22. Jinbu Y, Sato S , Nakao T, Nakao M, Tsukita S, Tsukita S , Ishikawa H. The role of ankyrin in shape and deformability change of human erythrocyte ghosts. Biochim Biophys Acta 1984;773:237-245. 23. Kay MMB, Bosman G , Lawrence C. Neurologic disease with acanthocytosis is associated with erythrocyte membrane transport and band 3 abnormalities.Clin b s 1989;37:547 A. 24. Bennett V.The membrane skeletonof human erythrocytesand its
implication for more complex cells. Annu Rev Biochem 1985;54304. 25. Tang TK, Qia Z, Let0 T, Marchesi VT,Benz EY Jr. Heterogeneity of mRNA and proteinproductsarisingfrom the protein 4.1 gene in erythroidand non-erythroidtissues. J Cell Biol1990;110617-624. 26. Winklemann JC, Chang JG, Tse WT, Scarpa AL, Marchesi VT, Forget BG. Full length sequence of the DNA for human erythroid beta-spectrin.J Biol Chem 1990;26511827-11832.
Mild cognitive impairment in the elderly: Predictors of dementia Charles Flicker, PhD; Steven H. Ferris, PhD; and Barry Reisberg, MD
Article abstract-We conducted full diagnostic evaluations, including a comprehensive cognitive assessment battery, of a group of 32 elderly subjects with a clinically identified mild cognitive impairment and a group of 32 age-matched and educationmatched normal subjects. The mildly impaired subjects performed significantly more poorly than the controls on tests of recent memory, remote memory, language function, concept formation, and visuospatial praxis. Follow-up evaluations of cognitive status 2 years later revealed clinically detectable cognitive decline relative to baseline in 23 (72%)of the mildly impaired subjects. Several of the objective psychological tests accurately discriminated at baseline between the decliners and nondecliners in the mildly impaired group. Among the 20 mildly impaired subjects with no complicating conditions, 16 exhibited cognitive deterioration between baseline and follow-up. These results suggest that most elderly subjects with mild cognitive deficits, as determined by clinical evaluation and objective psychological testing, will manifest the progressive mental deterioration characteristic of dementia and that psychometric predictors can be used to distinguish between benign and more significant underlying disorders in mildly impaired elderly subjects. NEUROLOGY 1991;41:1006-1009
Cognitive evaluation in the elderly is often conducted to determine whether the subject is likely to undergo significant cognitive deterioration. Clinical interviews with elderly subjects commonly yield objective evidence of mild cognitive impairment. Although such subjects can be identified as psychologically abnormal for their age group, the information derived from a clinical interview is usually not considered sufficient to classify such mildly impaired subjects as demented or nondemented.1*2In the assessment of such cases, psychometric tests that might distinguish between “benign senescent f~rgetfulness”~ and more significant conditions would be of great value. The identification of valid prognostic indices requires longitudinal study of mildly impaired subjects. The present investigation was a longitudinal study of a group of relatively unselected memory clinic patients who were classified as mildly impaired and possibly demented, and a group of age-matched and educationmatched normal controls. This paper compares the baseline and follow-up cognitive status of the two groups before and after a 2-year interval and examines
the potential of objective psychological test measures to serve as predictors of follow-up cognitive status. Methods. Subjects. The study population consisted of 32 elderly normal subjects and 32 mildly impaired subjects. The group with mild cognitive impairment consisted of 32 consecutive subjects screened at the Aging and Dementia Research Center of NYU Medical Center who received a Global Deterioration Scale (GDS)2rating of 3 and were administered the neuropsychological test battery. The 32 age- and education-matched controls were also selected from participants in longitudinal studies of aging at the Aging and Dementia Research Center. Informed consent was obtained from all subjects. The mean age of the elderly normal subjects was 70.9 1.3 years, and the mean age of the mildly impaired subjects was 71.3 f 1.4 years. The mean years of education of the elderly normals was 13.3 0.6 years, and the mean years of education of the mildly impaired subjects was 13.1 0.6 years. On the basis of the screening procedures described below, subjects were excluded from participation in the study if they had the following:a history of seizures, schizophrenia, mania, depression, alcoholism, drug abuse, head trauma with loss of consciousness for more than 1 hour, or any serious neurologic disorder; clinically significant focal neurologic deficits; or a
*
*
From the Aging and Dementia Research Center, New York University Medical Center, New York, NY. Supported in part by Grant #MH 40410 from the National Institute of Mental Health and by Grant #AG 03051 from the National Institute on Aging. Received July 30,1990. Accepted for publication in final form December 27,1990. L Address correspondenceand reprint requeststo Dr. Charles Flicker,Dept. Psychiatry (HN314), New York University Medical Center, 550 First Avenue, New York, NY 10016. 1006 NEUROLOGY 41 July 1991
