The Meyers Short Battery MSB

1
The Meyers Short Battery (MSB)

• John E. Meyers, Psy.D.
• Center for Neuroscience, Orthopedics Spine,
  Dakota Dunes, SD
• Martin L. Rohling, Ph.D.
• University of South Alabama, Mobile, AL

2
Table of Contents

• Philosophy of MSB
• Development of MSB
• Norms Development
• Sensitivity and Specificity
• LOC Dose Response
• Profiles
• Ecological Validity

3
Philosophy of MSB

• MSB began as much longer battery of tests.
• Using Discriminate Function Selected tests that
  discriminated Normal from TBI.
• Did original study years ago.

4
Philosophy of MSB

• Goal was to find the best/shortest battery
• Sensitive to Brain Injury
• Commonly used Tests, that most NPs know
• Originally a 6 hour battery cut to 2.5-3 hrs
• Mean for Litigants 2.89 SD .63 hrs (n 410)
• Mean for non litigants 2.86 SD .68 hrs (n
  941)
• Tests were selected not only for sensitivity but
  also ease of administration and scoring (i.e.
  Category vs WCST).

5
Testing Order for MSB

• Short WAIS-III
• Forced Choice (FC)
• Rey Complex Figure (RCFT) - Copy
• Animal Naming
• 3 min recall of RCFT
• COWA
• Dichotic Listening
• N. Am. Adult Reading (NAART)
• Sentence Repetition
• 30 min Recall of RCFT

• Recognition Trial of RCFT
• (Break offered)
• AVLT
• JOL
• Boston Naming
• Finger Tapping
• Finger Localization
• Trails A B
• Token Test
• AVLT 30 minute Recall
• AVLT Recognition Trial
• Booklet Category Test

6
Individual Tests in The MSB

• PICTURE COMPLETION
• DIGIT SYMBOL
• SIMILARITIES
• BLOCK DESIGN
• ARITHMETIC
• DIGIT SPAN
• INFORMATION
• Ward 7 Subtest (Pilgrim, Meyers, et al., 1999)

7
MSB Database

• Description of large patient sample
• (N 2939)
• Descriptive Statistics

• Note The individual with 0 years of education
  was from Mexico and had not completed a single
  year of education.

8
MSB Database

• Gender
• Female 1334
• Male 1605
• Handedness
• Right Handed 2583
• Left Handed 356

9
MSB Database

• Ethnicity
• African American 64
• Mixed Racial 41
• Caucasian 2715
• Asian 11
• Native American 55
• Hispanic 53

10
MSB Database
11
MSB Database
12
MSB Database
13
MSB DatabaseMental Health Diagnoses
14
MSB Database Normative Data

• Evaluating normative data, observed variation in
  test norms, with age education.
• AVLT normative data (Spreen Strauss, 1998)
• M 11.4 (SD 2.4) for Trial 6 (IR) Age 30-39
• Now, at age 40, M 10.4 (SD 2.7).
• Individual with 10 on this test 1 day before DOB,
  then tested again 1 day after DOB, score (i.e.,
  10) would improve from a 44T to 48T (using a
  linear transformation).
• Using the Heaton et al. (1991) then score would
  improve from Below Average to Average range
  following 1 day.
• Common problem w/ non-smoothed normative data.

15
MSB Normative Data

• Decided to smooth normative data
• Selected all subjects with validity scores
• 15 years or older
• 15 years selected as age for adult Trails A B.

16
MSB Normative Data

• Subgroup size n 1727.
• Mean age 45.7 (sd 20.7)
• Education 12.3 (sd 2.7).
• Gender 779 female 948 male
• 1543 RH (88) 184 LH (12).
• Ethnicity of sample
• 32 mixed 22 African America
• 1617 were Caucasian 2 Asian
• 27 Native American 27 Hispanic.

17
MSB Normative Data

• Regression equation used
• Raw score
• Variables, age, education, gender, hand, race
• Predict T-score previously using standard
  normative data.

18
MSB Regression Norms

• Process smoothes data
• Also adds dimension of age, ed, sex, hand,
  ethnicity adjustment in patient sample.
• Normal sample, these variables not always sign.
• But injured group, these variables take on
  important impact on scores.

