Research Square · Oct 27, 2026
Recovery in Place: A Multimodal Pilot Study of Neighborhood Context, Life-Space, and Social and Physical Functioning After Lower-Limb Fracture Surgery
By Chu CH, Abedi A, Julian P et al.
Background Older adults returning home after lower-limb fracture surgery or hip replacement recover within heterogeneous social and spatial contexts, yet neighborhood conditions are rarely integrated into post-discharge monitoring. Guided by the capability approach and the life-space concept, this study examined recovery as a situated process shaped by residential context, activity space, and changing social and physical function after discharge. Methods Eighteen community-dwelling older adults completed eight weeks of monitoring, contributing 1,008 person-days of multimodal sensor data and repeated biweekly clinical assessments. Clinical measures included the Social Isolation Scale, Oxford Hip Score, Oxford Knee Score, Timed Up and Go test, and 30-second chair-stand test. Residential and visited locations were linked to amenity, crime, and Canadian census-derived socioeconomic data. Generalized estimating equations tested associations between socio-spatial exposures and the Social Isolation Scale, Oxford Hip Score, and Oxford Knee Score. Predictive models compared sensor-only, socio-spatial-only, and combined feature sets for all clinical outcomes under random-shuffled and cross-person validation. Results Neighborhood material disadvantage was associated with lower social interaction in both residential and activity-space analyses. More location transitions per day were associated with higher social interaction after correction for multiple testing, while other location-derived mobility measures showed exploratory associations. Activity-space exposure to libraries and food establishments was associated with better hip function after correction for multiple testing, whereas similar amenity associations were not observed around the residence. Predictive performance varied by model, outcome, and validation setting. Prediction errors were generally higher under cross-person validation, and socio-spatial features improved prediction for some outcomes and models but not others. Conclusions These preliminary findings suggest that post-discharge recovery is shaped by the interaction between changing bodily capacity and the social and spatial environments that become accessible during recovery. Integrating residential and activity-space context may improve interpretation of remote-monitoring data, but larger longitudinal studies are required before clinical application.