Footwear and balance in older men.

Abstract

To the Editor:—We are writing this letter in reaction to the article by Robbins et al1 which, we feel, does little to illuminate mechanisms that relate footwear to the prevention and reduction of falling behavior in the elderly. The authors suggest that, while it is unwise to discount clinical observations, it is likewise dangerous to follow recommendations without objective confirmation. While we believe the authors conducted a well designed study, it is our opinion that the findings can not be validly extrapolated to gait on a level surface, but must be restricted to gait on a balance beam. At this time, we would like to focus only upon the issues that support this contention. Based upon the observed increase in the “balance failure frequency,” Robbins et al “prudently” conclude that the preferred shoe design includes thin, hard soles and that barefoot locomotion should simply be avoided by posturally unstable elderly individuals. We dispute these conclusions based upon the observations that “instability induced by even the most destabilizing shoe was not detectable by subjects walking freely on the floor of the laboratory but became apparent instantly upon mounting the beam” (p. 1093). Therefore, a direct and logical explanation of the findings of Robbins et al is that it was the experimental task, that is, beam-walking, and not the footwear that was the source of the instability. The direct influence of the measurement technique (balance beam) on the measured quantity (balance) violates a fundamental measurement principle. The authors were interested in assessing the effects of footwear on instability in elderly people and used a protocol that effectively induced instability. In addition, when walking on a beam that is narrower than the width of the forefoot, it is impossible to obtain a “normal” gait pattern. Since a beam of width 7.8 cm was used, we believe the subjects exhibited a gait pattern considerably different from the one they would normally adopt on a regular floor due to at least two effects: (1) most people do not walk with their feet directly in front of each other, and (2) during normal gait the feet are placed in a “toed-out” manner.2 In addition, elderly people may demonstrate a larger interfoot distance during the stance phase of gait compared with younger subjects. Thus, the wider the normal stance base, the greater the relative challenge offered by the beam, a factor that was not considered in the study. Similarly, no statistical control was included to consider the possible influence of shoe width (the range in shoe size was 7.0–10.5) in effectively providing a more stable platform on the narrow beam. This type of analysis may have provided more insightful data than the striking finding of a “significant relation between age and stability” that demonstrated a correlation coefficient of 0.0251! Over the 21-year age range studied (61 to 82 years), one could thus expect the oldest person to experience an extra half fall. Therefore, the data of Robbins et al indicate that a 40-year difference in age would be required to result in a difference of one fall. A similar argument could be made with regard to the significant relationship reported between subject height and stability, which demonstrated a correlation coefficient of 0.206. This adds to our impression that the method of testing is inadequate since, for all practical purposes, people of any age fall off the beam the same number of times. A good testing protocol would show a significant and meaningful difference in the stability of younger and older subjects. Robbins et al extrapolate their findings to the case of walking in a bathroom while wearing slippers. We believe that this is a selective example in which the authors attempt to relate their findings to the known incidence of falling in bathrooms. They do not, however, acknowledge that a hard bathroom floor effectively increases the stiffness of thin slippers, nor do they discuss why elderly people do not find carpeting to be hazardous (carpeting is certainly softer than a tiled surface). Thus, we conclude that the results of the study reported by Robbins et al should not have been broadly generalized but applied only to conditions in which elderly individuals are walking on a narrow beam. We suggest that those statements made by Robbins et al related to balance, in general, should have the phrase “on a balance beam” added as a suffix.