Abstract
Objective:
The clinical diagnosis of Alzheimer's disease is often based, at least in part, on poor cognitive test performance compared with normative values. The presence and extent of an ascertainment bias (omission of affected individuals) produced by such criteria when applied as early as possible in the course of the disease was examined.
Design:
Longitudinal study from 1979 to 2008.
Setting:
Washington University in St. Louis Alzheimer Disease Research Center.
Participants:
Of 78 individuals aged 65 to 101 years enrolled as healthy controls 55 later developed autopsy-confirmed AD; 23 remained cognitively healthy and did not have neuropathologic AD.
Main Outcome Measures:
Criteria for diagnosis of Alzheimer disease based on various cut-points (1.5, 1.0, and 0.5 standard deviations below the mean for robust test norms) for two standard psychometric measures from each of three cognitive domains (episodic memory, visusospatial ability, working memory) were applied to data from the first assessment associated with an independent clinical diagnosis of cognitive impairment for those who developed symptomatic AD and the last assessment for those who did not.
Results:
Areas under the curve from ROC analyses ranged from .71 to .49; sensitivities and specificities were unsatisfactory even after adjusting for age and education, using combinations of tests, or examining longitudinal decline prior to clinical diagnosis.
Conclusions:
Reliance on divergence from group normative values to determine initial cognitive decline caused by Alzheimer disease results in failure to include people in the initial symptomatic stage of the illness.
To implement effective treatment, when available, it is important to identify individuals as early as possible in the neuropathological course of Alzheimer's disease (AD). Clinical diagnosis often relies on comparison of performance on cognitive tests with group norms. In their seminal article Blessed, Tomlinson, and Roth1 reported that some cases had substantial neuropathology at autopsy but performed well on measures of cognition, an observation confirmed by others.2-5 Blessed and colleagues1 suggested “a certain amount of the change estimated by plaque counts may be accommodated within the reserve capacity of the cerebrum without causing manifest intellectual impairment” (p. 807). Thus, individuals with greater biological cerebral reserve could experience extensive neuropathology but not reach the threshold for expression of symptoms necessary for clinical diagnosis because of that reserve capacity.6 This hypothesis, later expanded to include cognitive reserve (e.g., experience and strategies),7 has been popular in research on AD.
Another obstacle to early diagnosis is ascertainment bias,8 which focuses on methodological issues related to determining the clinical diagnosis of dementia. For example, some popular methods for identifying dementia were designed to detect frank dementia and are too easy (ceiling effect) for high functioning individuals in the beginning stages of a dementing illness. In one meta-analysis, 6 of the 13 brain reserve studies used such mental status instruments as the only outcome measure.9 Even if instruments are sufficiently difficult, cut-points used to diagnose dementia are usually based on norms from samples that contain individuals with early cognitive impairment whose deficits have not yet been identified as dementia.10 This produces a cut-point that is too lenient, and all individuals will be further along in the disease process when they are identified. For example, in the Nun Study11 only 59% of those who met neuropathological criteria for AD also met clinical criteria for dementia, although almost all had some detectable cognitive deficit.
The purpose of this report was to assess whether there is ascertainment bias produced by test criteria similar to those often used for the clinical diagnosis of AD when they are applied as early in the course of the disease as possible. The sample included individuals from a longitudinal study who were cognitively healthy at enrollment but later developed autopsy-confirmed symptomatic AD. Sensitivity of test criteria was examined at the time of first clinical diagnosis of a deficit in cognitive function that reflected intraindividual change as elicited by interviews with informants and the individuals rather than by the person's relative standing based on robust group norms or deviation from predicted scores based on age and education. Although the primary focus was on the percentage of persons with symptomatic AD who could be identified using test scores, individuals from the same sample who did not develop clinical symptoms of cognitive deficits and who did not have a diagnosis of any dementing disease at autopsy were included as a control group. Longitudinal course was also examined using growth and survival analyses.
