Abstract

Large prospective studies of Alzheimer's disease (AD) have sought to understand the pathological evolution of AD and factors that may influence the rate of disease progression. Estimates of rates of cognitive change are available for 12 or 24 months, but not for shorter time frames (e.g., 3 or 6 months). Most clinical drug trials seeking to reduce or modify AD symptoms have been conducted over 12- or 24-week periods. As such, we aimed to characterize the performance of a group of healthy older adults, adults with amnestic mild cognitive impairment (aMCI), and adults with AD on the CogState battery of tests over short test–retest intervals. This study recruited 105 healthy older adults, 48 adults with aMCI, and 42 adults with AD from the Australian Imaging, Biomarkers, and Lifestyle study and administered the CogState battery monthly over 3 months. The CogState battery of tests showed high test–retest reliability and stability in all clinical groups when participants were assessed over 3 months. When considered at baseline, the CogState battery of tests was able to detect AD-related cognitive impairment. The data provide important estimates of the reliability, stability, and variability of each cognitive test in healthy older adults, adults with aMCI, and adults with AD. This may potentially be used to inform future estimates of cognitive change in clinical trials.

Introduction

Large prospective studies that include both biomarkers and clinical information are important to understand the pathogenesis of Alzheimer's disease (AD). Early data from these studies show that AD-related pathophysiological changes begin prior to the clinical diagnosis of AD (Bateman et al., 2012; Jack et al., 2009; Kantarci et al., 2012) with the clinical classification of amnestic mild cognitive impairment (aMCI) considered to be the most likely prodrome to AD (Albert et al., 2011; Rowe et al., 2008). There is also growing evidence that the decline in the cognitive domains central to AD also occurs in healthy individuals who show high levels of Aβ amyloid (Darby, Brodtmann, Pietrzak, et al., 2011; Lim, Ellis, Pietrzak, et al., 2012; Small et al., 2012) with one study showing progression over 36 months sufficient to warrant clinical criteria for aMCI (Villemagne et al., 2011). Such findings raise the prospect that pharmacotherapy designed to modify or even halt AD-related pathology could be tested in both the clinical and preclinical phases of AD (Francis, Nordberg, & Arnold, 2005; Sperling et al., 2011).

Although cognitive measures that are used to inform decisions about drug efficacy in clinical trials in AD are well accepted (e.g., ADAS-Cog; Rogers et al., 1998), there is also agreement that these same measures do not possess sufficient sensitivity to detect cognitive change in the earlier stages of the disease (Jelic, Kivipelto, & Winblad, 2006; Petersen et al., 2010). For example, data from the Alzheimer's Disease Neuroimaging Initiative (ADNI) study show that in healthy older adults and in adults with aMCI, performance on the ADAS-Cog is characterized by ceiling effects, range restriction, and negative skew (Petersen et al., 2010). These psychometric characteristics limit the sensitivity of the ADAS-Cog to any treatment-related change in mild AD and in aMCI (Petersen et al., 2010). Consequently, there is a need for cognitive instruments that will be sensitive to cognitive change in very early as well as established AD.

In a series of recent studies, we have found that tests from the CogState battery were sensitive to cognitive impairment (i.e., relative to matched controls) in mild to moderate AD, aMCI, and also in healthy older adults who carry putative AD biomarkers (Harel et al., 2011; Lim, Ellis, Harrington, et al., 2012; Lim, Pietrzak, Snyder, Darby, & Maruff, 2012). Of equivalent importance, in a different sample, we also found that performance on these same CogState tasks remained stable despite repeated administration in healthy older adults and was characterized by reliable decline over periods of 12 months or greater in aMCI (Harel et al., 2011). Furthermore, in patients with AD, performance on the verbal list learning task of the CogState battery declined over 1 year (Lim, Pietrzak, et al., 2012). Taken together, these data suggest that the same computerized battery of cognitive tests may be used to measure cognitive function repeatedly in older individuals with normal cognition and in patients with aMCI and AD. However, for use in AD groups, it has been necessary to simplify some (i.e., visual learning and associate learning; Harel et al., 2011) but not all (i.e., verbal learning; Lim, Pietrzak, et al., 2012) tests of memory in order to maximize their acceptability. Finally, the CogState battery has demonstrated sensitivity to cognitive improvement arising from treatment with current pharmacotherapies for AD (e.g., cholinesterase inhibitors), with clinical trial data showing that performance is improved after acute treatment with donepezil in healthy older adults (Pietrzak, Maruff, & Snyder, 2009; Snyder, Bednar, Cromer, & Maruff, 2005) and with daily dosing in AD (Jaeger, Hardemark, & Zettergren, 2011). There is a growing appreciation of the importance of understanding the dynamics and reliability of neuropsychological assessments used for repeated assessments (Duff, 2012; Heilbronner et al., 2010). However, as yet, there has been no direct comparison of the stability and reliability of these same tests between healthy older adults and those with MCI or AD assessed over the same time intervals.

The Australian Imaging, Biomarkers, and Lifestyle (AIBL) study is a prospective natural history study of over 1000 adults who are cognitively normal or have either a diagnosis of MCI or mild AD (Ellis et al., 2009). These individuals undergo extensive assessment using psychiatric, neuropsychological, neurological, neuroradiological, and lifestyle measures at 18-month intervals (Ellis et al., 2009; Rowe et al., 2010). The AIBL Rates of Change substudy (hereafter referred to as ROCS) was designed to leverage the care and attention used in recruiting, assessing and characterizing the subjects in AIBL, by taking a subset of each clinical group and assessing them repeatedly at short retest intervals using the CogState battery to determine the extent to which any change in cognitive function could be detected in individuals with different stages of AD over intervals of 1, 2, 3, 6, 12, and 18 months. As the ROCS study is now enrolled fully and complete to the 3-month assessment, these data can be used to examine the acceptability of the tests in healthy adults, and adults with aMCI and AD, as well as to examine the magnitude of differences in performance between these groups. Some clinical trials of putative cognitive enhancers in AD are also conducted over 3 months and these trials generally measure cognitive performance at baseline and then at multiple follow-up assessments (i.e., weeks 4, 8 and 12; Pietrzak et al., 2009; Rogers et al., 1998). Therefore, we also investigated the stability of performance on the battery over 12 weeks in each of these cognitive measures between groups. Data from this prospective study can provide estimates of the expected rate of change in cognitive function over 12 weeks, as well as estimates of associated error (i.e., test–retest reliability and stability of the different outcome measures). Such data can be useful for computing power in clinical trials conducted over the same time interval. Finally, by restricting our analyses to the very short term, the effects of disease-related variability would be minimized, thus allowing direct comparison of any differences in rates of change or stability of performance in the different stages of the disease.

The first aim of this study was to directly compare the performance on the CogState battery between healthy older adults, adults with aMCI and adults with AD who had completed 3 months of assessment in the ROCS. The first hypothesis was that for tests where performance could be compared directly, healthy older adults would perform better than adults with aMCI, who would in turn show performance better than adults with AD. The second aim was to determine the stability of performance on the CogState battery over the initial 12 weeks of the ROCS in which it had been administered four times. The second hypothesis was that performance measures on the CogState battery would be reliable and remain stable (i.e., unchanged) over the short test–retest period in healthy, aMCI, and AD groups. The third aim was to explore the estimates of variability in performance over time on the CogState battery and to compare these between the three clinical groups to determine whether they are different.