19
MSB Normative Data

• Therefore, regression equations change the data
  (using prior example)
• Female age 39
• Score 10 AVLT Trial 6 (IR) 45 (T score)
• Also, 1 day after DOB score 45 (T score)

20
MSB Normative Data
Because of the skewedness of the data
percentile scores were computed and transformed
to T Scores for this test.
21
MSB Normative Data
22
MSB Normative Data
23
MSB Data Children

• Same regression procedures used with children
  with similar results.
• Sample size for children N 348.

24
MSB Regression Equations
25
MSB Regression Equations
26
MSB Regression Equations
27
Norms for the Token TestAdult Token Test

• Token Test only instrument in battery for which
  regression procedure inappropriate.
• As a result, equipotential procedures used to
  generate T scores for Token.
• See hand-out for T score conversions.

28
MSB Recap

• Step 1 Took battery of well known NP Tests
• Tests with which most clinicians are familiar.
• Tests selected based on
• Utility
• ease of scoring
• assess wide array of functions
• Battery result several preliminary batteries.

29
MSB Recap (contd)

• Step 2 Large database of Ss collected
• N 2939
• Step 3 Examined norms for smoothing
• Data smoothed across battery for ages 6-99
• Separate norms for 6-14 and 15-99
• Adjusted for age, education, gender, ethnicity,
  handedness.

30
MSB Recap (contd)

• Step 4. Recalculate database with new norms
  (Step 3)
• Now on to Step 5
• Is this battery of tests valid?

31
MSB Step 5 Is this battery valid?

• Examine Reliability Validity of MSB
• Meyers, J.E., Rohling, M.L. (2004). Validation
  of the Meyers Short Battery on Mild TBI patients.
  Archives of Clinical Neuropsychology, 19,
  637-651.
• Study with 4 groups.

32
Validity of MSB

• 30 Medical Controls, in hospital for non CNS
  problem (i.e. ingrown toe nails) (Group 1)
• All community dwelling
• No Hx of LD, DD, Substance abuse, TBI, or Mental
  Health problem, or anything that would disqualify
  as Normal.
• M Age 38.6 years (SD 18.9) years.
• M Ed 13.4 years of education (SD 3.2).
• Gender 15 male, 15 female.
• Handedness 29 R handed 1 L handed.
• Ethnicity 29 Caucasian 1 Native American.

33
Validity of MSB

• Depressed Group (Group 2) 41 patients
• All on SSRI
• M Age 46.0 years (SD 15.0)
• M ed 13.5 yrs (SD 2.7)
• Gender 20 female 21 male
• Handedness 38 RH 3 LH
• Ethnicity 1 mixed 40 Caucasian.
• 29 of these completed MMPI-2
• L 52.1 (SD 11.4) 1 63.8 (SD 12.8)
• F 60.5 (SD 11.7) 2 70.8 (SD 14.5)
• K 50.2 (SD 10.2) 3 66.7 (SD 16.0)

34
Validity of MSB

• Chronic Pain 32 cases treated outpatient.
• None involved in litigation time of assessment,
  no previous litigation.
• Individuals injured in non-work-related accidents
  or on own farms
• Chose not to pursue workmans comp.

35
Validity of the MSBChronic Pain Group Continued

• M Age 40.72 years (SD 14.17)
• M Education 13.41 years (SD 2.06).
• Gender 20 females and 12 males
• Handedness 29 R handed 3 L handed
• Ethnicity 31 Caucasian 1 Native American.

36
Validity of MSB

• Group 4 consisted of 59 individuals with history
  of Traumatic Brain Injury (TBI).
• All individuals had been seen at the local
  hospital and rehabilitation unit and followed
  through rehab.
• All had identified loss of consciousness (LOC)
  that was 20 minutes or less, other data such as
  GCS and PTA were not always available however,
  LOC data were available for all participants.
• LOC was defined as the time to follow commands
  (e.g., Dikmen, et al., 1995 Volbrecht, et al.,
  2000).

37
Validity of MSB

• M Age 36.9 (SD 15.1)
• M Ed 12.6 years (SD 2.1).
• Time Post Injury 7.6 months (SD 11).
• Gender 14 female 43 were male
• Handedness 51 RH Dom 6 LH Dom
• Ethnicity 2 mixed, 1 Hispanic 54 Caucasian.

38
Validity of MSB Test scores obtained for each of
the study groups
39
Validity of MSB Test scores obtained for each of
the study groups
40
Validity of MSB

• Validity was assessed using a discriminant
  function analysis comparing the Non TBI
  participants with the TBI participants.
• The resulting function resulted in a 96.1
  correct classification rate with 98.9
  specificity and 90 sensitivity.

