Cognitive Function May Be Important for Fall Injury Prevention Trials

To the Editor: We read with interest the recent paper by Lamb et al.1 We found the process of identifying a common outcome data set for fall injury prevention for European trials interesting and informative and congratulate the investigators for their efforts. This work is somewhat reminiscent of the Frailty and Injuries: Cooperative Studies of Intervention Techniques collaborative effort,2 in which investigators moved the field of falls research forward in the early 1990s. Nevertheless, we were surprised by the committee's recommendations with regard to cognitive function. It was not clear whether this feature was purposely being left out altogether or if this was merely not being included as a target for reducing fall-related injuries. Either way, we wish to call attention to a growing body of research that links cognitive function, gait disturbances, and falls. Using dual-task paradigms, many studies have demonstrated that cognitive resources, in particular attention, are necessary for the proper maintenance of balance and gait and that subjects with an increased risk of falls display a heightened adverse response to dual tasking.3 For example, we found that an attention-demanding task does not affect the stride variability of healthy older adults but markedly increases stride variability, a marker of fall risk, in older adults prone to falls.4 A number of investigations have also demonstrated an association between cognitive function, gait instability, gait speed, and other measures related to fall risk.4,5 Moreover, “stops walking while talking” has been shown to predict falls in older adults,6 and executive function and other aspects of cognitive function are apparently altered in fallers and may also be predictive.4,7,8 Finally, studies on patients with Alzheimer's disease may offer further insight into the relationship between cognitive function and falls. These patients suffer from a highly increased fall risk with a fracture rate that is two to three times as high as that of the age- and sex-matched general population.9 This is likely due to deficits in cognitive function (perhaps they take inappropriate risks), because traditional aspects of motor function do not predict falls in this population. In addition, intervention programs that do not include a focus on cognitive function have not been successful in reducing fall risk in this population.10 It may, perhaps, be premature to target cognitive function as an outcome measure in trials designed to reduce fall-related injuries. Nonetheless, we suggest that there is already sufficient evidence to warrant inclusion of cognitive function measures, in particular attention and executive function, in any such trials, at least as a baseline covariate, to adequately characterize the study population and the propensity to fall. Financial Support: The authors are supported in part by National Institutes of Health Grants AG14100, HD39838, and AG08812. Author Contributions: JMH initiated the idea of writing this letter. GY wrote the initial draft, and both authors revised it. Sponsor's Role: NA.