Methods
Participants
Archival data from 78 participants (36 men, 2 African American, the remainder White) were available from a sample of 521 volunteers aged 65 to 101 years at entry enrolled between 1979 and 2008 in a longitudinal study of healthy aging and dementia at the Washington University Alzheimer's Disease Research Center (ADRC). At entry all were clinically evaluated to be cognitively healthy (Clinical Dementia Rating12 [CDR] = 0) with no potentially confounding neurological or psychiatric disorders. Of the 163 people who progressed to a CDR > 0 at their last assessment, 96 have died, 63 had autopsies; and 55 (included here as the AD group) had neuropathologically confirmed AD. The 8 omitted had other neuropathological diagnoses; 5 had a clinical diagnosis of symptomatic AD and 3 were diagnosed with a nonAD dementia, yielding a clinical diagnostic accuracy rate for AD of 92% (55 of 60 cases). The control group included the 23 people who were still CDR 0 at their last assessment prior to death and for whom no dementing disease was found at autopsy.
Sample demographics and mean scores on a brief cognitive screening measure13 are shown in Table 1. The percentage of people with an apolipoprotein E4 (APOE4) allele was comparable (p = .50) in the AD (26%) and normal (19%) groups. The Washington University Human Studies Committee approved all procedures. Data from these participants have appeared in other publications from the center.
Table 1.
Sample Characteristics
| AD | Control | ||
|---|---|---|---|
| Variable | (n = 55) | (n = 23) | P Value |
| Education, mean (SD), y | 13.9 (3.6) | 13.9 (3.8) | .98 |
| Age at entry, mean (SD),y | 81.0 (9.7) | 80.1 (9.2) | .72 |
| Age at last CDR 0, mean (SD), y | 85.5 (8.1) | 86.6 (8.4) | .59 |
| SBT at entry, mean (SD) | 1.6 (2.3) | 1.8 (2.3) | .72 |
| SBT at last CDR 0, mean (SD) | 2.0 (2.3) | 1.6 (2.2) | .57 |
| SBT at first CDR > 0, mean (SD) | 6.0 (6.1) |
SBT = Short Blessed Test13, a dementia screening test with scores (number of errors) that range from 0 (best) to 28 (worst).
Clinical Evaluation
The CDR staging and diagnostic evaluation is based on annual semistructured interviews with the participant and a knowledgeable collateral source (usually spouse or adult child), a health history, medication and depression inventories, an aphasia battery, and a neurological examination of the participant.14 Generally participants were seen by different clinicians from year to year, although the clinician was the same at the next visit for 17% (91/546) of the follow-ups. Clinicians did not have access to previous clinical evaluations or to previous and current psychometric test results. The research-trained clinician determined if any cognitive problems represented decline from former level of function for that individual and interfered to some degree with the person's ability to carry out accustomed activities. If so, a CDR > 0 was assigned. This procedure is highly successful (93%) in identifying autopsy-confirmed AD.15
Neuropathology
The neuropathological diagnosis of AD was made according to standard assessment procedures.15 To maintain consistency of diagnosis over time, given that participants were enrolled beginning in 1979, the criteria proposed by Khachaturian16 as modified by our group17 were used for diagnosis.
Psychometric Assessment
A psychometric battery assessing a broad spectrum of abilities was administered to all participants, usually a week or two after the annual clinical assessment. The standard tests examined here are ones often used in clinical diagnosis based on normative values and included two measures from each of three cognitive domains:18 episodic memory, speeded visuospatial ability, and working memory. The episodic memory tests were immediate recall of Logical Memory (verbatim scored according to the Russell19 criteria) and Associate Learning from the Wechsler Memory Scale (WMS).20 The visuospatial tests were Block Design and Digit Symbol from the Wechsler Adult Intelligence Scale (WAIS).21 The tests of working memory were WMS Digit Span Backward20 and Letter Fluency for S and P.22
Test scores were converted to z scores using as the reference (normative) group23 a sample of 310 people (M age = 74.5, SD = 8.6; M education = 14.8, SD = 3.2), including the 23 in the normal brain group reported here, who similarly were enrolled as CDR 0 during the same time period, had at least one annual follow-up, but never progressed to CDR > 0. The means and SDs for the variables were: Logical Memory, M = 8.87, SD = 2.81; Associate Learning, M = 13.42, SD = 3.53; Block Design, M = 30.05, SD = 8.63; Digit Symbol, M = 45.67; SD = 11.53; Digit Span Backward, M = 4.75, SD = 1.28; Letter Fluency, M = 29.41, SD = 9.73.
