ORIGINAL CONTRIBUTION

A 7 Minute Neurocognitive Screening Battery

Highly Sensitive to Alzheimer’s Disease
Paul R. Solomon, PhD; Aliina Hirschoff; Bridget Kelly; Mahri Relin; Michael Brush;
Richard D. DeVeaux, PhD; William W. Pendlebury, MD

Objective: To determine the validity and reliability of Results: Mean time of administration was 7 minutes 42
a rapidly administered neurocognitive screening bat- seconds. Mean scores for patients with AD and control
tery consisting of 4 brief tests (Enhanced Cued Recall, subjects on all 4 individual tests were significantly dif-
Temporal Orientation, Verbal Fluency, and Clock Draw- ferent (for each, P,.001). When the 4 tests were com-
ing) to distinguish between patients with probable Alz- bined in a logistic regression, the battery had a sensitiv-
heimer’s disease (AD) and healthy control subjects. ity of 100% and a specificity of 100%. A series of 1000
repeated random samples of 30 patients with AD and 30
Subjects: Sixty successive referrals to the Memory Dis- control subjects taken from the overall sample of 60 pa-
orders Clinic at Southwestern Vermont Medical Center, tients with AD and 60 control subjects had a mean sen-
Bennington, who were diagnosed as having probable AD sitivity of 92% and a mean specificity of 96%. The bat-
and 60 community-dwelling volunteers of comparable tery was equally sensitive to patients with mild AD as
age, sex distribution, and education. demonstrated by correctly classifying all 13 patients with
AD using Mini-Mental State Examination scores of 24 or
Design: Interrater and test-retest reliability, inter- higher. Neither age nor education was a statistically sig-
group comparisons between patients with AD and con- nificant factor when added as a covariate. Test-retest re-
trol subjects on the 4 individual tests, and determina- liabilities for individual tests ranged from 0.83 to 0.93.
tion of probability of dementia for patients with AD and Test-retest reliability for the entire battery was 0.91. In-
control subjects using the entire battery of tests. terrater reliability for the entire battery was 0.92.

Setting: Outpatient care. Conclusions: The 7 Minute Screen appears highly sen-
sitive to AD and may be useful in helping to make initial
Main Outcome Measure: Comparison of the prob- distinctions between patients experiencing cognitive
ability of dementia on the 7 Minute Screen with the cri- changes related to the normal aging process and those ex-
terion standard of clinical diagnosis established by periencing cognitive deficits related to dementing disor-
examination and laboratory studies. ders such as AD. It has reasonable interrater and test-
retest reliability, can be administered in a brief period, and
Secondary Outcome Measures: Test-retest and in- requires no clinical judgment and minimal training.
terrater reliability (correlation coefficients), time for ad-
ministration. Arch Neurol. 1998;55:349-355

T
From the Department of
HE INCREASING prevalence of tributing factor to the underdiagnosis of AD
Psychology (Dr Solomon and
Ms Hirschoff), the Program in Alzheimer’s disease (AD)1 is issues surrounding testing of mental sta-
Neuroscience (Dr Solomon and and the emerging treatments tus. Research indicates that primary care
Mss Hirschoff, Kelly, and Relin, for this disease2,3 suggest that physicians either do not use mental status
and Mr Brush), and the thereisanincreasingneedfor examinations (MSEs)9 or use MSEs that lack
Department of Mathematics accurate and easily administered screening sensitivity.6,10,11 Increasingly abbreviated
(Dr DeVeaux), Williams instruments.Ideally,theseinstrumentscould clinic visits may preclude the use of MSEs,
College, Williamstown, Mass; be administered quickly by nurses, physi- even if cognitive deficits are suspected. Fur-
the Departments of Pathology cian’s assistants, or other trained personnel thermore, currently available MSEs were
and Neurology (Dr Pendlebury),
University of Vermont College of
with the goal of distinguishing between cog- not designed to screen for or to diagnose
Medicine, Burlington; and the nitive changes seen in the normal aging pro- AD and should not be expected to possess
Memory Disorders Clinic, cess and cognitive decline consistent with those capabilities.11 For example, the Mini-
Southwestern Vermont Medical AD and other dementing diseases.4 Mental State Examination (MMSE),12 the
Center (Drs Solomon and Primary care physicians do not diag- most widely used MSE, was originally de-
Pendlebury), Bennington. nose AD in routine practice.5-8 One con- veloped to evaluate elderly psychiatric pa-

ARCH NEUROL / VOL 55, MAR 1998

349

©1998 American Medical Association. All rights reserved.

