Baseline cognitive function predicts rate of decline in basic-care abilities of individuals with dementia of the Alzheimer's type
Archives of Clinical Neuropsychology 22 (2007) 99–107
Baseline cognitive function predicts rate of decline in basic-care
abilities of individuals with dementia of the Alzheimer’s type
Timothy B. Atchison a,∗ , Paul J. Massman b,c , Rachelle S. Doody c
a Department of Psychology, Sociology, and Social Work, West Texas A&M University,
Box 60296, Canyon, TX 79016-0001, United States
b Department of Psychology, University of Houston, TX, United States
c Department of Neurology, Baylor College of Medicine, TX, United States
Accepted 2 November 2006
Abstract
Decline in basic self-care abilities is an important risk factor for institutionalization in individuals with dementia. The ability to
predict such decline would be of clinical importance in working with families of dementia patients. Research has suggested that
cognitive decline may precede loss of functional capacity. This paper utilized a large sample of probable Alzheimer’s disease patients
(N = 150) who were evaluated longitudinally to assess the pattern of neuropsychological functioning predictive of rapid decline in
self-care. The findings indicated that despite initial equality of Lawton Physical Self-Maintenance (PSM) scores, patients showing
rapid decline of PSM function displayed significantly more impaired performance on neuropsychological measures at diagnosis.
They also exhibited a statistically significant difference in the pattern of scores from patients who remained stable. The pattern
of the rapid declining group included more severe impairment in visual spatial skills, processing speed, and concept formation.
Difficulties in using individual patients’ cognitive profiles to make predictions about future rate of PSM decline are discussed.
© 2006 National Academy of Neuropsychology. Published by Elsevier Ltd. All rights reserved.
Keywords: Alzheimer’s disease; Neuropsychological assessment; Disease course; Self-care function
The ability to predict the rate of decline of cognitive functional abilities in Alzheimer’s disease (AD) patients is
of practical importance. Rapidly progressing patients require greater resources for management sooner, making the
early identification of such patients meaningful for both family and clinicians (Mortimer, Ebbitt, Jun, & Finch, 1992).
Individual rates of progress of cognitive deterioration in AD have been shown to vary significantly between patients
(Stern & Jacobs, 1995). However, factors associated with this variability are not well understood and attempts to predict
cognitive decline have had only limited success (Chan, Salmon, Butters, & Johnson, 1995). In part, the difficulty in
predicting cognitive decline may arise from the choice of measures available to index general cognitive function
(Atchison, Bradshaw, & Massman, 2004).
The expected decline in basic self-care may be an even more important question to families than cognitive decline.
The decline in independent activities of daily living (IADL) and physical self-maintenance (PSM) are important risk
factors for institutionalization in the elderly and in individuals with dementia (Steeman, Abraham, & Godderis, 1997).
∗ Corresponding author. Tel.: +1 806 651 2729; fax: +1 806 651 2728.
E-mail address: (T.B. Atchison).
0887-6177/$ – see front matter © 2006 National Academy of Neuropsychology. Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.acn.2006.11.006
100
T.B. Atchison et al. / Archives of Clinical Neuropsychology 22 (2007) 99–107
Functional loss is also a strong predictor of death in dementia patients independent of age of onset or dementia severity
(Bracco et al., 1994). Therefore, the ability to anticipate who is likely to decline quickly in IADL and PSM function
has clinical importance in planning patient care.
Research indicates that neuropsychological measures may have the ability to predict future functional capacity.
Nussbaum, Goreczny, and Haddad (1995) found that functional capacity was related to concurrent cognitive decline
and that widespread cortical involvement was most associated with this observed functional deterioration. This study
employed the Dementia Behavior Rating Scale as the measure of functional ability. This scale includes measures of
both physical self-maintenance and areas of higher cognitive ability (social interaction).
Other researchers have found that more specific aspects of brain function may be able to predict functional decline.
McCue, Rogers, and Goldstein (1990) reported that the Memory and Motor scales of a shortened version of the Luria
Nebraska battery showed significant concurrent prediction of instrumental self-care but not physical self-care items
on a performance assessment. They surmised that cognitive assessment may have a difficult time predicting physical
self-care (McCue et al., 1990). Conversely, Mortimer et al. (1992) found the AD patients with poorer performance
on nonverbal neuropsychological test (e.g., visual discrimination, figure copy) at entry into the study subsequently
exhibited faster progression of functional impairment (a combination of PSM scores and higher-level IADL scores). In
this study functional decline was also related to paranoid behavior, hallucinations, and activity disturbance (Mortimer
et al., 1992).
Other researchers reported that performance on the Block Design subtest of the WAIS-R as well as age, co-ordination,
and snout reflex predicted decline in PSM type activities in subcortical vascular dementia patients (Bennett et al., 2002).
These researchers concluded that both physical and cognitive impairments combine to affect ADLs. They also found
that more complex IADL decline was associated with paratonia (disorder of muscle tone), attention and memory tasks.
A recent study addressed the ecological validity of neuropsychological tests to predict IADL function in AD patients
(Faris, Harrell, Neumann, & Houtz, 2003). These researches found that a set of neuropsychological measures accounted
for 25 percent of the variance in IADLs in a family report measure and 50 percent of the variance in a performance-based
assessment of IADL-type activities. The only nonverbal measure in the set of neuropsychological measures was a figure
copy test and it showed a stronger relationship to the performance measure than the family report measure (Faris et al.,
2003). This study did not explore the relationship of PSM activities with neuropsychological measures. These studies
indicate strongly that cognitive neuropsychological measures are associated with current and future IADL activities.
These studies lend weaker support for the proposition that cognitive neuropsychological measures are associated with
current PSM function or can predict maintenance or loss of PSM-type activities in the future.
It has been suggested that patterns or profiles of neuropsychological test performance may assist in diagnosis or
predict course in neurological conditions better than linear combinations or single tests (Brinkman & Braun, 1984; (...truncated)