Beginning September 1, 2005, the funding agency required changes in three tests: WMS Logical Memory was replaced by WMS-R Logical Memory Story A;24 WAIS Digit Symbol was replaced by WAIS-R Digit Symbol;25 WMS Digit Span Backward was replaced by the WMS-R version.24 The last two changes were trivial and therefore are ignored here. The Logical Memory change was substantial (e.g., gist scoring, only one story) and affected data for 7 AD participants and 3 controls. Although smaller and with less follow-up than the one used for the other measures, the reference group used for z-score conversion for this test was the first assessment of 78 people (M age = 73.4) enrolled with CDR 0 since September, 2005, who remained CDR 0 throughout follow-up: M = 12.49, SD = 3.39.
Because performance on standard tests may vary with individual differences such as age or education, norms are often adjusted. For ease of clinical application, usually such adjusted norms are stratified by age decade, for example, or by blocking education into several ordinal categories.26 A more precise method is to calculate an individual's predicted score on the test from the person's exact age and years of education and then compute a residual to indicate how far the observed score is from the predicted one. The regression equations used to form such standardized residuals in this study were derived on the robust normative sample of 310 mentioned previously; there were no significant interactions of education with age.26 The standardized regression equations were as follows: Logical Memory = .26 education - .21 age; Associate Learning = .04 education - .18 age; Block Design = .27 education - .37 age; Digit Symbol = .12 education - .45 age; Digit Span Backward = .08 education - .13 age;Letter Fluency = .26 education - .005 age.
Statistical Analyses
Comparisons of the two groups were made with t tests for independent groups for the quantitative variables and the chi-square test of association for the presence of the APOE4 allele. The area under the curve (AUC) from ROC analyses was examined for each cognitive measure based on group norms and for age- amd education-corrected residuals using scores obtained at the time of the first CDR > 0 (AD group) or last assessment (control group). Classification accuracy of cut points based on z values of −1.5, −1.0, and −0.5 were examined for each measure individually and also based on whether or not any of the six measures fell below the cut point. Cox regression analyses were used to examine time to first CDR > 0 and its correlates in the AD group.
Longitudinal course was examined using a random coefficients model applied to each of the measures (SAS v9.1.3; PROC MIXED). The fixed effect was group (AD, control). A piecewise linear growth curve over time was connected at the last time of assessment prior to the first CDR > 0, which was the last assessment for the control group. Thus, any significant change in slope applies only to the AD group. Classification accuracy was calculated for a decline in performance of ≥ 0.5 SD based on the robust group norms between the first CDR > 0 and the assessment just prior to that for the AD group and the last two assessments for the control group.
Results
The AUCs from the ROC analyses using group norms and individual residuals were poor (Table 2) as was sensitivity (percentage of correct classifications in the AD group; Table 3). The commonly used cut-points on measures of episodic memory at 1.5 or 1.0 SD below the mean produced clearly unacceptable detection rates (23 to 44%), although specificity (accurate classification of the control group) was excellent to good (83 to 96%). Sensitivity decreased using the individual residuals; specificity increased slightly (Table 3).
Table 2.
Results from ROC Analyses Using Either Group Norms or Individual Residuals
| Group Norms | Residuals | |||
|---|---|---|---|---|
| Measure | AUC | 95% CI | AUC | 95% CI |
| Logical Memory | .71 | .59 - .83 | .71 | .58 - .83 |
| Associate Learning | .69 | .56 - .83 | .71 | .58 - .84 |
| Block Design | .64 | .49 - .79 | .72 | .58 - .86 |
| Digit Symbol | .49 | .34 - .64 | .50 | .35 - .65 |
| Digit Span Backward | .53 | .39 - .68 | .54 | .40 - .68 |
| Letter Fluency | .57 | .43 - .70 | .57 | .44 - .70 |
ROC = receiver operating characteristics; AUC = area under the curve; CI = confidence interval
Table 3.
