The Brief Cognitive Assessment Tool (BCAT):
A New Test Emphasizing Contextual Memory and Executive Functions
Drs. William Mansbach, Elizabeth MacDougall, & Andrew Rosenzweig
November 22, 2011
Dementia Facts
Over 65 population projected to double from 36 million in 2003 to 72 million
in 2030, and will increase from 12% to 20% of the total population
Over 85 population also projected to double from 4.7 million in 2003 to
9.6 million in 2030
Alzheimer’s disease (AD) affects over 5 million Americans, with people 85
and older at the highest risk
It is estimated that by 2050, 13 million Americans will have AD
Current prevalence rates of dementia in SNF and ALF is at or above 50%
Estimated cumulative costs associated with AD alone will exceed $20 trillion
between 2010 and 2050
Why a New Cognitive Screening Tool?
Mini-Mental State Examination (MMSE)
Short Test of Mental Status (STMS)
Montreal Cognitive Assessment (MoCA)
St. Louis University Mental Status Examination (SLUMS)
Brief Interview for Mental Status (BIMS)
Cognitive screening measures can be enhanced if they:
can be administered by paraprofessionals and clinicians alike;
can be completed in approximately 10-15 minutes;
can differentiate among MCI, mild dementia and moderate dementia;
contain a strong, multi-level verbal memory component;
have a broadly complex executive function component, and
predict IADL performance.
BCAT Reliability & Validity Study
Study Objectives
Validate a new cognitive test
Determine if the BCAT yields statistically robust Contextual
Memory and Executive Functions clusters
Determine if the BCAT is predictive of IADLs
BCAT Reliability & Validity Study
Method
111 Participants
Largely from ALF
IRB-approved study
Participants referred to Memory Center for comprehensive neuropsychological
evaluation
Psychologists expert in dementia made CDR DX classifications without reference
to BCAT scores
Demographics:
Ages 61-97, 88% Caucasian, 75% widowed, 75% 12/+ education
67% demented, 33% non-demented
BCAT Reliability & Validity Study
Method, continued
Select Demographic Characteristics and Diagnoses of Participants
Characteristics
Number
Percent
Male
31
29.8
Female
73
70.2
African-American
8
7.7
Caucasian
91
87.5
Missing
5
4.8
Gender
Race
BCAT Reliability & Validity Study
Method, continued
Select Demographic Characteristics and Diagnoses of Participants
Characteristics
Number
Percent
Single, Never Married
5
4.8
Married
24
23.1
Divorced
7
6.7
Widowed
67
64.4
Missing
1
1.0
Marital Status
BCAT Reliability & Validity Study
Method, continued
Select Demographic Characteristics and Diagnoses of Participants
Characteristics
Number
Percent
5-7
9
8.7
9-11
7
6.7
12
38
36.5
13-15
17
16.3
16
16
15.4
17-18
7
6.7
>18
9
8.7
1
1.0
Education
(years completed)
Missing
BCAT Reliability & Validity Study
Method, continued
Select Demographic Characteristics and Diagnoses of Participants
Characteristics
Number
Percent
No diagnosis
3
2.9
Mild Cognitive
Impairment
31
29.8
Mild Dementia
36
34.6
Moderate Dementia
30
28.8
Severe Dementia
4
3.8
Diagnosis
BCAT Reliability & Validity Study
Results
Average inter-item correlation for the BCAT was .36
Internal consistency reliability, as estimated by Cronbach’s α based on
standardized items, was .92.
Test-retest reliability was r = .99.
Convergent validity was demonstrated by the significant correlation of the
BCAT with the MMSE, rs= .90, p<.001, and with the STMS, r = .84, p<.001.
Discriminant validity was demonstrated by the absence of a relationship
between the BCAT and the GDS, rs= -.08 (n = 66; median GDS score = 5.0).
BCAT Reliability & Validity Study
Results, continued
Additional evidence of construct validity was demonstrated by the
significant correlation of the BCAT with several measures of functioning
and judgment, including the NAB Judgment scale
BCAT mean scores were significantly different across the three
diagnostic categories (i.e., MCI, mild dementia, moderate dementia),
F(2, 94) = 82.03, p<.001, ω2= .63
Post-hoc (Bonferroni) analysis revealed significant BCAT mean score
differences for MCI vs. mild dementia, p<.001, MCI vs. moderate dementia,
p<.01, and for mild dementia vs. moderate dementia, p<.001
BCAT Reliability & Validity Study
BCAT Scores by Diagnostic Category
MCI
Mild Dementia
Moderate Dementia
Participants
n = 31
n = 36
n = 30
BCAT mean score
38.81
28.19
18.57
BCAT standard deviation
5.3
6.53
6.56
95% confidence interval
for BCAT mean
36.86 – 40.75
25.99 – 30.40
16.12 – 21.02
BCAT Reliability & Validity Study
Results, continued
The predictive validity of the BCAT was addressed with analyses
of sensitivity, specificity, positive predictive value, and negative
predictive value.