41
Reliability of MSB

• Reflecting a general clinical sample,
• 63 persons with mixed diagnoses were assessed
  more than once, with the first testing at least
  six months post injury.
• Some in litigation, all passed all validity
  checks
• Group descriptive
• Age 38.38 years (SD 22.8) Ed 12.2 (SD 2.9)

42
Test ReTest Reliability

• First Testing Months post Injury 21.6 (SD
  22.8)
• ReTest post injury 40.7 months (SD 33.2).
• Time between testing 19.1 months (SD 16.6)
• range 2 to 91 months, and a median months
  difference of 13 months.
• Reliability of r .86

43
MSB Correlation with CT/MRI

• From larger database 534 subjects had CT/MRI
  data/Passed built in Validity Checks (will
  Discuss later)
• Age M 44.1 (SD 20.4)
• ED M 12.0 (SD 2.81)
• Coded 1 or 0 Based on CT/MRI lesion lobes

44
CT/MRI

• LF Left Frontal
• LPLeft Pariental
• LTLeft Temporal
• LOLeft Occipital

• RFRight Frontal
• RPRight Pariental
• RTRight Temporal
• RORight Occipital

45
Correlation CT/MRI with MSB
46
Correlation CT/MRI with MSB
47
Correlation CT/MRI with MSB
48
MSB

• The MSB
• Assesses wide variety of cognitive areas,
  corresponding with each lobe of the brain.
• Significant correlations
• Correlations were significant, but should not be
  used as localizing info.

49
Validity

• Meyers, J.E. Volbrecht, M. E. (2003). A
  validation of multiple malingering detection
  methods in a large clinical sample, Archives of
  Clinical Neuropsychology, 18, 261-276.
• Other publications

50
Internal Validity Check (0 FP Rate cutoff)
51
Internal Validity Checks

• A total of 796 participants in the study, ages
  ranged from 16 to 86, with education ranging from
  5 years to 23 years.

52
Internal Validity Checks 15 Groups

• Non litigant TBI group LOC lt1 hr
• Non litigant TBI group LOC gt1 hour lt24 hour
• Non litigant TBI group LOC gt24 hour lt 8 days
• Non litigant TBI group LOC gt 9 days
• Non litigant group Chronic pain
• Depressed group
• Litigants LOC lt 1 hour
• Litigants LOC gt 1 hour

• Litigants chronic pain
• Normals
• Institutionalized patients
• Noninstitutionalized patient failed no more
  than 1 validity check
• Noninstitutionalized and failed 2 or more
  validity checks (not in litigation).
• Noninstitutionalized and failed 2 or more
  validity checks and in litigation.
• Informed actors (portraying role of a malingerer).

53
Internal Validity Checks

• It was found that the 9 neuropsychological tests
  (when used together) were able to correctly
  identify litigant and nonlitigating groups. This
  method showed 83 sensitivity and 100
  specificity.

54
Internal Validity Check

• This method showed 83 sensitivity and 100
  specificity. A 0 false positive rate was found.
  

55
Domains used by the MSB(N 936)

• Attention and Working Memory
• Forced Choice
• Digit Span
• Sentence Repetition
• Animal Naming
• AVLT 1

• Processing Speed/Mental Flexibility
• Digit Symbol
• Dichotic Both
• Trails A
• Trails B
• RCFT Time
• COWA

56
Domains used by the MSB(N 936)

• Verbal Reasoning
• Arithmetic
• Similarities
• Information
• Dichotic Right
• Dichotic Left
• Boston Naming
• Token Test

• Visual Reasoning
• Picture Completion
• Block Design
• JOL
• Category
• RCFT Copy

57
Domains used by the MSB

• Verbal Memory
• AVLT Total
• AVLT Immediate
• AVLT Delayed
• AVLT Recognition

• Visual Memory
• RCFT Immediate
• RCFT Delayed
• RCFT Recognition

58
Domains used by the MSB (N 936)

• Motor and Sensory
• Finger Tapping Dominant Hand
• Finger Tapping Non-Dominant Hand
• Finger Localization Dominant Hand
• Finger Localization Non-Dominant Hand

59
Domain Consistency

• N 936
• Passed all Validity Checks
• No missing data
• Not involved in Litigation
• Calculated Domain Means
• Calculated Regression Equation to predict each
  Domain Mean based on other Domains