Methods

Subjects

The sample method of recruitment, inclusion/exclusion criteria, and measures used in the AIBL study has been described in detail previously (Ellis et al., 2009; Rowe et al., 2010). As the ROCS sample was drawn from the AIBL group, these same criteria applied. In particular, participants were excluded if they met clinical criteria for anxiety or depression (Ellis et al., 2009). The selection criteria for enrollment into ROCS were (a) consensus classification as either being healthy (with or without subjective memory impairment, hereafter termed healthy older adults, HA), adults with aMCI, or AD (Ellis et al., 2009), (b) existing or planned magnetic resonance imaging and Pittsburgh Compound B (PiB)-positron emission tomography (PET) brain scans, (c) complete baseline neuropsychological and psychiatric ratings, (d) a willingness to allow serial assessments, and (e) ability to perform computerized cognitive tasks. Physical limitations were not part of the exclusion criteria; however, if participants had physical limitations (e.g., unable to use a mouse, vision problems) and were unable to perform the computerized cognitive tasks as a result, they were excluded from ROCS. Healthy older adults were classified with subjective memory impairment if they answered “yes” to the question “Do you have difficulty with your memory?” At the completion of the pilot phase of the study, ROCS consisted of 205 enrolled subjects and all subjects had completed their first 3 months assessment (baseline, 1-month follow-up, 2-month follow-up, and 3-month follow-up). All patients with AD were taking acetylcholinesterase medications and/or memantine, but no patients in the aMCI group were taking any such medications. Clinical and demographic characteristics of the healthy older, aMCI, and AD groups are shown in Table 1.

Table 1.

Demographic and clinical characteristics of each clinical group

 HA (N = 115) aMCI (N = 47) AD (N = 43) 
% Female 57.39% 52.08% 48.84% 
% APOE ɛ4 21.05% 36.84% 70.27% 
Age 73.62 (6.86) 78.87 (6.92) 79.57 (6.61)** 
Years of education 9–12 9–12 9–12 
Premorbid IQ 108.21 (7.08) 109.69 (5.84) 103.00 (10.56)** 
MMSE 29.06 (1.18) 27.95 (1.33) 22.16 (4.38) 
CDR-SB 0.04 (0.14) 0.57 (0.70) 4.09 (2.60) 
HADS-D 2.46 (2.15) 3.11 (2.20) 3.82 (2.60)** 
HADS-A 4.05 (2.80) 4.05 (2.52) 4.73 (3.34) 
CVLT-II Total 52.60 (9.49) 36.19 (8.42) 23.06 (9.63) 
CVLT-II Delayed 12.01 (3.27) 6.35 (3.95) 1.79 (2.50) 
Stroop C/D 2.33 (0.68) 2.68 (0.95) 3.20 (1.22) 
BNT 28.53 (1.60) 27.35 (3.08) 22.50 (5.58) 
 HA (N = 115) aMCI (N = 47) AD (N = 43) 
% Female 57.39% 52.08% 48.84% 
% APOE ɛ4 21.05% 36.84% 70.27% 
Age 73.62 (6.86) 78.87 (6.92) 79.57 (6.61)** 
Years of education 9–12 9–12 9–12 
Premorbid IQ 108.21 (7.08) 109.69 (5.84) 103.00 (10.56)** 
MMSE 29.06 (1.18) 27.95 (1.33) 22.16 (4.38) 
CDR-SB 0.04 (0.14) 0.57 (0.70) 4.09 (2.60) 
HADS-D 2.46 (2.15) 3.11 (2.20) 3.82 (2.60)** 
HADS-A 4.05 (2.80) 4.05 (2.52) 4.73 (3.34) 
CVLT-II Total 52.60 (9.49) 36.19 (8.42) 23.06 (9.63) 
CVLT-II Delayed 12.01 (3.27) 6.35 (3.95) 1.79 (2.50) 
Stroop C/D 2.33 (0.68) 2.68 (0.95) 3.20 (1.22) 
BNT 28.53 (1.60) 27.35 (3.08) 22.50 (5.58) 

Notes: HA = healthy older adult; aMCI = amnestic mild cognitive impairment; AD = Alzheimer's disease; MMSE = Mini Mental State Examination; CDR-SB = Clinical Dementia Rating Scale, Sum of Boxes Score; HADS-D = Hospital Anxiety and Depression Scale, Depression Subscale; HADS-A = Hospital Anxiety and Depression Scale, Anxiety Subscale; CVLT-II = California Verbal Learning Test, Second edition; Stroop C/D = Stroop Interference (colors/dots ratio); BNT = Boston Naming Task (No Cue). Raw scores have been reported for all neuropsychological and clinical tests.

**Group differences were significant, p < .001.

Measures

The following cognitive measures were selected because they are brief to administer, can be given repeatedly without eliciting practice effects, have demonstrated ability to detect AD-related memory impairment, and because the tasks have been shown to be sensitive to cognitive change associated with existing and novel therapies for AD (Darby, Brodtmann, Pietrzak, et al., 2011; Fredrickson et al., 2010; Hammers et al., 2011). Table 2 summarizes the outcome measure for each task.

Table 2.

Group mean (SD) for each performance measure in each clinical group and summary of the results of the LMM analyses