Downloaded From: http://archneur.jamanetwork.com/ by a Tulane University User on 05/12/2015
 SUBJECTS AND METHODS 7 MINUTE SCREEN

The 7 Minute Screen consists of 4 tests representing 4 cog-

HEALTHY CONTROL SUBJECTS nitive areas typically compromised in AD: (1) memory,
(2) verbal fluency, (3) visuospatial and visuoconstruc-
Sixty community-dwelling elderly individuals were re- tion, and (4) orientation for time. Each test (or modifica-
cruited through newspaper advertisements and a local health tion of the test) was selected because previous research
maintenance organization. A medical history was ob- showed that it had a high degree of sensitivity to AD, could
tained from each subject, including current medication use be rapidly administered by personnel with little training,
and incidence of head trauma, stroke, mental illness, men- and could be scored objectively.
tal retardation, or life-threatening illness during the last 5
years. None had a history of psychiatric or neurological dis- ENHANCED CUED RECALL
order. None was taking antidepressant or other psychoac-
tive medications. All claimed to be independent in activi- This test, which was initially described in a longer form by
ties of daily living, including shopping, transportation, and Buschke and colleagues,25 takes advantage of the finding that
managing finances. Cognitive screening was accom- healthy elderly control subjects benefit from mnemonic strat-
plished in all subjects by administering the BIMC. A ran- egies that facilitate the storage and retrieval of information (eg,
dom sample of 30 subjects underwent more extensive neu- reminder cues), whereas patients with AD show significantly
ropsychological testing, including the Wechsler Memory less benefit from these strategies. This test requires the sub-
Scale–Revised (Logical Memory I and II, Visual Reproduc- ject to recall 16 items. The items are presented 4 at a time on
tion I and II) and the MMSE. Test procedures were com- 4 individual cards. While looking at the card, the subject is
pleted after informed written consent had been obtained. asked to identify the picture on the card that best fits with the
semantic cue given by the examiner (eg, examiner: “There is
PATIENTS WITH AD a piece of furniture on the card, what is it?” subject: “A desk”).
When all 4 objects have been successfully identified with the
All patients with AD were seen at the Memory Disorders semantic cue, the card is removed and immediate recall test-
Clinic at Southwestern Vermont Medical Center, Benning- ing is performed. The examiner provides the cue and the sub-
ton (an affiliated clinic of the Massachusetts Alzheimer’s Dis- ject names the picture just observed (eg, “I just showed you
ease Research Center, Boston). The 60 patients represented a piece of furniture, what was it?” “A desk.”). If the subject
consecutive referrals to the clinic and met the National In- misses 1 or more items, all items on the card are presented a
stitute of Neurological Disorders and Stroke–Alzheimer’s Dis- second time. This repetition is intended to produce the same
ease and Related Disorders Association diagnostic criteria for depth of encoding in all subjects. No card is repeated more
AD33 based on (1) neurological, medical, psychiatric, and/or than twice. If a subject makes an error on the second presen-
social examinations; (2) standard laboratory studies; (3) com- tation, the examiner provides the correct response and con-
puted tomographic scans; (4) neuropsychological evalua- tinues to the next card. After all 4 cards are presented, the sub-
tions; and (5) history from a caregiver indicating at least a ject is distracted by reciting the months of the year backward.
1-year history of progressive cognitive decline. Results from The subject is then asked to free recall as many of the pictures
the 7 Minute Screen did not contribute to the diagnosis. as possible. When the subject cannot recall any additional

tients and has been criticized both for its level of sensitiv- We selected a battery of tests based on recent devel-
ity13 and specificity14,15 as well as the influence that opments in cognitive neuropsychology and behavioral neu-
education has on performance.16,17 The Blessed Informa- rology that have begun to elucidate the differences between
tion Memory Concentration Test (BIMC)18 is also widely cognitive changes in the normal aging processs and those
used but has been criticized for failing to sample a num- that accompany AD. There is general agreement that learn-
ber of cognitive functions (eg, language and visuospatial ing and memory tasks are the most useful diagnostic tests
abilities) commonly deficient in people with AD.10,19 Other for the detection of AD.22-24 However, the 3-item recall on
batteries have been shown to be useful in diagnosing the MMSE has proved not to be sensitive enough.11 We chose
AD20 or monitoring the progression of the disease,21 but the Enhanced Cued Recall task25 because it has been shown
each of these is too long and/or requires too much clini- to distinguish between memory loss that accompanies the
cal judgment to be useful as a screening tool in a pri- normal aging process and memory deficits present in pa-
mary care setting. tients with AD.24,25 Orientation for date and time is a highly
To address these issues, we sought to develop a reliable measure of cognitive status, although it lacks sen-
mental status screening instrument that had the follow- sitivity for AD as it is commonly used in clinical practice.
ing characteristics: (1) it could be rapidly administered We used the Benton modification of this test26 to provide
by a technician (since physician time for administering greater sensitivity to AD.27 Also, visuoconstructive deficits
MSEs is greatly limited), (2) it required no clinical judg- are common to patients with AD and constitute a nonver-
ment and little training to use, (3) it took advantage of bal area of cognition. A number of investigators28-30 have
the evolving understanding of the cognitive differences shown that clock drawing is sensitive to AD. In addition,
between AD and the normal aging process, and (4) it verbal fluency has been shown to discriminate between pa-
was capable of reliably distinguishing AD from cogni- tients with AD and healthy elderly control subjects.31,32 The
tive deficit changes associated with the normal aging combination of these measures is brief, and taken together
process. can be administered in less than 10 minutes.

