Stair-Fall Risk Parameters in a Controlled Gait Laboratory Environment and Real (Domestic) Houses: A Prospective...

Falling on stairs is a major health hazard for older people. Risk factors for stair falls have been identified, but these are mostly examined in controlled biomechanics/gait laboratory environments, on experimental stairs with a given set of step dimensions. It remains unknown whether the conclusions drawn from these controlled environments would apply to the negotiation of other domestic staircases with different dimensions in real houses where people live.

The aim of this paper is to investigate whether selected biomechanical stepping behavior determined through stair gait parameters such as foot clearance, foot contact length and cadence are maintained when the staircase dimensions are different in real houses.

Twenty-five older adults (>65 years) walked on a custom-made seven-step laboratory staircase. Older adults were classified into two groups (fallers and non-fallers) based on recent fall history. Among the 25 participants, 13 people had at least one fall, trip, or slip in the last six months and they were assigned to the fallers group; 12 people did not experience any fall in the last six months, so they were assigned to the non-fallers group. In addition, these participants walked on the stairs in three different real exemplar houses wearing a novel instrumented shoe sensor system that could measure the above stair gait parameters. MATLAB was used to extract fall risk parameters from the collected data. One-way ANOVA was used to compare fall risk parameters on the different staircases. In addition, the laboratory-based fall risk parameters were compared to those derived from the real house stairs.

There was a significant difference in selected stair-fall biomechanical risk factors among the house and laboratory staircases. The fall risk group comparisons suggest that high-risk fallers implemented a biomechanically riskier strategy that could increase overall falling risk.

The significant differences due to the main effects of the fallers and non-fallers groups were obtained. For example, when ascending, the fallers group had less foot clearance on the entry (p = 0.016) and middle steps (p = 0.003); in addition, they had more foot clearance variability on the entry steps (p = 0.003). This suggests that the fallers group in this present study did not adopt more conservative stepping strategies during stair ascent compared to low-risk older adults. By showing less foot clearance and more variability in foot clearance, the risk for a trip would be increased.