This article was downloaded by: [McGill University Library] On: 09 February 2015, At: 13:21 Publisher: Routledge Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK
Journal of Clinical and Experimental Neuropsychology Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/ncen19
Cued Recall and Early Identification of Dementia H. Tuokko Beattie
a b
a
a
, R. Vernon-wilkinson , J. Weir & B. L.
a
a
Clinic for Alzheimer Disease and Related Disorders , University Hospital-UBC Site , Vancouver b
Clinic for Alzheimer Disease and Related Disorders , University Hospital-UBC Site , 2211 Wesbrook Mall, Vancouver, B.C., Canada , V6T 2B5 Published online: 04 Jan 2008.
To cite this article: H. Tuokko , R. Vernon-wilkinson , J. Weir & B. L. Beattie (1991) Cued Recall and Early Identification of Dementia, Journal of Clinical and Experimental Neuropsychology, 13:6, 871-879, DOI: 10.1080/01688639108405104 To link to this article: http://dx.doi.org/10.1080/01688639108405104
PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sublicensing, systematic supply, or distribution in any form to anyone is expressly
Downloaded by [McGill University Library] at 13:21 09 February 2015
forbidden. Terms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions
Journal of Clinical and Experimental Neuropsychology 1991, Vol. 13, NO. 6, pp. 871-879
0 168-8634/91/1306-0871$3 .MI Q Swets & Zeitlinger
Cued Recall and Early Identification of Dementia*
Downloaded by [McGill University Library] at 13:21 09 February 2015
H. Tuokko, R. Vernon-Wilkinson, J. Weir, and B.L. Beattie Clinic for Alzheimer Disease and Related Disorders, University Hospital-UBC Site, Vancouver
ABSTRACT The early detection of dementia carries implications for clinical management for patients and their families and is of utmost concern if an effective pharmacological treatment is to be found. The utility of an enhanced cued recall paradigm for predicting dementia in a group of elderly subjects was examined. Forty-five subjects referred for clinical evaluation who did not meet the criteria for dementia at initial assessment were reassessed 12-18 months later. Eighteen of these subjects were diagnosed as having possible or probable Alzheimer Disease at reassessment and the diagnostic status of 27 remained unchanged. At initial assessment the ‘change’ group performed more poorly than the ‘no change’ group on measures of retrieval, acquisition and retention derived from the cued recall procedure. As would be expected, the performance of the ‘no change’ group remained stable over time whereas the performance of the ‘change’ group deteriorated, resembling the performance of a group of subjects with mild possible or probable Alzheimer disease. When the sensitivity and specificity of the memory variables were examined, the retrieval measure (i.e., free recall) appeared most useful as an early predictor of dementia. Continued longitudinal evaluations of subjects with questionable dementia are needed to address more fully the natural history of early memory changes associated with dementia.
It h a s been estimated that between 5 and 20 % of persons over the age o f 65 years show evidence of dementing disorders (Mortimer, Schulman, & French, 1981) with devastating costs t o the persons afflicted, their families a n d society. W i t h the ever increasing proportion of elderly in the population, the early diagnosis o f dementia has become an important objective. Early detection carries implications f o r clinical management of patients and their families i n order t o plan appropriately for management of personal affairs and future care. The establishment of behavioral or educational programs designed to teach methods f o r cir-
* The authors would like to thank the patients who took part in this research study and their families. We gratefully acknowledge support for Janine Weir from the Alzheimer Society of British Columbia. Requests for reprints can be directed to Holly Tuokko, Ph.D.. Clinic for Alzheimer Disease and Related Disorders, University Hospital-UBC Site, 221 1 Wesbrook Mall, Vancouver, B.C. Canada, V6T 2BS. Accepted for publication: February 5, 1991.
Downloaded by [McGill University Library] at 13:21 09 February 2015
872
H. TUOKKO ET AL.
cumventing cognitive limitations has implications for improving the quality of life for patients and their families and may prove most useful early in the course of the disorder. The choice of medication for patient management may also be affected, as some drugs, known to affect cognitive functioning adversely, could be avoided in patients with early dementia. Lastly, early detection is of utmost concern if pharmacological treatment for dementia is to be offered when it is most likely to be of benefit. Neuropathological and neurochemical changes evident in some forms of dementia (e.g., Alzheimer disease) may be present before dementia is clearly expressed and early intervention is of great importance if progression of the disease is to be prevented or delayed. In order for a diagnosis of dementia to be made according to the presently available criteria (DSM-III-R: American Psychiatric Association, 1980; ICD-9: World Health Organization, 1978; NINCDS-ADRDA: McKhann et al., 1984), memory impairment must be present together with at least some other areas of cognitive impairment. Memory impairment is central to the diagnosis of dementia and is considered to be one of the earliest manifestations. The Cued Recall procedure for memory assessment described by Buschke in 1984, and Tuokko and Crockett in 1989, coordinates encoding and retrieval of new information for effective cued recall by using a search procedure to control encoding. Without the use of such a procedure, memory deficits found may be due to inefficient use of strategies, impaired attention, reduced processing capacity or impairment of other cognitive processes that limit learning and memory (Grober & Buschke, 1987). Only by ensuring that appropriate processing has been carried out, can observed deficits be attributed solely to memory. The purpose of the present study was to examine the utility of the enhanced cued recall paradigm described by Buschke in 1984, for predicting subsequent dementia in a group of patients who did not meet the criteria for dementia at initial assessment. It was hypothesized that patients who would subsequently meet the criteria for dementia would show worse performances on this measure of genuine memory at initial assessment than patients who remained “not demented” at reassessment. It was further expected that the patients who went on to dement would show a deterioration in their memory functioning over time whereas this was not expected for those who remained “not demented”. The performance of the patients subsequently diagnosed with dementia on this measure at reassessment was not expected to differ from that seen in other patients with a diagnosis of mild dementia. Finally, the sensitivity and specificity of this measure as a predictor of dementia was evaluated in relation to the prevalence rate for the development of dementia after an initial diagnosis of “not demented” in our clinic. METHOD Subjects Subjects in this study were referred by community physicians to the Clinic for Alzheimer Disease and Related Disorders at University Hospital-UBC Site in Vancouver. Canada, an
Downloaded by [McGill University Library] at 13:21 09 February 2015
EARLY IDENTIFICATION
873
