The primary purpose of this study was to examine the association between longitudinal changes in domain-specific cognitive functions with longitudinal change in IADLs. A strength of this study was that it followed older adults whose cognitive functioning was well characterized by detailed neuropsychological testing over an average of five years. Results showed that declines in both MEM and EXEC confer unique and additive effects upon everyday function. Thus, an individual who experiences a decline over time in memory would also be expected to show a concomitant decline in everyday function. Similarly, an individual who shows a decline in executive function would likely also show a decline in everyday function. Since declines in each cognitive domain have independent associations with change in everyday function, individuals who show change in both memory and executive functioning would be expected to show an even greater decline in everyday function than individuals experiencing a decline in either one of the cognitive domains alone.
Very few previous studies have examined longitudinal relationships between cognition and everyday function and so this study represents an important extension of previous cross-sectional research. As previous cross-sectional studies (i.e. (Bell-McGinty et al., 2002
; Cahn-Weiner et al., 2002
) have suggested, the present study further confirms that executive dysfunction has important ramifications on an individual’s functional capacities. Importantly, the present study suggests that longitudinal decline
in executive functions is associated with declines in everyday abilities. In particular, the present study suggests that change in those executive functions related to working memory, behavioral initiation and regulation, strategy generation, and abstract thinking and concept formation are associated with changes in daily function.
Perhaps somewhat more controversial, but also supported by cross-sectional studies (i.e. (Farias et al., 2004
; Jefferson et al., 2008
), the current study also shows that memory abilities make important contributions to a person’s functional capacities. Change in memory conferred its own effect on change in IADLs and this effect remained strong even when change in executive function was jointly included in the model. Such findings help to explain other recent findings that show individuals with MCI, many of whom have cognitive deficits confined to memory, demonstrate declines in everyday functioning (Tomaszewski Farias et al., In Press
We are aware of only one prior study that examined concurrent change in specific cognitive domains and change in everyday function in older adults. Previously Royall and colleagues (Royall, Palmer, Chiodo, & Polk, 2005
) found that change in executive function, but not change in memory, was independently associated with change in functional impairment. In contrast, the current study suggests that the effect of change in memory on change in IADLs is not entirely mediated by changes in executive function. The differences in results across the two studies may, in part, be the result of differences in the measurement properties of the different scales used in each study. In the present study we used measures of memory and executive function that were specifically designed to have similar measurement properties (i.e. similar reliability and sensitivity across a broad spectrum of ability level, and linear measurement properties such that neither scale has appreciable floor or ceiling effects (Mungas et al., 2003
). The use of psychometrically matched measures in the current study allows us to draw more confident conclusions about domain-specific cognitive effects on everyday function. Another potential reason for the difference in results between the current study and that of Royall and colleagues is that participants in the latter study were largely cognitively normal at baseline, whereas the sample in the present study represented greater cognitive diversity, including those with cognitive impairment and frank dementia. Thus, certain cognitive changes may be selectively important depending on baseline status: change in executive function may be more important in predicting change in normal older adults, whereas memory change likely becomes particularly important in predicting functional decline in MCI and dementia. In some support of this hypothesis, when we repeated the primary analysis using only those with cognitive impairment (MCI or dementia) only the association between change in MEM and change in IADLs reached statistical significance. Alternatively when analysis only included the normals, change in EXEC became associated with change in IADLs, although change in MEM was still also independently associated with change in IADLs (data not shown). Additionally, we observed that the EXEC scale showed more change in the normals than the MEM scale (see ) suggesting that the EXEC domain is probably more sensitive to the effects of normal aging. This finding is consistent with other literature which suggests that declines in executive functioning are associated with normal aging, and may reflect some loss in the integrity of white matter connection which are vulnerable to cerebrovascular disease (Kramer et al., 2007
In examining the magnitude of the relationship between the two cognitive domains and everyday function we found that the overlapping variance between MEM and EXEC with IADLs ranged from 48% to 52%. Such findings suggest fairly strong relationships between these domains. In a recent review article (N Chaytor & Schmitter-Edgecombe, 2003
) the authors concluded that cross-sectional relationships between neuropsychological tests and measures of everyday functioning are primarily in the moderate range, often in the 18% to 20% range (N.Chaytor, Schmitter-Edgecombe, & Burr, 2006
). Thus, longitudinal relationships among cognition and everyday function may be stronger than cross-sectional relationships. Further research is need to confirm this preliminary finding but if it proves to hold true it could have important clinical relevance and suggest serial neuropsychological testing maybe particularly useful.
In the current study we specifically selected a sample with broad variability of cognition and everyday function, ranging from fully cognitively normal to moderately impaired (at baseline). The assumption is that correspondingly broad variability of brain pathology will underlie this behavioral variability. We did not focus on separate analyses for normals, MCI, and demented cases because this inherently reduces variability and decreases sample size, and ultimately does not assess the continuous effects of pathology across its full range. Further, important for longitudinal studies like the present one, separate subgroup analyses also provides limited information about how functional limitations progress from normal cognition to severely impaired (Kraemer et al., 2000
The current study does have a number of limitations that deserve mention. A degree of caution about the generalizability of the results is warranted. The participants of this longitudinal study were as a whole, well educated and comprised of a clinical sample, primarily recruited from memory disorders clinics where AD is the predominant disease (selection bias). As such, our results may differ from studies utilizing older adults out in the community who are not actively seeking treatment.
The current study focused on two cognitive domains, memory and executive function because prior studies had identified these domains as especially important to daily function. The particular executive function scale used in this study was derived primarily, although not exclusively, from working memory and verbal fluency tests. Both verbal fluency and working memory are commonly considered measures of select aspects of executive functioning, and both have been linked to frontal lobe functions (for recent reviews see (Cabeza & Nyberg, 2000
; Henry & Crawford, 2004
). However, other executive functions not covered by this composite are also likely to make important contributions to everyday function. For example, measures of novel problem solving and practical judgment are likely to be particularly relevant to everyday functioning but were not included in the current study. Further research on which aspects of executive functioning are particularly important to functional abilities will be important. Other noncognitive/behavioral variables including depression can also play an important role in everyday function but were unfortunately not available in the current study. Additionally, while the BRDRS has advantages as a measure of everyday function (it has been correlated with postmortem pathological brain changes, and it is very simple to administer and time efficient), it is also has limitations because it is a fairly gross measure of everyday function. Also the use of informant-based ratings of everyday function offers both costs and benefits. Use of an informant or proxy to rate an individual’s everyday functioning has been shown to be useful in differentiating individuals with dementia from healthy elders (DeBettignies, Mahurin, & Pirozzolo, 1990
; Isella et al., 2006
; A.F. Jorm & Jacomb, 1989
; A.F. Jorm & Korten, 1988
; Kemp, Brodaty, Pond, & Luscombe, 2002
; Seltzer, Vasterling, Mathias, & Brennan, 2001
), in predicting who will go on to show further decline (A. F. Jorm, Christensen, Jacomb, Korten, & Mackinnon, 2001
), and in predicting incident dementia (Daly et al., 2000
; Harwood, Hope, & Jacoby, 1997
). A disadvantage of informant report is that it is subject to reporter bias.
Currently there is limited knowledge about the course and determinants of late life functional impairment, something which carries with it tremendous personal and social cost. This is the first study to show that longitudinal declines in both memory and executive functions are independently related to decline in everyday function in older adults. In conjunction with other findings, the current results provide further evidence that impairment and decline in memory and executive function play critical roles leading to functional disability in older adults.