severe cardiac, pulmonary, cerebrovascular (eg, stroke), metabolic, or hematologic disorder. Procedures. The subject evaluation procedure consisted of a medical evaluation, a neurologic examination (conductedby a neurologist),a psychiatric evaluation (conducted or directly supervised by a psychiatrist), CT of the head, and a cognitive evaluation. The medical evaluation included a physical examination, medical history, serum electrolytes, serum glucose, serum enzymes, corpuscular and differential blood counts, a urinalysis, and an ECG. The psychiatric evaluation included administration of the GDS? a 7-point scale that rates the subject's global cognitive status according to clearly defined criteria based on subjective complaints and objective evidence. Subjects with a GDS rating of l or 2 are functionally and cognitively unimpaired, the two groups differing in that subjects rated GDS 2 have subjective complaints of memory loss. Subjects with a GDS rating between 3 and 7 exhibit objective evidence of cognitive deficits, and the degree of cognitive impairment displayed by the five groups increases with GDS from mild, to moderate, moderately severe, severe, and very severe. Subjects with a GDS rating of 3, classified as mildly impaired but not demented, have exhibited at least two of the following symptoms: (1)getting lost when traveling to an unfamiliar location, (2) decline in work performance apparent to coworkers, (3) word- and name-finding deficit apparent to intimates, (4) relatively little retention of material read in a passage or book, ( 5 ) decreased facility remembering the names of newly introducedpeople, (6) losing or misplacing an object of value, or (7) a concentration deficit apparent upon clinical testing. Memory deficit in these subjects is demonstrable but not superficially apparent. By contrast, subjects with a GDS of 4 exhibit readily apparent deficits which manifest as (1)decreased knowledge of current and recent events, (2) a concentration deficit on serial subtraction tasks; (3) decreased ability to travel, handle finances, or perform complex tasks; and (4) possibly deficient memory of their own personal history. The neuropsychologicalevaluation, describedin detail previously: was designed to provide a comprehensive,high-resolution profile of the differences in cognitive abilities among elderly subjects. The assessment battery included tests of immediate memory, verbal recall, visuospatial recall, visual recognition memory, remote memory, language function, concept formation, visuospatial praxis, visuoperceptualfunction, and psychomotor speed. Follow-up GDS was obtained approximately 2 years after the initial evaluation. The actual mean follow-up interval was 2.21 +. 0.04 years in the normal elderly and 2.11 +. 0.09 years in the mildly impaired subjects. Subjects who refused or were unable to return to the clinic were evaluated by telephone. The telephone evaluation consisted of telephone interviews with the subject, the subject's physician, and any relative, caregiver, health aide, social worker, or medical personnel who could provide additional or confirmatory information about the subject.
Results. Some of the cross-sectional cognitive test results presented here have been previously reported for an overlapping group of subject^.^-^ As can be seen in table 1,subjects with clinically identified mild cognitive impairment performed significantly more poorly than elderly normals on tests of recent memory, remote memory, language function (most tests), concept formation, and visuospatial praxis. Between-group differences were not significant on tests of immediate memory, visuoperceptual function, and some of the tests of language function and psychomotor speed.
Table 1. Cognitive test performance of elderly normal subjects and mildly impaired subjects Elderly Mildly normals impaired (mean f SEM) (mean f SEM) Immediate Memory Digit Span Delayed Spatial Recall (0-W Verbal Recall Paragraphs Shopping List Viuospatial Recall Delayed Spatial Recall
6.5 f 0.2 11.9 ? 0.1
6.0 ? 0.3 11.6 ? 0.3
8.3 & 0.6 38.4 f 1.7
3.9 f 0.4. 20.4 ? 1.6'
0.5
5.3 f 0.9*
9.0 ? 0.4
4.3 f 0.5*
19.1 f 1.5 1.62 f 0.14
15.0 f 1.1' 1.10 f 0.14*
15.1 f 0.6
11.2 f 0.8.
64.2 f 2.0 14.5 f 0.7 16.4 f 0.8 18.7 f 0.6 18.8 & 0.4 60.7 f 0.6 11.3 k 0.2
52.5 f 3.0' 9.7 f 0.7* 11.8 & 1.2* 16.7 f 0.7* 16.4 f 0.7* 58.6 f 1.2 10.6 f 0.4
34.2 & 2.2
27.2
47.0 f 2.3
30.0 f 3.1'
27.4 f 1.0
24.4 f 1.4
71.6 f 3.1
63.0 f 2.9
408 & 36 512 f 32
561 f 68* 543 ? 42
9.1
&
(30-WC)
Misplaced Objects Visual Recognition Memory Visual Recognition Span Facial Recognition Remote Memory Remote Memory Questionnaire ~guage
Vocabulary Category Retrieval Object Naming Object Function Recall Object Name Recognition Object Function Recognition Object Identification Concept Formation Object Sorting Visuoapatial Praxis Digit Symbol visuoperceptual Function Road Map Psychomotor Speed Finger Tapping Driving Test Release Travel
&
1.9*
* Significantly merent from group mean of elderly normals,p < 0.05, based on two-tailed t test.