60
Domain Means Correlations All were Significant (
p lt .001 )
61
Domains Regression Equations

• Attention Working Memory
• Verbal Reasoning .315
• Verbal Memory .273
• Processing Speed .193
• Constant 10.972

62
Domains Regression Equations

• Processing Speed/ Mental Flexibility
• Verbal Reasoning .401
• Visual Reasoning .284
• Attention Working Memory .230
• Constant 2.434

63
Domains Regression Equations

• Verbal Reasoning
• Processing Speed .361
• Attention Working Memory .354
• Visual Reasoning .243
• Constant 2.500

64
Domains Regression Equations

• Visual Reasoning
• Visual Memory .322
• Process Speed/Mental Flex .213
• Verbal Reasoning .208
• Constant 11.813

65
Domains Regression Equations

• Verbal Memory
• Attention Working Memory .738
• Visual Memory .388
• Constant -7.615

66
Domains Regression Equations

• Visual Memory
• Visual Reasoning .698
• Verbal Memory .311
• Processing Speed .091
• Constant -5.517

67
Regression
68
Review So Far

• Took a battery of well known tests
• Developed Norms
• Identified Validity, Reliability, Sensitivity and
  Specificity.
• Internal Validity Checks and Internal Consistency

69
The Meyers Short Battery, as Seen through the
Lens of the Rohling Interpretive Method

• John E. Meyers, Psy.D.
• Center for Neuroscience, Orthopedics Spine,
  Dakota Dunes, SD
• Martin L. Rohling, Ph.D.
• University of South Alabama, Mobile, AL

70
Neuropsychologist as Diagnostician

• Most clinicians creatively approach assessment
  based upon relevant knowledge in cognitive
  neuroscience, clinical neurology, and test
  development
• 85 of neuropsychologists recently surveyed
  administer a flexible battery.
• Consequently, the omnibus index scores
  generated from fixed batteries serve little
  purpose.

71
The Dilemma

• Insurance reimbursements are forcing the
  reconsideration of index scores from fixed
  batteries.
• Evidence based medicine
• Recent court rulings (e.g., Daubert, Kumho)
• These forces require the presence of empirical
  support for diagnostic decisions based upon
  neuropsychological testing.

72
So Why Consider Index Scores?

• When diagnosis is essentially yes or no,
  actuarial procedures consistently outperform
  clinical decision-making and are more efficient
• However, the omnibus index scores from fixed
  batteries are all weve got with established
  rates of accuracy for identifying brain damage.

73
Tough Im not about to start giving the HRB
because

• It takes too long to give.
• Its too psychometrically messy.
• The HRB does not allow separation of
  neurocognitive constructs that can be delineated
  by newer tests.
• Its too old.
• Norms are out-of-date, as are test materials.
• However, the HRB remains the gold standard, as
  far as actuarial procedures for
  neuropsychological assessment.

74
The RIM Solution

• Wouldnt it be cool if, for these yes no
  evaluations, we could generate a meaningful
  summary index based upon the scores generated
  from a flexible battery?

75
If such an index existed, it would have to be

• As useful as common fixed battery indices.
• HII, AIR, GNDS
• Able to incorporate different measures without
  losing substantial utility.
• Practical for everyday clinical use.
• Easy to cross-validate across samples.

76
RIM Based on Meta-AnalysisWhat Is Meta-Analysis

• A statistical procedure for combining data across
  empirical studies.
• Increases the signal-to-noise ratio by combining
  true score variance while distributing error
  variance.
• Facilitates the investigation of causality and
  methodology that cannot be accomplished by any
  single empirical design.

77
Effect Size Calculations

• Cohens d
• (Me Mc) / (SDpooled)
• Glasss D
• (Me Mc) / (SDc)
• Average effect sizes across studies to get more
  valid and reliable results.

78
Sample RIM Summary Graph
79
(No Transcript)
80
RIM Steps to Generate Summary Scores Interpret
Data

• Design administer battery.
• Estimate premorbid general ability.
• Convert test scores to a common metric.
• Assign scores to domains.

• Calculate domain M, sd, n.
• Calculate test battery means (TBM).
• Calculate p for heterogeneity.
• Determine categories of impairment.

81
RIM Steps to Generate Summary Scores Interpret
Data

• Determine of test impaired.
• Calculate ES for all domains and TBMs.
• Calculate CI.
• Determine the upper limit of performance.