outpatient diagnostic service. Referrals were typically made when a patient or a family member reported a change in memory functioning. Each subject underwent thorough physical, psychiatric, and neurological examinations, a psychological assessment, and a social work interview. The information derived from each evaluation (e.g., physical, psychiatric, neurological, psychological and social work) was specific to that discipline, and was collated for the purpose of diagnosing dementia. Thus, all team members contributed to the diagnostic process. However, determinations of presence and degree of impairment were not based on individual test scores (e.g.. Mini-Mental State) but rather on a group consensus making use of history and daily observation information obtained from collateral informants as well as the observations of individual team members. The Functional Rating Scale (FRS: Tuokko & Crockett, 1989; 1991; Crockett, Tuokko, Parks, & Koch, 1989) was used as a vehicle for summarizing the agreed-upon levels of patient functioning. The FRS is a multidimensional scale which permits differentiation in levels of functioning from healthy (1) to severely impaired (5) on eight dimensions: memory; social/ occupationkommunity functioning; home and hobbies; problem solving; personal care; affect; language and orientation. In keeping with the standard criteria for diagnosing dementia (e.g., DSM-111-R, ICD-9/10), dementia was diagnosed in the presence of: (1) at least mild memory impairment, (2) at least mild impairment in some area of daily functioning, (3) at least mild impairment in some other area of cognitive functioning, (4) memory impairment present for at least six months, ( 5 ) no clouding of consciousness. The NINCDSADRDA criteria (McKhann et al., 1984) were applied when a diagnosis of dementia was made to determine research classification regarding Alzheimer disease (i.e., unlikely, possible, probable). Three groups of subjects were selected for this study; a group of subjects who did not meet criteria for dementia over time (i.e., ‘no change’ group); a group of subjects who initially did not meet criteria for dementia but, on reassessment, met dementia criteria (Le., ‘change’ group); and a group of subjects diagnosed with mild dementia at initial assessment (i.e., ‘mild dementia’ group). The latter group was included to determine whether or not the performance of the ‘change’ group on reassessment resembled that seen in other mildly demented individuals. Forty-five subjects did not meet the criteria for dementia (DSM-III-R, ICD-9/10, NINCDS-ADRDA), a determined by the multidisciplinary team, on initial assessment and were reassessed 12-18 months later. At reassessment, no change in diagnostic status was seen for 27 of the subjects and 18 were diagnosed with possible or probable Alzheimer disease (AD) in accordance with the NINCDS-ADRDA criteria (McKhann et al., 1984). The “mild dementia” group was comprised of 117 subjects diagnosed as having possible or probable-AD at initial assessment whose average rating on the Functional Rating Scale was 3.0 (Mild) or less. The means and standard deviations for age and education and the gender distribution for each of these groups are shown in Table 1. Analyses performed to determine whether the ‘change’ and ‘no change’ groups differed on these variables yielded nonsignificant results (age: t(1,43) = 1.27, p < .210; education: r(1,43) = 1.42,p< ,162; gender: chisquare= .8437,df= l , p < .35). The ‘change’ group, that is, subjects who subsequently met the criteria for dementia did not differ from the“milddementia”group onthese variables (age: ~(1,133)= .08,p< .934; education: t(1,133) = -.69, p < ,493); gender: chi square = ,19089, df = 1, p < ,662).
Procedure All subjects were administered the Cued Recall procedure for memory assessment described by Buschke (1984) and modified by Tuokko and Crockett (1989). This procedure involved having subjects scan an array of 12 pictures of common objects (e.g., bed, gun). The subject was given a semantic category cue (e.g., piece of furniture) and asked to point to and name that item. This was done for each of the 12 items. After the array had been presented, scanned and named, the subject was asked to count backward from 100 for 60 s
874
H. TUOKKO ET AL.
Table 1. Age, Education and Gender Distributions for the “Change”, “No Change” and “Mild Dementia” Groups.
N AGE -M
-SD
CHANGE
NO CHANGE
MILD DEMENTIA
18
27
117
70.6 7.1
67.4 8.8
71.5 8.2
11.2 3.5
12.6 3.5
11.5 5.4
8 10
17 10
42 75
Downloaded by [McGill University Library] at 13:21 09 February 2015
EDUCATION
-M -SD GENDER
-M -F
as a distraction task. Subsequently, the subject was asked to engage in free recall of as many of the items as possible. After a period of 20 s during which no items were recalled, semantic category cues were used to prompt recall of those items not previously recalled. Any items not recalled during either free or cued recall conditions were presented again as part of the entire array, and the subject was asked to locate and name only those items not recalled under either free or cued recall conditions. If all items were recalled under free or cued recall conditions, the pictures were not presented again and the subject proceeded to the second recall trial. On the second trial, the subject was asked, once again, to recall as many of the items (i.e.. 12) as possible. As in the Fist recall trial. those items not recalled under the free recall condition were prompted with semantic category cues. A search of the 12 pictures by semantic category was, once again, undertaken for items not recalled under either recall condition. In instances where a subject misidentified an item (e.g.. pointed to bus in response to semantic cue “something for carrying”), the subject was asked to continue the search for another example of that category and given positive feedback for selecting the correct item. If an item was misnamed, the subject was asked to provide another name for that item and given positive feedback for selecting the correct name. If the subject was unable to provide the correct name, the examiner provided the correct response and had the subject locate and name the item to ensure subject was able to process this information. A third recall trial was conducted in the same fashion. A remote recall trial including the free and cued recall conditions described for trials 1, 2 and 3 was administered after the subject engaged in nonmemory tasks for 15 minutes. Three scores were derived from this procedure: (1) a retrieval score -the number of items retrieved without cuing over the three learning trials (Free Recall Trial 1+2+3; maximum = 36); (2) an acquisition score - the number of items acquired over the three trials (Free + Cued Recall Trial 1+2+3; maximum = 36); and (3) a retention score - the total number of items recalled on the Delayed Recall trial (Free Recall + Cued Recall Trial 4; maximum = 12).
Analyses T tests, both paired and unpaired as appropriate, were used to evaluate the differences between the groups on the three memory measures. An evaluation of the statistical power
875
EARLY IDENTIFICATION
or the ability of the measure to have an 80% likelihood of detecting a statistically significant change at the .05 level was carried out (Cohen, 1988). The effect size, or difference in mean performance between the groups, was expressed in terms of standard deviations. Effect sizes of 1.74, 1.78 and 1.32 standard deviations were calculated for the retrieval, acquisition and retention variables, respectively. This suggests that, despite the relatively small sample sizes for the “change” and “no change” groups, the magnitude of the effects observed was sufficient to maintain adequate power.