Follow-up clinical evaluation of the mildly impaired group revealed an increase in GDS rating in 23 of 32 subjects. Among the normal elderly subjects, only four exhibited GDS ratings of 3 or more at follow-up. Although complete diagnostic evaluations were not conducted a t follow-up on all subjects, the available information from the 23 decliners in the mildly impaired group was consistent with diagnoses of probable Alzheimer's disease (n = 16), vascular or mixed dementia (n = 5 ) , brain granuloma (n = l),and leukemia with CNS involvement (n = 1).Examination of the baseline cognitive test data revealed that four of the objective psychological tests yielded significantly lower scores in the mildly impaired subjects who subsequently declined in comparison to the nondecliners (tabIe 2). These four tests were the shopping list task,1° a test of verbal recall; the misplaced objects task," a test of visuospatial recall; and two of the language tests-the object function recognition task7 and the object identification task.7 Classification analyses (table 2) demonstrated that these tests have high sensitivity and specificity in discriminating between mildly impaired subjects who will and will not undergo a significant decline in cognitive status July 1991NEUROLOGY 41 1007
Table 2. Predictors of decline in subjects with mild cognitive impairment
I Test Shopping List Misplaced Objects Object Function Recognition Object Identification
Nondecliners (N = 9) (mean k SEM)
28.5 rt 6.2 k 61.4 rt 11.6 rt
Decliners (N = 23) (mean k SEMI
t
Sensitivity
Specificity
18.8 k 1.6 3.7 ? 0.5 56.6 ? 1.6 9.9 k 0.6
2.43 2.13 2.48 2.51
90.0% 70.0% 85.7% 57.1%
94.7% 93.3% 100.0% 100.0%
3.4 1.2 0.5 0.2
over a 2-year follow-up interval. These tests elicited lower scores from decliners regardless of whether they had complicating medical conditions.
Discussion. The results of this study suggest that objective psychological tests can be used to discriminate between mildly impaired elderly subjects who are likely to undergo significant cognitive deterioration over a 2year interval and those whose prognosis is relatively benign. Four of the tests in the assessment battery administered to these subjects exhibited high sensitivity and specificity in distinguishing between decliners and nondecliners. It would thus appear that the combination of a careful, structured clinical interview and the appropriate neuropsychological tests can potentially provide elderly subjects of questionablecognitive status with an accurate prognosis of their future level of functioning. Berg et aln conducted baseline and 1-year follow-up evaluations of 43 mildly demented subjects. After 1year, a clinicallysignificantloss of cognitive status was detectable in half of the subjects followed. Baseline performance of the decliners was significantly worse than the nondecliners on tests of immediate memory (digit span), recent memory, remote memory, visuospatialpraxis (digit symbol), language (object naming), verbal fluency, and psychomotor speed. The broader array of psychometric predictors identified by Berg et al,I2 in comparison to the present study, is perhaps attributable to the greater severity of their subjects’ cognitive dysfunction. Their patients’ disorder was classified as more advanced than “questionable dementia,” which is characterized by mild deficits in recent memory only. For example, on the digit symbol test, which was administered at baseline in both studies, the mildly impaired subjects of the present study scored over 50% higher than the mild dementia patients of the Berg et a1 study. In the present study, only certain tests of verbal recall, visuospatial recall, and language proved to be useful predictors of cognitive decline in mildly impaired subjects. That tests of recent visuospatial and verbal recall should be most sensitive to incipient dementia is not surprising, since these deficits are most reliably associated with the histopathologically confirmed diagnosis of Alzheimer’s disease.13J4The shopping list task in particular, a list recall task using the selective reminding procedure,15 has been established as reliably sensitive to cross-sectional differences, longitudinal changes, and pharmacologic effects in this population. The two language tests, in which subjects pointed out 1008 NEUROLOGY 41 July 1991
items after being supplied with a name or a category, were similar to receptive vocabulary tests and did not require word production. It is revealing that these two tests, although they had perfect specificity in identifying the decliners,were the only language tasks that were insensitive to the cross-sectional differences between the subjects with mild cognitive impairment and the normal subjects. Mildly impaired subjects with abnormally low scores on these two tasks invariably underwent further cognitive deterioration. Previous crosssectional studies have revealed significant impairments on these tests in more advanced stages of dementia? The results of the present study also strongly suggest that objective evidenceof two or more of the characteristic symptoms of possible incipient dementia,as obtained in a clinical interview, can be a reliable predictor of future cognitive deterioration in the elderly. Of 20 mildly impaired subjects with no apparent medical basis for their cognitive difficulties, 16 were classified as demented 2 years later. Similarly, Rubin et all6 have recently reported that 10 of 15 subjects with “questionabledementia” exhibited significant cognitive decline or a diagnosis of Alzheimer’s disease when followed up over a 34month interval. Postmortem studies of patients in this questionable dementia stage reveal the histopathologic signs characteristic of Alzheimer’s disease.”’Like these patients,18 the mildly impaired subjects in our study performed significantly more poorly than normal elderly subjects on most of the neuropsychological tests with established sensitivity to Alzheimer’s disease. The high probability of their continued cognitive decline implies a correspondingly high frequency of underlying pathologic processes. The data clearly argue that such borderline cases should not be classified as normal. Although further research is needed, it would appear that the combined information derived from in-depth clinical evaluations and objective psychological tests can perhaps yield very accurate prognostic information, and possibly also lead to improved diagnosis of elderly subjects with mild cognitive impairment. Acknowledgments The authors thank Dr. Jacob Cohen for consultation on statistical analyses and Grete Greene for assistance in manuscript preparation.