Sensitivity and Specificity Using Different Criteria
| Measure and Cut Point | Group Norms | Residuals | ||
|---|---|---|---|---|
| AD | Control | AD | Control | |
| Logical Memory | ||||
| −1.5 SD | 35 (28-37) | 91 (77-98) | 23 (17-25) | 96 (83-99) |
| −1.0 SD | 44 (37-48) | 87 (71-95) | 39 (32-40) | 96 (83-99) |
| −0.5 SD | 62 (54-68) | 61 (44-76) | 52 (45-56) | 83 (66-93) |
| Associate Learning | ||||
| −1.5 SD | 23 (17-25) | 96(83-99) | 8 (4-8) | 100 (92-100) |
| −1.0 SD | 35 (28-40) | 83 (67-93) | 29 (22-33) | 87 (72-95) |
| −0.5 SD | 62 (54-69) | 70 (53-83) | 46 (38-52) | 74 (57-86) |
| Block Design | ||||
| −1.5 SD | 20 (14-25) | 86 (73-95) | 11 (6-13) | 96 (85-99) |
| −1.0 SD | 34 (26-38) | 86 (71-95) | 20 (14-25) | 86 (73-95) |
| −0.5 SD | 48 (39-55) | 64 (47-78) | 36 (29-41) | 86 (71-95) |
| Digit Symbol | ||||
| −1.5 SD | 32 (24-39) | 71 (55-85) | 9 (4-14) | 86 (75-94) |
| −1.0 SD | 49 (41-57) | 43 (27-60) | 30 (22-36) | 76 (60-88) |
| −0.5 SD | 65 (58-73) | 33 (19-50) | 49 (40-57) | 57 (40-73) |
| Digit Span Backward | ||||
| −1.5 SD | 2 (0-4) | 96 (92-99) | 2 (0-4) | 96 (92-99) |
| −1.0 SD | 23 (16-28) | 79 (63-90) | 17 (12-21) | 87 (74-95) |
| −0.5 SD | 60 (52-67) | 48 (32-64) | 29 (22-35) | 70 (54-83) |
| Letter Fluency | ||||
| −1.5 SD | 15 (10-18) | 91 (79-98) | 13 (9-13) | 100 (90-100) |
| −1.0 SD | 38 (31-43) | 74 (57-87) | 36 (29-40) | 83 (67-93) |
| −0.5 SD | 55 (48-62) | 52 (63-82) | 51 (44-58) | 52 (36-68) |
| Any measure | ||||
| −1.5 SD | 66 (58-71) | 65 (48-80) | 40 (33-44) | 83 (66-93) |
| −1.0 SD | 81 (76-88) | 35 (21-49) | 72 (66-79) | 61 (44-75) |
| −0.5 SD | 94 (91-98) | 13 (5-21) | 87 (82-93) | 26 (14-39) |
Sensitivity = percentage correct classifications in AD group; specificity = percentage correct classifications in control group; 95% confidence intervals are shown in parentheses.
Better sensitivity was obtained when a deficit was determined to occur on any one or more of the six tests (last section, Table 3), but specificity deteriorated. For example, 81% of the AD group had at least one value 1.0 SD below the mean using group norms but only 35% of the control group were correctly classified. Although specificity improved (65 %) if one required a 1.0 SD deficit on at least two measures, sensitivity declined to 62%. In about one fourth of the AD group the cognitive area(s) affected did not include episodic memory for 22% (group norms) to 26% (residuals) of the AD gtoup.
Median time to a CDR > 0 for the AD group was 5.0 years (95% CI: 3.6 - 6.2). When age at entry, APOE4 status, education, and z scores at entry on the six cognitive measures were entered as the sole covariate in Cox regression analyses, four were significant (Table 4): age, APOE4 status, Logical Memory, and Digit Symbol. Only Digit Symbol made a unique contribution when these covariates were entered simultaneously (Table 4),. It appears that the variance in time to CDR > 0 explained by age, APOE4 status, and initial score on Logical Memory represented shared variance. Age at entry was correlated −.37 with APOE4 status, −.34 with Logical Memory, and −.72 with Digit Symbol.
Table 4.