 Outcome measure Group Baseline (mean [SD]) 1 month (mean [SD]) 2 months (mean) 3 months (mean) FTime ([df] FFGroup ([df] FFInteraction ([df] F
DET Speed (log10 milliseconds) HA 2.56 (0.10) 2.60 (0.18) 2.58 (0.18) 2.58 (0.19) (2,492) 2.98 (2,492) 3.67* (4,492) 1.23 
aMCI 2.59 (0.12) 2.61 (0.18) 2.60 (0.18) 2.59 (0.18)    
AD 2.64 (0.15) 2.61 (0.19) 2.62 (0.19) 2.64 (0.19)    
IDN Speed (log10 milliseconds) HA 2.73 (0.06) 2.77 (0.18) 2.77 (0.18) 2.77 (0.19) (2,492) 2.05 (2,492) 3.60* (4,492) 1.74 
aMCI 2.77 (0.09) 2.76 (0.18) 2.75 (0.18) 2.74 (0.19)    
AD 2.85 (0.11) 2.78 (0.19) 2.77 (0.19) 2.76 (0.19)    
OCL Accuracy (arcsine proportion correct) HA 1.00 (0.09) 0.99 (0.12) 1.00 (0.12) 1.02 (0.12) (2,490) 0.04 (2,490) 33.65** (4,490) 0.78 
aMCI 0.94 (0.10) 0.94 (0.12) 0.93 (0.12) 0.93 (0.12)    
AD 0.82 (0.11) 0.87 (0.12) 0.86 (0.12) 0.85 (0.12)    
OBK Accuracy (arcsine proportion correct) HA 1.37 (0.16) 1.30 (0.16) 1.34 (0.16) 1.36 (0.16) (2,473) 2.28 (2,473) 35.85** (4,473) 1.63 
aMCI 1.23 (0.20) 1.24 (0.16) 1.26 (0.15) 1.35 (0.15)    
AD 0.86 (0.30) 1.10 (0.20) 1.10 (0.20) 1.07 (0.20)    
CPAL Total errors HA 31.66 (30.98) 36.42 (25.69) 35.55 (25.72) 30.22 (25.71) (2,482) 1.82 (2,482) 27.63** (4,482) 0.72 
aMCI 75.13 (42.13) 61.23 (25.62) 54.10 (25.54) 59.52 (25.66)    
AD 67.82 (28.39) 48.16 (25.94) 40.04 (25.94) 39.77 (25.96)    
ISLT Total Total words recalled HA 25.05 (4.50) 23.42 (3.95) 23.57 (3.93) 23.52 (3.93) (2,491) 0.48 (2,491) 50.85** (4,491) 0.13 
aMCI 17.61 (5.26) 21.29 (3.85) 21.50 (3.82) 21.61 (3.78)    
AD 11.00 (5.16) 16.42 (5.30) 17.09 (5.31) 17.26 (5.31)    
ISLT Delay Total words recalled HA 8.41 (2.32) 7.60 (1.92) 7.63 (1.91) 7.70 (1.91) (2,481) 1.28 (2,481) 50.17** (4,481) 0.62 
aMCI 4.53 (3.09) 6.32 (1.85) 5.78 (1.84) 5.90 (1.84)    
AD 1.15 (1.40) 2.84 (4.03) 2.26 (4.05) 2.53 (4.05)    
 Outcome measure Group Baseline (mean [SD]) 1 month (mean [SD]) 2 months (mean) 3 months (mean) FTime ([df] FFGroup ([df] FFInteraction ([df] F
DET Speed (log10 milliseconds) HA 2.56 (0.10) 2.60 (0.18) 2.58 (0.18) 2.58 (0.19) (2,492) 2.98 (2,492) 3.67* (4,492) 1.23 
aMCI 2.59 (0.12) 2.61 (0.18) 2.60 (0.18) 2.59 (0.18)    
AD 2.64 (0.15) 2.61 (0.19) 2.62 (0.19) 2.64 (0.19)    
IDN Speed (log10 milliseconds) HA 2.73 (0.06) 2.77 (0.18) 2.77 (0.18) 2.77 (0.19) (2,492) 2.05 (2,492) 3.60* (4,492) 1.74 
aMCI 2.77 (0.09) 2.76 (0.18) 2.75 (0.18) 2.74 (0.19)    
AD 2.85 (0.11) 2.78 (0.19) 2.77 (0.19) 2.76 (0.19)    
OCL Accuracy (arcsine proportion correct) HA 1.00 (0.09) 0.99 (0.12) 1.00 (0.12) 1.02 (0.12) (2,490) 0.04 (2,490) 33.65** (4,490) 0.78 
aMCI 0.94 (0.10) 0.94 (0.12) 0.93 (0.12) 0.93 (0.12)    
AD 0.82 (0.11) 0.87 (0.12) 0.86 (0.12) 0.85 (0.12)    
OBK Accuracy (arcsine proportion correct) HA 1.37 (0.16) 1.30 (0.16) 1.34 (0.16) 1.36 (0.16) (2,473) 2.28 (2,473) 35.85** (4,473) 1.63 
aMCI 1.23 (0.20) 1.24 (0.16) 1.26 (0.15) 1.35 (0.15)    
AD 0.86 (0.30) 1.10 (0.20) 1.10 (0.20) 1.07 (0.20)    
CPAL Total errors HA 31.66 (30.98) 36.42 (25.69) 35.55 (25.72) 30.22 (25.71) (2,482) 1.82 (2,482) 27.63** (4,482) 0.72 
aMCI 75.13 (42.13) 61.23 (25.62) 54.10 (25.54) 59.52 (25.66)    
AD 67.82 (28.39) 48.16 (25.94) 40.04 (25.94) 39.77 (25.96)    
ISLT Total Total words recalled HA 25.05 (4.50) 23.42 (3.95) 23.57 (3.93) 23.52 (3.93) (2,491) 0.48 (2,491) 50.85** (4,491) 0.13 
aMCI 17.61 (5.26) 21.29 (3.85) 21.50 (3.82) 21.61 (3.78)    
AD 11.00 (5.16) 16.42 (5.30) 17.09 (5.31) 17.26 (5.31)    
ISLT Delay Total words recalled HA 8.41 (2.32) 7.60 (1.92) 7.63 (1.91) 7.70 (1.91) (2,481) 1.28 (2,481) 50.17** (4,481) 0.62 
aMCI 4.53 (3.09) 6.32 (1.85) 5.78 (1.84) 5.90 (1.84)    
AD 1.15 (1.40) 2.84 (4.03) 2.26 (4.05) 2.53 (4.05)    

Notes: HA = healthy older adult; aMCI = amnestic mild cognitive impairment; AD = Alzheimer's disease. The AD group received an easier version of the OCL and CPAL tasks than did MCI and HA groups; DET = Detection; IDN = Identification; OCL = One Card Learning; OBK = One Back; CPAL = Continuous Paired Associate Learning; ISLT = International Shopping List Test.

*p < .05.

**p < .001.

Detection task

The Detection (DET) task is a measure of simple reaction time shown to measure psychomotor function. In this task, subjects must attend to the card in the center of the screen and response to the question: “Has the card turned over”? Subjects were instructed to press the “Yes” key as soon as the card turned face up. The face of the card was always the same generic joker card. The task ended after 35 correct trials had been recorded. Trials on which anticipatory responses occurred were excluded, and another trial was given so that all subjects completed the 35 trials. The primary performance measure for this task was reaction time in milliseconds (speed), which was normalized using a logarithmic base 10 (log10) transformation.

Identification task

The Identification (IDN) task is a measure of choice reaction time shown to measure visual attention. In this task, the participant must attend to the card in the center of the screen and respond to the question “Is the card red”? Participants were required to press the “Yes” key if it is and the “No” key if it is not. The face of the cards displayed were either red or black joker cards in equivalent numbers in random order. These cards were different to the generic joker card used in the DET task. The task ended after 30 correct trials. Trials on which anticipatory responses occurred were excluded, and another trial was given so that all participants completed the 30 trials. The primary performance measure for this task was reaction time in milliseconds (speed), which was normalized using a log10 transformation.

One Card Learning task

The One Card Learning (OCL) task is a continuous visual recognition learning task that assesses visual learning within a pattern separation model. In this task, the participant must attend to the card in the center of the screen and respond to the question “have you seen this card before in this task”? If the answer was yes, participants were instructed to press the “Yes” key, and the “No” key if the answer was no. Normal playing cards were displayed (without joker cards). In this task, six cards are drawn at random from the deck and are repeated throughout the task. These six cards are interspersed with distractors (non-repeating cards). The task ends after 42 trials, without rescheduling for postanticipatory correct trials. This version of the task was administered to the HA and aMCI groups. For the AD group, a simpler version of the task was used. In this version, only four target cards were interspersed with distractors. The primary performance measure for this task was the proportion of correct answers (accuracy), which was normalized using an arcsine square-root transformation.

