Normal Aging and Executive Functions in - CiteSeerX

International Psychogeriah-ics, Vol, 14, No. 2, 2002, pp. 139-159 02002 International Psychogeriatric Association

Normal Aging and Executive Functions in “Old-Old’’Community Dwellers: Poor Performance Is Not an Inevitable Outcome OLIVIER PIGUET, DAVIDA. GRAYSON, G. ANTHONY BROE,ROBYNL. TATE, HAYLEY P. BENNETT, TANYA C. LYE, HELEN CREASEY, AND LLOYDRIDLEY ABSTRACT. Background: Studies on normal aging and cognitive functioning commonly describe early and more pronounced age-related changes in executive functions (EFs) compared to other cognitive abilities. Two of the three most common neurodegenerative disorders associated with aging (vascular dementia [VaD] and extrapyramidal [EPI-related dementia) show executive dysfunctions in their clinical presentation; and these cognitive deficits are not uncommon in the third one: Alzheimer’s disease (AD). Methods: Nine EF tests (yielding 12 measures) were administered to 123 randomly selected community dwellers, aged 81 years and over, with the view to determine the effect of age on performance. Markers of AD, VaD, and EP-related dementia, as well as sociodemographic and psychological variables, were selected and their contribution to EF performance was investigated. Results: Multiple linear regression analyses revealed the greatest contribution to EF scores from the markers of AD and estimated IQ but not from the markers of VaD and EP-related dementia or from age. Conclusions:These findings suggest that chronological age acts as a proxy variable mediating the impact of other factors such as subclinical signs of neurodegenerative disorders and that it has little independent contribution to make. They also indicate the importance of cognitive abilities supported by posterior cortical circuits in EF problem resolution. This study demonstrates that cognitive decline is not an ineluctable process that is associated with “normal” aging but rather represents, in many cases, a byproduct of neurodegenerative disorders, albeit themselves highly age-related. KEYWORDS: Normal aging; dementia; cognition; executive functions; 80 years and over

From the Centre for Education &Research on Ageing at Concord Repatriation General Hospital, The University of Sydney, Sydney, Australia (0.Piguet, PhD; T. C. Lye, PhD; and H. Creasey, FRACP), Prince of Wales Medical Research Institute, Randwick, Australia (0.Piguet; G. A. Broe, FRACP; and H. P. Bennett, PhD), Department of Psychology @. A. Grayson, PhD) and Rehabilitation Studies Unit, Department of Medicine @. L. Tate, PhD), The

University of Sydney, Sydney, Australia, and Department of Radiology, Concord Repatriation General Hospital, Concord, Australia (L. Ridley, FRANZCR). Offprints.Requests for offprints should be directed to Olivier Piguet, PhD, Centre for Education & Research on Ageing, Concord Hospital C25, Concord NSW 2139, Australia. email: olivierp @psych.usyd.edu.au 139

140

Aging processes affect the brain and its functions. Changes in brain size, neurotransmitters, regional blood flow, and glucose metabolism, as well as occurrence of neuropathological changes normally associated with neurodegenerative disorders, have all been described in neurologically intact older adults, but large interindividual differences exist. At any age, these changes appear to affect more predominantly the frontal region compared to other brain regions (West, 1996). Cognitively, executive functions (EFs), which have been consistently associated with the prefrontal region (Stuss & Benson, 1986), appear to show greater a g e related decline compared to other cognitive domains such as language, memory, and perception (for example, see Mittenberg et al., 1989). EFs encompass the capacity to program, regulate, and verify the unfolding of complex behaviors (Luria, 1973) as well as concept formation, reasoning, and cognitive flexibility. These cognitive processes (or abilities) are required for the effective formulation and implementation of new plans of actions. Decline in concept formation and abstraction, reduced mental flexibility, and increased time required to shift mental sets, to adapt to novel situations, or to solve new or unusual problems are all reported changes in EFs that have been associated with aging. These deficits have been reported even in participants as young as 65 years (Daigneault & Braun, 1993) and appear to increase rapidly in the eighth decade of life (Albert et al., 1990;Daigneault & Braun, 1993;Mittenberg et al., 1989; West, 1996). Thus, there is evidence supporting a comparatively greater sensitivity of the prefrontal brain region and its associated cognitive abilities to the aging process (West, 1996). However, the validity of this International Psychogeriatrics, 14(2), June 2002

0. Piguet et al.

“frontal aging hypothesis” has been challenged (Greenwood, 2000) and recent studies have further questioned the selectivity, uniformity, and extent of ageassociated EF on neuropsychological examination. For example, Crawford and colleagues (2000) reported equivocal results with poor performance found on some (Stroop test, Tower of London, modified card sorting test) but not other (verbal fluency, cognitive estimates, use of common objects) measures of EFs (Crawford et al., 2000). Alternative explanations for this observed decline have been suggested, such as cross-sectional designs and cohort effects emphasizing differences between groups (Schaie, 1990), poor control of the impact of sociodemographic variables and medical history on EF performance, particularly important in the older age segment (Wecker et al., 2000), and failure to take into account variables (e.g., vascular risk factors) implicated in the development of neurodegenerative disorders (Crawford et al., 2000; Lowe & Reynolds, 1999). Ritchie and colleagues (Ritchie, 1998; Ritchie et al., 1996) have further indicated that, more generally, a large portion of the cognitive impairment in elderly individuals can be explained by the presence of pathological processes rather than being the result of “normal” aging processes. Their findings support the view that broader factors need to be envisaged as potentially significant contributors to cognitive performance, because these may indeed explain agerelated changes reported in some studies. The presence of cognitive changes associated with aging processes raises the question of the relation and the transition between “normal” aging and dementia. Age remains the single most important risk factor contributing to the

Executive Functions in “Old-Old’’Community Dwellers

development of dementia (van Duijn, 1996). Epidemiological studies generally indicate a prevalence of dementia in community studies at around 5% at age 65, rising exponentially to affect between 30% and 40% of the general population over the age of 85 years (Jorm et a]., 1987). Beyond that age, the pattern is less robust and the curve appears to plateau. Incidence of dementia also rises significantly between the age of 65 and 85 years to reach approximately 5% in those over age 85 (Jorm & Jolley, 1998; Waite et al., 2001b). In Western societies, Alzheimer’s disease (AD) is the most common neurodegenerative disorder affecting individuals over the age of 65 years and represents up to two thirds of all dementia cases (Katzman & Kawas, 1994). Vascular dementia (VaD) accounts for 1%-30%of all dementia cases (Nyenhuis & Gorelick, 1998), but recent prevalence studies on dementia with Lewy bodies (DLB) indicate that this disease may be much more common than previously found. McKeith (1999) reported a modal frequency of 15%-20% of DLB cases (ranging from 12% to 36% in a review of autopsy series), which would make it the second most common form of dementia. Other less common forms are dementia associated with Parkinson’s disease (PD) with a prevalence in affected patients ranging between 3% and 90% and a modal tendency around 20%-25%(Cibb, 1989); and frontotemporal dementia, which may account for up to 20% of all dementia cases (Brun & Gustafson, 1999). Altogether, AD, dementia associated with extrapyramidal (EP) disorders (DLB or PD), and VaD account for about 90% of all cases of dementia. These dementia syndromes are clinically distinct. AD is characterized by an insidious onset with little or no neurological