• Conduct t tests.
• Conduct a between-subjects ANOVA.
• Power analyses.
• Sort scores.
• Display summary statistics.

82
RIM Steps to Generate Summary Scores Interpret
Data

• Assess battery validity.
• Examine influence of psychopathology
• Use OTBM, DTBM, ITBM to determine if
  impairment exists.

• Determine current strengths and weaknesses.
• Examine non-cognitive domains.
• Explore for Type II errors.
• Examine sorted T-scores.

83
RIM Studies Related to MSB

• TBI Dose Response MSB HRB.
• Rohling, M. L., Meyers, J. E., Millis, S.
  (2003). Neuropsychological impairment following
  TBI A dose response analysis. The Clinical
  Neuropsychologist, 17, 289-302.
• RIM Analysis of the HRB.
• Rohling, M. L., Williamson, D. J., Miller, L. S.,
  Adams, R. (2003). Using the Halstead Reitan
  Battery to diagnose brain damage A comparison of
  the predictive power of traditional techniques to
  Rohlings Interpretive Method. The Clinical
  Neuropsychologist, 17, 531-544.
• Return to Work after Injury - MSB data used.
• Rohling, Meyers, Blanton (in submission).

84
Study 1 TBI Dose Response

• Summary statistics to generate ESs from
• Dikmen, S. S., Machamer, J. E., Winn, H. R.,
  Temkin, N. R. (1995). Neuropsychological outcome
  at 1-yr post head injury. Neuropsychology, 9,
  80-90.
• Raw data requested, but not accessible.
  Therefore, meta-analysis used to calculate ESs
  from published summary statistics.

85
Study 1 TBI Dose Response

• Initial analysis presented at NAN (2000) was
  based on n 150 and came from
• Volbrecht, M. E., Meyers, J. E.,
  Kaster-Bundgaard, J. (2000). Neuropsychological
  outcome of head injury using a short battery.
  Archives of Clinical Neuropsychology, 15,
  251-265.
• Later, raw data accessed with larger sample (n
  317).
• Symptom validity used to clean sample (n 68)
• ESs calculated for each subject averaged.
• Overall Test Battery Mean (OTBM)

86
Study 1 TBI Dose Response

• Dikmen et al.s (1995) - TBI assessed 1 yr post
• Divided into 6 severity of injury groups based on
  Time to follow verbal commands.
• (1) lt 1 hr, (2) 1-23 hrs, (3) 1-6 d, (4) 7-13 d,
  (5) 14-28 d, (6) gt 28 d.
• ESs generated as part of a prior meta-analysis
  (Binder, Rohling, Larrabee, 1997).
• Halstead Impairment Index used as measure of
  global severity of cognitive impairment.

87
Study 1 TBI Dose Response

• Meyers data classified using same 6 severity
  groups as Dikmen et al. (1995).
• Flexible battery administered with number of
  tests varying across patients.
• Average test time was 2.6 hrs.
• Max. number of measures examined 26.
• Only 4 measures used by both researchers.

88
Study 1 OTBM Regressed using HRB MSB Summary
Stats

• Mostly severely impaired group differences
  between the two samples.
• Dikmen et al. estimated for untestable pts.
• Meyers excluded untestable pts.
• Regression across 5 severity groups found high
  degree of agreement between samples.
• R .97
• Slope .92 (NS from 1)
• Intercept 4.7 (NS from 0)

89
TBI Demographics Dikmen Meyers

• Dikmen Sample
• Age 29.9 (12.5)
• Ed 12.0 (2.3)
• Sex 72 male
• T since TBI 1 yr
• Barona IQ 95.8 (est.)

• Meyers Sample
• Age 32.8 (14.7)
• Ed 12.0 (2.4)
• Sex 64 male
• T since TBI 3 yrs
• Barona IQ 94.3
• NART IQ 97.3

90
Dikmen ESs Meyers T Scores
91
Combined Dikmen Meyers Estimates ES, T,
Difference
92
Study 1 HRB TBI Severity

• Correlation .98 for OTBM HII across 6 groups
  for Dikmen et al. (1995) sample.
• Slopes -.039
• Intercept 1.98
• T for Imp. high (43.0)
• Later to be used in cross-validation study with
  the Oklahoma HRB sample

93
Domains of Cognition Combined Dikmen Meyers
94
TBI Dose Response Summary

• The OTBM from the MSB is as sensitive to degrees
  of impairment as is the HRB.
• Practical advantages to the MSB-RIM
• lt 3 hrs to administer vs. 8 for HRB.
• Domain scores generated for interpretation.
• Better to detect suboptimal performance.
• Easy to view graphics for greater Dx accuracy.
• Indices robust across multiple studies.