One-Back Task

The One-Back (OBK) task is a task of working memory and attention. Similar in presentation to the OCL task, participants must attend to the card in the center of the screen and respond to the question “is this card the same as that on the immediately previous trial”? If the answer was yes, participants were instructed to press the “Yes” key, and the “No” key if the answer was no. The task ends after 30 correct trials. A correct but post-anticipatory response led to scheduling of an extra trial. The primary performance measure for this task was the proportion of correct answers (accuracy), which was normalized using an arcsine square-root transformation.

Continuous Paired Associate Learning task

The Continuous Paired Associate Learning (CPAL) task is a measure of visual learning and episodic memory. In this task, subjects must learn a series of associations between a set of simple shapes and their associated location. In HA and aMCI groups, six pattern–location associations must be learned. For the mild AD group, four pattern–location associations must be learned. There are two parts to the CPAL, although subjects are not made aware that the stages are different. In the presentation phase of the task, the pattern appears at the location and the subject is required to acknowledge that they have seen the pattern by touching the location at which it appears. At this stage of the task, the subject will also see that there are two locations at which no target appears (distractor locations). Patterns are presented in random order; however, once presented the pattern remains at the same location throughout the task. In the learning phase of the task, subjects must place each of the four patterns in their correct locations. They must do this in six trials. For the first trial, one of the patterns is presented in the center location and the subject is required to remember the location at which it was shown. They indicate the location by touching it. If they touch an incorrect location, a visual and audible signal occurs (a red cross appears over the location and a buzzer sound is presented). The subject is then required to choose a second location. This process continues until the subject finds the location that has been paired with the pattern. Once the pattern has been associated with the location, the next pattern is presented in the center location and the process continues again. This repeats until the subject has correctly placed all of the targets in their correct locations. Once all patterns have been placed correctly, the second trial begins. In the second trial, the patterns remain in the same locations, but their order of presentation in the center of the screen is different to that of the first trial (randomized). The process of placing each target in the correct location proceeds as it did in the first trial. When the second trial is complete, the same process is repeated for trials 3–6. The primary performance measure for this task was the total number of errors made.

International Shopping List Test

The International Shopping List Test (ISLT) is a four trial (three learning trials and one delayed recall trial) verbal list learning test in which individuals are instructed to remember a list of 12 items that they need to obtain from their local store. To ensure that items on the shopping list are relevant to Australian test-takers, a large pool of common shopping list items (128 items) were rated by 30 Australian healthy adults using a web-based survey to indicate the ease (i.e., “very difficult,” “difficult,” “easy,” or “very easy”) with which each shopping list item could be obtained locally (Lim et al., 2009). In this study, a pool of 10 alternate forms was generated. The items on each list have been reported previously (Lim, Pietrzak, et al., 2012). For each assessment, for each participant, one of the 10 lists was chosen at random by the computer software, with the condition that if it has been selected previously for a participant, it would not be selected again. For each list, the order in which items were presented was randomized between participants by the computer software, with the order of the items remaining constant across the three learning trials. During each assessment, the computer presented the words to the examiner one at a time at a rate of 1 word per 2 s. The participant was instructed to “try and remember as many items on the shopping list as possible.” The examiner then read each item to the participant as they appeared on the screen. The computer screen was never visible to the participant. Once all 12 words had been read to the participant, they were instructed to recall as many items from the list as possible with the statement “tell me as many items on the shopping list as you can remember.” The list of 12 words was displayed on the computer screen and as the participant recalled each item, the examiner marked the item by clicking on the relevant checkbox. If words were repeated, the checkbox was clicked again. Another checkbox was clicked if the participant said a word that was not on the original list (i.e., an intrusion). When the participant indicated that no more items could be recalled, the trial was stopped and the same process was repeated two more times. For the delayed recall trial, participants were asked to recall as many items as possible from the initial list after a delay of approximately 30 min, during which the participant was administered other cognitive tasks (i.e., DET, IDN, OCL, OBK, and CPAL).

Procedure

All subjects were enrolled in the AIBL study before selection for ROCS began. Once clearance by institutional ethics and research committees of Austin Health, St Vincent's Health, Hollywood Private Hospital and Edith Cowan University were granted, the study coordinator (KH) contacted subjects by telephone, explained the study to them, and invited them to participate. If they agreed to participate in the study, they were informed that the assessments were brief and that these could occur at their home, a location near their house that was convenient, or they could attend the research unit (Mental Health Research Institute). Individual assessors organized a mutually suitable time for the visit. All study visit times were held constant by raters, and the actual date of follow-up assessments were allowed to vary by 1 week. Participants had also agreed to undergo at least one PET neuroimaging scan using PiB in the future, although at the time of this report, the imaging aspect had not been completed for all participants.

Assessors used to gather neuropsychological data were trained on administration of the CogState battery in accord with standard protocols. In order to facilitate communication and encourage subjects to identify with the study, each assessor was assigned to conduct the repeated assessments on the same subjects and organized home visits directly with each subject.

Data Analysis

Performance on the three assessments over 3 months was then compared between groups using a series of 3 (group) × 4 (assessment) linear mixed model (LMM) analyses of covariance (ANCOVAs) with age, premorbid IQ, depressive symptomatology, and baseline scores entered as covariates (as group comparisons of demographics identified these as being significantly different between the three groups), and participants treated as a random factor. For each test, the magnitude of difference between the aMCI and the AD group to the HA group was expressed using Cohen's d. As simpler versions of the CPAL and OCL tasks were administered to the AD groups, performance on these tasks was not compared statistically between the HA and AD groups. Average measure intra-class correlation coefficients (ICCs) were used to compute the test–retest reliability of each ROCS battery performance measure over the four assessment periods, in both the total group and each clinical classification group separately.

Finally, to allow the appreciation of the change that will occur on each of the outcome measures in the CogState battery over 3 months, each outcome measure from the CogState battery was submitted to an LMM ANCOVA, with age, premorbid IQ, depressive symptomatology, and baseline scores entered as covariates, which allowed computation of the slope of change in performance and associated standard error of this change between the first and last assessment sessions, that is, between baseline and month 3, for each clinical classification group. Within-subject standard deviations (WSDs) were also calculated for each clinical classification group, and Hartley's Fmax test (Hartley, 1950) was used to compare variances between each clinical classification group. These estimates of group mean change and within-subject variability over 3 months were used to derive practice-corrected confidence intervals (CIs) for reliable change indices (RCIs) for each measure in each clinical group (Hinton-Bayre, 2010). Criterion values for change scores used commonly in neuropsychology were computed; specifically, change of greater than 1 (i.e., 85% CI, one-tailed) or 1.65 SD (i.e., 95% CI, one-tailed) over 3 months. For example, the group mean change for ISLT delayed recall in individuals with AD is −0.16, with a WSD of 0.90. A decline in performance greater than the 85% CI (or 1 SD) over 3 months is computed by the following equation 

formula
which yields a score of −1.43. Thus, in AD, a decline of more than 2 words on the ISLT delayed recall over 3 months would be considered clinically meaningful (Jacobson & Truax, 1991). These values are computed for each task in each of the clinical groups.