References 1. Hughes CP, Berg L, Danziger WL, Cohen LA, Martin RL. A new clinical scale for the staging of dementia. Br J Psychiatry
1982;140566-572. 2. Reisberg B, Ferris SH, de Leon MJ, Crook T. The global deterioration scale for assessment of primary degenerative dementia. Am J Psychiatry 1982;1391136-1139. 3. Kral VA. Senescent forgetfulness: benign and malignant. Can Med Assoc J 1962;86:257-260. 4. Flicker C, Serby M, Ferns SH. Scopolamine effects on memory, language, visuospatial praxis, and psychomotor speed. Psychopharmacology (Berlin) 1990;100:243-250. 5. Flicker C, Ferris SH, Crook T, Bartus RT. The effects of agingand dementia on concept formation as measured on an object-sorting task. Dev Neuropsychol1986;2:65-72. 6. Flicker C, Ferris SH, Crook T, Bartus RT. A visual recognition memory test for the assessment of cognitive function in aging and dementia. Exp Aging Res 1987;13:127-132. 7. Flicker C, Ferris SH, Crook T , Bartus RT. Implications of memory and language dysfunction in the naming deficit of senile dementia. Brain Lang 1987;31:187-200. 8. Flicker C, Ferris SH, Crook T, Reisberg B, Bartus RT. Equivalent spatial rotation deficits in normal aging and Alzheimer’sdisease.J Clin Exp Neuropsychol1988;10387-399. 9. Flicker C, Ferris SH, Crook T, Bartus RT. Age differences in the vulnerability of facial recognition memory to proactive interference. Exp Aging Res 1989;15:189-194.
10.McCarthy M, Ferris SH, Clark E, Crook T. Acquisition and retention of categorized material in normal aging and senile dementia. Exp Aging Res 1981;7:127-135. 11. Crook T, Ferris SH, McCarthy M. The misplaced-objects task a brief test for memory dysfunction in the aged. J Am Geriatr SOC 1979:27:284-287. 12. Berg L,Danziger WL, Storandt M, et al. Predictive features in mild senile dementia of t h e Alzheimer type. Neurology 198434563-569. 13. Sim A, Sussman I. Alzheimer’s disease: its natural history and differential diagnosis. J Nerv Ment Dis 1962;135:489-499. 14. Neary D, Snowden JS, Bowen DM, et al. Neuropsychological syndromes in presenile dementia due to cerebral atrophy. J Neurol Neurosurg Psychiatry 1986;49:163-174. 15. Buschke H.Selective reminding for analysis of memory and learning. J Verb Learn Verb Behav 1973;12:543-550. 16. Rubin EH,Morris JC, Grant EA, Vendegna T. Very mild senile dementia of the Alzheimer type. I. Clinical assessment. Arch Neurol 1989;46379-382. 17. Morris JC, McKeel DW, Storandt M, et al. Very mild Alzheimer’s disease: informant-based clinical, psychometric, and pathologic distinction from normal aging. Neurology 1991;41:469-478. 18. Storandt M, H ill RD. Very mild senile dementiaof the Alzheimer type. II. Psychometric test performance.Arch New1 1989;46:383-386.
July 1991 NEUROLOGY 41 lo09
Mild cognitive impairment in the elderly: Predictors of dementia Charles Flicker, Steven H. Ferris and Barry Reisberg Neurology 1991;41;1006 DOI 10.1212/WNL.41.7.1006 This information is current as of July 1, 1991 Updated Information & Services
including high resolution figures, can be found at: http://www.neurology.org/content/41/7/1006.full.html
Citations
This article has been cited by 52 HighWire-hosted articles: http://www.neurology.org/content/41/7/1006.full.html##otherarticles
Permissions & Licensing
Information about reproducing this article in parts (figures,tables) or in its entirety can be found online at: http://www.neurology.org/misc/about.xhtml#permissions
Reprints
Information about ordering reprints can be found online: http://www.neurology.org/misc/addir.xhtml#reprintsus
Neurology ® is the official journal of the American Academy of Neurology. Published continuously since 1951, it is now a weekly with 48 issues per year. Copyright © 1991 by the American Academy of Neurology. All rights reserved. Print ISSN: 0028-3878. Online ISSN: 1526-632X.