Regression Coefficients for Covariates Entered Singly and Simultaneously in Cox Regression Analyses in the AD Group
| Covariate | Entered Singly |
Entered Simultaneously |
||||
|---|---|---|---|---|---|---|
| B | SEB | P Value | B | SEB | P Value | |
| Age | .07 | .02 | <.001 | .01 | .03 | .75 |
| APOE4† | −.83 | .39 | .02 | −.38 | .39 | .34 |
| Education | .00 | .04 | .94 | .04 | .39 | .86 |
| Logical Memory |
−.40 | .15 | .009 | −.04 | .21 | .85 |
| Associate Learning |
−.18 | .14 | .21 | −.13 | .22 | .56 |
| Block Design |
−.26 | .19 | .16 | .10 | .22 | .64 |
| Digit Symbol |
−.75 | .16 | <.001 | −.95 | .32 | .003 |
| Digit Span Backward |
−.26 | .16 | .09 | −.03 | .22 | .88 |
| Letter Fluency |
−.05 | .14. | .73 | .20 | .21 | .34 |
APOE4 status coded 1 if E4 allele was present, 0 if absent.
The longitudinal slopes on the six cognitive measures (Table 5) of the two control group were similar to those of the AD group before clinical diagnosis; differences between the slopes approached significance (p = .12) only for Logical Memory. The significant inflection points indicate a change in the slopes of the AD group at the time of diagnosis on all measures.
Table 5.
Annual Rate of Change in Control Group and in D Group Before and After Clinical Diagnosis of Dementia
| Slope (SE) | Slope (SE) | Slope (SE) | Inflection Point |
|
|---|---|---|---|---|
| Measure (z) |
Control | AD Before | AD After | P Value |
| Logical Memory |
.00 (.02) | −.04 (.01) | −.24 (.03) | <.001 |
| Associate Learning |
−.02 (.03) | .00 (.02) | −.14 (.02) | <.001 |
| Block Design |
−.02 (.02) | −.04 (.02) | −.24 (.04) | <.001 |
| Digit Symbol |
−.08 (.02) | −.09 (.02) | −.24 (.03) | <.001 |
| Digit Span Backward |
−.02 (.02) | −.01 (.02) | −.11 (.02) | <.001 |
| Letter Fluency |
.00 (.01) | −.01 (.01) | −.23 (.03) | <.001 |
As shown in Table 6, declines of ≥ 0.5 SD between the assessment at the time of the first CDR > 0 and the prior assessment in the AD group produced poor sensitivity (≤ 50%) for the individual measures. It improved (89%) if the criterion was a decline of ≥ 0.5 SD on any of the six measures, but a similar percentage (85%) of those in the control group also had such a decline between their last two assessments, χ2 (1) = 0.23, p = .63.
Table 6.
Sensitivity and Specificity Based on 0.5 SD Decline Between Two Assessments in AD Control Groups and One-Year Test-Retest Reliabilities in the Robust Norms Reference Group23
| Measure | Sensitivity | Specificity | Reliability |
|---|---|---|---|
| Logical Memory | 50 (43 - 57) | 55 (37-72) | .69 |
| Associate Learning | 39 (32-43) | 85 (68-94) | .67 |
| Block Design | 30 (22-33) | 90 (73-97) | .80 |
| Digit Symbol | 33 (25-40) | 68 (51-83) | .84 |
| Digit Span Backward | 44 (37-50) | 70 (52-84) | .56 |
| Letter Fluency | 49 (42-54) | 70 (51-84) | .79 |
| Any measure | 89 (86-94) | 15 (6-28) |
Sensitivity = percentage correct classifications in AD group; specificity = percentage correct classifications in control group; 95% confidence intervals (CI) are shown in parentheses.
Discussion
Divergence from group normative values to determine initial cognitive decline in AD is not very effective even with robust norms on measures with sufficient range so as to avoid ceiling effects. At the time individuals were identified clinically as no longer cognitively healthy, commonly used cut-points such as −1.5 or −1.0 SD on measures of episodic memory detected only 23 to 44% of those later confirmed to have neuropathological AD. Adjusting for age and education was somewhat useful in preventing cognitively intact people being labeled as demented, particularly on one of the speeded visuospatial measures. Sensitivity, however, decreased, making it more difficult to identify those with symptomatic AD by lowering the cut-point for older people, who are also more likely to have AD.
The “best” balance between sensitivity (.72) and specificity (.61) was obtained when a −1.0 SD criterion based on residuals was met for any of the six measures. Even using information from any three of the cognitive domains affected by AD, robust norms, and adjusting for age and education, a diagnosis based on objective test performance at that time was wrong almost a third of the time.