141

signs. The hallmark of the disease is a p r e gressive cognitive decline affecting predominantly new learning and memory. Aspects of cognition mediated by the temporal and parietal regions such as language disturbance (naming or wordfinding difficulty) and disruption in visuoconstructional abilities are also commonly affected (McKhann et al., 1984). Basic attention and immediate memory, on the other hand, remain comparatively intact. Deficits in EFs are prominent in a subset of patients early in the disease (I-afleche&Albert, 1995;Reid et al., 1996), although they generally tend to occur somewhat later in the course of the disease. Deficits in EFs have been explained by frontal lobe pathology (Double et al., 1996) or by a breakdown in connections between the anterior and posterior ass@ ciation cortices (Collette et a]., 1999). The label of VaD is applied in the presence of a cognitive decline occurring against a background of cerebrovascular disease and when a temporal association between the two disorders can be established (Nyenhuis & Gorelick, 1998). The contribution of many risk factors (e.g., hypertension, diabetes, previous history of stroke, atrial fibrillation, cardiovascular disease, transient ischemic attack [TJA], and cigarette smoking) is well documented. Vascular events responsible for the dementing process are varied and factors t o take into account are the location of the vascular lesions and the volume of brain destruction (Markesbery, 1998). Focal neurological signs and a stepwise progression may be observed (e.g., after large-vessel events), but these may be transient and asymmetric and the course progressive with small-vessel disease. Gait impairment, dysarthria, and incontinence may also be present. The type and severity of

142

the cognitive deficits associated with VaD are dependent in part upon the location and the extent of the underlying vascular pathology. Nevertheless, a pattern emerges with consistent reports of impairments in verbal fluency, response inhibition, and mental set shifting, all characteristic of frontal systems abnormalities. Executive aspects of memory (organization and strategic encoding of information and retrieval efficiency), as well as constructional problem solving, have also been reported to be abnormal (Freeman et al., 2000). At postmortem, DLB patients exhibit cortical and subcortical neuropathological changes. Lewy bodies are seen diffusely across the cortex affecting more predominantly the frontal region, anterio r cingulate, and temporal cortices, more rarely the hippocampus (McKeith, 1999). Subcortically, the pattern of distribution of Lewy bodies is very similar to that seen in PD but to a lesser degree. In addition to the core clinical features of fluctuating cognition, visual hallucinations, and spontaneous motor features of parkinsonism, the presentation may further include repeated falls, syncope, or increased sensitivity to neuroleptics. With the involvement of frontosubcortical and nigrostriatal circuits in DLB, its associated cognitive deficits resemble those seen in PD. They are characterized by a slowness of ideation and a prominent breakdown in EFs such as problem-solving difficulties, difficulty with planning and concept formation, deficit in shifting set, poor mental flexibility, and decline in abstraction Fitvan, 1999). Deficits are also commonly reported in some aspects of working memory (McKeith, 1999) and executive aspects of memory (Bondi et al., 1993). Patients with DLB or PD may further International Psychogeriatrics, 14(2), June 2002

0. Piguet et al.

exhibit visuoperceptual and visuoconstructive deficits that are not entirely explained by the EF disturbance (Bondi et al., 1993; McKeith, 1999). In summary, important issues exist in relation to the study of EFs associated with the aging process. (a) Although there is a suggestion that EFs may show a n earlier and steeper decline than other cognitive abilities with aging, this cognitive change is not uniform and may be accounted for, at least in part, by other factors such as sociodemographic variables, medical history, or the presence of risk factors for the development of neurodegenerative diseases. (b) Studies on cognitive aging, with rare exceptions, have failed to investigate cognition in the oldest segment of t h e population (i.e., over the age of 80 years). (c) Further, the relation and transition between normal and pathological aging remain somewhat unclear. (d) Deficits in EFs are the central features in two (i.e., EP-related dementia and VaD) out of the three most common neurodegenerative diseases associated with aging and are not uncommon in the third one (i.e., AD). Given these findings and given that common neurodegenerative diseases may be preceded by a “silent” clinical period that can be as long as 10, 15, o r even 30 years for PD, VaD, and AD, respectively (Braak & Braak, 1996; Gibb & Luthert, 1994; Skoog, 1994), it is not inappropriate to propose that signs associated with these diseases will contribute t o the cognitive decline in the predementia phase of the disease. Given also that the prevalence and incidence of neurodegenerative diseases increase with age, one can also reasonably assume that their impact will be greater in the oldest segment of the population.

Executive Functions in “Old-Old”Community Dwellers

The aims of this study are t o explore the relations between EF performance, on the one hand, and markers of common neurodegenerative brain illnesses of aging and relevant sociodemographic variables (with particular attention given t o chronological age), on t h e other, in the oldest segment of the general population. From this, three specific hypotheses are proposed: 1 . Within the sets of markers of common brain illnesses associated with aging, markers of VaD and EP-related dementia will have the greatest impact on EF performance compared to those associated with AD; 2. Sociodemographic variables will contribute significantly to the performance of EF tasks, particularly education and general cognitive functioning; 3. The independent contribution of chronological age to EF performance will be reduced (or even removed) once the contributions of markers of common brain illnesses of aging have been taken into account. In other words, age will act as a proxy variable for these markers.

METHODS Participants This study took place within the context of the Sydney Older Persons Study, which is a longitudinal study of 630 randomly selected elderly community dwellers aged 75 years and over at the time of recruitment in 1991 (Creasey et al., 2001; Waite et al., 2001a). The unique inclusion criterion for participation in the current study was the capacity to give informed consent independently or with the assistance of a relative if necessary. In order to maintain the broadest

143

possible panorama of community-living persons’ cognitive abilities in this age segment, no other a priori medical exclusion criteria were applied. Participants were excluded when they were deemed too cognitively or physically impaired, to prevent potential distress from highlevel cognitive testing or from magnetic resonance imaging (MRI) procedures included in the study protocol. At t h e time of t h e 6-year review (between 1997 and 1999), 299 of the original 630 participants were potentially available, and of these, 167 agreed t o t h e MRI study (56%). A further 44 participants were excluded: l died before testing could be completed, 14 participants changed their mind, and 13 subsequently declined because of poor health. Three participants lived too far from the testing center and could not be seen. Finally, in order t o avoid complications associated with the MRI strong magnetic fields and metal parts in the body, eight participants who had a cardiac pacemaker inserted had to be excluded and anothe r five declined t o participate because they were claustrophobic and unable t o undergo t h e MRI procedure. Thus, 123 individuals (65 men and 58 women) took part in the current study. The participant demographic characteristics are presented in Table 1. Examination of t h e range of Mini-Mental State Examination (MMSE) scores (Folstein et a]., 1975) revealed four participants with a score below 21 points (13, 18, 19, and 19 points, respectively). Because the emphasis of the study was on a community sample, these participants were not excluded in order t o maintain a more accurate reflection of t h e cognitive functioning of this age group.