Downloaded by [McGill University Library] at 13:21 09 February 2015
RESULTS The means and standard deviations of the memory variables obtained at initial ( T l) and repeat (T2) assessment for the “no change” and “change” groups are presented in Table 2. The “no change” and “change” groups were compared on retrieval, acquisition and retention using t-tests at T1. Significant differences between the groups were found on all three variables (retrieval-t(1,43) = 3.26, p < .002; acquisitiont (1,43) = 3.03, p < .004; retention-t(1,43) = 2.41, p < .020) with the “no change” group showing higher initial performances. The “change” group showed significant deterioration betweenT1 and T2 on all three measures (retrieval-t(1,17) = 3.64, p < .002; acquisition-t(l,l7) = 3.78, p < .002; retention-t(1,17) = 2.85. p < .011) whereas the “no change” group did not (retrieval-t(1,26) = -.65, p < S25; acquisition-t(1,26) = -.68, p < S05;retention-t(1,26) = .OO, p = 1.00). When performance on the memory variables at T2 for the “change” group was compared with the initial performance for the “mild dementia” group, no significant differences were found (retrieval-t(1,133)= .62, p < S34; acquisition-t(1,133) = .42,p < .673; retention-t(1,133) = -1.14, p < .256). The contribution of the Cued Recall procedure as a predictor of dementia was assessed by determining the sensitivity, specificity and predictive value of each variable. Sensitivity and specificity were calculated using cut-off scores two standard deviations below the mean derived from a group of 66 community
Table 2. Means and Standard Deviations (SD) on Average Retrieval, Acquisition and Retention at Initial (Tl) and Repeat (T2) Assessment. ~
CHANGE
NO CHANGE
MILD DEMENTIA
-Tl - T2
1 . 5 (1.9)
9.2 (1.6) 9.4 (1.9)
5.4 (1.9)
6.0 (2.0)
Acquisition - T1
8.9 (1.6) 7.3 (2.5)
10.2 (1.3) 10.3 (1.4)
6.8 (1.9)
- T2 - T1 - T2
1 1 . 1 (1.4) 8.0 (4.5)
11.8 (0.6) 11.8 (0.7)
8.5 (4.4)
Retrieval
Retention
876
H. TUOKKO ET AL.
Table 3. Sensitivity, Specificity and Predictive Values for Retrieval, Acquisition and Retention. CUT-’ OFF MEASURE SCORES SENSITIVITY SPECIFICITY _
_
~
~
~
Retrieval
8.4
72%
77%
62%
84%
Acquisition
9.6
72
74
59
84
11.6
39
85
57
73
Retention Downloaded by [McGill University Library] at 13:21 09 February 2015
~
PREDICTIVE VALUESb FOR THE PRESENCE OF DISEASE NO DISEASE
Note: a 2 standard deviations below the mean derived from a group of 66 community dwelling seniors who were functioning independently without difficulty based on a prevalence rate of 34% for the development of dementia in our Clinic sample
dwelling seniors who were functioning independently without difficulty. Positive and negative predictive values for each variable were obtained according to the principles of Bayes Theorem (Branconnier, Cole, Spera, & Devitt, 1982; Satz, Van Gorp, Soper, & Mitrushina, 1989). Calculations were based on a prevalence rate, in our clinic population, of 34% for the development of dementia after an initial diagnosis of “not demented”. The value of each variable as a predictor of dementia is described in Table 3. This indicates that retrieval was the best a priori detector of dementia. The positive predictive value of 62% suggests accurate prediction for the development of a dementia at a rate considerably higher than the estimated prevalence rate in our Clinic sample (i.e., 34%). Thus, by using the test result, an accurate prediction for the development of a dementia would be achieved 62% of the time whereas without the test (i.e., base rate prediction alone), only 34% would be accurately predicted. Similarly, the negative prediction value (i.e., prediction of who will not go on to become demented) was substantially higher than the estimated rate in our clinic sample (i.e., 84 vs. 66%).
DISCUSSION Our finding, that scores derived from the Cued Recall procedure for memory assessment were useful for predicting the development of a dementia, is similar to that reported by Fuld, Blau, Aronson, & Dickson, (1987) and Katzman et al., (1989). Fuld et al. followed a group of normal elderly longitudinally and found another measure of memory functioning (Fuld Object Memory Test) to be related to the presence of senile plaques at autopsy. Katzman et al. followed 434 presumably nondemented aged volunteers over five years. Fifty-six (an incidence of 3.53 per
Downloaded by [McGill University Library] at 13:21 09 February 2015
EARLY IDENTIFICATION
877
100 person years at risk) developed a progressive dementia. The best predictor of the development of AD was a poor score (5-8 errors) on the Information-MemoryConcentration Test (Blessed et al., 1968). The 16% of the cohort with scores in this range developed AD at a rate of over 12% per year. For each of these measures, examinations of their usefulness for predicting dementia in new samples of individuals (i.e., cross-validation) would be of great value. Storandt and Hill (1989) compared individuals with very mild or questionable dementia, mild dementia and healthy aged participants on a variety of psychometric measures. Levels of severity of dementia were defined by Clinical Dementia Rating (CDR) scores of .5, 1 , and 0, respectively. The Wechsler Memory ScaleLogical Memory subtest made the largest contribution to the discrimination between the mildly demented and healthy aged groups. Unfortunately, there was extensive overlap between performances of the questionable group with both the healthy and mildly demented groups, and, although it was assumed that the questionable group were very mild ADS, longitudinal findings were not presented. A companion article by Rubin, Morris, Grant, & Vendegna, (1989) reported that 11/16 subjects with CDR scores of .5 progressed to definite dementia but no significant differences at initial assessment in performance were identified. Since the measures used in these studies assessed global memory functioning without controlling for encoding, it is possible that they lacked the sensitivity to identify differences. The use of a more sensitive tool, such as enhanced cued recall, appears necessary to identify very early decrements in functioning. Our study differed from these others in the type of memory assessment procedure and in some other notable ways. All of our subjects were initially referred for dementia evaluation. Our prevalence rate was determined on the basis of experience in our Clinic which indicated that 21/61 nondemented patients at initial assessment went on to meet the criteria for dementia within a 12-18 month time interval. This prevalence rate will differ markedly as a function of the kind of setting in which evaluations are conducted (i.e., selection bias), hence the prediction values for the measure would differ as well. Another related issue is the enormous variability in rate of deterioration in dementias. It is possible that members of our “no change” group, may, indeed, be in the early stages of a dementing disorder but will not go on the meet the criteria for some time. This may affect the determination of the prevalence rates fur deterioration as well as the composition of the groups. Of interest is the incidental observation that subjects, including those with mild dementia, performed relatively well on the delayed recall trial. This appears inconsistent with other reports of rapid forgetting on delayed memory measures in dementia of the Alzheimer type (e.g. Butters, et al., 1988) but may have been reflective of differences in the procedures used (i.e., Wechsler Memory Scale versus enhanced cued recall). Since the retention score used in this study reflected the total number of items recalled (i.e., Free + Cued Recall) on the delayed recall trial, it remained possible that a decline in Free Recall between the third learning trial and the delayed recall trial may have occurred and may have been a sensitive