In addition to failing to detect symptomatic AD as early as possible for treatment purposes, AD research samples determined by such normative criteria do not contain people in the initial symptomatic stage of the disease. Similar to the problem of contamination of putatively normal comparison samples pointed out previously, 10, 27 samples are also biased if they omit people in the beginning phases of AD. Depending on the purpose of the research, results may be misleading. For example, it would be difficult to identify the sequence of pathological processes that occur at disease onset if individuals in the very early stage are absent.
The survival analyses suggested that a speeded visuospatial test (Digit Symbol) is a stronger predictor of time to clinical detection of AD than age, apoE status, or episodic memory. Block Design, another visuospatial test, did not demonstrate the same effect. Digit Symbol reflects perceptual speed,28 whereas Block Design, although timed, is primarily a measure of spatial ability. It is possible that perceptual speed is affected very early by the pathological process, but the comparable rates of decline on Digit Symbol in the AD and control groups prior to clinical diagnosis of AD would argue against that interpretation. Alternatively, because processing speed is strongly related to fluid intelligence,29 perhaps it would be a better indicator of brain or cognitive reserve than education, which was unrelated to either the development of AD or the time to clinical diagnosis in this sample.
We have argued30 that it is necessary to consider the person's longitudinal course rather than emphasizing current standing in a normative group. Although the present results illustrate the limitations of emphasis on the person's current standing, it is not clear that applying a longitudinal approach to standard psychometric tests provides a satisfactory alternative. The linear rate of decline in symptomatic AD (Table 5) is substantial over the entire course of the disease, but it is actually an accelerating rate of decline.23 For standard cognitive measures such as the ones used here, the annual decline is less in the early symptomatic phases of the disease than it is later,31 making it difficult to differentiate cognitive decline from measurement error. As shown in Table 6, Logical Memory, for example, dropped 0.5 SD in about half of both groups. Test-retest stability may be inadequate for a number of standard measures.32 To examine that possibility, we calculated 1-year test-retest reliabilities (Table 6) in the reference sample23 of cognitively healthy individuals using values obtained at their second and third assessments so as to avoid the practice effect often seen after the first assessment. Half the measures had poor (<.70) test-retest stability.
Although this study has a number of strengths (e.g., autopsy-confirmation of diagnosis, initial healthy status of all participants, robust norms and statistical correction for age and education), it is a study of research volunteers, which limits generalization. A larger sample, especially of the autopsy-verified control group, would increase statistical power, which might allow detection of memory decline prior to clinical diagnosis. The cognitive measures described here are widely used, but they represent a limited set of available instruments; further, they do not include measures of delayed list-learning recall or of executive function, attention, and working memory developed since this longitudinal research began. If cognitive measures are to be used in the diagnosis of AD when it first occurs, more sensitive, reliable measures of memory and other cognitive domains affected by the disease are needed in order to avoid the ascertainment bias that currently affects many research samples.
Clinicians must also carefully assess changes in function as determined by history or collateral report, especially in initially high functioning individuals. Comparatively little attention has been paid to the development of measures of subtle changes in function for either clinical practice or research. The AD833 is one such attempt; it is a brief (8 questions) informant interview that is sensitive to early features of cognitive change and is highly correlated with cognitive and functional components of the CDR and with neuropsychological testing.34
Acknowledgements
The authors wish to thank the Dan W. McKeel, Jr., M.D. and the Washington University Alzheimer Disease Research Center Neuropathology Core for the neuropathologic diagnoses and the Genetics Core, Alison Goate, DPhil, Director, for APOE status.
This study was supported by grants P01 AG03991 and P50 AG05681 from the National Institute on Aging (NIH).