ARCH NEUROL / VOL 55, MAR 1998

350

©1998 American Medical Association. All rights reserved.

Downloaded From: http://archneur.jamanetwork.com/ by a Tulane University User on 05/12/2015
 pictures, the examiner provides the category cues for the re- hands to “twenty to four.” Our standard procedure is to
maining items (cued recall). The score is the total number provide the subject with a pen and blank sheet of paper
of items recalled in both free and cued recall (range, 0-16). and say, “I want you to draw a clock with all the numbers
on it. Make it large.” After this is completed the examiner
CATEGORY FLUENCY states, “Now draw the hands set at twenty to four.” There
has been some variation in the literature regarding which
The Category Fluency test requires that the subject gen- time to use. Several studies suggest that “ten after eleven”
erate as many words as possible in a fixed time period. This is optimal35,36 whereas others favor “two forty-five,”29,37 or
task has been shown to be sensitive to AD.31,32,34 Subjects “twenty to four.”29 There are also several scoring meth-
were asked to generate exemplars from the semantic cat- ods.38 We have developed a simplified version of that used
egory “animals” in 1 minute. The total number of animals by Freedman et al.28 This system requires the examiner to
named produces the score. record the presence of 7 attributes (eg, all 12 numbers are
present). This can be accomplished as the patient com-
BENTON TEMPORAL ORIENTATION TEST pletes the clock drawing.

The Benton Temporal Orientation Test26 uses a graduated NEUROPSYCHOLOGICAL BATTERY

scoring system to assess the patient’s knowledge of month,
date, year, day of the week, and time of day. Unlike the ori- The neuropsychological battery given to all patients seen
entation scales on the MMSE or BIMC, in this variation the at the Memory Disorders Clinic consists of the following:
degree of error is scored. Patients who miss the date by 1 the North American Adult Reading Test, Wechsler
day receive a minimal deduction compared with patients Memory Scale–Revised (verbal and visual memory, imme-
who miss the date by 15 days or the year by 3 years. The diate and delayed, and digit span), Delayed Word Recall,22
scoring system is (1) month—5 error points for each month Boston Naming Test (15-item version), Wechsler Adult
off to a maximum of 30, (2) year—10 error points for each Intelligence Scale–Revised (Block Design), Alzheimer’s
year off to a maximum of 60, (3) date—1 error point for Disease Assessment Scale, MMSE, Geriatric Depression
each date off to a maximum of 15, (4) day of week—1 er- Scale, Cognitive Rating Scale, Global Deterioration Scale,
ror point for each day to a maximum of 3, and (4) time—1 and BIMC.
error point for each 30-minute deviation to a maximum of
5. The maximum total error score is 113. We should note STATISTICAL METHODS
that although scores on this test were not used to diag-
nose AD, temporal orientation is included in the MMSE and Associations between neuropsychological test scores and
Alzheimer Disease Assessment Scale-Cognitive that were demographic variables, interrater reliability, and test-
used in clinical diagnosis. retest reliabilities were calculated using the Pearson corre-
lation coefficient. Differences on individual tests were
CLOCK DRAWING evaluated using the Student t test. Logistic regression
analysis was used to determine the degree to which the
Our version of this widely used test requires the patient to battery discriminated between control subjects and
draw the face of a clock with all the numbers and set the patients with AD.

RESULTS highly reliable as was the overall score on the battery.

Individual test-retest reliabilities were Category Flu-
DEMOGRAPHICS ency (r=0.83), Clock Drawing (r=0.84), Benton Orien-
tation (r=0.93), and Enhanced Cued Recall (r=0.92). The
There were no differences in mean age or education (years overall test-retest reliability using the predicted prob-
of formal schooling) between patients with AD and con- ability of dementia from the logistic regression analysis
trol subjects (for each, t,1 and P..05; Table 1). There of the battery was r=0.91.
was also no significant difference in the ratio of male- Interrater reliability was accomplished by having 2
female subjects (x2,1; P..05; Table 1). raters score the same testing session for 25 randomly se-
lected patients with AD and 25 randomly selected con-
NEUROPSYCHOLOGICAL BATTERY trol subjects. Interrater reliability for the entire battery
was r=0.93. Because the only subscale in the battery that
Table 2 summarizes the neuropsychological data from requires any judgment to score is Clock Drawing, we had
patients with AD and control subjects. In all cases, statis- 2 raters score all the clocks for control subjects and pa-
tical analysis (t tests) indicated that the patients with AD tients with AD. The interrater reliability was r=0.92.
performed significantly worse than control subjects.
INDIVIDUAL TESTS
RELIABILITY
Figure 1 shows the distribution of scores for each of
Test-retest reliability was evaluated in 25 randomly se- the individual tests for patients with AD and control sub-
lected patients with AD and 25 randomly selected con- jects. To determine if patients with AD and control sub-
trol subjects by readministering the 7 Minute Screen 1 jects differed significantly on individual tests, we per-
to 2 months after initial administration. All tests were formed t tests between the mean test scores. Mean scores

ARCH NEUROL / VOL 55, MAR 1998

351

©1998 American Medical Association. All rights reserved.