Downloaded by [McGill University Library] at 13:21 09 February 2015
878
H. TUOKKO ET AL.
indicator of early dementia. To assess this possibility, a percent savings measure (i.e., [Free Recall delayed uial/Free Recall Trial 31 x 100) was calculated for each subject. No difference between the groups (i.e., ‘change’, ‘no change’) was found in terms of mean percent savings (t(1,43) = -.91, p < .365). Eight of the 27 (29.6%) of the subjects in the ‘no change’ group recalled fewer items on the delayed recall trial than they did on the third learning trial (i.e., less than 100% savings) in comparison to 5 of the 18 (27.7%) in the ‘change’ group. These findings support the contention that the enhanced cued recall procedure accounts for the discrepancy in findings between our study and others. Our finding that the retrieval variable yielded the highest predictive values is in keeping with our previous work using the Cued Recall procedure in discriminating between normal and memory impaired elderly (Tuokko & Crockett. 1989). However, Grober and Buschke (1987), using a 16-item version of this procedure, have found acquisition to be a better measure for identifying dementia, particularly early in the course of the disease. These differences in findings may be due to a ceiling effect for our 12-item measure. Grober and Buschke (1987) reported that demented patients were often able to demonstrate acquisition of 12 items through effective cued recall, but less than the lower bound of 15 items attained by normal elderly. Additional longitudinal studies of individuals who become demented over time are needed to clarify these issues. With such studies it will be possible to address more fully the natural history of early dementia and which specific measures prove to be the most reliable indicators of dementia.
REFERENCES American Psychiatric Association (1980). Diagnostic and statistical manual of mental disorders (3rd. ed.) Washington, DC: Author. Branconnier, R.J., Cole, J.O., Spera, K.F., & Devitt. D.R. (1982). Recall and recognition as diagnostic indices of malignant memory loss in senile dementia: A Bayesian analysis. Experimental Aging Research, 8 , 189-193. Buschke, H. (1984). Cued recall in amnesia. Journal of Clinical Neuropsychology, 6,433440. Butters, N., Salmon, D.P., Cullum, C.M., Cairns, P., Troster, A.I., & Jacobs, D. (1988). Differentiation of amnesic and demented patients with the Wechsler Memory ScaleRevised. Special Issue: Initial validity studies of the Wechsler Memory Scale-Revised. The Clinical Neuropsychologist, 2 , 133-148. Cohen, J. (1988). Statistical power analysis for the behavioral sciences. Hillsdale, NJ: Lawrence Erlbaum Associates. Crockett, D., Tuokko, H., Koch. W. & Parks, R. (1989). The assessment of everyday functioning using the Present Functioning Questionnaire and the Functional Rating Scale in elderly samples. Clinical Gerontologist. 8.3-25. Fuld, P.A., Blau, A.D., Aronson, M.K., L Dickson, D. (1987). Clinicopathological correlations suggestive of a preclinical phase of Alzheimer type dementia. Journal of Clinical and Experimental Neuropsychology, 9,39 (Abstract). Grober, E., & Buschke, H. (1987). Genuine memory deficits in dementia. Developmental Neuropsychology, 3, 13-36. Katzrnan, R., Aronson, M., Fuld, P., Kawas, C., Brown, T., Morgenstern, H.. Frishman,
Downloaded by [McGill University Library] at 13:21 09 February 2015
EARLY IDENTIFICATION
879
W., Gidez, L., Eder, H., & Ooi, W.L. (1989). Development of dementing illnesses in an 80- year-old volunteer cohort. Annals of Neurology, 25,317-324. McKhann, G., Drachman, D., Folstein, M.. Katzman, R., Price, D. & Stadlan, E.M.(1984). Clinical diagnosis of Alzheimer’s disease: Report of the NINCDS-ADRDA work group under the auspices of the Department of Health and Human Services Task Force on Alzheimer’s Disease. Neurology, 34, 939-944. Mortimer, J.A., Schulman, L.M., & French, L.R. (1981). Epidemiology of dementing illnesses. In J.A. Mortimer & L.M. Schulman (Eds.), The epidemiology of dementia (pp. 3-23). New York: Oxford University Press. Rubin, E.H., Morns, J.C., Grant, E.A. & Vendegna, T. (1989). Very mild senile dementia of the Alzheimer type. I. Clinical Assessment. Archives of Neurology, 46, 379-382. Satz, P., Van Gorp, W.G., Soper, H.V. & Mitrushina, M. (1987). A WAIS-R marker for dementia of the Alzheimer type? An empirical and statistical induction test. Journal of Clinical and Experimental Neuropsychology, 9,767-774. Storarrdt, M., & Hill, R.D. (1989). Very mild senile dementia of the Alzheimer type. II. Psychometric test performance. Archives of Neurology, 46, 383-386. Tuokko, H., & Crockett, D. (1989). Cued recall and memory disorders in dementia. Journal of Clinical and Experimental Neuropsychology , 11,278-294. Tuokko, H., & Crockett, D. (1991). Assessment of everyday functioning in normal and malignant memory disordered elderly. In D. Tupper & K. Cicerone (Eds.), The neuropsychology of everyday life: Issues in development and rehabilitation (pp. 135182). Boston: Kluwer Academic Publisher. World Health Organization (1978). Mental Disorders: Glossary and guide to their classification in accordance with the Ninth Revision of the International Classification of Diseases. Geneva: Author.