Results

Comparison of Performance Between Clinical Classification Groups and Across Assessments

The LMM indicated that performance of healthy older adults, adults with aMCI, and adults with AD differed significantly on the DET, IDN, OBK, and ISLT tasks. However, no significant differences between assessments were identified for any of the measures and no groups by assessment interaction terms were statistically significant (Table 2). Compared with controls, the magnitude of impairment observed in the aMCI group (averaged over assessments) was large for the CPAL (d = 1.53), OCL (d = 0.94), OBK (d = 0.60), ISLT total recall (d = 0.82), and ISLT delayed recall (d = 1.36) tasks and small for the DET (d = 0.20) and IDN (d = 0.10) tasks. Compared with controls, the magnitude of impairment observed in the AD group (averaged over assessments) were large for the OBK (d = 2.40), ISLT total recall (d = 2.52), and ISLT delayed recall (d = 2.62) tasks and moderate for the DET (d = 0.62) and IDN (d = 0.53) tasks.

Test–Retest Reliability in Each Clinical Classification Group

The ICCs for each outcome measure for each group are shown in Table 3. When considered according to clinical classification, all measures from the ROCS battery demonstrated high (i.e., r > .70), test–retest reliability and these estimates were equivalent between the clinical groups (Table 3). The highest ICCs were observed for the overall group for the ISLT total and delayed recall scores, CPAL total errors, and accuracy of performance on the OBK task.

Table 3.

Summary of average measure ICCs, two-way random effects model, on each CogState performance measure in each clinical group across four assessments

Measure Overall (N = 205) HA (N = 115) aMCI (N = 47) AD (N = 43) 
DET Speed 0.79** 0.93** 0.92** 0.70** 
IDN Speed 0.76** 0.92** 0.95** 0.71** 
OCL Accuracy 0.87** 0.77** 0.82** 0.68** 
OBK Accuracy 0.93** 0.75** 0.79** 0.93** 
CPAL Total Errors 0.91** 0.87** 0.88** 0.81** 
ISLT Total Recall 0.96** 0.86** 0.86** 0.95** 
ISLT Delayed Recall 0.97** 0.87** 0.92** 0.82** 
Measure Overall (N = 205) HA (N = 115) aMCI (N = 47) AD (N = 43) 
DET Speed 0.79** 0.93** 0.92** 0.70** 
IDN Speed 0.76** 0.92** 0.95** 0.71** 
OCL Accuracy 0.87** 0.77** 0.82** 0.68** 
OBK Accuracy 0.93** 0.75** 0.79** 0.93** 
CPAL Total Errors 0.91** 0.87** 0.88** 0.81** 
ISLT Total Recall 0.96** 0.86** 0.86** 0.95** 
ISLT Delayed Recall 0.97** 0.87** 0.92** 0.82** 

Notes: HA = healthy older adult; aMCI = amnestic mild cognitive impairment; AD = Alzheimer's disease; DET = Detection; IDN = Identification; OCL = One Card Learning; OBK = One Back; CPAL = Continuous Paired Associate Learning; ISLT = International Shopping List Test.

**p < .001.

Comparison of Variance in Change Scores and Data for RCI

The group mean change, standard error of this change, and WSDs between the first and the last assessment were computed for each measure of the ROCS battery, for each clinical classification group (Table 4). For each clinical classification group, no significant difference in mean change was observed for any of the measures of the ROCS battery. Statistically significant differences in WSDs were observed between the HA and AD groups for the accuracy of performance on the DET and IDN tasks and between the HA and aMCI groups for the CPAL errors (Table 4). Table 4 also provides the practice-corrected 85% CI and 95% CI for RCIs computed from each cognitive outcome measure.

Table 4.

Mean (SD) of difference and WSD between baseline and month 3 for each clinical group

 Group Group mean change Group SD of change WSD Practice-corrected CIs
 
85% CI (z = −1.00) 95% CI (z = −1.65) 
DET HA −0.01 0.11 0.06 −0.08 −0.14 
aMCI −0.01 0.07 0.07 −0.10 −0.16 
AD −0.06 0.32 0.28* −0.46 −0.71 
IDN HA 0.01 0.11 0.03 −0.03 −0.06 
aMCI −0.01 0.07 0.03 −0.04 −0.07 
AD −0.06 0.32 0.29* −0.47 −0.74 
OCL HA 0.02 0.11 0.08 −0.09 −0.17 
aMCI −0.01 0.07 0.07 −0.11 −0.17 
AD −0.01 0.06 0.10 −0.15 −0.24 
OBK HA 0.03 0.11 0.14 −0.17 −0.30 
aMCI 0.05 0.14 0.15 −0.16 −0.30 
AD −0.01 0.26 0.14 −0.21 −0.34 
CPAL HA −3.08 19.95 14.80 17.85 31.45 
aMCI −0.96 38.52 23.71* 32.57 54.36 
AD −4.22 26.51 19.19 22.91 40.54 
ISLT Total HA 0.05 3.32 2.83 −3.95 −6.55 
aMCI 0.17 4.36 3.31 −4.51 −7.55 
AD 0.46 4.99 2.34 −2.85 −5.00 
ISLT Delay HA 0.05 1.72 1.41 −1.94 −3.24 
aMCI −0.22 2.56 1.50 −2.34 −3.72 
AD −0.16 0.91 0.90 −1.43 −2.26 
 Group Group mean change Group SD of change WSD Practice-corrected CIs
 
85% CI (z = −1.00) 95% CI (z = −1.65) 
DET HA −0.01 0.11 0.06 −0.08 −0.14 
aMCI −0.01 0.07 0.07 −0.10 −0.16 
AD −0.06 0.32 0.28* −0.46 −0.71 
IDN HA 0.01 0.11 0.03 −0.03 −0.06 
aMCI −0.01 0.07 0.03 −0.04 −0.07 
AD −0.06 0.32 0.29* −0.47 −0.74 
OCL HA 0.02 0.11 0.08 −0.09 −0.17 
aMCI −0.01 0.07 0.07 −0.11 −0.17 
AD −0.01 0.06 0.10 −0.15 −0.24 
OBK HA 0.03 0.11 0.14 −0.17 −0.30 
aMCI 0.05 0.14 0.15 −0.16 −0.30 
AD −0.01 0.26 0.14 −0.21 −0.34 
CPAL HA −3.08 19.95 14.80 17.85 31.45 
aMCI −0.96 38.52 23.71* 32.57 54.36 
AD −4.22 26.51 19.19 22.91 40.54 
ISLT Total HA 0.05 3.32 2.83 −3.95 −6.55 
aMCI 0.17 4.36 3.31 −4.51 −7.55 
AD 0.46 4.99 2.34 −2.85 −5.00 
ISLT Delay HA 0.05 1.72 1.41 −1.94 −3.24 
aMCI −0.22 2.56 1.50 −2.34 −3.72 
AD −0.16 0.91 0.90 −1.43 −2.26 

Notes: WSD = within-subject standard deviation; HA = healthy older adult; aMCI = amnestic mild cognitive impairment; AD = Alzheimer's disease; DET = Detection; IDN = Identification; OCL = One Card Learning; OBK = One Back; CPAL = Continuous Paired Associate Learning; ISLT = International Shopping List Test.

*p < .05.