Downloaded From: http://archneur.jamanetwork.com/ by a Tulane University User on 05/12/2015
 Table 1. Demographic Characteristics Table 2. Neuropsychological Test Results*

Patients With Healthy Patients With Alzheimer’s Healthy Control

Alzheimer’s Disease Control Disease, Mean (SD) Subjects, Mean (SD)
Characteristic (n=60) Subjects Tests (n=60) (n=30) t
Highest education, mean y 13.3 14.4 MMSE 21 (7.8) 28.7 (2.1) 5.31
Highest educational level, range 8-20 6-23 WMS-R
Mean age, y (range) 77.6 (66-89) 77.5 (67-91) LM-1 4.8 (3.8) 22.7 (7.1) 15.4
Sex, M/F 20/40 21/39 LM-2 0.3 (4.6) 18.7 (9.8) 12.3
VR-1 8.1 (5.4) 32.4 (6.6) 18.9
VR-2 4.1 (3.9) 28.3 (9.8) 16.9

and t values are shown in Table 3. As the table shows, *MMSE indicates Mini-Mental State Examination; WMS-R, Wechsler
Memory Scale–Revised; LM-1, Logical Memory-1; LM-2, Logical Memory-2;
there were significant differences on all measures (for each, VR-1, Visual Recall-1; and VR-2, Visual Recall-2. For each, P,.001.
P,.001).
logistic regression model fit to them was used to predict
LOGISTIC REGRESSION the status of the 30 patients with AD and the 30 control
subjects not included in the samples. Repeating this ran-
To determine the degree to which the battery discrimi- dom selection 1000 times, using r=0.1 and r=0.9 as cut-
nated between control subjects and patients with AD, a off points, and conservatively considering all “diagnosis-
logistic regression model deferred cases” to be misclassifications, the 7 Minute

S D
Screen classified 92% of patients with AD correctly and
pi
log =b0+b1*ECRi+b2*CFi+b3*BTOi+b4*CDi+ei 96% of control subjects correctly (Table 4). These num-
1−pi bers represent the percentages correctly predicted from
was estimated using the 4 tests from the screening battery only the 30 patients and 30 control subjects not in-
asthepredictorvariables.Becauseoftheclearnon-normality cluded in the model building phase.
of the data from the battery, discriminant analysis was re- To determine how well the model would predict dis-
jected as a possible alternative method. Specifically, the ease in patients with less severe AD, patients with AD who
following model was estimated where ECR indicates En- had MMSE scores below 21 were excluded, leaving 35
hanced Cued Recall; CF, Category Fluency; BTO, Benton of 60 patients. The resulting sensitivity and specificity
Temporal Orientation; and CD, Clock Drawing. were 98% using 0.1 and 0.9 as cutoff values. A model us-
The response ing only the 13 patients with scores of 24 or higher per-

S D
formed equally well, with a sensitivity of 98% and a speci-
pi ficity of 100%.
log
1−pi We also calculated the positive and negative pre-
dictive values for different base rates of AD using the sen-
can be viewed as the natural logarithm of the odds in fa- sitivity and specificity rates from the 1000 random
vor of having AD. For clarity, we then transformed this samples. Using hypothetical population base rates of 5%,
response back to the probability of having the disease, 10%, 20%, and 50% produced positive predictive values
pi. We classified someone as likely to have AD if pi.0.7 between 54% and 95% and negative predictive values be-
and unlikely to have AD if pi,0.3. We categorized pa- tween 92% and 99% (Table 5).
tients with 0.3#pi#0.7 (pi indicating probability of AD) Two of the predictors (Enhanced Cued Recall and
as requiring further testing (diagnosis deferred). Clock Drawing) in the model in the first equation were
Estimating the model on the 120 patients (60 con- not statistically significant at the a=.05 level using the t
trol subjects and 60 patients with AD) gave the follow- statistic. A stepwise procedure resulted in a model drop-
ing model, with SEs of the estimated coefficients given ping the clock score with all 3 remaining predictors sta-
within parentheses: tistically significant. Moreover, in the 1000 randomly re-

