Author(s): VAN ROIE, E., MEULEMANS, L., SEGHERS, J., DELECLUSE, C., Institution: KU LEUVEN, Country: BELGIUM, Abstract-ID: 2029

INTRODUCTION:
A good stair-climbing (SC) ability is crucial for independent living in older adults. A simple formula that estimates the mean power needed to ascend a flight of stairs in a predetermined time (i.e., total ascent duration) is easy to implement in practice, but lacks information on the actual utilized power per step. One might thus expect that sensor-derived SC power (i.e., actual utilized power) differs from SC power estimated through a simple duration-based equation (i.e., minimal power required for a given duration), questioning whether both methodologies actually measure a similar construct and whether both are equally sensitive for age-related changes. Therefore, the objectives of this study were (1) to compare equation-derived and sensor-derived SC power data; and (2) to explore the potential of both methods to detect age-related differences across the full adult lifespan.
METHODS:
Three-hundred eighteen participants (men: N = 156, women: N = 162; age 19-85 years) performed a 6-step SC test and two methodologies were used to estimate mean SC power: (1) a body-fixed sensor with automated detection of power production per step (Psensor), and (2) a mathematic equation based on timed ascent duration, body mass and stair height (Pformula). Linear mixed models were used to investigate the differences on SC power between both methods and their possible interaction with age and sex. Intraclass correlation coefficients (ICC 2,1) and Bland-Altman plot were used to better understand the relationship between Psensor and Pformula.
RESULTS:
Pformula was 210.4W lower than Psensor, lower in women versus men and in older versus young adults (p<0.001). The difference between Pformula and Psensor was greater in individuals with better performance (i.e., men and young adults) (p<0.001), indicating a ceiling effect of the formula in well-functioning and younger individuals. Likewise, ICC’s between both methodologies showed poor reliability in people aged <65 years (0.087-0.363) and moderate to good reliability in people aged ≥65 years (0.453-0.780).
CONCLUSION:
Participants with better SC performance are able to largely overshoot the minimal power required to ascend the stairs in a certain duration. This makes the sensor more sensitive to identify early age-related differences compared to the formula. However, further optimization of the methodology is necessary for large-scale implementation, e.g. by developing automated detection of SC events outside of the laboratory. It is also important to realize that a simpler approach to measure SC power, i.e. by only measuring time, can be just as valuable compared to the more complex sensor-based methodology in older individuals with limited performance. Future research should investigate whether higher values of SC power can postpone the development of future negative health outcomes. In addition, research should focus on identifying the threshold value of SC power below which functional impairments are likely to occur.