate author conflict-of-interest declaration.

Do not conclude that real-world wearable gait monitoring is ready to guide treatment decisions. This study validated measurement algorithms against a reference system, not patient outcomes, and stride length was often inaccurate, especially in slow or short walking bouts.