S D
sampled runs, Enhanced Cued Recall was positive (the
p
log = 35.59 −1.303 ECR−1.378 CF+3.298 BTO−0.838 CD wrong sign) only 39 times, indicating an approximate P
1−p value of .04. Clock Drawing was positive 240 times. Given
(17.8) (1.11) (0.78) (1.37) (1.19)
the a priori evidence of the diagnostic value of the Clock
This model resulted in a sensitivity of 100% and a Drawing score and the behavior of the cross-validation ex-
specificity of 100% because the patients with AD all had ercise, we left in all 4 predictors for the final model.
pi.0.76 and all control subjects had pi,0.3. In fact, 58 Other variables may have predictive ability. A model
of the 60 control subjects had pi,0.1, and 58 of the 60 using age, years of education, and sex of patient was also
patients with AD had pi.0.9. Using 0.1 and 0.9 as cut- considered. Since sex showed no predictive ability, a sec-
off points resulted in a sensitivity and a specificity of 96%. ond model with only age and years of education was es-
Figure 2 shows the distribution of probabilities for pa- timated. Using a probability of disease of only 0.5 as a
tients with AD and control subjects. cutoff probability for diagnosis, the model had a sensi-
To test the robustness of the model, 2 different strat- tivity of only 57% (34/60) and a specificity of 65% (39/
egies were used. First, a sample of 30 patients with AD 60). Probabilities for the patients with AD ranged from
and 30 control subjects was chosen at random, and the 0.29 to 0.83 (mean, 0.52), while the control subjects

ARCH NEUROL / VOL 55, MAR 1998

352

©1998 American Medical Association. All rights reserved.

Downloaded From: http://archneur.jamanetwork.com/ by a Tulane University User on 05/12/2015
 60 ECR 30 Category Fluency
AD
Control

50 25

40 20

30 15

20 10

10 5

0 0
Frequency

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 0 1 2 1 4 3 6 4 8 6 10 7 12 9 14 10 16 12 18 14 20+

60 Benton Temporal Orientation 30 Clock Drawing

50 25

40 20

30 15

20 10

10 5

0 0
0 1 2 3 4 5 6 7 8 9 10+ 0 1 2 3 4 5 6 7
Score

Figure 1. Frequency distributions of individual test scores for patients with Alzheimer’s disease (AD) and healthy control subjects. ECR indicates Enhanced Cued
Recall.

ranged from 0.28 to 0.70 (mean, 0.47). Thus, while age healthy subjects; (3) there is a high degree of test-retest
and years of education have some ability to predict AD, reliability and interrater reliability; (4) the ability of the
it is weak. Additionally, when these terms were added battery to discriminate patients with AD from control sub-
to the model above as covariates, neither age, educa- jects is not affected by age, sex, or education; and (5) the
tion, nor sex was statistically significant. battery shows high sensitivity and specificity in patients
with very mild, mild, and moderate disease.
ADMINISTRATION TIME The ability of the battery to discriminate between
patients with AD and control subjects appears to be high.
The mean time to complete the test for all subjects was This, however, may not be surprising given the compo-
7 minutes and 42 seconds (range, 6 minutes and 40 sec- nents of the test. Recent developments in the neuropsy-
onds to 11 minutes and 32 seconds). Patients with AD chological features of AD have identified a number of sen-
took somewhat longer than control subjects, with a mean sitive cognitive tests. The present battery attempted to
of 7 minutes and 57 seconds (range, 6 minutes and 51 take advantage of this research by truncating and com-
seconds to 11 minutes and 32 seconds). The control sub- bining these tests. For example, a longer version of the
jects took a mean of 7 minutes and 27 seconds (range, 6 Enhanced Cued Recall test originally described by Grober
minutes and 40 seconds to 9 minutes and 43 seconds). and Buschke24 compared a group of patients with AD with
a group of community-dwelling control subjects and
COMMENT found 94% sensitivity and 99% specificity. In this sample,
however, the patients with dementia appeared more im-
This study provides initial reliability and validity data for paired than in the present sample and the control sub-
an empirically derived brief screening battery to iden- jects were very high functioning (eg, mean intelligence
tify patients with AD. The data from the present study quotient, 120.5). This test was developed to take advan-
suggest that (1) the battery is highly sensitive, 100% in tage of the well-documented finding in cognitive psy-
the initial sample and 92% in a series of random samples chology that either self-generated or experimenter-
taken from the overall group, in detecting patients with generated encoding retrieval cues facilitate subsequent
AD; (2) the battery shows a high degree of specificity, recall in healthy aged subjects.24 The initial work by
100% in the initial sample and 96% in a series of ran- Grober and Buschke24 indicated that 3 trials of 16 items
dom samples taken from the overall group, in detecting were necessary to discriminate between patients with AD

ARCH NEUROL / VOL 55, MAR 1998

353

©1998 American Medical Association. All rights reserved.

