Associations of Laboratory-Based Gait Speed and Daily-Life Gait Speed With Frailty, Sarcopenia, and Life-Space Mobility in Community-Dwelling Older Adults.

OBJECTIVES

Although daily-life gait speed (DGS) may better reflect real-world walking than laboratory- or clinic-based gait speed (LGS), it is unclear whether DGS provides superior discriminatory value for clinical outcomes. We compared LGS and DGS on their discriminative abilities concerning the risk of prefrailty/frailty, probable sarcopenia, and restricted life-space mobility (RLSM) in community-dwelling middle-aged and older adults.

DESIGN

Cross-sectional study.

SETTING AND

PARTICIPANTS

A sample of 1025 participants (mean [SD], 63 [7.3]years) completed a survey and fitness assessment, from which LGS, DGS, FRAIL (fatigue, resistance, ambulation, illness, and loss of weight) score, handgrip strength, and life-space assessment were measured.

METHODS

LGS was measured by the 10-m gait speed test and average DGS was computed from walking trips recorded in a 14-day travel diary on a geospatial tracking mobile application. Participants with a FRAIL score ≥1, a weak handgrip strength (<18 kg for women and <28 kg for men), and a life-space assessment <60 points were classified as having prefrailty/frailty, probable sarcopenia, and RLSM, respectively.

RESULTS

LGS and DGS were weakly correlated (Spearman rho = 0.22), and their difference (mean ±2 SD) was 0.33 ± 0.48 m/s. Across outcomes, LGS showed higher areas under the curve (AUCs) (0.63 to 0.74) than did DGS (AUCs, 0.59 to 0.73). Adding DGS to LGS did not significantly increase the AUCs for the discrimination of all outcomes (differences in AUCs, 0.00 to 0.01; P values, 0.15 to 0.70).

CONCLUSIONS AND IMPLICATIONS: In a sample of generally well-functioning adults, although LGS and DGS were not interchangeable, sufficient discrimination accuracy for the prefrailty/frailty, probable sarcopenia, and RLSM outcomes could be achieved with LGS alone. These findings suggest that routine laboratory gait speed assessment may be sufficient for risk stratification in community-dwelling adults.