144

Measures and Procedure Measures of Executive Functions. Seven commonly used clinical tasks and two computerized tests were administered, yielding 12 EF variables. The clinical tasks selected were those that were “user friendly” and of short duration. These tasks were also chosen t o minimize peripheral sensory deficits. For these reasons, other commonly used clinical tests of EFs such as t h e Wisconsin Card Sorting Test (Heaton et al., 1993) o r the Stroop test (Golden, 1978) were omitted. The following tests were administered: (a) the Wechsler Adult Intelligence Scale-Revised (WAIS-R) Similarities Subtest (Wechsler, 1981); (b) the New Tower of London (New TOL; Hanes, 1997); (c) Semantic fluency (Animal category; Spreen & Strauss, 1991); (d) Phonemic fluency (FAS; Benton & Hamsher, 1983); (e) the Ruff Figural Fluency Task (RFFT; Ruff, 1988); ( f ) the Oral Trail Making Test-Part B (Oral Trails; Kaye et al., 1990); and (9) the WAIS-R-Digit Span Backward subtest @SB; Wechsler, 1981). In addition to these clinical tasks, portions of two computerized tests that require working memory input were included: the NBack Colour 1- and 2-Back tasks (Cohen et al., 1993) and t h e abbreviated California Computerized Assessment Package (CalCAP; Miller, 1996). Measures of Common Brain Illnesses of Old Age. Three sets were constructed to measure the impact of the most common neurodegenerative diseases of aging on EF performance: 1. Cognitive-AD: This set comprised measures as much as possible free of executive components tapping into the cognitive domains generally affected by AD: (a) new learning and memory (Wechsler Memory International Psychogeriatrics, 14(2), June 2002

0. Piguet et al.

Scale-Revised Logical Memory subtest [Wechsler, 1987]), @) language (Boston Naming Test [Kaplan et al., 1983]), and (c) visuoperceptual abilities (Judgment of Line Orienta-

tion [Benton et al., 19831). In order to sample beyond cognitive tests and to strengthen the independence between the set of CognitiveAD and the EF measures, the memory component of the Clinical Dementia Rating (CDR-m; Hughes et al., 1982), gathered from an informant and independent of testing, was also included. 2. Extrapyramidal (EP): This set comprised three measures reflecting EP change: (a) a composite EP score including the sum of 10 separate measures: resting tremor, glabella tap, nuchal rigidity, limb rigidity, cogwheeling, general bradykinesia, reduced arm swing, impairment in fine finger movements, postural flexion, and flexed gait (Waite et al., 2001a), (b) a measure of timed walk over a 5-meter distance (controlled for the presence of lower limb arthritis), and (c) a measure of alternate finger tapping speed for both hands over 15 seconds. 3. Vascular: This set comprised three measures reflecting vascular-related changes: First, a clinical global arteriopathy score that measured both the presence of noncerebrovascular disease and known vascular risk factors derived from the International Classification of Diseases-9th revision (KD9) diagnoses (World Health Organization, 1978). These variables, which were scored dichotomously (0 = absent or 1 = present) and summed, included diabetes, hypertension, atrial fibrillation, heart disease, claudication, hypercholesterolemia, and smoking history. This last variable was measured in number of cigarettes per annum and rescaled to a 0-1 score in keeping with the other measures. The second score of the vascular set was a l i f e long measure of history of stroke or


145

TABLE 1. Demographic Characteristics for the Study Participants (N = 123) Age, years Years of education Estimated IQ Mini-Mental State Examination total score Informant Clinical Dementia Rating (n = 119) TIA following the ICD-9 diagnosis categories. Finally, white matter changes on brain MRI in the periventricular and deep white matter regions were visually rated. Size of periventricular hyperintensities (PVH) was recorded in three regions: frontal caps, lateral bands, and occipital caps (Fazekas et al., 1987; Scheltens et al., 1998). Deep white matter hyperintensities (DWMH) were rated for size in four categories (3 mm or smaller, 4-10 mm, larger than 10 mm, and large confluent lesion/infarct) in four regions defined on the horizontal plane as follows: (a) anterior, defined as the region anterior to the foramen of Monroe, (b) posterior, defined as the area posterior to the splenium of the corpus callosum, (c) temporal, defined as the region around (or inferior to) the temporal horns of the lateral ventricles, and (d) intermediate, defined as any remaining regions not comprised in the other areas (Barber et al., 1999). For each region, a global composite score taking into account the number of lesions and the lesion size was computed.

Couariates. In addition t o these three sets, several variables likely t o have an effect on cognitive functioning were also selected. These variables were grouped as follows: (a) Sociodemographic, comprising level of education, occupation, and estimated cognitive functioning (derived from the number of errors on the National Adult Reading Test (Nelson & Willison, 1991);

Mean

SD

Range

85.49 10.36 105.5 26.59 1.58

3.09 2.11 5.7 2.83 0.74

81-97 6-19 83- 128 13-30 1-5

(b) Psychological, comprising Life Satisfaction Index (LSI; Neugarten et al., 196l), Center for Epidemiological Studies-Depression scale (CES-D; Radloff, 1977), Drive and Control Index (Elsass & Kinsella, 1989), and Intimate Bond Measure (Wilhelm & Parker, 1988); and (c) Miscellaneous, comprising hearing loss, decreased visual acuity, and use of medications with possible central nervous system (CNS) disturbance side effects. Procedure. Testing occurred over two sessions, 4 weeks apart on average, the first one at the participant’s home and the second at the local general hospital. In addition to the cognitive tests, the participants received a detailed medical interview, covering medication use, medical history, and a medical and neurological examination. Tests were adrriinistered in t h e same order for each participant. Within 1 month of the second session, participants underwent cerebral MRI scanning. MRIs were carried out on a 1.5-tesla scanner (G. E. Signa). Two axial MRI image sequences were collected. First, an FSE2 sequence (repetition time/echo time [TR/TE] 3600/108 ms) was obtained with images 5-mm thick and 2.5-mm gap with a 22-cm field of view and a 256 x 256 matrix. Second, a FLAIR sequence (TR/TE 10000/140 ms) was obtained with contiguous 4 m m thickness images, a 22-cm field of view, and a 256 x 256 matrix. Hard copies of the sequences were obtained and

146

images were also archived on electronic media (CD-ROM or DAT). Finally, collateral information was independently collected from an informant nominated by each participant for the following instruments: CDR, Drive and Control Index, and the Intimate Bond Measure. The medical data collected (such as medications and medical history) were reviewed by an experienced geriatrician/neurologist within 1 week of the visit. This physician subsequently made all the clinical diagnoses of (chronic) systemic and neurodegenerative diseases and allocated all relevant ICD-9 diagnoses. The hard copies of the MEU films were examined by one experienced radiologist for the presence of PVH and DWMH. Thirty-two protocols were rated twice giving rise t o intrarater reliability coefficients ranging between .79 and .93 for the PVH ratings and between .61 and .99 for the DWMH ratings, with the exception of the posterior region, which gave rise to a coefficient of .50 and .56 for the categories 3 mm o r less and large confluent lesion, respectively. These lower coefficients were due to the restricted range in scores. These reliability measures were found t o be in keeping with Scheltens and colleagues’ (1998) review paper. Statistical Analyses. The associations between t h e EF task measures and the defined sets of variables (measures of common brain illnesses of old age and covariates) were investigated using a multiple linear regression model. The assumption behind these analyses is that a causal relation exists between each EF measure and the sets of independent variables (illnesses and covariates). The individual contribution of the predictors to each EF score was first determined. In a second step, International Psychogeriatrics, 14(2), June 2002