References
- 1.Blessed G, Tomlinson BE, Roth M. The association between quantitative measures of dementia and of senile changes in the cerebral grey matter of elderly subjects. Brit J Psychiat. 1968;114:797–811. doi: 10.1192/bjp.114.512.797. [DOI] [PubMed] [Google Scholar]
- 2.Troncoso JC, Cataldo M, Nixon RA, et al. Neuropathology of preclinical and clinical late-onset Alzheimer's disease. Ann Neurol. 1998;43:673–676. doi: 10.1002/ana.410430519. [DOI] [PubMed] [Google Scholar]
- 3.Schmitt FA, Davis DG, Wekstein DR, et al. “Preclinical” AD revisited: Neuropathology of cognitively normal older adults. Neurology. 2000;55:370–376. doi: 10.1212/wnl.55.3.370. [DOI] [PubMed] [Google Scholar]
- 4.Knopman DS, Parisi JE, Salviati A, et al. Neuropathology of cognitively normal elderly. J Neuropathol Exp Neurol. 2003;62:1087–1095. doi: 10.1093/jnen/62.11.1087. [DOI] [PubMed] [Google Scholar]
- 5.Price JL, McKeel DW, Jr, Buckles VD, et al. Neuropathology of nondemented aging: Presumptive evidence for preclinical Alzheimer disease. Neurobiol Aging. 2009;30:1026–1036. doi: 10.1016/j.neurobiolaging.2009.04.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Satz P. Brain reserve capacity on symptom onset after brain injury: A formulation and review of the evidence for threshold theory. Neuropsychology. 1991;7:273–295. [Google Scholar]
- 7.Stern Y. What is cognitive reserve? Theory and research application of the reserve concept. J Int Neuropsychol Soc. 2002;8:448–460. [PubMed] [Google Scholar]
- 8.Tuokko H, Garrett DD, McDowell I, et al. Cognitive decline in high-functioning older adults: reserve or ascertainment bias? Aging & Mental Health. 2003;7:259–270. doi: 10.1080/1360786031000120750. [DOI] [PubMed] [Google Scholar]
- 9.Valenzuela MJ, Sachdev P. Brain reserve and cognitive decline: a nonparametric systematic review. Psychol Med. 2006;36:1065–1073. doi: 10.1017/S0033291706007744. [DOI] [PubMed] [Google Scholar]
- 10.Sliwinski M, Lipton RB, Buschke, et al. The effects of preclinical dementia on estimates of normal cognitive functioning in aging. J Gerontol B Psychol Sci. 1996;51B:P217–P225. doi: 10.1093/geronb/51b.4.p217. [DOI] [PubMed] [Google Scholar]
- 11.Riley KP, Snowdon DA, Desrosiers MF, et al. Early life linguistic ability, late life cognitive function, and neuropathology: findings from the Nun Study. Neurobiol Aging. 2005;26:341–347. doi: 10.1016/j.neurobiolaging.2004.06.019. [DOI] [PubMed] [Google Scholar]
- 12.Morris JC. The clinical dementia rating (CDR): current version and scoring rules. Neurology. 1993;43:2412–2414. doi: 10.1212/wnl.43.11.2412-a. [DOI] [PubMed] [Google Scholar]
- 13.Katzman R, Brown T, Fuld P, et al. Validation of a short orientation-memory-concentration test of cognitive impairment. Am J Psychiatry. 1983;140:734–739. doi: 10.1176/ajp.140.6.734. [DOI] [PubMed] [Google Scholar]
- 14.Morris JC, Storandt M, McKeel DW, Jr, et al. Cerebral amyloid deposition and diffuse plaques in “normal” aging: Evidence for presymptomatic and very mild Alzheimer's disease. Neurology. 1996;1996;46:707–719. doi: 10.1212/wnl.46.3.707. [DOI] [PubMed] [Google Scholar]
- 15.Berg L, McKeel DW, Jr., Miller JP, et al. Clinicopathologic studies in cognitively healthy aging and Alzheimer disease: Relation of histologic markers to dementia severity, age, sex, and apolipoprotein E genotype. Arch Neurol. 1998;55:326–335. doi: 10.1001/archneur.55.3.326. [DOI] [PubMed] [Google Scholar]
- 16.Khachaturian ZS. Diagnosis of Alzheimer's disease. Arch Neurol. 1985;42:1097–105. doi: 10.1001/archneur.1985.04060100083029. [DOI] [PubMed] [Google Scholar]