Downloaded From: http://archneur.jamanetwork.com/ by a Tulane University User on 05/12/2015
 Table 3. Results of Individual Tests* Table 4. Mean Predictions From Logistic
Regression Analysis for 1000 Iterations
Patients With
Alzheimer’s Healthy Control Patients With Healthy Control
Disease, Subjects, Diagnosis Alzheimer’s Disease, % Subjects, %
Mean (SD) Mean (SD)
Diagnosed correctly 92 96
Tests (n=60) (n=60) t
Diagnosis deferred 3 3
Enhanced Cued Recall 6.8 (0.7) 15.9 (0.4) 13.9 Diagnosed incorrectly 5 1
Category Fluency 8.8 (0.5) 19.0 (0.7) 11.9
Benton Orientation Test 37.9 (4.2) 0.4 (0.1) 8.8
Clock Drawing 3.2 (0.3) 6.3 (0.1) 11.6
Table 5. Positive and Negative Predictive Values
*For each, P,.001.
Base Positive Negative
60
AD Rate Predictive Value, % Predictive Value, %
Control
5 55 99
50 10 72 99
20 85 98
50 96 92
40
Frequency

30 for sensitivity and specificity for the entire sample, but ap-
pears to be more accurate for less impaired patients. For
20
example, we were able to detect 13 of 13 patients with
MMSE scores above 24, a group of patients who would
generally be considered to be mildly demented. These pa-
10
tients would have been classified as within normal limits
by the MMSE using the 23/24 cutoff criterion.
0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 One ongoing criticism of the MMSE is its sensitiv-
Probability of Dementia ity to education.16,17 For example, Anthony et al39 found
Figure 2. Probability of cognitive impairment as predicted from the logistic that for patients with less than an eighth-grade educa-
regression analysis for patients with Alzheimer’s disease (AD) and healthy tion, specificity decreased from 82% to 63%. Similarly,
control subjects. O’Connor et al14 reported that false negatives are much
more likely to occur in patients with high educational
and control subjects, but the data from the present study levels. The present battery does not appear to be as sen-
suggest that 1 trial is sufficient. The time savings of us- sitive to education. Adding years of education to the lo-
ing only 1 trial makes the test appropriate for a short gistic regression analysis as a covariate did not signifi-
screening battery. Similarly, recent data suggest that ver- cantly affect the predictions.
bal fluency,31,32 clock drawing,28 and orientation (using The MMSE has also been shown to be sensitive to
the Benton26 scoring method) all have a high degree of age. Most of the age-related changes begin at 55 to 60
sensitivity and specificity for AD. Given the high de- years, and dramatically accelerate over the age of 75 to
grees of individual sensitivity and specificity of these tests, 80 years.11 The present battery does not appear to show
it is not surprising that combining them identifies pa- as much sensitivity to age. Adding age to the logistic re-
tients with AD. gression analysis as a covariate did not significantly af-
Because the Enhanced Cued Recall test alone ap- fect the predictions.
pears to yield relatively good sensitivity and specificity, Reliability in the present battery appears reason-
it may be tempting to use this single test as a screening ably high after a 1- to 2-month interval. Test-retest reli-
instrument. There are, however, 2 arguments against this. abilities for individual tests ranged between r=0.83 and
First, the diagnosis of AD requires cognitive deficits in r=0.93. Test-retest reliability for the entire battery was
at least 2 areas. Second, the Enhanced Cued Recall test r=0.91 and interrater reliability for the entire battery was
is not as sensitive or specific as the entire battery. Using r=0.93. These reliabilities would seem to compare fa-
a cutoff point of 15/16 items correct, this test correctly vorably with the MMSE where test-retest reliabilities have
classifies 59 of 60 patients with AD and 54 of 60 control typically ranged from 0.80 to 0.95 for short intervals (min-
subjects. utes, hours, and days),11 but significantly lower for in-
The sensitivity and specificity levels of the present bat- tervals of 1 to 2 months (r=0.50).40,41
tery seem to compare favorably with the most commonly It is likely that the degree of sensitivity and speci-
used tests of mental status in elderly patients. A recent re- ficity reported for this battery is higher than what would
view of the MMSE11 found that the majority (about 75%) be seen in day-to-day practice. The patients with demen-
of studies using the 23/24 cutoff points reported sensitiv- tia in the present study were all referrals to a memory
ity in the 80% to 90% range. However, this range de- disorders clinic and were all suspected to have cognitive
creases substantially (44%-68%) when the group with de- impairment. Ultimately, the utility of this screening test
mentia is less impaired (ie, mean MMSE score .20). The will be determined by how well it discriminates in pri-
present battery was well within the range of the MMSE mary care settings. Nevertheless, the data from our study

ARCH NEUROL / VOL 55, MAR 1998

354

©1998 American Medical Association. All rights reserved.