0. Piguet et al.

the variables found to be univariately significant were entered together in the equation and their concurrent effect was examined. Nonsignificant univariate predictors were discarded and not considered for further analyses in t h e multivariate models. The multivariate model was refined by discarding any variable whose partial coefficient was nonsignificant before being retested. To ensure that t h e simultaneous removal of several such nonsignificant variables from an initial multivariate model was acceptable, t h e refined model was also tested against the initial model with a n increment test. These steps were repeated until either (a) all t h e remaining variables contributed significantly to t h e model o r (b) the increment test became significant indicating that the last discarded nonsignificant variables could not be removed and their contributions were necessary t o t h e model. In this instance, the “best” nonsignificant predictors (i.e., the predictors closest to reaching significance level) were reentered in the model one by one in their order of significance. The multivariate model was retested after each addition until the increment test became nonsignificant, at which point the process was stopped. A significance level of .05 was adopted for all the multivariate analyses. The final multivariate model obtained therefore corresponds t o t h e minimal number of individually significant variables, within t h e given set available, required t o significantly explain the score under investigation. Analyses were conducted on the EF scores in t h e following order: (a) with the common brain illnesses associated with aging; (b) with the sets of covariates; before combining (a) and (b).


RESULTS Descriptive Statistics The descriptive statistics for the EF measures, the markers of common brain illnesses associated with aging, and the covariates are presented in Tables 2, 3, and 4, respectively. For ease of interpretation of the multivariate analyses, all the EF scores were arbitrarily oriented in the same direction so that higher scores represented poorer performance. Thus, whenever necessary, scores were inverted by subtracting the participants’ scores on an EF measure from the highest score obtained on this measure within the sample. Cognitive-AD. On these measures, most participants appeared cognitively preserved. On the CDR-m, only 11 participants obtained a rating of “mild,” “moderate,” or “severe” impairment (8, 2, and 1 participants, respectively). On the direct measures of cognition, overall group performance was within normal limits based on the Mayo Older American Normative Studies norms (Ivnik et al., 1996; Smith et al., 1997). On the Logical Memory test, two participants obtained a better delayed recall score compared to their immediate recall, as reflected by the negative value on the percent forgetting score. Using a cutoff score of 2 standard deviations (SO)below the group means defining an impaired performance in this sample, seven participants scored below this limit. EP Measures. The composite EP score, which reflected the sum of 10 components, was positively skewed with 117 participants (95%) obtaining a score lower than 3. Given the high prevalence of lower limb arthritis in this age group (Waite et al., 200la), the time to complete a 5-meter returned

147

walk was corrected for the presence and severity of lower limb arthritis in both the hip and knee joints as a potential factor of gait slowing using a linear regression model. This procedure gave rise to a timed walk free of arthritis based on the following equation: corrected timed walk = timed walk 7.529412 x lower limb arthritis score. For the alternate finger tapping, the mean scores of both hands were recorded. Because t h e procedure required participants to move their fingers only (with the hand resting on the table for support), movements were not affected by possible upper limb arthritis and correction was therefore not necessary. Vascular Measures. On average, study participants exhibited fewer than two vascular risk factors, detected cluring the clinical interview. In contrast, 45% of participants reported, or were diagnosed with, some lifetime history of stroke o r TIA. MRI investigations revealed white matter abnormalities to be extremely common in this sample. All the participants exhibited PVH, with a comparable severity in all three subregions. In contrast, as shown in Table 3, a gradient of severity was present for the DWMH. These lesions were most severe in the intermediate region and only 5%of participants were found to be lesion free. The severity decreased in the other regions associated with an increase in the percentage of participants free of lesions: 18%,23%,and 64% for the anterior, posterior, and temporal regions, respectively. Psychological, Well-Being, and Behavioral Measures. Study participants perceived themselves in good psychological health. Only six participants (less than 5%of the sample) obtained a score of 12 or more (out of 20) on the LSI

148

and one participant endorsed 17/20 items of depressive symptomatology on the CESD. Further, informant ratings of personality or behavior changes over the previous 3 years were very small. On the Drive and Control Index, two participants were rated as much more apathetic (i.e., greater loss of drive), and a mild loss of control was reported in four participants. On the Intimate Bond Measure, most participants were rated highly on both t h e Affection and t h e Control scales. In other words, participants were not perceived as showing less affection towards other people or becoming more controlling of other persons’ activities. Miscellaneous Variables. The severity of sensory deficits was rated on a 4point scale (“no,” “mild,” “moderate,” or “severe” impairment) with 17 participants being rated with “moderate” or “severe” visual defect (12 and 5, respectively) and 21 with “moderate” or “severe” hearing deficit (20 and 1, respectively). Use of medication with potential CNS side effects was recorded on a 2point scale (i.e., use/no use). The means can be interpreted as the prevalence of use in this sample. In other words, 12%of participants were using prescribed benzodiazepines and 1 1% tricyclic antidepressants, but no one used stimulants, antiparkinsonian medications, or other sedatives. Ninety-two participants were free of these medications and, among those taking drugs, 18 were using one medication only, 10 using two drugs, and 3 participants three drugs.

Multiple Linear Regression Analyses The multivariate models for each EF measure are presented in Table 5. For ease of reading, predictors t h a t became nonsignificant on all 12 EF International Psychogeriatrics, 14(2), June 2002

0. Piguet et a!.

measures following t h e univariate analyses were removed from the table. Greyed-out cells indicate t h e predictors that failed t o bring a significant univariate contribution t o t h e EF scores. Empty cells represent nonsignificant contributions of predictors to the EF scores when considered in t h e multivariate models. These multivariate models comprised between one and five predictors each, explaining up t o 48% of the EF score variance. Detailed inspection of these multivariate models revealed that the markers associated with AD (Cognitive-AD) had the largest group contribution. Only two EF scores did not receive a significant contribution from this set of markers. In addition, five scores received multiple contributions within this set. N o other sets of markers associated with common illnesses of aging (VaD and EP-related dementia) contributed consistently to the EF scores. Among the other sets of predictors, the sociodemographic group of variables had the second largest number of contributions (10 in total) but unevenly in that estimated IQ contributed to seven scores compared to two scores for age and one for years of education. The other sets were present only sporadically in the multivariate models. The results from the univariate analyses are not presented here because they are not the focus of this study, but some findings are worth reporting. Univariately, age contributed significantly to all 12 EF scores. Similarly, the markers associated with EP-related dementia showed univariate significant contributions to between 8 and 10 EF scores. These results were in stark contrast to the vascular predictors, which appeared to lack explanatory power even when their


effect was considered one at a time, the “best performer” (clinical arteriopathy score) contributing to 4 out of 12 possible EF scores.