implication for more complex cells. Annu Rev Biochem 1985;54304. 25. Tang TK, Qia Z, Let0 T, Marchesi VT,Benz EY Jr. Heterogeneity of mRNA and proteinproductsarisingfrom the protein 4.1 gene in erythroidand non-erythroidtissues. J Cell Biol1990;110617-624. 26. Winklemann JC, Chang JG, Tse WT, Scarpa AL, Marchesi VT, Forget BG. Full length sequence of the DNA for human erythroid beta-spectrin.J Biol Chem 1990;26511827-11832.
Mild cognitive impairment in the elderly: Predictors of dementia Charles Flicker, PhD; Steven H. Ferris, PhD; and Barry Reisberg, MD
Article abstract-We conducted full diagnostic evaluations, including a comprehensive cognitive assessment battery, of a group of 32 elderly subjects with a clinically identified mild cognitive impairment and a group of 32 age-matched and educationmatched normal subjects. The mildly impaired subjects performed significantly more poorly than the controls on tests of recent memory, remote memory, language function, concept formation, and visuospatial praxis. Follow-up evaluations of cognitive status 2 years later revealed clinically detectable cognitive decline relative to baseline in 23 (72%)of the mildly impaired subjects. Several of the objective psychological tests accurately discriminated at baseline between the decliners and nondecliners in the mildly impaired group. Among the 20 mildly impaired subjects with no complicating conditions, 16 exhibited cognitive deterioration between baseline and follow-up. These results suggest that most elderly subjects with mild cognitive deficits, as determined by clinical evaluation and objective psychological testing, will manifest the progressive mental deterioration characteristic of dementia and that psychometric predictors can be used to distinguish between benign and more significant underlying disorders in mildly impaired elderly subjects. NEUROLOGY 1991;41:1006-1009
Cognitive evaluation in the elderly is often conducted to determine whether the subject is likely to undergo significant cognitive deterioration. Clinical interviews with elderly subjects commonly yield objective evidence of mild cognitive impairment. Although such subjects can be identified as psychologically abnormal for their age group, the information derived from a clinical interview is usually not considered sufficient to classify such mildly impaired subjects as demented or nondemented.1*2In the assessment of such cases, psychometric tests that might distinguish between “benign senescent f~rgetfulness”~ and more significant conditions would be of great value. The identification of valid prognostic indices requires longitudinal study of mildly impaired subjects. The present investigation was a longitudinal study of a group of relatively unselected memory clinic patients who were classified as mildly impaired and possibly demented, and a group of age-matched and educationmatched normal controls. This paper compares the baseline and follow-up cognitive status of the two groups before and after a 2-year interval and examines
the potential of objective psychological test measures to serve as predictors of follow-up cognitive status. Methods. Subjects. The study population consisted of 32 elderly normal subjects and 32 mildly impaired subjects. The group with mild cognitive impairment consisted of 32 consecutive subjects screened at the Aging and Dementia Research Center of NYU Medical Center who received a Global Deterioration Scale (GDS)2rating of 3 and were administered the neuropsychological test battery. The 32 age- and education-matched controls were also selected from participants in longitudinal studies of aging at the Aging and Dementia Research Center. Informed consent was obtained from all subjects. The mean age of the elderly normal subjects was 70.9 1.3 years, and the mean age of the mildly impaired subjects was 71.3 f 1.4 years. The mean years of education of the elderly normals was 13.3 0.6 years, and the mean years of education of the mildly impaired subjects was 13.1 0.6 years. On the basis of the screening procedures described below, subjects were excluded from participation in the study if they had the following:a history of seizures, schizophrenia, mania, depression, alcoholism, drug abuse, head trauma with loss of consciousness for more than 1 hour, or any serious neurologic disorder; clinically significant focal neurologic deficits; or a
*
*
From the Aging and Dementia Research Center, New York University Medical Center, New York, NY. Supported in part by Grant #MH 40410 from the National Institute of Mental Health and by Grant #AG 03051 from the National Institute on Aging. Received July 30,1990. Accepted for publication in final form December 27,1990. L Address correspondenceand reprint requeststo Dr. Charles Flicker,Dept. Psychiatry (HN314), New York University Medical Center, 550 First Avenue, New York, NY 10016. 1006 NEUROLOGY 41 July 1991