95
Study 2 HRB Validation of RIM

• Sherer et al. (1993) published a cross-validation
  of the GNDS (Reitan Wolfson, 1991), using the
  University of Oklahoma Health Science Center
  database generated from 1977 to 1989.
• Vanderploeg et al. (1997) responded to Reitan
  Wolfsons (1995) statement that age and
  education dont matter with the same dataset.

96
Study 2 HRB Sample from Okla.

• Originally 114 patients.
• 73 Brain Damaged patients (TBI, CVA, etc.).
• 41 Pseudoneurological pts (i.e., psychiatric).
• Role of research on pseudoneurological controls
  looking at cut scores for most accurate
  classification of patients to groups.
• These were predominantly psychiatric cases with
  schizophrenia, depression, anxiety dx, etc.

97
What is the Halstead Impairment Index (HII)?

• A measure of overall neurocognitive functioning
  designed by Retain for the HRB.
• Variables rated from 0 (normal) to 1 (impaired)
• Number of variables 7
• Range 0 (normal ) to 1.0 (severe)
• HII a rounded ratio to the nearest 10ths of the
  of scores impaired by total number of scores.
• Tests Category, TPT (time, mem, loc), Rhythm,
  Speech, Tapping (dom).
• Cutoff score for impairment recommended .30

98
What is the Average Impairment Rating (AIR)?

• Russell (1970) measure of global cognitive
  functioning designed to improve on HII.
• Each variable rated 0 (normal) to 3 (severe)
• Number of variables is flexible
• Range from 0 (normal ) to 6 (profound)
• 0.5 intervals
• Measures normed on this scale by Russell (1970).
  Tables to convert scores to AIRs
• Cutoff score for impairment 1.30

99
What is the Global Neurological Deficit Scale
(GNDS)?

• Retain Wolfson (1993) - measure of cognition
  designed to improve HII AIR.
• Each variable rated 0 (normal) to 3 (severe)
• Performance Level Variables (n 19)
• Pathognomic Sign Variables (n 13)
• Pattern Variables (n 2)
• Anterior/Posterior
• Right/Left Laterality
• Range (n 34) 0 to 102.
• Cutoff Recommended 28.

100
Study 2 OTBM vs. HII

• Heaton et al.s (1991) HRB norms used to get OTBM
• T Score (M50, sd10)
• OTBM Correlation with HII -.79
• (p lt .0001)
• 62 variance accounted
• Over predicts low
• Under predicts high

101
Study 2 OTBM vs. GNDS

• OTBM Correlates with GNDS -.87
• 76 variance accounted
• OTBM neither under nor over predicts across the
  range of the GNDS
• Intercept for impairment as defined by Reitan
  Wolfson (GNDS 29) is a T Score 46.0

102
Study 2 OTBMs Relationship to Other Global
Indices
103
Study 2 Diagnostic Classification Using the HII
104
Study 2 Cross-Validation of RIM using the HRB in
2 Samples

• Regressed Dikmen Meyers TBI data
• Generated a predicted HII for pts in OK dataset.
• Correlation actual predicted HII .95
• Sense .60, Spec .77, PPV .78, NPV .59
• Overall Correct Classification 71
• Predicted HII from MSBs OTBM more accurate
  indicator of impairment than actual HII.

105
Study 2 Summary of Results

• The RIM as good a predictor of cognitive
  functioning as validated indices of HRB.
• (i.e., HII, AIR, GNDS).
• RIMs diagnostic accuracy as good or better than
  HII, AIR, GNDS in discriminating abnormal
  cognition.

106
Study 3 Return to Work after Injury

• Three main hypotheses using MSB-RIM
• OTBM will predict return to work level
• Cognitive domain that will be most predictive
  will be executive function
• Adding the Patient Competency Rating Scale will
  improve work prediction
• PCRS is by Prigatano (1985)

107
Study 3 ANOVA Results for OTBM
108
Logistic Regression using OTBM
109
Study 3 Summary of OTBM

• Significant OTBM differences between return to
  work groups.
• Logistic results Disabled /Unemployed not able to
  separate.
• Logistic results Below/At Previous not able to
  separate.
• Collapsed groups result in 71 correct, well
  above base rate of 52 correct.