DISCUSSION Current research has demonstrated the preservation of various aspects of cognition such as language, visuospatial abilities, and even aspects of memory with aging. However, the frontal lobesjfrontal systems and the cognitive functions they support (i.e., EFs) have been claimed to be the most sensitive to the effect of the aging process, showing first signs of decline. This study was designed t o examine the impact of markers associated with common neurodegenerative illnesses of aging on EFs in participants over the age of 80 years and to determine, more specifically, the unique contribution of age to performance. In this context, it was hypothesized that the markers associated with VaD and EP-related dementia, and known to have a negative impact on frontal systems, would have the greatest impact on EF performance. In turn, it was expected that the markers associated with AD would have the smallest contribution to EF performance. The results show an inverse pattern of contributions: greater for AD markers and smaller for the other two sets. These findings are in contrast with the current literature on EFs in demented patients, which consistently reports significant cognitive decline, predominantly in EFs, for VaD and EPrelated dementia compared to normal controls. Similar findings are reported in studies on EF task performance in nondemented, elderly participants, when the role of vascular risk factors is taken into account.

149

Early deficits in EFs have been described in some AD patients, but in the majority of cases, cognitive deficits involve memory, language, and praxis. The large and consistent contribution from AD markers indicates that cognitive abilities mediated by discrete posterior cortical circuits appear to play a critical role in EF-type problem resolution. A disruption of these systems may result in the relay of inaccurate information to the frontal systems via the posterior-anterior connection fibers (superior longitudinal fasciculus and fronto-occipital fasciculus) and give rise to faulty response sets. This disruption in the flow of information between the posterior and anterior brain systems has been put forward as one possible explanation for the emergence of impairments in EFs in AD patients (Collette et al., 1999), although others have implicated direct frontal lobe involvement (Amieva et al., 1998). In light of the literature on EFs and neurodegenerative disorders, the small predictive component showed by the markers associated with VaD and EPrelated dementia is unexpected. Although markers of EP-related dementia generally showed consistent univariate contributions, no specific contribution to the EF scores emerges in the multivariate models. Markers of VaD lacked explanatory power in the univariate analyses. This was despite objective evidence of changes on brain MRI, changes that have been previously associated with deficits in EFs in small-vessel disease and VaD studies. It is possible that the effect of these markers on cognitive performance may be doserelated provided a certain threshold is reached (E3oone et al., 1992) or they may need to be accompanied by other systems dysfunctions giving rise to

0. Piguet e t al.

150

TABLE 2. Descriptive Statistics for the Executive Function Measures WAIS-R Digit Span Backwards Ruff Figural Fluency Test Total WAIS-R Similarities Phonemic Fluency (FAS) Total Semantic Fluency (Animals) Total New Tower of London highest level Oral Trail Making Test efficiency rate CalCAP 1 efficiency rate CalCAP 2 efficiency rate CalCAP 3 efficiency rate N-Back Colour 1-Back efficiency rate N-Back Colour 2-Back efficiency rate

Mean

SD

Range

5.89 40.54 16.06 40.20 26.67 4.88 13.14 0.14 0.44 0.58 36.79 89.14

0.90 13.67 7.01 12.61 4.41 1.59 19.98 0.15 0.23 0.18 37.60 39.68

3-8 12-77 1-28 8-68 9-38 1-9 0-100 0-0.80 0-1.05 0.20-0.96 0- 187.50 12.50-287.50

Note. All scores are similarly oriented so that a higher score represents a poorer performance. CalCAP = California Computerized Assessment Package; WAIS-R = Wechsler Adult Intelligence Scale-Revised.

TABLE 3. Descriptive Statistics for the Sets of Markers of Common Brain Illnesses Associated With Aging Variable Cognitive-AD Clinical Dementia Rating-memory item Logical Memory % loss Boston Naming Test Judgment of Line Orientation Extrapyramidal Timed walk Composite extrapyramidal score Alternate finger tapping Vascular Clinical arteriopathy score History of stroke/TIAs Deep white matter lesionsa Anterior region Temporal region Intermediate region Posterior region Periventricular hyperintensitiesa Frontal caps Occipital caps Lateral bands

Mean

SD

1.57 34.30 13.20 3.19

0.77 27.08 7.62 2.52

14.11 1.06 68.14

7.15 0.88 20.06

1.75 0.45

1.21 0.50

7.96 1.52 20.56 10.65

10.36 3.15 16.53 13.30

5.82 7.82 4.73

2.84 5.65 3.17

Range 1-5 -15.38-100 0-33 0-1 2 5-5 1 0-4.13 1.5-98.5 0-5 0-1 0-60 0-1 7 0-64 0-60 1-1 7.5 1-27.5 1-20

Note. Higher scores indicate poorer performance, greater impairment, or more severe lesions. AD = Alzheimer’s disease; TlAs = transient ischemic attacks. an = 115 because data were lost for 3 participants and 5 claustrophobic participants refused the procedure.

International Psychogeriatrics, 14(2), J u n e 2002

Executive Functions in "Old-Old" Communily Dwellers

151

TABLE 4. Descriptive Statistics for the Psychological Well-Being, Behavioral, and Miscellaneous Measures Variable (Range of Scores) Psychological well-being Life Satisfaction Index (0-20) CES-D (0-60) Behavioral changea Drive Index (0-9) Control Index (0-10) IBM Affection subscale (0-12) IBM Control subscale (0-12) Sensory deficits Hearing Vision Medications Anticholinergics Antiemetics Antihistamines Antipsychotics Tricyclic antidepressants Benzodiazepines Other sedatives Anticonvulsants Antiparkinsonians (dopaminergics) Antiparkinsonians (anticholinergics) Narcotics Stimulants

Mean

SD

Range

High Score

5.96 7.42

3.09 8.29

0-14 0-52

Less satisfaction More depressed

0.74 0.25 2.01 0.56

1.26 0.59 2.09 1.17

0-8 0-4 0-1 1 0-7

Loss of drive Loss of control Less caring More controlling

1.65 1.61

0.78 0.83

1-4 1-4

Impairment Impairment

0.02 0.03 0.02 0.02 0.1 1 0.12 0 0.01 0 0 0.06 0

0.13 0.18 0.15 0.13 0.3 1 0.33 0 0.09 0 0 0.23 0

0- 1 0- 1 0-1 0- 1 0-1 0- 1 0-0 0-1 0-0 0-0 0-1 0-0

Use Use Use Use Use Use Use Use Use Use Use Use

Nofe. CES-D = Center for Epidemiological Studies-Depression questionnaire; IBM = Interpersonal Bond Measure. "n = 119 because 4 participants refused to nominate an informant.

an additive process (Waite et al., 2001b). Also, the current results provide little support for a differential effect of periventricular versus deep white matter changes on EF performance despite suggestion of a different pathogenesis (Barber et al., 1999). This suggests that, in this age group, white matter hyperintensities may be the result of a process not related to pathological vascular changes or that the relative importance of known vascular risk factors may change with age (Breteler et al., 1994). It is also possible that MRI may be too sensitive an instrument, detecting small lesions that are not clinically significant, unlike computed

tomographic scans (Barber et al., 1999; Breteler et al., 1994; Skoog et al., 1996). This lack of activation from the markers of VaD applies to the vascular set as a whole (i.e., markers based on clinical information as well as the objective measures), and therefore unlikely to be due to MRI measurement error. As hypothesized, the variable age was a significant contributor to only 2 of t.he 12 multivariate models despite exhibiting significant univariate contributions to all EF scores. This supports the position that the unique contribution of chronological age to EF performance is considerably reduced once the contributions of