severe cardiac, pulmonary, cerebrovascular (eg, stroke), metabolic, or hematologic disorder. Procedures. The subject evaluation procedure consisted of a medical evaluation, a neurologic examination (conductedby a neurologist),a psychiatric evaluation (conducted or directly supervised by a psychiatrist), CT of the head, and a cognitive evaluation. The medical evaluation included a physical examination, medical history, serum electrolytes, serum glucose, serum enzymes, corpuscular and differential blood counts, a urinalysis, and an ECG. The psychiatric evaluation included administration of the GDS? a 7-point scale that rates the subject's global cognitive status according to clearly defined criteria based on subjective complaints and objective evidence. Subjects with a GDS rating of l or 2 are functionally and cognitively unimpaired, the two groups differing in that subjects rated GDS 2 have subjective complaints of memory loss. Subjects with a GDS rating between 3 and 7 exhibit objective evidence of cognitive deficits, and the degree of cognitive impairment displayed by the five groups increases with GDS from mild, to moderate, moderately severe, severe, and very severe. Subjects with a GDS rating of 3, classified as mildly impaired but not demented, have exhibited at least two of the following symptoms: (1)getting lost when traveling to an unfamiliar location, (2) decline in work performance apparent to coworkers, (3) word- and name-finding deficit apparent to intimates, (4) relatively little retention of material read in a passage or book, ( 5 ) decreased facility remembering the names of newly introducedpeople, (6) losing or misplacing an object of value, or (7) a concentration deficit apparent upon clinical testing. Memory deficit in these subjects is demonstrable but not superficially apparent. By contrast, subjects with a GDS of 4 exhibit readily apparent deficits which manifest as (1)decreased knowledge of current and recent events, (2) a concentration deficit on serial subtraction tasks; (3) decreased ability to travel, handle finances, or perform complex tasks; and (4) possibly deficient memory of their own personal history. The neuropsychologicalevaluation, describedin detail previously: was designed to provide a comprehensive,high-resolution profile of the differences in cognitive abilities among elderly subjects. The assessment battery included tests of immediate memory, verbal recall, visuospatial recall, visual recognition memory, remote memory, language function, concept formation, visuospatial praxis, visuoperceptualfunction, and psychomotor speed. Follow-up GDS was obtained approximately 2 years after the initial evaluation. The actual mean follow-up interval was 2.21 +. 0.04 years in the normal elderly and 2.11 +. 0.09 years in the mildly impaired subjects. Subjects who refused or were unable to return to the clinic were evaluated by telephone. The telephone evaluation consisted of telephone interviews with the subject, the subject's physician, and any relative, caregiver, health aide, social worker, or medical personnel who could provide additional or confirmatory information about the subject.
Results. Some of the cross-sectional cognitive test results presented here have been previously reported for an overlapping group of subject^.^-^ As can be seen in table 1,subjects with clinically identified mild cognitive impairment performed significantly more poorly than elderly normals on tests of recent memory, remote memory, language function (most tests), concept formation, and visuospatial praxis. Between-group differences were not significant on tests of immediate memory, visuoperceptual function, and some of the tests of language function and psychomotor speed.
Table 1. Cognitive test performance of elderly normal subjects and mildly impaired subjects Elderly Mildly normals impaired (mean f SEM) (mean f SEM) Immediate Memory Digit Span Delayed Spatial Recall (0-W Verbal Recall Paragraphs Shopping List Viuospatial Recall Delayed Spatial Recall
6.5 f 0.2 11.9 ? 0.1
6.0 ? 0.3 11.6 ? 0.3
8.3 & 0.6 38.4 f 1.7
3.9 f 0.4. 20.4 ? 1.6'
0.5
5.3 f 0.9*
9.0 ? 0.4
4.3 f 0.5*
19.1 f 1.5 1.62 f 0.14
15.0 f 1.1' 1.10 f 0.14*
15.1 f 0.6
11.2 f 0.8.
64.2 f 2.0 14.5 f 0.7 16.4 f 0.8 18.7 f 0.6 18.8 & 0.4 60.7 f 0.6 11.3 k 0.2
52.5 f 3.0' 9.7 f 0.7* 11.8 & 1.2* 16.7 f 0.7* 16.4 f 0.7* 58.6 f 1.2 10.6 f 0.4
34.2 & 2.2
27.2
47.0 f 2.3
30.0 f 3.1'
27.4 f 1.0
24.4 f 1.4
71.6 f 3.1
63.0 f 2.9
408 & 36 512 f 32
561 f 68* 543 ? 42
9.1
&
(30-WC)
Misplaced Objects Visual Recognition Memory Visual Recognition Span Facial Recognition Remote Memory Remote Memory Questionnaire ~guage
Vocabulary Category Retrieval Object Naming Object Function Recall Object Name Recognition Object Function Recognition Object Identification Concept Formation Object Sorting Visuoapatial Praxis Digit Symbol visuoperceptual Function Road Map Psychomotor Speed Finger Tapping Driving Test Release Travel
&
1.9*
* Significantly merent from group mean of elderly normals,p < 0.05, based on two-tailed t test.