Discussion

The first hypothesis that healthy older adults would perform better than adults with aMCI, who would in turn perform better than adults with AD was supported. In general, performance on each measure from the CogState battery was worse in AD than in controls, although the magnitude of impairment was larger for the measures of episodic memory (ISLT total and delayed recall) than it was for the measures of attention and psychomotor function (IDN and DET). As the AD group performed simpler versions of the CPAL and OCL tasks, group mean performance on these tasks was not compared directly between AD and HA groups. For the aMCI group, performance was worse than controls only for the measures of episodic and working memory. Performance of the aMCI group on measures of psychomotor function and attention was equivalent to controls. The finding that both patients with AD and aMCI performed worse than controls on the ISLT total and delayed recall, CPAL and OCL is consistent with previous studies conducted in different samples (Harel et al., 2011; Lim, Ellis, Harrington, et al., 2012; Lim, Pietrzak, et al., 2012). Furthermore, we observed that the magnitude of impairment of performance on these measures of learning and memory was less in aMCI than in AD. This is consistent with disease models that propose aMCI to represent a disease stage intermediate between healthy aging and dementia (Albert et al., 2011).

The second hypothesis that performance on the CogState battery would be reliable and remain stable over the 12 weeks in healthy adult, aMCI, and AD groups was supported for all of the tasks. Each of the outcome measures from the CogState battery of tests showed high test–retest reliability and the strength of these estimates of reliability did not change with increasing disease severity. Additionally, the statistical analyses showed that average performance for each of the groups remained constant across the 12-week period on each of the CogState outcome measures. Although the omnibus analyses showed no change in performance over time for any of the clinical groups, inspection of mean performance across assessments for the AD group suggests that performance improved from the first to the second assessment on tests of immediate and delayed verbal memory (ISLT), visual associate learning (CPAL), and working memory (OBK) tasks. The magnitude of this improvement from baseline to the 3-month assessment is summarized in Table 4. In the AD group, words recalled on the ISLT increased by half a word, errors on the CPAL decreased by four errors, and when expressed as change from baseline, rather than as group mean at each assessment, accuracy of performance on the OBK did not change. We have previously reported that small improvements from the first to second assessments can occur for these same tasks in healthy older and younger adults as participants become more familiarized with the tests (Collie, Maruff, Darby, & McStephen, 2003). However, in these studies, the data for the first assessment was taken from the first time individuals performed the test (i.e., individuals had no previous practice with the test). In the current study, all individuals had completed a single practice assessment prior to the baseline assessment report here. Thus, when considered with our previous findings, there should have been no improvement in the AD group from the baseline to the second assessment. It is possible that the non-significant improvement in performance on the memory tasks in the AD group occurred because more than one practice session is required in order to achieve stable performance. This hypothesis will be tested by the extent to which performance on this battery remains stable at the 6-month reassessment in the AD group. Importantly, despite these small improvements in performance on the ISLT and CPAL, the reliability of performance on each of these measures remained very high (∼0.8, Table 4) in the AD group.

In healthy older adults, the stability of performance indicates that despite being given at relatively short retest intervals, performance on the battery did not improve with repeated administration (i.e., no practice effect). This stability of performance is consistent with that observed previously in healthy adults assessed repeatedly over longer time intervals (e.g., 1 year; Fredrickson et al., 2010). The stability of performance in the healthy older adults group supports the conclusion that the absence of any improvement in performance also reflects the absence of practice effects in the aMCI and AD groups. The stability of performance observed for the aMCI and AD groups on all of the CogState tasks indicates also that there was no decline in any aspect of cognition in the AD group over the 12-week test–retest period. The finding that performance in AD does not decline over 12 weeks is consistent with data from the placebo groups in recent clinical trials conducted over the same time period in which no decline in cognitive function has been observed for the ADAS-Cog (Salloway et al., 2009), other paper and pencil neuropsychological tests (e.g., the Neuropsychological Test Battery; Faux et al., 2010), or with some of the CogState measures studied here.

The third aim was to explore estimates of variability in 3-month change scores for the ROCS CogState measures in the three clinical groups. In general, within-individual variability over time was equivalent between the three groups for the majority of the outcome measures. The only exceptions to this were that variability within individuals over time was greater in the aMCI group than controls for CPAL and that within-individual variability over time was greater in the AD than controls for the DET and IDN tasks. An increase in within-individual variability indicates that performance on the specific outcome measure varies more from assessment to assessment. However, in general, the data show that in aMCI and AD, variability in performance on the cognitive tests over time does not increase despite the lower levels of performance on these same tests.

The estimates of variability and expected mean change over time in the three groups studied here (i.e., Table 4) can be used to guide the design of clinical trials seeking to measure the effect of a pharmacotherapeutic intervention in groups of AD, aMCI, or even healthy older adults. For example, the change from baseline to week 12 on verbal memory (i.e., ISLT total recall) in the AD group was a very slight increase in words recalled of 0.46 with a standard deviation of 4.99 for this change. Thus, to plan a trial where the improvement in performance under treatment was, for example, d = 0.4 (i.e., an improvement of performance on the ISLT total recall of 2 words), then 100 patients would need to be assigned randomly to the placebo and treatment groups (assuming that there are two groups, the equal ratio of assignment to each group, Type I error of 0.05, and sample size required for statistical significance; Faul, Erdfelder, Lang, & Buchner, 2007). For the OCL, the same expected effect size (and, therefore, the same sample size) would require the accuracy of performance on this task to improve from 0.82 at baseline to 0.85 at week 12 (refer to Table 4) in order for an effect to be rendered statistically significant. It is important to note that for both tests, the magnitude of improvement associated with this effect size would not increase the level of performance of the AD group to that of the MCI group. The practice-corrected RCIs presented in Table 4 can be used by neuropsychologists to guide decisions about the presence of change of cognitive function in healthy older adults or in adults who meet clinical criteria for aMCI or AD. For example, the 85% CI for the ISLT delayed recall indicates that in the aMCI group, a reduction of 3 words over 3 months would be clinically meaningful. Similarly, in the AD group, an increase in 23 errors on the CPAL task over 3 months would be clinically significant at 85% CI. Further, the magnitude of change required for a decline in performance to be considered clinically meaningful in the different groups is relatively similar for each task.

Taken together, the results of this study show that performance in the AD and aMCI groups did not decline over the 3-month period and that this stability is consistent with that observed on other cognitive assessments in recent industry sponsored clinical trials (Faux et al., 2010; Salloway et al., 2009). This suggests that clinical trials of putative cognitive enhancing drugs should be designed to detect improvement in cognition in the treatment group compared with placebo. This approach is different to those where the beneficial effects of cognitively enhancing drugs are indicated by stabile cognitive function in the treated group while performance in the placebo group declines (Darby, Brodtmann, Woodward, Budge, & Maruff, 2011). Alternatively, trials of drugs designed to ameliorate AD-related cognitive decline will have to be conducted over time intervals longer than 3 months if the CogState battery is used to measure cognition. In a previous 12-month study of mild to moderate AD, we observed a moderate decline (d = 0.60, or 2.5 words) on the ISLT total recall measure (Lim, Pietrzak, et al., 2012). Similarly, we observed a decline in performance over 12 months on the OCL in patients with aMCI (Maruff et al., 2004). However, future analysis of ROCS data will provide estimates of the stability of performance on this cognitive test battery over 9, 12, and 18 month periods in the clinical groups studied here.

