| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
| ||||||||||||||
LETTER TO THE EDITOR |
Department of Internal Medicine Section on Gerontology and Geriatric Medicine Wake Forest University School of Medicine Winston-Salem, North Carolina
Address correspondence to Hal H. Atkinson, MD, Department of Internal Medicine, Section on Gerontology and Geriatric Medicine, Sticht Center on Aging, Wake Forest University School of Medicine, Medical Center Blvd., Winston-Salem, NC 27157. E-mail: hatkinso{at}wfubmc.edu
To the Editor:
We appreciate the commentary provided by Dr. Royall regarding our recent work demonstrating that baseline measures of global cognitive function and executive control function (ECF) are associated with subsequent declines in gait speed (1). Dr. Royall points out the very important consideration that baseline cognitive function is likely one point in a trajectory of cognitive change for the individual. It is certainly possible that the change in the shortened version of the Executive Interview (2) (EXIT 15) over time may yet predict change in gait speed, but we have two primary reasons for assessing only baseline cognitive function in our analyses: 1) a measure of cognitive function at a single point in time has value for the clinician to assess an individual's risk of subsequent physical decline, and 2) the EXIT 15 and CLOX1 (3) were only obtained during Year 3 of the Health, Aging and Body Composition Study, thus it was not feasible for us to address change in these measures. Although gait is not traditionally considered a primary cognitive task (in contrast to most Instrumental Activities of Daily Living (IADLs) in Dr. Royall's work (4,5), we also speculate that associations would exist between change in ECF and gait speed.
Our discussion briefly outlines possible reasons for the attenuation of the association of the EXIT 15 and CLOX1 with gait speed decline. Dr. Royall's letter points out one of the possible three-variable problems that we may be encountering—conditions, such as depressive symptoms, might be responsible for longitudinal changes in cognitive function that directly influence gait speed. In this case, ECF would be a mediator of the relationship between conditions such as depressive symptoms and subsequent gait speed. However, there are a couple of other possible explanations involving three-variable problems that we can discuss using depressive symptoms to illustrate: 1) It is possible that there is a true confounding relationship; that is, ECF may be associated with depressive symptoms and depressive symptoms associated with gait speed decline, when there really is no direct relationship between ECF and gait speed. 2) It is possible that depressive symptoms and poorer ECF are both indicators of an underlying disease process that may be directly responsible for gait speed decline; for example, lesions in the frontal lobe or subcortical vascular disease (6), which might affect both ECF and physical performance directly. We favor either Dr. Royall's suggestion that ECF is a partial mediator between covariates and physical decline or that ECF and the covariates are indicators of a common underlying disease process that directly affects physical performance. Unfortunately, all of these three-variable problems are indistinguishable in a statistical model. Further longitudinal studies, mechanistic studies, and interventional trials of the effects of improving cognitive function on gait will provide the best support for a direction of causality in the relationship between cognitive function and physical performance decline.
References
| ||||||||||||||
| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