Follow-up clinical evaluation of the mildly impaired group revealed an increase in GDS rating in 23 of 32 subjects. Among the normal elderly subjects, only four exhibited GDS ratings of 3 or more at follow-up. Although complete diagnostic evaluations were not conducted a t follow-up on all subjects, the available information from the 23 decliners in the mildly impaired group was consistent with diagnoses of probable Alzheimer's disease (n = 16), vascular or mixed dementia (n = 5 ) , brain granuloma (n = l),and leukemia with CNS involvement (n = 1).Examination of the baseline cognitive test data revealed that four of the objective psychological tests yielded significantly lower scores in the mildly impaired subjects who subsequently declined in comparison to the nondecliners (tabIe 2). These four tests were the shopping list task,1° a test of verbal recall; the misplaced objects task," a test of visuospatial recall; and two of the language tests-the object function recognition task7 and the object identification task.7 Classification analyses (table 2) demonstrated that these tests have high sensitivity and specificity in discriminating between mildly impaired subjects who will and will not undergo a significant decline in cognitive status July 1991NEUROLOGY 41 1007
Table 2. Predictors of decline in subjects with mild cognitive impairment
I Test Shopping List Misplaced Objects Object Function Recognition Object Identification
Nondecliners (N = 9) (mean k SEM)
28.5 rt 6.2 k 61.4 rt 11.6 rt
Decliners (N = 23) (mean k SEMI
t
Sensitivity
Specificity
18.8 k 1.6 3.7 ? 0.5 56.6 ? 1.6 9.9 k 0.6
2.43 2.13 2.48 2.51
90.0% 70.0% 85.7% 57.1%
94.7% 93.3% 100.0% 100.0%
3.4 1.2 0.5 0.2
over a 2-year follow-up interval. These tests elicited lower scores from decliners regardless of whether they had complicating medical conditions.
Discussion. The results of this study suggest that objective psychological tests can be used to discriminate between mildly impaired elderly subjects who are likely to undergo significant cognitive deterioration over a 2year interval and those whose prognosis is relatively benign. Four of the tests in the assessment battery administered to these subjects exhibited high sensitivity and specificity in distinguishing between decliners and nondecliners. It would thus appear that the combination of a careful, structured clinical interview and the appropriate neuropsychological tests can potentially provide elderly subjects of questionablecognitive status with an accurate prognosis of their future level of functioning. Berg et aln conducted baseline and 1-year follow-up evaluations of 43 mildly demented subjects. After 1year, a clinicallysignificantloss of cognitive status was detectable in half of the subjects followed. Baseline performance of the decliners was significantly worse than the nondecliners on tests of immediate memory (digit span), recent memory, remote memory, visuospatialpraxis (digit symbol), language (object naming), verbal fluency, and psychomotor speed. The broader array of psychometric predictors identified by Berg et al,I2 in comparison to the present study, is perhaps attributable to the greater severity of their subjects’ cognitive dysfunction. Their patients’ disorder was classified as more advanced than “questionable dementia,” which is characterized by mild deficits in recent memory only. For example, on the digit symbol test, which was administered at baseline in both studies, the mildly impaired subjects of the present study scored over 50% higher than the mild dementia patients of the Berg et a1 study. In the present study, only certain tests of verbal recall, visuospatial recall, and language proved to be useful predictors of cognitive decline in mildly impaired subjects. That tests of recent visuospatial and verbal recall should be most sensitive to incipient dementia is not surprising, since these deficits are most reliably associated with the histopathologically confirmed diagnosis of Alzheimer’s disease.13J4The shopping list task in particular, a list recall task using the selective reminding procedure,15 has been established as reliably sensitive to cross-sectional differences, longitudinal changes, and pharmacologic effects in this population. The two language tests, in which subjects pointed out 1008 NEUROLOGY 41 July 1991
items after being supplied with a name or a category, were similar to receptive vocabulary tests and did not require word production. It is revealing that these two tests, although they had perfect specificity in identifying the decliners,were the only language tasks that were insensitive to the cross-sectional differences between the subjects with mild cognitive impairment and the normal subjects. Mildly impaired subjects with abnormally low scores on these two tasks invariably underwent further cognitive deterioration. Previous crosssectional studies have revealed significant impairments on these tests in more advanced stages of dementia? The results of the present study also strongly suggest that objective evidenceof two or more of the characteristic symptoms of possible incipient dementia,as obtained in a clinical interview, can be a reliable predictor of future cognitive deterioration in the elderly. Of 20 mildly impaired subjects with no apparent medical basis for their cognitive difficulties, 16 were classified as demented 2 years later. Similarly, Rubin et all6 have recently reported that 10 of 15 subjects with “questionabledementia” exhibited significant cognitive decline or a diagnosis of Alzheimer’s disease when followed up over a 34month interval. Postmortem studies of patients in this questionable dementia stage reveal the histopathologic signs characteristic of Alzheimer’s disease.”’Like these patients,18 the mildly impaired subjects in our study performed significantly more poorly than normal elderly subjects on most of the neuropsychological tests with established sensitivity to Alzheimer’s disease. The high probability of their continued cognitive decline implies a correspondingly high frequency of underlying pathologic processes. The data clearly argue that such borderline cases should not be classified as normal. Although further research is needed, it would appear that the combined information derived from in-depth clinical evaluations and objective psychological tests can perhaps yield very accurate prognostic information, and possibly also lead to improved diagnosis of elderly subjects with mild cognitive impairment. Acknowledgments The authors thank Dr. Jacob Cohen for consultation on statistical analyses and Grete Greene for assistance in manuscript preparation.