152

markers of common brain illnesses of aging, and other variables, are taken into account. This finding is in clear contrast with many studies on EFs and aging that report a significant and consistent negative association between these two variables (Daigneault & Braun, 1993; Mittenberg et al., 1989; Parkin & Lawrence, 1994), although some recent studies have questioned the importance of its contribution to performance (Lowe & Reynolds, 1999; Wecker et al., 2000). Interestingly, none of the studies reporting a negative association controlled for the presence of signs associated with neurodegenerative disorders, except for Mittenberg and colleagues (1989), who screened for “age-correlated illness,” which appeared to include systemic illnesses or declared dementing processes. This suggests that, in these studies, age may have in fact reflected the underlying effect of other processes commonly associated with aging rather than measured chronological age exclusively. Thus, it appears that in the current study, age is acting as a proxy variable mediating the impact of the variables associated with abnormal processes (but more common with advancing age) and has little i n d e pendent contribution to make. The final hypothesis was that EF performance would receive significant contributions from sociodemographic variables reflecting general intellectual performance. Of these variables, only estimated IQ was present, contributing to seven multivariate models. Level of education did not play a role in the multivariate analyses, suggesting that its contribution was entirely accounted for by other similar measures (e.g., estimated 1Q). This finding is consistent with the view that tasks of EFs are closely related to measures of general intelligence, International Psychogeriatrics, 14(2), June 2002

0. Piguet et al.

particularly fluid intelligence (Burgess, 1997). It also demonstrates the importance of sociodemographic variables on cognitive functioning and the importance of controlling for their effects when interpreting test scores. Interestingly, and unlike recent reports (Ritchie et al., 1996), EF performance in this sample appears not to have been affected by behavioral or psychological changes. Also, not surprisingly, is the indication that the presence of medication with known CNS side effects needs to be ascertained. The large contribution of vision to the two N-Back scores is puzzling. Some authors have proposed that age-related changes in cognitive and sensory functioning share an underlying common process and reflect an index of the physiological integrity of the aging brain (Lindenberger & Baltes, 1994; Salthouse et al., 1996). These age-related differences would be the expression of this third common factor. However, this does not explain the absence of contribution of this variable to the other EF measures requiring t h e manipulation of visual information (e.g., RFFT or New TOL). Compared to other studies on cognitive functioning in old age, this study is unique in that no particular a priori exclusion criteria were applied (apart from overt dementia) in order to obtain a more accurate panorama of the (physical and cognitive) health of community dwellers over the age of 80 years. One general difficulty in studying this age group is that of sample representativeness. In this study, obviously, the frailest and most cognitively impaired individuals from this cohort were absent because of their incapacity o r unwillingness to consent to the MRI procedure. Also missing from this study are participants who may have exhibited dementia


markers earlier in life and therefore may have had increased morbidity and mortality (Bennett et al., 1996; Ben Shlomo & Marmot, 1995; Katzman et al., 1994). However, the study sample remains r e p resentative of the individuals who remain independent-living community dwellers exhibiting aging-related illnesses. Unlike other studies, this sample comprised an (almost) equal number of women and men. Although this may seem surprising, it is the result of the sampling procedures used in the Sydney Older Persons Study at recruitment where an equal number of men and women were selected. (See Waite et al., 2001a, for a detailed descrip tion of the sampling procedures.) This indicates that the mortality and dropout rates are similar for men and women in this age group at 6 years and that the gender difference commonly reported takes place earlier in life. Another potential limitation of this study is related t o the Cognitive-AD predictors (i.e., used as markers of AD). Apart from the CDR-m, which represented an independent assessment of the participant’s memory functioning, all the measures within this set were cognitive tests. Every possible step was taken to select measures that were as much a s possible independent from the measures of EFs, in order to avoid circularity. However, one may expect to see a closer relation among cognitive tests than between cognitive and noncognitive measures, and, ideally, noncognitive measures of posterior brain integrity would have been preferable. Whether this difficulty can be completely avoided remains uncertain inasmuch as an appreciation of cognitive constructs (such as posterior brain systems in this study) will almost necessarily involve the use of cognitive

153

instruments. This was not an issue for the other sets of predictors (vascular and EP) because these comprised “medical” measures only. Royal1 (2000) suggested that EF deficits were strong predictors of decline in instrumental activities of daily living and institutionalization. The inclusion of measures of functional status may be a complernentary approach t o establish the determinants of EFs, and that independently from other cognitive domains. This study indicates that further research on the role of vascular and EP variables on cognition, and on EFs, in the absence of dementia is warranted. The cognitive profiles associated with these disease processes are now relatively well known when the disease is severe enough to reach the stage of dementia. Waite and colleagues (2001a) recently demonstrated that the presence of EP or vascular deficits on their own had no effect on the incidence of dementia after 3 years. However, they showed that these deficits lead to a significantly greater incidence of dementia when associated with cognitive deficits, suggesting an additive process. This additive process was particularly pronounced in the presence of EP deficits. There is also a need for further research on the construct of EFs. The current results show that there were different patterns of contributions t o the EF scores. This provides support for a multicomponent view of EF composed of a complex array of different cognitive abilities rather than one common underlying ability. In the context of the theories of aging and the continuum between normal and pathological decline, the current results support the view that cognitive decline is not necessarily an obligatory outcome

154

International Psychogeriatrics, 14(2), June 2002

0. Piguet et al.

Executive Functions in “Old-Old’’ Community Dwellers

155

156

associated with aging. The unique contribution of chronological age to EF performance is reduced considerably when its impact is considered simultaneously with the markers of neurodegenerative disorders. This also indicates that loss of cognitive “fuel,” for example reflected by a poor performance on tasks of EFs seen in old participants, may not be part of the normal aging process but is more likely to be associated with underlying agerelated pathological processes. Aging should not be seen as synonymous with poor cognitive performance, and preserved complex cognitive functioning in this age group is a very reasonable expectation t o have. These findings therefore support the position that cognitive decline is not an ineluctable process that is associated with “normal” aging but rather represents, in many cases, a byproduct of neurodegenerative disorders, albeit themselves highly agerelated.

