Digital Mobility Outcomes to Describe Real-World Walking During Recovery from a Hip Fracture—The Mobilise-D Perspective

Background/Objectives: The burden of disease caused by fragility fractures, and hip fractures in particular, is significant both from a personal and a societal perspective. Digital technologies such as wearable devices now allow a real-world assessment of walking and physical mobility, describing temporal and spatial measures. Built on a transparent technical validation, the Mobilise-D consortium validated a set of digital mobility outcomes (DMOs) alongside a longitudinal Clinical Validation Study (CVS). This article provides a description of DMOs along with supervised mobility assessment and patient-reported outcomes collected at entry of the CVS study in a sample of home-dwelling participants with a proximal femoral fracture (PFF) recruited within one year from surgery. Methods: Participants who underwent surgical treatment for a low-energy PFF were recruited from five sites in three high-income European countries. The baseline visit encompassed a clinical assessment (~3 hours) and subsequent 7-day measurement of real-world mobility with a single wearable device. We categorized participants into four recovery phases according to the assessment time after surgery: acute phase (≤14 days after surgery, hospital stay), post-acute phase (15-42 days after surgery, mostly rehabilitation), extended recovery phase (43 – 182 days post-surgery, typically post-discharge at home) and long-term recovery phase (183 – 365 days post-surgery). All variables were reported descriptively for each phase using mean and standard deviation or median and 25th-75th percentile depending on their distribution. Results: We enrolled 564 PFF participants (66% women). The mean age was 77.5 (SD 9.6) years. Almost 90% of all participants (n=505) had a valid digital mobility assessment (wear time of >12 hours per day and at least 3 valid days). DMO domains such as amount (e.g. number of steps), pace (e.g. walking speed) and pattern (bout distribution) demonstrated large differences across phases. Conclusions: The observed variation in walking amount, pace and pattern across recovery phases indicate that DMOs provide a granular analysis of real-world mobility of hip fracture survivors. When confirmed by longitudinal analyses, including results on minimal important differences, the use of selected DMOs will provide a novel approach for monitoring, predictive modeling, prognosis, stratification and evaluation of clinical trials and hip fracture services.