Journal of Clinical and Experimental Neuropsychology Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/ncen19
Cued Recall and Early Identification of Dementia H. Tuokko Beattie
a b
a
a
, R. Vernon-wilkinson , J. Weir & B. L.
a
a
Clinic for Alzheimer Disease and Related Disorders , University Hospital-UBC Site , Vancouver b
Clinic for Alzheimer Disease and Related Disorders , University Hospital-UBC Site , 2211 Wesbrook Mall, Vancouver, B.C., Canada , V6T 2B5 Published online: 04 Jan 2008.
To cite this article: H. Tuokko , R. Vernon-wilkinson , J. Weir & B. L. Beattie (1991) Cued Recall and Early Identification of Dementia, Journal of Clinical and Experimental Neuropsychology, 13:6, 871-879, DOI: 10.1080/01688639108405104 To link to this article: http://dx.doi.org/10.1080/01688639108405104
PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sublicensing, systematic supply, or distribution in any form to anyone is expressly
Downloaded by [McGill University Library] at 13:21 09 February 2015
forbidden. Terms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions
Journal of Clinical and Experimental Neuropsychology 1991, Vol. 13, NO. 6, pp. 871-879
0 168-8634/91/1306-0871$3 .MI Q Swets & Zeitlinger
Cued Recall and Early Identification of Dementia*
Downloaded by [McGill University Library] at 13:21 09 February 2015
H. Tuokko, R. Vernon-Wilkinson, J. Weir, and B.L. Beattie Clinic for Alzheimer Disease and Related Disorders, University Hospital-UBC Site, Vancouver
ABSTRACT The early detection of dementia carries implications for clinical management for patients and their families and is of utmost concern if an effective pharmacological treatment is to be found. The utility of an enhanced cued recall paradigm for predicting dementia in a group of elderly subjects was examined. Forty-five subjects referred for clinical evaluation who did not meet the criteria for dementia at initial assessment were reassessed 12-18 months later. Eighteen of these subjects were diagnosed as having possible or probable Alzheimer Disease at reassessment and the diagnostic status of 27 remained unchanged. At initial assessment the ‘change’ group performed more poorly than the ‘no change’ group on measures of retrieval, acquisition and retention derived from the cued recall procedure. As would be expected, the performance of the ‘no change’ group remained stable over time whereas the performance of the ‘change’ group deteriorated, resembling the performance of a group of subjects with mild possible or probable Alzheimer disease. When the sensitivity and specificity of the memory variables were examined, the retrieval measure (i.e., free recall) appeared most useful as an early predictor of dementia. Continued longitudinal evaluations of subjects with questionable dementia are needed to address more fully the natural history of early memory changes associated with dementia.
It h a s been estimated that between 5 and 20 % of persons over the age o f 65 years show evidence of dementing disorders (Mortimer, Schulman, & French, 1981) with devastating costs t o the persons afflicted, their families a n d society. W i t h the ever increasing proportion of elderly in the population, the early diagnosis o f dementia has become an important objective. Early detection carries implications f o r clinical management of patients and their families i n order t o plan appropriately for management of personal affairs and future care. The establishment of behavioral or educational programs designed to teach methods f o r cir-
* The authors would like to thank the patients who took part in this research study and their families. We gratefully acknowledge support for Janine Weir from the Alzheimer Society of British Columbia. Requests for reprints can be directed to Holly Tuokko, Ph.D.. Clinic for Alzheimer Disease and Related Disorders, University Hospital-UBC Site, 221 1 Wesbrook Mall, Vancouver, B.C. Canada, V6T 2BS. Accepted for publication: February 5, 1991.
Downloaded by [McGill University Library] at 13:21 09 February 2015
872
H. TUOKKO ET AL.
cumventing cognitive limitations has implications for improving the quality of life for patients and their families and may prove most useful early in the course of the disorder. The choice of medication for patient management may also be affected, as some drugs, known to affect cognitive functioning adversely, could be avoided in patients with early dementia. Lastly, early detection is of utmost concern if pharmacological treatment for dementia is to be offered when it is most likely to be of benefit. Neuropathological and neurochemical changes evident in some forms of dementia (e.g., Alzheimer disease) may be present before dementia is clearly expressed and early intervention is of great importance if progression of the disease is to be prevented or delayed. In order for a diagnosis of dementia to be made according to the presently available criteria (DSM-III-R: American Psychiatric Association, 1980; ICD-9: World Health Organization, 1978; NINCDS-ADRDA: McKhann et al., 1984), memory impairment must be present together with at least some other areas of cognitive impairment. Memory impairment is central to the diagnosis of dementia and is considered to be one of the earliest manifestations. The Cued Recall procedure for memory assessment described by Buschke in 1984, and Tuokko and Crockett in 1989, coordinates encoding and retrieval of new information for effective cued recall by using a search procedure to control encoding. Without the use of such a procedure, memory deficits found may be due to inefficient use of strategies, impaired attention, reduced processing capacity or impairment of other cognitive processes that limit learning and memory (Grober & Buschke, 1987). Only by ensuring that appropriate processing has been carried out, can observed deficits be attributed solely to memory. The purpose of the present study was to examine the utility of the enhanced cued recall paradigm described by Buschke in 1984, for predicting subsequent dementia in a group of patients who did not meet the criteria for dementia at initial assessment. It was hypothesized that patients who would subsequently meet the criteria for dementia would show worse performances on this measure of genuine memory at initial assessment than patients who remained “not demented” at reassessment. It was further expected that the patients who went on to dement would show a deterioration in their memory functioning over time whereas this was not expected for those who remained “not demented”. The performance of the patients subsequently diagnosed with dementia on this measure at reassessment was not expected to differ from that seen in other patients with a diagnosis of mild dementia. Finally, the sensitivity and specificity of this measure as a predictor of dementia was evaluated in relation to the prevalence rate for the development of dementia after an initial diagnosis of “not demented” in our clinic. METHOD Subjects Subjects in this study were referred by community physicians to the Clinic for Alzheimer Disease and Related Disorders at University Hospital-UBC Site in Vancouver. Canada, an
Downloaded by [McGill University Library] at 13:21 09 February 2015
EARLY IDENTIFICATION
873
