From laboratory to daily life assessment: a systematic review and meta-analysis of gait speed discrepancies in healthy older adults

Background: Gait speed is widely recognized as a key indicator of functional decline in older adults. Digital health technologies have extended gait speed assessment from the laboratory to daily life, but assessment methodologies and resulting values vary substantially between laboratory-based and real-world conditions, complicating clinical interpretation. To address this gap, this study summarized the instruments and data-processing approaches used to assess real-world gait speed, compared real-world and supervised gait speed metrics in healthy older adults, and examined age- and sex-related differences in real-world gait speed. Methods: A systematic review and meta-analysis were conducted following PRISMA guidelines. A comprehensive search of PubMed, Scopus, and Web of Science was performed for studies up to October 2025, utilizing search terms combining “real-world gait speed”, “wearable sensors”, and “older adults”. Results: A total of 11 studies met the eligibility criteria. The pooled mean difference between laboratory and real-world gait speed was 0.22 m/s (95% CI: 0.09–0.35, p < 0.001), indicating consistently higher walking speed under supervised laboratory conditions. Meta-regression analyses showed that age, sex, and monitoring duration were not significant moderators of this difference. Aggregated real-world gait speed data from 106,156 healthy older adults ranged from 0.81 to 1.28 m/s. Age-related differences in real-world gait speed varied substantially across studies. Within the same age cohorts, women tend to walk more slowly than men. Conclusion: Laboratory-measured gait speed was consistently higher than its real-world counterpart. Substantial heterogeneity was observed in age-related patterns of real-world gait speed across studies, highlighting the need for more standardized measurement approaches to facilitate future research on real-world mobility.