REFERENCES Albert, M. S., Wolfe, J., &Lafleche, G. (1990). Differences in abstraction ability with age. Psychology & Aging, 5, 94100. Amieva, H., Lafont, S., Auriacombe, S., Rainville, C., Orgozozo, J.-M., et al. (1998). Analysis of error types in the Trail Making Test evidences an inhibitory deficit in dementia of the Alzheimer type. Journal of Clinical & Experimental Neuropsychology, 20, 280-285. Barber, R., Scheltens, P., Gholkar, A., Ballard, C., McKeith, l., et al. (1999). White matter lesions on magnetic resonance imaging in dementia with Lewy bodies, Alzheimer’s disease, vascular dementia and normal aging. Journal o f Neurology, Neurosurgery & Psychiatry, 67, 66-72. Bennett, D. A., Beckett, L. A., Murray, A. M., Shannon, K. M., Goetz, C. G., et al. International Psychogeriatrics, 14(2), June 2002

0. Piguet et al.

(1996). Prevalence of parkinsonian signs and associated mortality in a community population of older people. New England Journal o f Medicine, 334, 71-76. Ben Shlomo, Y., & Marmot, M. G. (1995). Survival and cause of death in a cohort of patients with parkinsonism: Possible clues to aetiology? Journal of Neurology, Neurosurgery & Psychiatry, 58, 293-299. Benton, A. L., & Hamsher, K. (1983). Multilingual aphasia examination. Iowa City, IA: AJA Associates. Benton, A. L., Hamsher, K. D., Varney, N. R., & Spreen, 0. (1983). Contributions to neuropsychological assessment. New York: Oxford University Press. Bondi, M. W., Kaszniak, A. W., Bayles, K. A., & Vance, K. T. (1993). Contributions of frontal system dysfunction to memory and perceptual abilities in Parkinson’s disease. Neuropsychology, 7, 89-102. Boone, K. B., Miller, B. L., Lesser, I. M., Mehringer, M., Hill-Gutierrez, E., et al. (1992). Neuropsychological correlates of white-matter lesions in healthy elderly subjects: A threshold effect. Archives o f Neurolog, 49,549-554. Braak, H., & Braak, E. (1996). Development of Alzheimer-related neurofibrillary changes in the neocortex inversely recapitulates cortical myelogenesis. Acta Neuropathologica, 92, 197-201. Breteler, M. M., van Swieten, J. C., Bots, M. L., Grobbee, D. E., Claus, J. J., et al. (1994). Cerebral white matter lesions, vascular risk factors, and cognitive function in a population-based study: The Rotterdam Study. Neurology, 44, 1246-1252. Brun, A., & Gustafson, L. (1999). Clinical and pathological aspects of frontotemporal dementia. In B. L. Miller & J. L. Cummings (Eds.), The human frontal lobes functions and disorders (pp. 349-369). New York: The Guilford Press. Burgess, P. W. (1997). Theory and methodology in executive function research. In P. Rabbitt (Ed.), Methodology o f frontal and executive function (pp. 81-116). Hove, England: Psychology Press.


Cohen, J. D., Forman, S. D., Braver, T. S., Casey, B. J., Servan-Schreiber, D., et al. (1993). Activation of the prefrontal cortex in a nonspatial working memory task with functional MRI. Human Brain Mapping, 1, 293-304. Collette, F., Van der Linden, M., &Salmon, E. (1999). Executive dysfunction in Alzheimer’s disease. Cortex, 35, 57-72. Crawford, J. R., Bryan, J., Luszcz, M. A., Obonsawin, M. C., & Stewart, L. (2000). The executive decline hypothesis of cognitive aging: Do executive deficits qualify as differential deficits and d o they mediate age-related memory decline? Aging, Neuropsychology, & Cognition, 7, 9-3 1. Creasey, H., Waite, L. M., Grayson, D. A., Bennett, H. P., Dent, O., et al. (2001). The impact of the neurodegenerative disorders on ageing: An overview of the Sydney Older Persons Study. Australasian Journal on Ageing, 20, 11-16. Daigneault, S., & Braun, C. M. J. (1993). Working memory and the Self-ordered Pointing Task: Further evidence of early prefrontal decline in normal aging. Journal of Clinical & Experimental Neuropsychology, 15, 881-895. Double, K. L., Halliday, G. M., Kril, J. J., Harasty, J. A., Cullen, K., et al. (1996). Topography of brain atrophy during normal aging and Alzheimer’s disease. Neurobiology o f Aging, I 7, 5 13-521. Elsass, L., & Kinsella, G. (1989). Development of a scale for measuring behaviour change following closed head injury. In V. Anderson & M. Bailey (Eds.), Theory and function-Bridging the gap. Proceedings of the Fourteenth Annual Brain Impairment Conference (pp. 124-131). Melbourne, Australia: Australian Society for the Study of Brain Impairment. Fazekas, F., Chawluk, J. B., Alavi, A., Hurtig, H. I., & Zimmerman, R. A. (1987). MR signal abnormalities at 1.5 T in Alzheimer’s dementia and normal aging. American Journal o f Roentgenoloa,149, 351-356.

157

Folstein, M. F., Folstein, S. E., & McHugh, P. R. (1975). “Mini-mental state”: A practical method for grading the cognitive state of patients for the clinician. ./ournal o f Psychiatric Research, 12, 189-198. Freeman, R. Q., Giovannetti, T., Lamar, M., Cloud, B. S., Stern, R. A., et al. (2000). Visuoconstructional problems in dementia: Contribution of executive systems functions. Neuropsychology, 14, 415-426. Gibb, W. R. (1989). Dementia and Parkinson’s disease. British Journal o f Psychiat ~ y154, , 596-614. Gibb, W. R., &Luthert, P. J. (1994). Dementia in Parkinson’s disease and Lewy body disease. In A. Burns & R. Levy (Eds.), Dementia (pp. 719-737). London: Chapman & Hall. Golden, C. J. (1978). Stroop Color and Word Test. Chicago: Stoelting. Greenwood, P. M. (2000). The frontal aging hypothesis evaluated. Journal of the International Neuropsychological Society, 6, 705-726. Hanes, K. R. (1997). New Tower o f London (2nd ed.). Carlton, Australia: Vanguard. Heaton, R. K., Chelune, G. J., Talley, J. L., Kay, G. G., &Curtis, G. (1993). Wisconsin Card Sorting Test (WCST) manual revised and expanded. Odessa, FL: I’sychological Assessment Resources. Hughes, C. P., Berg, L., Danziger, W. L., Coben, L. A., &Martin, R. L. (1982). A new clinical scale for the staging of dementia. British Journal of Psychiatry, 140, 566-572. Ivnik, R. J., Malec, J. F., Smith, G. E., & Tangalos, E. G. (1996). Neuropsychological tests’ norms above age 55: COWAT, BNT, MAE Token, WRAT-R Reading, AMNART, STROOP, TMT, and JLO. The Clinical Neuropsychologist, 10, 262-278. Jorm, A. F., & Jolley, D. (1998). The incidence of dementia: A meta-analysis. Neurology, 51, 728-733. Jorm, A. F., Korten, A. E., &Henderson, A. S. (1987). The prevalence of dementia: A quantitative integration of the literature. Acta Psychiatrica Scandinavica, 76, 465-479.