Downloaded From: http://archneur.jamanetwork.com/ by a Tulane University User on 05/12/2015
 suggest that the screening battery may have sufficient pre- Papadakis M, eds. Current Medical Diagnosis and Treatment. 33rd ed. East Nor-
walk, Conn: Appleton & Lange.
dictive power to be useful in this type of setting. Studies 8. Valenciano R, Morera A, Rodriguez-Perera F, Fernandez-Lopez L. Diagnosis of
are currently under way to determine if this is the case. dementia in primary care. Arch Neurobiol. 1989;52:174-177.
A number of questions remain regarding the pres- 9. Somerfield M, Weisman C, Ury C, Chase W, Folstein M. Physician practices in
the diagnosis of dementing disorders. J Am Geriatr Soc. 1991;39:172-175.
ent battery. Although the battery appears sensitive to AD, 10. Zec RF. Neuropsychological functioning in Alzheimer’s disease. In: Parks RW,
we do not know if it will be sensitive to other dementing Zec RF, Wilson RS, eds. Neuropsychology of Alzheimer’s Disease and Other De-
disorders. Similarly, we do not know if it will be useful mentias. New York, NY: Oxford University Press Inc; 1993:3-80.
11. Tombaugh TN, McIntyre NJ. The Mini-Mental State Examination: a comprehen-
in distinguishing AD from other dementing disorders. One sive review. J Am Geriatr Soc. 1992;40:922-935.
distinction that may be important if the battery is to be 12. Folstein MF, Folstein SE, McHugh PR. ‘Mini-Mental State’: a practical method
for grading cognitive states of patients for the clinician. J Psychiatr Res. 1975;
useful as a screening instrument is its ability to distin- 12:189-198.
guish between AD and clinical levels of depression. Based 13. Glasako D, Klauber MR, Hofstetter CR, Salmon DP, Lasker B, Thal LJ. The Mini-
on what is known about some of the individual tests, there Mental State Examination in the early diagnosis of Alzheimer’s disease. Arch Neu-
rol. 1990;47:49-52.
is reason to believe that this may be the case. For ex- 14. O’Connor DW, Pollitt PA, Hyde JB, et al. The reliability and validity of the Mini-
ample, the type of memory test included in the battery Mental State in a British community survey. J Psychiatr Res. 1989;23:87-96.
has been useful in making this distinction.23 15. Hodges JR, Salmon DP, Butters N. Differential impairment of semantic and epi-
sodic memory in Alzheimer’s and Huntington’s diseases: a controlled prospec-
In summary, the 7 Minute Screen appears to be highly tive study. J Neurol Neurosurg Psychiatry. 1990;53:1089-1095.
sensitive and specific in its ability to discriminate between 16. Kluger A, Ferris SH. Scales for the assessment of Alzheimer’s disease. Psychiatr
patients with AD and community-dwelling control sub- Clin North Am. 1991;14:309-326.
17. Uhlmann RF, Larson EB. Effect of education on the Mini-Mental State Examina-
jects. It can be rapidly administered and scored and has ad- tion as a screening test for dementia. J Am Geriatr Soc. 1991;39:876-880.
equate reliability. Although a number of questions remain 18. Blessed G, Tomlinson BE, Roth M. The association between quantitative mea-
sures of dementia and of senile change in the cerebral gray matter of elderly sub-
regarding the battery’s ability to distinguish between other jects. Br J Psychiatry. 1968;114:797-811.
forms of dementia, screening batteries such as this ulti- 19. Kazniak AW, Poon LW, Riege W. Assessing memory deficits: an information-
mately may be useful in helping make initial distinctions processing approach. In: Poon LW, ed. Clinical Memory Assessment of Older
Adults. Washington, DC: American Psychological Association; 1986:277-284.
between cognitive changes that accompany the normal ag- 20. Mattis S. Mental status examination for organic mental syndrome in the elderly
ing process and those that are suggestive of dementing dis- patient. In: Bellack R, Karasu B, eds. Geriatric Psychiatry. New York, NY: Grune
orders, such as AD. This may lead to earlier diagnosis and & Stratton Inc; 1976:77-121.
21. Rosen WG, Mohs RC, Davis KL. A new rating scale for Alzheimer’s disease. Am
better management. J Psychiatry. 1984;141:1356-1364.
22. Knopman DS, Ryberg S. A verbal memory test with high predictive accuracy for
dementia of the Alzheimer type. Arch Neurol. 1989;46:141-145.
Accepted for publication June 11, 1997. 23. LaRue A. Aging and Neuropsychological Assessment. New York, NY: Plenum
Press; 1992.
This work was supported by grant AG 05134-08S2 from 24. Grober E, Buschke H. Geniune memory deficits in dementia. Dev Neuropsychol.
the National Institute on Aging, Bethesda, Md, a grant from 1987;3:13-36.
the Howard Hughes Medical Foundation, Chevy Chase, Md, 25. Grober E, Buschke H, Crystal H, Bang S, Dresner R. Screening for dementia by