outpatient diagnostic service. Referrals were typically made when a patient or a family member reported a change in memory functioning. Each subject underwent thorough physical, psychiatric, and neurological examinations, a psychological assessment, and a social work interview. The information derived from each evaluation (e.g., physical, psychiatric, neurological, psychological and social work) was specific to that discipline, and was collated for the purpose of diagnosing dementia. Thus, all team members contributed to the diagnostic process. However, determinations of presence and degree of impairment were not based on individual test scores (e.g.. Mini-Mental State) but rather on a group consensus making use of history and daily observation information obtained from collateral informants as well as the observations of individual team members. The Functional Rating Scale (FRS: Tuokko & Crockett, 1989; 1991; Crockett, Tuokko, Parks, & Koch, 1989) was used as a vehicle for summarizing the agreed-upon levels of patient functioning. The FRS is a multidimensional scale which permits differentiation in levels of functioning from healthy (1) to severely impaired (5) on eight dimensions: memory; social/ occupationkommunity functioning; home and hobbies; problem solving; personal care; affect; language and orientation. In keeping with the standard criteria for diagnosing dementia (e.g., DSM-111-R, ICD-9/10), dementia was diagnosed in the presence of: (1) at least mild memory impairment, (2) at least mild impairment in some area of daily functioning, (3) at least mild impairment in some other area of cognitive functioning, (4) memory impairment present for at least six months, ( 5 ) no clouding of consciousness. The NINCDSADRDA criteria (McKhann et al., 1984) were applied when a diagnosis of dementia was made to determine research classification regarding Alzheimer disease (i.e., unlikely, possible, probable). Three groups of subjects were selected for this study; a group of subjects who did not meet criteria for dementia over time (i.e., ‘no change’ group); a group of subjects who initially did not meet criteria for dementia but, on reassessment, met dementia criteria (Le., ‘change’ group); and a group of subjects diagnosed with mild dementia at initial assessment (i.e., ‘mild dementia’ group). The latter group was included to determine whether or not the performance of the ‘change’ group on reassessment resembled that seen in other mildly demented individuals. Forty-five subjects did not meet the criteria for dementia (DSM-III-R, ICD-9/10, NINCDS-ADRDA), a determined by the multidisciplinary team, on initial assessment and were reassessed 12-18 months later. At reassessment, no change in diagnostic status was seen for 27 of the subjects and 18 were diagnosed with possible or probable Alzheimer disease (AD) in accordance with the NINCDS-ADRDA criteria (McKhann et al., 1984). The “mild dementia” group was comprised of 117 subjects diagnosed as having possible or probable-AD at initial assessment whose average rating on the Functional Rating Scale was 3.0 (Mild) or less. The means and standard deviations for age and education and the gender distribution for each of these groups are shown in Table 1. Analyses performed to determine whether the ‘change’ and ‘no change’ groups differed on these variables yielded nonsignificant results (age: t(1,43) = 1.27, p < .210; education: r(1,43) = 1.42,p< ,162; gender: chisquare= .8437,df= l , p < .35). The ‘change’ group, that is, subjects who subsequently met the criteria for dementia did not differ from the“milddementia”group onthese variables (age: ~(1,133)= .08,p< .934; education: t(1,133) = -.69, p < ,493); gender: chi square = ,19089, df = 1, p < ,662).
Procedure All subjects were administered the Cued Recall procedure for memory assessment described by Buschke (1984) and modified by Tuokko and Crockett (1989). This procedure involved having subjects scan an array of 12 pictures of common objects (e.g., bed, gun). The subject was given a semantic category cue (e.g., piece of furniture) and asked to point to and name that item. This was done for each of the 12 items. After the array had been presented, scanned and named, the subject was asked to count backward from 100 for 60 s
874
H. TUOKKO ET AL.
Table 1. Age, Education and Gender Distributions for the “Change”, “No Change” and “Mild Dementia” Groups.
N AGE -M
-SD
CHANGE
NO CHANGE
MILD DEMENTIA
18
27
117
70.6 7.1
67.4 8.8
71.5 8.2
11.2 3.5
12.6 3.5
11.5 5.4
8 10
17 10
42 75
Downloaded by [McGill University Library] at 13:21 09 February 2015
EDUCATION
-M -SD GENDER
-M -F
as a distraction task. Subsequently, the subject was asked to engage in free recall of as many of the items as possible. After a period of 20 s during which no items were recalled, semantic category cues were used to prompt recall of those items not previously recalled. Any items not recalled during either free or cued recall conditions were presented again as part of the entire array, and the subject was asked to locate and name only those items not recalled under either free or cued recall conditions. If all items were recalled under free or cued recall conditions, the pictures were not presented again and the subject proceeded to the second recall trial. On the second trial, the subject was asked, once again, to recall as many of the items (i.e.. 12) as possible. As in the Fist recall trial. those items not recalled under the free recall condition were prompted with semantic category cues. A search of the 12 pictures by semantic category was, once again, undertaken for items not recalled under either recall condition. In instances where a subject misidentified an item (e.g.. pointed to bus in response to semantic cue “something for carrying”), the subject was asked to continue the search for another example of that category and given positive feedback for selecting the correct item. If an item was misnamed, the subject was asked to provide another name for that item and given positive feedback for selecting the correct name. If the subject was unable to provide the correct name, the examiner provided the correct response and had the subject locate and name the item to ensure subject was able to process this information. A third recall trial was conducted in the same fashion. A remote recall trial including the free and cued recall conditions described for trials 1, 2 and 3 was administered after the subject engaged in nonmemory tasks for 15 minutes. Three scores were derived from this procedure: (1) a retrieval score -the number of items retrieved without cuing over the three learning trials (Free Recall Trial 1+2+3; maximum = 36); (2) an acquisition score - the number of items acquired over the three trials (Free + Cued Recall Trial 1+2+3; maximum = 36); and (3) a retention score - the total number of items recalled on the Delayed Recall trial (Free Recall + Cued Recall Trial 4; maximum = 12).
Analyses T tests, both paired and unpaired as appropriate, were used to evaluate the differences between the groups on the three memory measures. An evaluation of the statistical power
875
EARLY IDENTIFICATION
or the ability of the measure to have an 80% likelihood of detecting a statistically significant change at the .05 level was carried out (Cohen, 1988). The effect size, or difference in mean performance between the groups, was expressed in terms of standard deviations. Effect sizes of 1.74, 1.78 and 1.32 standard deviations were calculated for the retrieval, acquisition and retention variables, respectively. This suggests that, despite the relatively small sample sizes for the “change” and “no change” groups, the magnitude of the effects observed was sufficient to maintain adequate power.