158

Kaplan, E. F., Goodglass, J., & Weintraub, S. (1983). The Boston Naming Test (2nd ed.). Philadelphia: Lea & Febiger. Katzman, R., Hill, L. R., Yu, E. S., Wang, Z. Y , Booth, A., et al. (1994). The malignancy of dementia. Predictors of mortality in clinically diagnosed dementia in a population survey of Shanghai, China. Archives of Neuroloa, 51, 1220-1225. Katzman, R., & Kawas, C. (1994). The epidemiology of dementia and Alzheimer disease. In R. D. Terry, R. Katzman, & K. L. Bick (Eds.), Alzheimer disease (pp. 105-122). New York: Raven Press. Kaye, K., Grigsby, J., Robbins, L. J., & Korzun, B. (1990). Prediction of independent functioning and behaviour problems in geriatric patients. Journal o f the American Geriatrics Society, 38, 1304-1310. Lafleche, G., &Albert, M. S. (1995). Executive function deficits in mild Alzheimer’s disease. Neuropsychology, 9, 313-320. Lindenberger, U., & Baltes, P. B. (1994). Sensory functioning and intelligence in old age: A strong connection. Psychology & Aging, 9, 339-355. Litvan, I. (1999). Extrapyramidal disorders and frontal lobe function. In B. L. Miller & J. L. Cummings (Eds.), The human frontal lobes (pp. 402-421). New York: The Guilford Press. Lowe, P. A., & Reynolds, C. R. (1999). Age, gender, and education may have little influence on error patterns in the assessment of set-shifting and rule induction among normal elderly. Archives o f Clinical Neuropsychology, 14, 303-315. Luria, A. (1973). The working brain. New York: Basic Books. Markesbery, W. R. (1998). Vascular dementia. In W. R. Markesbery (Ed.), Neuropathology o f dementing disorders @p. 293-311). London: Arnold. McKeith, I. G. C . (1999). Lewy body disorders. In B. L. Miller & J. L. Cummings (Eds.), The human frontal lobes (pp. 422-435). New York: The Guilford Press. International Psychogeriatrics, 14(2), June 2002

0. Piguet et al.

McKhann, G., Drachman, D., Folstein, M., Katzman, R., Price, D., et al. (1984). Clinical diagnosis of Alzheimer’s disease; Report of the NINCDS-ADRDA Work Group under the auspices of Department of Health and Human Services Task Force on Alzheimer’s Disease. Neurology, 34, 939-944. Miller, E. N. (1996). CalCAP California Computerized Assessment Package manual. Los Angeles: Norland Software. Mittenberg, W., Seidenberg, M., O’Leary, D. S., & DiGiulio, D. V. (1989). Changes in cerebral functioning associated with normal aging. Journal o f Clinical & Experimental Neuropsychology, 1I , 918-932. Nelson, H. E., & Willison, J. (1991). National Adult Reading Test (NART): Test manual. (2nd ed.). Windsor, England: NFER Nelson. Neugarten, B. L., Havighurst, R. J., &Tobin, S. S. (1961). The measurement of life satisfaction. Journal of Gerontology, 16, 134-143. Nyenhuis, D. L., & Gorelick, P. B. (1998). Vascular dementia: A contemporary review of epidemiology, diagnosis, prevention, and treatment. Journal o f the American Geriatrics Society, 46, 1437-1448. Parkin, A. J., & Lawrence, A. (1994). A dissociation in the relation between memory tasks and frontal lobe tests in the normal elderly. Neuropsychologia, 32, 1523-1532. Radloff, L. S. (1977). The CESD Scale: A selfreport depression scale for research in the general population. Applied Psychological Measurement, 1, 385-401. Reid, W., Broe, G., Creasey, H., Grayson, D., McCusker, E., et al. (1996). Age at onset and pattern of neuropsychological impairment in mild early-stage Alzheimer disease. A study of a community-based population. Archives o f Neurology, 53, 10561061. Ritchie, K. (1998). Establishing the limits of normal cerebral ageing and senile dementias. British Journal o f Psychiatry, 173, 97-101.

Executive Functions in “Old-Old”Community Dweliers

Ritchie, K., Leibovici, D., Ledesert, B., &Touchon, J. (1996). A typology of sub-clinical senescent cognitive disorder. British Journal of Psychiatry, 168, 470-476. Royall, D. R. (2000). Executive cognitive impairment: A novel perspective on dementia. Neuroepidemiolog, 19,293-299. Ruff, R. M. (1988). Ruff Figural Fluency Test administration manual. San Diego, CA: Neuropsychological Resources. Salthouse, T. A., Hancock, H. E., Meinz, E. J., & Hambrick, D. Z. (1996). Interrelations of age, visual acuity, and cognitive functioning. Journals o f Gerontology Series B-Psychological Sciences & Social Sciences, 51B, P317-P330. Schaie, K. W. (1990). Intellectual development in adulthood. In J. E. Birren & K. W. Schaie (Eds.), Handbook o f the psychology o f aging (3rd ed., pp. 291-309). San Diego, CA: Academic Press. Scheltens, P., Erkinjuntti, T., Leys, D., Wahlund, L.O., Inzitari, D., et al. (1998). White matter changes on CT and MRI: An overview of visual rating scales. European Task Force on Age-Related White Matter Changes. European Neurology, 39, 80-89. Skoog, I. (1994). Risk factors for vascular dementia: A review. Dementia, 5, 137-144. Skoog, I., Berg, S., Johansson, B., Palmertz, B., &Andreasson, L. A. (1996). The influence of white matter lesions on neuropsychological functioning in demented and nondemented 85yearalds. Acta Neurologica Scandinavica, 93, 142-148. Smith, G. E., Wong, J. S., Ivnik, R. J., &Malec, J. F. (1997). Mayo’s Older American Normative Studies: Separate norms for WMS-R Logical Memory stories. Assessment, 4, 79-86. Spreen, O., & Strauss, E. (1991). A compendium o f neuropsychological tests. New York: Oxford University Press. Stuss, D. T., & Benson, D. F. (1986). The frontal lobes. New York: Raven Press.

159

van Duijn, C. M. (1996). Epidemiology of the dementias: Recent developments and new approaches. Journal o f Neurology, Neurosurgery & Psychiatry, 60, 478-488. Waite, L. M., Broe, G. A., Grayson, D. A., & Creasey, H. (2001a). Preclinical syndromes predict dementia: The Sydney Older Persons Study. Journal o f Neurolog, Neurosurgery &Psychiatry, 71,296-302. Waite, L. M., Broe, G . A., Grayson, D. A., & Creasey, H. (2001b). The incidence of dementia in an Australian community population: The Sydney Older Persons Study. International Journal o f Geriatric PsychiaPy, IS, 680-689. Wechsler, D. (1981). WAIS-R manual. New York: The Psychological Corporation. Wechsler, D. (1987). WMS-R manual. New York: The Psychological Corporation. Wecker, N. S., Kramer, J. H., Wisniewski, A., Delis, D. C., & Kaplan, E. (2000). Age effects on executive ability. Neuropsycholog, 14, 409-414. West, R. L. (1996). An application of prefrontal cortex function theory to cognitive aging. Psychological Bulletin, 120, 272-292. Wilhelm, K., & Parker, G. (1988). The development of a measure of intimate bonds. Psychological Medicine, 18, 225-234. World Health Organization. (1978). Mental disorders: Glossary and guide to fheir classification in accordance with the ninth revision o f the International Classification o f Diseases. Geneva, Switzerland: Author.

Acknowledgments. This research was supported in part by a grant from the National Health and Medical Research Council of Australia and an Infrastructure Stream C grant from the Department of Health of New South Wales, Australia.