memory testing. Neurology. 1988;38:900-903.
and a grant from the Essel Foundation, Mamaronek, NY. 26. Benton AL. Contributions to Neuropsychological Assessment. New York, NY: Ox-
All the materials necessary for administering and scor- ford University Press Inc; 1983.
27. Winogrand JR, Fisk AA. Alzheimer’s disease: assessment of functional status.
ing the 7 Minute Screen are available on request and at no J Am Geriatr Soc. 1983;31:780-785.
charge from Paul R. Solomon, PhD. Distribution of these ma- 28. Freedman M, Leach L, Kaplan E, Winocur G, Shulman KI, Delis D. Clock Drawing: A
terials is supported by Janssen Pharmaceutica Research, Neuropsychological Analysis. New York, NY: Oxford University Press Inc; 1994.
29. Sunderland T, Hill JL, Melow AM, et al. Clock drawing in Alzheimer’s disease: a
Titusville, NJ. novel measure of dementia severity. J Am Geriatr Soc. 1989;37:725-729.
We are grateful to Felicity Adams, Vivian Calvo, and 30. Wolf-Klein GP, Silvertsone FA, Levy AP, Brod, Brever. Screening for Alzheimer’s
MaryEllen Groccia for their help in collecting the data, to Judy disease by clock drawing. J Am Geriatr Soc. 1989;37:730-734.
31. Monsch AU, Bondi MW, Butters N, Salmon DP, Katzman R, Thal LJ. Compari-
Kessler for preparing the figures, and to Anthony Giuliano, son of verbal fluency tasks in detection of dementia of the Alzheimer type. Arch
PhD, David Knopman, MD, and Martin Samuels, MD, for their Neurol. 1992;49:1253-1257.
32. Rosen W. Verbal fluency in aging and dementia. J Clin Neuropsychol. 1980;2:
helpful comments on an earlier version of the manuscript. 135-146.
Reprints: Paul R. Solomon, PhD, Bronfman Science Cen- 33. McKhann GD, Drachman D, Folstein M, Katzman R, Price D, Stadlan EM. Clini-
ter, Williams College, Williamstown, MA 01267 (e-mail: cal diagnosis of Alzheimer’s disease: report of the NINCDS-ADRDA Work Group
under the auspices of Department of Health and Human Services Task Force on
[email protected]). Alzheimer’s disease. Neurology. 1984;34:939-944.
34. Huff FJ, Corkins S, Growdon JH. Semantic impairment and anomia in Alzhei-
mer’s disease. Brain Lang. 1986;28:235-249.
REFERENCES 35. Rouleau I, Salmon DP, Butters N, Kennedy C, McGuire K. Quantitative and quali-
tative analyses of clock drawings in Alzheimer’s and Huntington’s diseases. Brain
Cogn. 1992;18:70-87.
1. Evans DA, Funkenstein HH, Albert MS. Prevalence of Alzheimer’s disease in a 36. Tuokko H, Hadjistravropoulos T, Miller JA, Beattie BL. The clock test: a sensitive
community population of older persons. JAMA. 1989;262:2551-2556. measure to differentiate normal elderly from those with Alzheimer’s disease.
2. Knapp MJ, Knopman DS, Solomon PR, Pendlebury WW, Davis CS, Gracon SI, J Am Geriatr Soc. 1992;40:579-584.
for the Tacrine Study Group. A 30-week randomized controlled trial of high- 37. Huntzinger JA, Rosse RB, Schwartz BL, Ross LA, Deutsch SI. Clock drawing in
dose Tacrine in patients with Alzheimer’s disease. JAMA. 1994;271:985-991. the screening assessment of cognitive impairment in an ambulatory care set-
3. Rogers SL, Friedhoff LT, and the Donepezil Study Group. The efficacy and safety ting: a preliminary report. Gen Hosp Psychiatry. 1992;14:142-144.
of donepezil in patients with Alzheimer’s disease: results of a US multicenter, 38. Libon DJ, Malamut BA, Swenson R, Sands LP, Cloud BS. Further analysis of clock
randomized, double blind, placebo controlled trial. Dementia. 1996;7:293-303. drawings among demented and nondemented older subjects. Arch Clin Neurol.
4. Wallin A, Blennow K. Clinical diagnosis of Alzheimer’s disease by primary care 1996;11:193-205.
physicians and specialists. Acta Neurol Scand. 1992;39:26-31. 39. Anthony JC, LaResche L, Niaz U, VonKorff M, Folstein M. Limits of the ‘Mini-
5. Iliffe H, Haines A, Gallivan S, Booroff A, Goldenberg E, Morgan P. Assessment Mental State’ as a screening test for dementia and delirium among hospital pa-
of elderly people in general practice, I: social circumstances and general state. tients. Psychol Med. 1982;12:397-408.
Br J Gen Pract. 1991;41:9-12. 40. Mitrushina M, Satz P. Reliability and validity of the Mini-Mental State Exam in
6. McClean S. Assessing dementia, I: difficulties, definitions and differential diag- neurologically intact elderly. J Clin Psychol. 1991;47:537-543.
nosis. Aust N Z J Psychiatry. 1987;21:142-174. 41. Olin JT, Zelinski M. The 12-month reliability of the Mini-Mental State Examina-
7. Resnick N. Geriatric medicine and the elderly patient. In: Tierney L, McPhee S, tion. Psychol Assess. 1991;3:427-432.

ARCH NEUROL / VOL 55, MAR 1998

355

©1998 American Medical Association. All rights reserved.

Downloaded From: http://archneur.jamanetwork.com/ by a Tulane University User on 05/12/2015