Downloaded by [McGill University Library] at 13:21 09 February 2015
RESULTS The means and standard deviations of the memory variables obtained at initial ( T l) and repeat (T2) assessment for the “no change” and “change” groups are presented in Table 2. The “no change” and “change” groups were compared on retrieval, acquisition and retention using t-tests at T1. Significant differences between the groups were found on all three variables (retrieval-t(1,43) = 3.26, p < .002; acquisitiont (1,43) = 3.03, p < .004; retention-t(1,43) = 2.41, p < .020) with the “no change” group showing higher initial performances. The “change” group showed significant deterioration betweenT1 and T2 on all three measures (retrieval-t(1,17) = 3.64, p < .002; acquisition-t(l,l7) = 3.78, p < .002; retention-t(1,17) = 2.85. p < .011) whereas the “no change” group did not (retrieval-t(1,26) = -.65, p < S25; acquisition-t(1,26) = -.68, p < S05;retention-t(1,26) = .OO, p = 1.00). When performance on the memory variables at T2 for the “change” group was compared with the initial performance for the “mild dementia” group, no significant differences were found (retrieval-t(1,133)= .62, p < S34; acquisition-t(1,133) = .42,p < .673; retention-t(1,133) = -1.14, p < .256). The contribution of the Cued Recall procedure as a predictor of dementia was assessed by determining the sensitivity, specificity and predictive value of each variable. Sensitivity and specificity were calculated using cut-off scores two standard deviations below the mean derived from a group of 66 community
Table 2. Means and Standard Deviations (SD) on Average Retrieval, Acquisition and Retention at Initial (Tl) and Repeat (T2) Assessment. ~
CHANGE
NO CHANGE
MILD DEMENTIA
-Tl - T2
1 . 5 (1.9)
9.2 (1.6) 9.4 (1.9)
5.4 (1.9)
6.0 (2.0)
Acquisition - T1
8.9 (1.6) 7.3 (2.5)
10.2 (1.3) 10.3 (1.4)
6.8 (1.9)
- T2 - T1 - T2
1 1 . 1 (1.4) 8.0 (4.5)
11.8 (0.6) 11.8 (0.7)
8.5 (4.4)
Retrieval
Retention
876
H. TUOKKO ET AL.
Table 3. Sensitivity, Specificity and Predictive Values for Retrieval, Acquisition and Retention. CUT-’ OFF MEASURE SCORES SENSITIVITY SPECIFICITY _
_
~
~
~
Retrieval
8.4
72%
77%
62%
84%
Acquisition
9.6
72
74
59
84
11.6
39
85
57
73
Retention Downloaded by [McGill University Library] at 13:21 09 February 2015
~
PREDICTIVE VALUESb FOR THE PRESENCE OF DISEASE NO DISEASE
Note: a 2 standard deviations below the mean derived from a group of 66 community dwelling seniors who were functioning independently without difficulty based on a prevalence rate of 34% for the development of dementia in our Clinic sample
dwelling seniors who were functioning independently without difficulty. Positive and negative predictive values for each variable were obtained according to the principles of Bayes Theorem (Branconnier, Cole, Spera, & Devitt, 1982; Satz, Van Gorp, Soper, & Mitrushina, 1989). Calculations were based on a prevalence rate, in our clinic population, of 34% for the development of dementia after an initial diagnosis of “not demented”. The value of each variable as a predictor of dementia is described in Table 3. This indicates that retrieval was the best a priori detector of dementia. The positive predictive value of 62% suggests accurate prediction for the development of a dementia at a rate considerably higher than the estimated prevalence rate in our Clinic sample (i.e., 34%). Thus, by using the test result, an accurate prediction for the development of a dementia would be achieved 62% of the time whereas without the test (i.e., base rate prediction alone), only 34% would be accurately predicted. Similarly, the negative prediction value (i.e., prediction of who will not go on to become demented) was substantially higher than the estimated rate in our clinic sample (i.e., 84 vs. 66%).
DISCUSSION Our finding, that scores derived from the Cued Recall procedure for memory assessment were useful for predicting the development of a dementia, is similar to that reported by Fuld, Blau, Aronson, & Dickson, (1987) and Katzman et al., (1989). Fuld et al. followed a group of normal elderly longitudinally and found another measure of memory functioning (Fuld Object Memory Test) to be related to the presence of senile plaques at autopsy. Katzman et al. followed 434 presumably nondemented aged volunteers over five years. Fifty-six (an incidence of 3.53 per
Downloaded by [McGill University Library] at 13:21 09 February 2015
EARLY IDENTIFICATION
877
100 person years at risk) developed a progressive dementia. The best predictor of the development of AD was a poor score (5-8 errors) on the Information-MemoryConcentration Test (Blessed et al., 1968). The 16% of the cohort with scores in this range developed AD at a rate of over 12% per year. For each of these measures, examinations of their usefulness for predicting dementia in new samples of individuals (i.e., cross-validation) would be of great value. Storandt and Hill (1989) compared individuals with very mild or questionable dementia, mild dementia and healthy aged participants on a variety of psychometric measures. Levels of severity of dementia were defined by Clinical Dementia Rating (CDR) scores of .5, 1 , and 0, respectively. The Wechsler Memory ScaleLogical Memory subtest made the largest contribution to the discrimination between the mildly demented and healthy aged groups. Unfortunately, there was extensive overlap between performances of the questionable group with both the healthy and mildly demented groups, and, although it was assumed that the questionable group were very mild ADS, longitudinal findings were not presented. A companion article by Rubin, Morris, Grant, & Vendegna, (1989) reported that 11/16 subjects with CDR scores of .5 progressed to definite dementia but no significant differences at initial assessment in performance were identified. Since the measures used in these studies assessed global memory functioning without controlling for encoding, it is possible that they lacked the sensitivity to identify differences. The use of a more sensitive tool, such as enhanced cued recall, appears necessary to identify very early decrements in functioning. Our study differed from these others in the type of memory assessment procedure and in some other notable ways. All of our subjects were initially referred for dementia evaluation. Our prevalence rate was determined on the basis of experience in our Clinic which indicated that 21/61 nondemented patients at initial assessment went on to meet the criteria for dementia within a 12-18 month time interval. This prevalence rate will differ markedly as a function of the kind of setting in which evaluations are conducted (i.e., selection bias), hence the prediction values for the measure would differ as well. Another related issue is the enormous variability in rate of deterioration in dementias. It is possible that members of our “no change” group, may, indeed, be in the early stages of a dementing disorder but will not go on the meet the criteria for some time. This may affect the determination of the prevalence rates fur deterioration as well as the composition of the groups. Of interest is the incidental observation that subjects, including those with mild dementia, performed relatively well on the delayed recall trial. This appears inconsistent with other reports of rapid forgetting on delayed memory measures in dementia of the Alzheimer type (e.g. Butters, et al., 1988) but may have been reflective of differences in the procedures used (i.e., Wechsler Memory Scale versus enhanced cued recall). Since the retention score used in this study reflected the total number of items recalled (i.e., Free + Cued Recall) on the delayed recall trial, it remained possible that a decline in Free Recall between the third learning trial and the delayed recall trial may have occurred and may have been a sensitive