When interpreting the results of this study, it is important to note that the AIBL-ROCS study is not an epidemiological sample. In the recruitment of healthy older adults, participants were highly educated and had few existing or untreated medical or psychiatric illnesses. Further, the selection of MCI groups was biased toward the inclusion of individuals with aMCI. As such, it would be important for these findings to be replicated in individuals in population-based studies, such as the Mayo Clinic Study of Aging (Roberts et al., 2008), where it is possible that the estimates of variability in cognitive performance may be greater than that observed here. Despite this, the data presented here are important in that they are consistent with recent recommendations for estimates of stability and reliability to be reported for cognitive tests often used to measure change in cognitive function in individuals (Duff, 2012) and show that when considered over a short test–retest interval of 3 months, each of the CogState measure has good reliability and stability, and when considered at a single time-point, is able to detect AD-related cognitive impairment.

Funding

Funding for the study was provided by AstraZeneca Pharmaceuticals LP, the CSIRO Flagship Collaboration Fund and the Science and Industry Endowment Fund (SIEF) in partnership with Edith Cowan University (ECU), Mental Health Research institute (MHRI), Alzheimer's Australia (AA), National Ageing Research Institute (NARI), Austin Health, CogState Ltd, Hollywood Private Hospital, Sir Charles Gardner Hospital. The study also receives funding from the National Health and Medical Research Council (NHMRC), the Dementia Collaborative Research Centres program (DCRC), The McCusker Alzheimer's Research Foundation, and Operational Infrastructure Support from the Government of Victoria.

Conflict of Interest

PM and JJ are a full-time employees of CogState Ltd, the company that provided the CogState battery. DD is a scientific consultant to CogState Ltd and is a shareholder of stock in CogState Ltd. TA is a full-time employee of AstraZeneca Pharmaceuticals LP, the company that funded the ROCS study. JJ was a full-time employee of AstraZeneca until May 2012. TA is a shareholder of stocks in AstraZeneca. AS was a full-time employee of AstraZeneca until 2009 and now works as a paid consultant to pharmaceutical companies including AstraZeneca.

Acknowledgements

The ROCS team wishes to thank the participants in the ROCS study for their commitment and dedication to helping advance research into the early detection and causation of AD and the clinicians who referred patients to the study.

References

Albert
M. S.
DeKosky
S. T.
Dickson
D.
Dubois
B.
Feldman
H. H.
Fox
N. C.
, et al.  . 
The diagnosis of mild cognitive impairment due to Alzheimer's disease: Recommendations from the National Institute on Aging and Alzheimer's Association workgroup
Alzheimer's and Dementia
 , 
2011
, vol. 
7
 
3
(pg. 
270
-
279
)
Bateman
R. J.
Xiong
C.
Benzinger
T. L. S.
Fagan
A. M.
Goate
A.
Fox
N. C.
, et al.  . 
Clinical and biomarker changes in dominantly inherited Alzheimer's disease
The New England Journal of Medicine
 , 
2012
, vol. 
367
 
9
(pg. 
795
-
804
)
Collie
A.
Maruff
P.
Darby
D. G.
McStephen
M.
The effects of practice on the cognitive test performance of neurologically normal individuals assessed at brief test-retest intervals
Journal of the International Neuropsychological Society
 , 
2003
, vol. 
9
 (pg. 
419
-
428
)
Darby
D.
Brodtmann
A.
Pietrzak
R. H.
Fredrickson
J.
Woodward
M.
Villemagne
V.
, et al.  . 
Episodic memory decline predicts cortical amyloid status in community-dwelling older adults
Journal of Alzheimer's Disease
 , 
2011
, vol. 
27
 
3
(pg. 
627
-
637
)
Darby
D.
Brodtmann
A.
Woodward
M.
Budge
M.
Maruff
P.
Using cognitive decline in novel trial designs for primary prevention and early disease-modifying therapy trials of Alzheimer's disease
International Psychogeriatrics
 , 
2011
, vol. 
23
 
9
(pg. 
1376
-
1385
)
Duff
K.
Evidence-based indicators of neuropsychological change in the individual patient: Relevant concepts and methods
Archives of Clinical Neuropsychology
 , 
2012
, vol. 
27
 (pg. 
248
-
261
)
Ellis
K. A.
Bush
A. I.
Darby
D.
De Fazio
D.
Foster
J.
Hudson
P.
, et al.  . 
The Australian Imaging, Biomarkers and Lifestyle (AIBL) study of aging: Methodology and baseline characteristics of 1112 individuals recruited for a longitudinal study of Alzheimer's disease
International Psychogeriatrics
 , 
2009
, vol. 
21
 
4
(pg. 
672
-
687
)
Faul
F.
Erdfelder
E.
Lang
A. G.
Buchner
A.
G*Power 3: A flexible statistical power analysis program for the social, behavioral, and biomedical sciences
Behavior Research Methods
 , 
2007
, vol. 
39
 
2
(pg. 
175
-
191
)
Faux
N. G.
Ritchie
C. W.
Gunn
A.
Rembach
A.
Tsatsanis
A.
Bedo
J.
, et al.  . 
PBT2 rapidly improves cognition in Alzheimer's disease: Additional phase II analyses
Journal of Alzheimer's Disease
 , 
2010
, vol. 
20
 
2
(pg. 
509
-
516
)
Francis
P. T.
Nordberg
A.
Arnold
S. E.
A preclinical view of cholinesterase inhibitors in neuroprotection: Do they provide more than symptomatic benefits in Alzheimer's disease?
Trends in Pharmacological Sciences
 , 
2005
, vol. 
26
 
2
(pg. 
104
-
111
)
Fredrickson
J.
Maruff
P.
Woodward
M.
Moore
L.
Fredrickson
A.
Sach
J.
, et al.  . 
Evaluation of the usability of a brief computerized cognitive screening test in older people for epidemiological studies
Neuroepidemiology
 , 
2010
, vol. 
34
 (pg. 
65
-
75
)
Hammers
D.
Spurgeon
E.
Ryan
K.
Persad
C.
Heidebrink
J.
Barbas
N.
, et al.  . 
Reliability of repeated cognitive assessment of dementia using a brief computerized battery
American Journal of Alzheimer's Disease and Other Dementias
 , 
2011
, vol. 
26
 
4
(pg. 
326
-
333
)
Harel
B. T.
Darby
D.
Pietrzak
R. H.
Ellis
K. A.
Snyder
P. J.
Maruff
P.
Examining the nature of impairment in visual paired associate learning in amnestic mild cognitive impairment
Neuropsychology
 , 
2011
, vol. 
25
 
6
(pg. 
752
-
762
)
Hartley
H. O.
The use of range in analysis of variance
Biometrika
 , 
1950
, vol. 
37
 (pg. 
271
-
280
)
Heilbronner
R. L.
Sweet
J. J.
Attix
D. K.
Krull
K. R.
Henry
G. K.
Hart
R. P.
Official position of the American Academy of Clinical Neuropsychology on serial neuropsychological assessments: The utility and challenges of repeat test administration in clinical and forensic contexts
The Clinical Neuropsychologist
 , 
2010
, vol. 
24
 (pg. 
1267
-
1278
)
Hinton-Bayre
A. D.
Deriving reliable change statistics from test-retest normative data: Comparison of models and mathematical expressions
Archives of Clinical Neuropsychology
 , 
2010
, vol. 
25
 (pg. 
244
-
256
)
Jack
C. R.
Lowe
V. J.
Weigand
S. D.
Wiste
H. J.
Senjem
M. L.
Knopman
D. S.
, et al.  . 
Serial PIB and MRI in normal, mild cognitive impairment and Alzheimer's disease: Implications for sequence of pathological events in Alzheimer's disease
Brain
 , 
2009
, vol. 
132
 (pg. 
1355
-
1365
)
Jacobson
N. S.
Truax
P.
Clinical significance: A statistical approach to defining meaningful change in psychotherapy research
Journal of Consulting and Clinical Psychology
 , 
1991
, vol. 
59
 