The area under the ROC curve for the BCAT was .95, p<.001,
95% CI [.91, 1.00].
A BCAT cut-score of 38 yielded maximum sensitivity, with a PPV
of .91, indicating a 91% likelihood of correctly diagnosing dementia.
The NPV of .96 indicates a 96% probability of correctly determining
that one does not have dementia.
BCAT Reliability & Validity Study
Results, continued
Regression analyses provided further evidence of the predictive validity
of the BCAT. The BCAT total score was a significant predictor of basic
and instrumental ADLs, as measured by the PSMS (ß= -.41, p< .001),
and the Lawton IADL scale (ß= .63, p< .001)
The 3 “clusters” were found to be predictive of diagnostic category and
the IADL measure
BCAT Reliability & Validity Study
Predictive Utility of Several BCAT Cut Scores
Cut Scorea
Sensitivityb
Specificityb
PPVb
NPVb
35/36
.91 (.82-.97)
.84 (.66-.94)
.93 (.83-.97)
.81 (.63-.92)
36/37
.96 (.87-.99)
.81 (.62-.92)
.92 (.82-.97)
.92 (.82-.97)
37/38
.99 (.91-1.00)
.77 (.58-.90)
.91 (.81-.96)
.96 (.78-1.00)
38/39
.99 (.91-1.00)
.61 (.42-.78)
.85 (.75-.92)
.95 (.73-1.00)
39/40
.99 (.91-1.00)
.52 (.33-.69)
.82 (.72-.89)
.94 (.69-1.00)
a BCAT
b
cut-scores to classify participants into dementia /MCI diagnostic categories
95% confidence interval in parentheses
BCAT Reliability & Validity Study | Principal Component Analysis: 3 Clusters
Summary of exploratory principal components analysis results for the BCAT (N = 104)
Rotated Factor Loadings
Item
1
Story Recognition
.76
Delayed Story Recall
.74
Immediate Story Recall
.64
Orientation
.62
2
3
Mental Control #1
Verbal Traits
.81
Mental Control #2
.63
Judgment
.60
Digits Backward
.55
Arithmetic Reasoning
.55
Immediate Word-List Recall
.75
Naming
.67
Letter List
.61
Language Repetition
BCAT: The 3 Clusters
1. Contextual Memory
Immediate Story Recall
Delayed Story Recall
Story Recognition
Orientation
example
•
•
•
•
“Carol borrowed $10 from her brother Jack last week. She couldn’t
pay him back because she bought a delicious ice cream cone at
the circus.”
BCAT: The 3 Clusters
2. Executive Functions
Verbal Trails (OTMT)
Mental Control (days of the week, backward)
Judgment
Arithmetic Reasoning
Digits Backward
example
•
•
•
•
•
“Suppose you have a 1 PM appointment with your doctor. It takes
45 minutes to get there. What time is the latest you can leave to
get there at 1 PM?”
BCAT: The 3 Clusters
2. Executive Functions
Verbal Trails (OTMT)
Mental Control (days of the week, backward)
Judgment
Arithmetic Reasoning
Digits Backward
example
•
•
•
•
•
“You have $25 to spend at the grocery store. You buy milk for $3.
You buy 2 apples for a $1. How much money do you have left?”
BCAT: The 3 Clusters
3. Attentional Capacity
Immediate Word List
Naming
Letter List
example
•
•
•
Banana…
Justice…
Sara…
Bridge…
BCAT Reliability & Validity Study
Discussion
Contextual Memory: utility of Story Recall, especially Story Recognition
Immediate Word List recall: attention versus memory
Executive Functions: utility of OTMT (Verbal Trails)
Clinical considerations:
• “Cut” score
• Caution in generalizing
• Screening results require confirmation
The Brief Cognitive Assessment Tool (BCAT) Key Characteristics
Can be administered by professionals and paraprofessionals
21 items, 50-point scale
Can be administered in 10-15 minutes
Has a “cut” score separating dementia from Mild Cognitive Impairment (MCI)
Has scores ranges for MCI, mild dementia, & moderate dementia
Contains a multi-level verbal memory component
Contains a broadly complex executive functions component
Predicts Instrumental Activities of Daily Living (IADL)
The BCAT website has an automated scoring program
Introducing the BCAT Website – www.thebcat.com
Interpreting BCAT Scores
“Cut” score separating dementia from Mild Cognitive
Impairment is 37/38
How to interpret the Contextual Memory Cluster
How to interpret the Executive Functions Cluster
How to interpret the Attentional Capacity Cluster
Published BCAT Research
Mansbach, W. E., MacDougall, E. E., & Rosenzweig, A. S.
(in press). The Brief Cognitive Assessment Tool
(BCAT): A new test emphasizing contextual memory,
executive functions, attentional capacity, and the
prediction of instrumental activities of daily living.
Journal of Clinical and Experimental Neuropsychology.
William E. Mansbach, Ph.D.
Founder and CEO, Mansbach Health Tools, LLC
Chief Operating Officer, MedOptions
wmansbach @ thebcat.com
(443) 824-4208