Related Experiment Video
Updated: Sep 19, 2026
Video Movement Analysis Using Smartphones (ViMAS): A Pilot Study
Published on: March 14, 2017
Preinjury Smartphone-Derived Mobility Metrics Are Associated With Discharge Location After Lower Extremity Fracture
Ian P Marshall1, Patrick J McGlone1, Janse T Schermerhorn1
1Department of Orthopaedic Surgery, Walter Reed National Military Medical Center, Bethesda, Maryland.
Background:
Early prediction of nonhome discharge after lower extremity fracture treatment may improve care coordination and reduce costs. While preinjury functional status is an established predictor, it is commonly assessed subjectively. We hypothesized that preinjury Apple Health mobility metrics, including step count, walking speed, step length, walking asymmetry, and double support time, would be associated with discharge disposition after fracture management.
Methods:
We conducted a retrospective cohort study of adults with lower extremity fractures. Preinjury Apple Health mobility data within 26 weeks of injury were analyzed. The primary outcome was discharge home vs. nonhome. Multivariable logistic regression assessed associations between mobility metrics and discharge location, adjusting for age (<65 vs. ≥65 years) and isolated vs. nonisolated fractures.
Results:
The cohort included 249 patients (median age, 39 years; 58% male). Overall, 179 (72%) were discharged home, and 70 (28%) to nonhome facilities. Patients discharged home were younger and more likely to have isolated fractures (p < 0.001 for both). In the adjusted model (area under the receiver operating characteristic curve, 0.80; 95% confidence interval [CI], 0.74-0.86), higher preinjury step count was strongly associated with discharge disposition. Compared with <2,500 steps/day, 2,500 to 5,000 steps/day (OR, 2.73; p = 0.03) and >5,000 steps/day (OR, 4.21; p = 0.001) were associated with increased odds of home discharge. Walking asymmetry <10% (OR, 3.21; p = 0.01) and faster walking speed (OR per 0.1 m/s, 1.31; p = 0.01) were also associated with higher odds of home discharge. Patients <65 years with isolated injuries and >5,000 steps/day had a 94% probability of home discharge (95% CI, 90%-98%) compared with 8% (95% CI, 1%-16%; p < 0.001) among those ≥65 years with nonisolated injuries and <2,500 steps/day.
Conclusions:
Preinjury smartphone mobility data were independently associated with discharge disposition after lower extremity fracture treatment. These objective measures may improve perioperative risk stratification and discharge planning in orthopaedic trauma populations.
Level Of Evidence:
Prognostic Level III. See Instructions for Authors for a complete description of levels of evidence.