1
(pg. 
12
-
19
)
Jaeger
J.
Hardemark
H. G.
Zettergren
A.
Does repeated daily testing improve measurement sensitivity to the cognitive effects of donepezil in mild to moderate Alzheimer's disease?
2011
Paper presented at the Alzheimer's Association International Conference
Paris
Jelic
V.
Kivipelto
M.
Winblad
B.
Clinical trials in mild cognitive impairment: Lessons for the future
Journal of Neurology, Neurosurgery, and Psychiatry
 , 
2006
, vol. 
77
 
4
(pg. 
429
-
438
)
Kantarci
K.
Lowe
V.
Przybelski
S. A.
Weigand
S. D.
Senjem
M. L.
Ivnik
R. J.
, et al.  . 
APOE modifies the association between Aβ load and cognition in cognitively normal older adults
Neurology
 , 
2012
, vol. 
78
 
4
(pg. 
232
-
240
)
Lim
Y. Y.
Ellis
K. A.
Harrington
K.
Ames
D.
Martins
R. N.
Masters
C. L.
, et al.  . 
Use of the CogState Brief Battery in the assessment of Alzheimer's disease related cognitive impairment in the Australian Imaging, Biomarker and Lifestyle (AIBL) study
Journal of Clinical and Experimental Neuropsychology
 , 
2012
, vol. 
34
 
4
(pg. 
345
-
358
)
Lim
Y. Y.
Ellis
K. A.
Pietrzak
R. H.
Ames
D.
Darby
D.
Harrington
K.
, et al.  . 
Stronger effect of amyloid load than APOE genotype on cognitive decline in healthy older adults
Neurology
 , 
2012
, vol. 
79
 (pg. 
1645
-
1652
)
Lim
Y. Y.
Pietrzak
R. H.
Snyder
P. J.
Darby
D.
Maruff
P.
Preliminary data on the effect of culture on the assessment of Alzheimer's disease related verbal memory impairment with the International Shopping List Test (ISLT)
Archives of Clinical Neuropsychology
 , 
2012
, vol. 
27
 
2
(pg. 
136
-
147
)
Lim
Y. Y.
Prang
K. H.
Cysique
L.
Pietrzak
R. H.
Snyder
P. J.
Maruff
P.
A method for cross-cultural adaptation of a verbal memory assessment
Behavior Research Methods
 , 
2009
, vol. 
41
 
4
(pg. 
1190
-
1200
)
Maruff
P.
Collie
A.
Darby
D.
Weaver-Cargin
J.
Masters
C.
Currie
J.
Subtle memory decline over 12 months in mild cognitive impairment
Dementia and Geriatric Cognitive Disorders
 , 
2004
, vol. 
18
 (pg. 
342
-
348
)
Petersen
R. C.
Aisen
P. S.
Beckett
L. A.
Donohue
M. C.
Gamst
A. C.
Harvey
D. J.
, et al.  . 
Alzheimer's Disease Neuroimaging Initiative (ADNI): Clinical characterization
Neurology
 , 
2010
, vol. 
74
 (pg. 
201
-
209
)
Pietrzak
R. H.
Maruff
P.
Snyder
P. J.
Methodological improvements in quantifying cognitive change in clinical trials: An example with single-dose administration of Donepezil
Journal of Nutrition, Health and Aging
 , 
2009
, vol. 
13
 
3
(pg. 
268
-
273
)
Roberts
R. O.
Geda
Y. E.
Knopman
D. S.
Cha
R. H.
Pankratz
V. S.
Boeve
B. F.
, et al.  . 
The Mayo Clinic Study of Aging: Design and sampling, participation, baseline measures and sample characteristics
Neuroepidemiology
 , 
2008
, vol. 
30
 
1
(pg. 
58
-
69
)
Rogers
S. L.
Doody
R. S.
Mohs
R. C.
Friedhoff
L. T.
the Donepezil Study Group
Donepezil improves cognition and global function in Alzheimer disease: A 15-week, double-blind, placebo-controlled study
Archives of Internal Medicine
 , 
1998
, vol. 
158
 
9
(pg. 
1021
-
1031
)
Rowe
C. C.
Ackermann
U.
Browne
W.
Mulligan
R.
Pike
K. L.
O'Keefe
G.
, et al.  . 
Imaging of amyloid β in Alzheimer's disease with 18F-BAY94-9172, a novel PET tracer: Proof of mechanism
The Lancet Neurology
 , 
2008
, vol. 
7
 
2
(pg. 
129
-
135
)
Rowe
C. C.
Ellis
K. A.
Rimajova
M.
Bourgeat
P.
Pike
K. E.
Jones
G.
, et al.  . 
Amyloid imaging results from the Australian Imaging, Biomarkers and Lifestyle (AIBL) study of aging
Neurobiology of Aging
 , 
2010
, vol. 
31
 (pg. 
1275
-
1283
)
Salloway
S.
Sperling
R.
Gilman
S.
Fox
N. C.
Blennow
K.
Raskind
M.
, et al.  . 
A phase 2 multiple ascending dose trial of bapineuzumab in mild to moderate Alzheimer disease
Neurology
 , 
2009
, vol. 
73
 
24
(pg. 
2061
-
2070
)
Small
G. W.
Siddarth
P.
Kepe
V.
Ercoli
L. M.
Burggren
A. C.
Bookheimer
S. Y.
, et al.  . 
Prediction of cognitive decline by positron emission tomography of brain amyloid and tau
Archives of Neurology
 , 
2012
, vol. 
69
 
2
(pg. 
215
-
222
)
Snyder
P.
Bednar
M. M.
Cromer
J. R.
Maruff
P.
Reversal of scopolamine-induced deficits with a single dose of donepezil, an acetylcholinesterase inhibitor
Alzheimer's and Dementia
 , 
2005
, vol. 
1
 
2
(pg. 
126
-
135
)
Sperling
R. A.
Aisen
P. S.
Beckett
L. A.
Bennett
D. A.
Craft
S.
Fagan
A. M.
, et al.  . 
Toward defining the preclinical stages of Alzheimer's disease: Recommendations from the National Institute on Aging and the Alzheimer's Association workgroup
Alzheimer's and Dementia
 , 
2011
, vol. 
7
 
3
(pg. 
280
-
292
)
Villemagne
V. L.
Pike
K. E.
Chételat
G.
Ellis
K. A.
Mulligan
R. S.
Bourgeat
P.
, et al.  . 
Longitudinal assessment of Aβ and cognition in aging and Alzheimer disease
Annals of Neurology
 , 
2011
, vol. 
69
 
1
(pg. 
181
-
192
)