110
Study 3 Results of Domain Analysis

• Executive function not the most predictive.
• Most of the Logistic results variance cared by
  Perceptual Organization followed by Attention
  Working Memory.
• Using Cognitive Domains
• OTBM increases Correct from 71 to 74.
• Incremental validity of PCRS is very low.
• 7 of the variance.

111
Study 3 Results of Domain Analysis

• By including premorbid variables, increases
  diagnostic accuracy most helpful being
• Premorbid IQ, level of occupation, education
• Including acute measures also increases accuracy
  most helpful being
• LOC group
• Time since injury

112
Interpreting the MSB

• Profiles What the data is like, type of injury
• RIM How bad is it (OTBM)
• Domains What deficits
• Rehab Individual Scales

113
Comparisons

• Normals are normal hospital controls, were seen
  for some malady other than neurological (i.e. in
  grown toe nails) (Blue Line)
• Hospital Patients with identified brief LOC, by
  witness report of seconds to minutes. Mild TBI
  with brief LOC (Black Line) All reporting some
  Post Concussive Syndrome Characteristics
  including Headache
• Moderate TBI 1 hr to 24 hrs LOC, Documented in
  medical records (Red Line)
• Severe TBI 8 days or more LOC, Documented in
  medical records (Green Line)

114
Traumatic Brain Injury
115
Correlations between Profiles

• Correlations are with the Mild TBI Group
• Notice consistency with the TBI groups
• Correlations (My rule of thumb .30 )
• Configurations (My rule of thumb .59)
• Group N
• OTBM

116
Profiles

• Similar types of injuries produce similar type
  profiles
• i.e. Left Learning Disability, Left Hemisphere
  originating Seizure Disorder and MS.
• Why?
• Because all three have / can have impairment in
  Left hemisphere
• Similar type impairments produce similar patterns
  of performance
• Dissimilar injuries produce dissimilar patterns

117
Example

• Learning Disability (Black Line)
• Epilepsy (Red Line) Left hemisphere originating
• Left Temporal Lobectomy (Blue line)
• RCVA (Green Line)
• Notice that the RCVA group is different from the
  other groups

118
Similar and Dissimilar profiles
119
Case Study BBMVA

• Age 19
• Ed 12
• Male
• Right Handed
• Caucasian
• Single Vehicle Accident
• Not Forensic

• NART 57 T
• Barona 50 T
• Short Form WAIS-III
• 7 subtest
• VIQ 110
• PIQ 117
• FIQ 113

120
Case Study BB

• LOC at scene described by other passenger
  Knocked Out for couple of minutes he didnt wake
  up when I shook him a little, and then he started
  to groan a little bit.
• PTA Ambulance records Confused, but could
  answer his name
• ER Able to answer questions with repeated
  Questions, some mild confusion
• Whole time from accident to only mild confusion
  less than one hour

121
Case Study BB
122
Case Study BB

• Step 1 Check for Validity
• Validity Checks
• Internal Consistency
• Step 2 Profile
• Is the pattern of performance similar to what you
  expect for a TBI

123
Case Study BB

• History
• NO LD, ADD
• No Previous Head Injury
• College Freshman (first semester)

• Internal Validity Checks
• Passed All 8
• And Motor performance

124
Case Study BB

• Internal Consistency Check
• (Must be Greater than 15 to be clinically
  significant). He passed all
• Estimated Actual Difference
• Attention Wk Mem 48 47 1
• Proc Speed/Ment Flex 49 49 0
• Verbal Reasoning 50 52 -2
• Visual Reasoning 50 52 -2
• Verbal Memory 47 41 7
• Visual Memory 48 52 -4

125
Case Study BB
126
Case Study BB
127
Case Study BB

• From this you can see
• 1. Objective method of assessing validity and
  internal validity
• 2. Objective method of assessing the
  characteristics of the profile. To see if the
  pattern of scores is as expected

128
Summary of MSB

• Check Internal Validity Check
• Check the expected level of performance
• Profiles tells what performance is similar to
  (.30 correlation, .59 Configuration)
• Impaired Domains tell what type of deficits
• Individual test scores useful for designing
  cognitive rehab

129
Any ?

• Questions