Downloaded by [McGill University Library] at 13:21 09 February 2015
878
H. TUOKKO ET AL.
indicator of early dementia. To assess this possibility, a percent savings measure (i.e., [Free Recall delayed uial/Free Recall Trial 31 x 100) was calculated for each subject. No difference between the groups (i.e., ‘change’, ‘no change’) was found in terms of mean percent savings (t(1,43) = -.91, p < .365). Eight of the 27 (29.6%) of the subjects in the ‘no change’ group recalled fewer items on the delayed recall trial than they did on the third learning trial (i.e., less than 100% savings) in comparison to 5 of the 18 (27.7%) in the ‘change’ group. These findings support the contention that the enhanced cued recall procedure accounts for the discrepancy in findings between our study and others. Our finding that the retrieval variable yielded the highest predictive values is in keeping with our previous work using the Cued Recall procedure in discriminating between normal and memory impaired elderly (Tuokko & Crockett. 1989). However, Grober and Buschke (1987), using a 16-item version of this procedure, have found acquisition to be a better measure for identifying dementia, particularly early in the course of the disease. These differences in findings may be due to a ceiling effect for our 12-item measure. Grober and Buschke (1987) reported that demented patients were often able to demonstrate acquisition of 12 items through effective cued recall, but less than the lower bound of 15 items attained by normal elderly. Additional longitudinal studies of individuals who become demented over time are needed to clarify these issues. With such studies it will be possible to address more fully the natural history of early dementia and which specific measures prove to be the most reliable indicators of dementia.
REFERENCES American Psychiatric Association (1980). Diagnostic and statistical manual of mental disorders (3rd. ed.) Washington, DC: Author. Branconnier, R.J., Cole, J.O., Spera, K.F., & Devitt. D.R. (1982). Recall and recognition as diagnostic indices of malignant memory loss in senile dementia: A Bayesian analysis. Experimental Aging Research, 8 , 189-193. Buschke, H. (1984). Cued recall in amnesia. Journal of Clinical Neuropsychology, 6,433440. Butters, N., Salmon, D.P., Cullum, C.M., Cairns, P., Troster, A.I., & Jacobs, D. (1988). Differentiation of amnesic and demented patients with the Wechsler Memory ScaleRevised. Special Issue: Initial validity studies of the Wechsler Memory Scale-Revised. The Clinical Neuropsychologist, 2 , 133-148. Cohen, J. (1988). Statistical power analysis for the behavioral sciences. Hillsdale, NJ: Lawrence Erlbaum Associates. Crockett, D., Tuokko, H., Koch. W. & Parks, R. (1989). The assessment of everyday functioning using the Present Functioning Questionnaire and the Functional Rating Scale in elderly samples. Clinical Gerontologist. 8.3-25. Fuld, P.A., Blau, A.D., Aronson, M.K., L Dickson, D. (1987). Clinicopathological correlations suggestive of a preclinical phase of Alzheimer type dementia. Journal of Clinical and Experimental Neuropsychology, 9,39 (Abstract). Grober, E., & Buschke, H. (1987). Genuine memory deficits in dementia. Developmental Neuropsychology, 3, 13-36. Katzrnan, R., Aronson, M., Fuld, P., Kawas, C., Brown, T., Morgenstern, H.. Frishman,
Downloaded by [McGill University Library] at 13:21 09 February 2015
EARLY IDENTIFICATION
879
W., Gidez, L., Eder, H., & Ooi, W.L. (1989). Development of dementing illnesses in an 80- year-old volunteer cohort. Annals of Neurology, 25,317-324. McKhann, G., Drachman, D., Folstein, M.. Katzman, R., Price, D. & Stadlan, E.M.(1984). Clinical diagnosis of Alzheimer’s disease: Report of the NINCDS-ADRDA work group under the auspices of the Department of Health and Human Services Task Force on Alzheimer’s Disease. Neurology, 34, 939-944. Mortimer, J.A., Schulman, L.M., & French, L.R. (1981). Epidemiology of dementing illnesses. In J.A. Mortimer & L.M. Schulman (Eds.), The epidemiology of dementia (pp. 3-23). New York: Oxford University Press. Rubin, E.H., Morns, J.C., Grant, E.A. & Vendegna, T. (1989). Very mild senile dementia of the Alzheimer type. I. Clinical Assessment. Archives of Neurology, 46, 379-382. Satz, P., Van Gorp, W.G., Soper, H.V. & Mitrushina, M. (1987). A WAIS-R marker for dementia of the Alzheimer type? An empirical and statistical induction test. Journal of Clinical and Experimental Neuropsychology, 9,767-774. Storarrdt, M., & Hill, R.D. (1989). Very mild senile dementia of the Alzheimer type. II. Psychometric test performance. Archives of Neurology, 46, 383-386. Tuokko, H., & Crockett, D. (1989). Cued recall and memory disorders in dementia. Journal of Clinical and Experimental Neuropsychology , 11,278-294. Tuokko, H., & Crockett, D. (1991). Assessment of everyday functioning in normal and malignant memory disordered elderly. In D. Tupper & K. Cicerone (Eds.), The neuropsychology of everyday life: Issues in development and rehabilitation (pp. 135182). Boston: Kluwer Academic Publisher. World Health Organization (1978). Mental Disorders: Glossary and guide to their classification in accordance with the Ninth Revision of the International Classification of Diseases. Geneva: Author.