References 1. Hughes CP, Berg L, Danziger WL, Cohen LA, Martin RL. A new clinical scale for the staging of dementia. Br J Psychiatry
1982;140566-572. 2. Reisberg B, Ferris SH, de Leon MJ, Crook T. The global deterioration scale for assessment of primary degenerative dementia. Am J Psychiatry 1982;1391136-1139. 3. Kral VA. Senescent forgetfulness: benign and malignant. Can Med Assoc J 1962;86:257-260. 4. Flicker C, Serby M, Ferns SH. Scopolamine effects on memory, language, visuospatial praxis, and psychomotor speed. Psychopharmacology (Berlin) 1990;100:243-250. 5. Flicker C, Ferris SH, Crook T, Bartus RT. The effects of agingand dementia on concept formation as measured on an object-sorting task. Dev Neuropsychol1986;2:65-72. 6. Flicker C, Ferris SH, Crook T, Bartus RT. A visual recognition memory test for the assessment of cognitive function in aging and dementia. Exp Aging Res 1987;13:127-132. 7. Flicker C, Ferris SH, Crook T , Bartus RT. Implications of memory and language dysfunction in the naming deficit of senile dementia. Brain Lang 1987;31:187-200. 8. Flicker C, Ferris SH, Crook T, Reisberg B, Bartus RT. Equivalent spatial rotation deficits in normal aging and Alzheimer’sdisease.J Clin Exp Neuropsychol1988;10387-399. 9. Flicker C, Ferris SH, Crook T, Bartus RT. Age differences in the vulnerability of facial recognition memory to proactive interference. Exp Aging Res 1989;15:189-194.
10.McCarthy M, Ferris SH, Clark E, Crook T. Acquisition and retention of categorized material in normal aging and senile dementia. Exp Aging Res 1981;7:127-135. 11. Crook T, Ferris SH, McCarthy M. The misplaced-objects task a brief test for memory dysfunction in the aged. J Am Geriatr SOC 1979:27:284-287. 12. Berg L,Danziger WL, Storandt M, et al. Predictive features in mild senile dementia of t h e Alzheimer type. Neurology 198434563-569. 13. Sim A, Sussman I. Alzheimer’s disease: its natural history and differential diagnosis. J Nerv Ment Dis 1962;135:489-499. 14. Neary D, Snowden JS, Bowen DM, et al. Neuropsychological syndromes in presenile dementia due to cerebral atrophy. J Neurol Neurosurg Psychiatry 1986;49:163-174. 15. Buschke H.Selective reminding for analysis of memory and learning. J Verb Learn Verb Behav 1973;12:543-550. 16. Rubin EH,Morris JC, Grant EA, Vendegna T. Very mild senile dementia of the Alzheimer type. I. Clinical assessment. Arch Neurol 1989;46379-382. 17. Morris JC, McKeel DW, Storandt M, et al. Very mild Alzheimer’s disease: informant-based clinical, psychometric, and pathologic distinction from normal aging. Neurology 1991;41:469-478. 18. Storandt M, H ill RD. Very mild senile dementiaof the Alzheimer type. II. Psychometric test performance.Arch New1 1989;46:383-386.
July 1991 NEUROLOGY 41 lo09
Mild cognitive impairment in the elderly: Predictors of dementia Charles Flicker, Steven H. Ferris and Barry Reisberg Neurology 1991;41;1006 DOI 10.1212/WNL.41.7.1006 This information is current as of July 1, 1991 Updated Information & Services
including high resolution figures, can be found at: http://www.neurology.org/content/41/7/1006.full.html
Citations
This article has been cited by 52 HighWire-hosted articles: http://www.neurology.org/content/41/7/1006.full.html##otherarticles
Permissions & Licensing
Information about reproducing this article in parts (figures,tables) or in its entirety can be found online at: http://www.neurology.org/misc/about.xhtml#permissions
Reprints
Information about ordering reprints can be found online: http://www.neurology.org/misc/addir.xhtml#reprintsus
Neurology ® is the official journal of the American Academy of Neurology. Published continuously since 1951, it is now a weekly with 48 issues per year. Copyright © 1991 by the American Academy of Neurology. All rights reserved. Print ISSN: 0028-3878. Online ISSN: 1526-632X.