- 17.McKeel DW, Jr., Price JL, Miller JP, et al. Neuropathologic criteria for diagnosing Alzheimer disease in persons with pure dementia of the Alzheimer type. J Neuropathol Exp Neurol. 2004;63:1028–1037. doi: 10.1093/jnen/63.10.1028. [DOI] [PubMed] [Google Scholar]
- 18.Johnson DK, Storandt M, Morris JC, et al. Cognitive profiles in dementia: Alzheimer disease versus nondemented aging. Neurology. 2008;71:1783–1789. doi: 10.1212/01.wnl.0000335972.35970.70. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Russell EW. A multiple scoring method for the assessment of complex memory functions. J Consult Clin Psych. 1975;43:800–809. [Google Scholar]
- 20.Wechsler D, Stone CP. Manual: Wechsler Memory Scale. Psychological Corporation; New York: 1973. [Google Scholar]
- 21.Wechsler D. Manual: Wechsler Adult Intelligence Scale. Psychological Corporation; NewYork: 1955. [Google Scholar]
- 22.Thurstone LL, Thurstone LG. Examiner manual for the SRA Primary Mental Abilities Test. Science Research Associates; Chicago: 1949. [Google Scholar]
- 23.Johnson DK, Storandt M, Morris JC, et al. Longitudinal study of the transition from healthy aging to Alzheimer's disease. Arch Neurol. 2009;66:1254–1259. doi: 10.1001/archneurol.2009.158. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Wechsler D. Manual: Wechsler Memory Scale-Revised. Psychological Corporation; San Antonio, TX: 1987. [Google Scholar]
- 25.Wechsler D. Manual: Wechsler Adult Intelligence Scale - Revised. Psychological Corporation; New York: 1981. [Google Scholar]
- 26.Kittner SJ, White LR, Farmer ME, et al. Methodological issues in screening for dementia: The problem of education adjustment. J. chron Dis. 1986;39:163–170. doi: 10.1016/0021-9681(86)90019-6. [DOI] [PubMed] [Google Scholar]
- 27.Burgmans S, van Boxtel MPJ, Vuurman EFPM, et al. The prevalence of cortical gray matter atrophy may be overestimated in the healthy aging brain. Neuropsychology. 2009;23:541–550. doi: 10.1037/a0016161. [DOI] [PubMed] [Google Scholar]
- 28.Park DC, Smith AD, Lautensch G, et al. Mediators of long-term memory performance across the life span. Psychol Aging. 1996;11:621–637. doi: 10.1037//0882-7974.11.4.621. [DOI] [PubMed] [Google Scholar]
- 29.Kail R, Salthouse TA. Processing speed as a mental capacity. Acta Psychol. 1994;86:199–225. doi: 10.1016/0001-6918(94)90003-5. [DOI] [PubMed] [Google Scholar]
- 30.Morris JC, Storandt M. Detecting early-stage Alzheimer's disease in MCI and preMCI: The value of informants. In: Jucker M, Beyreuther K, Haass C, Nitsch R, Christen Y, editors. Alzheimer: 100 years and beyond, research and perspectives in Alzheimer's disease. Springer-Verlag; Berlin: 2006. [Google Scholar]
- 31.Storandt M, Grant EA, Miller JP, et al. Rates of progression in mild cognitive impairment and early Alzheimer's disease. Neurology. 2002;59:1034–1041. doi: 10.1212/wnl.59.7.1034. [DOI] [PubMed] [Google Scholar]
- 32.Morris JC, Edland S, Clark C, et al. The consortium to establish a registry for Alzheimer's disease (CERAD). Part IV. Rates of cognitive change in the longitudinal assessment of probably Alzheimer's disease. Neurology. 1993;43:2457–2465. doi: 10.1212/wnl.43.12.2457. [DOI] [PubMed] [Google Scholar]
- 33.Galvin JE, Roe C, Coats M, et al. The AD8: A brief informant interview to detect dementia. Neurology. 2005;65:559–564. doi: 10.1212/01.wnl.0000172958.95282.2a. [DOI] [PubMed] [Google Scholar]
- 34.Galvin JE, Roe CM, Xiong C, et al. The validity and reliability of the AD8 informant interview for dementia. Neurology. 2006;67:1942–1948. doi: 10.1212/01.wnl.0000247042.15547.eb. [DOI] [PubMed] [Google Scholar]