Estimating Premorbid Intelligence
Determining Change in Cognition Following Brain Injury
Richard A. Lanham, Jr., PhD and Thomas Misukanis, PhD
Brain Injury Source, Pediatric Issue, Volume 3, No.
3, Summer 1999
Introduction
Neuropsychology, a specialty
discipline within clinical psychology, is devoted to
the study of brain-behavior relationships, the assessment
of neurobehavioral impairment and its effects on functioning
and methods of treatment. In cases of traumatic brain
injury (TBI), where an individual is suspected of having
experienced a disruption in cognitive functioning, a
referral for neuropsychological evaluation often will
be made in order to determine the nature and extent
of any cognitive changes that may have occurred. Such
information is valuable for tracking the course of recovery,
guiding treatment and compensatory strategies and assisting
in discharge planning. This article discusses an essential
part of the neuropsychological evaluation: the methods
and associated problems of estimating premorbid level
of functioning against which post-injury test results
are compared in an effort to establish the resulting
cognitive impairments of the TBI.
Methods of Evaluation
In evaluating an individual
with a suspected brain injury, the neuropsychologist
makes use of a variety of psychometric tests. These
tests either have been specifically developed to assess
for cerebral impairment (e.g. Halstead-Reitan Neuropsychological
Test Battery, HRB) or have been routinely employed in
such a manner as to allow for the determination of deficient
cognitive functioning (e.g. Wechsler Adult Intelligence
Scale, now in its third edition). Without some standard
of comparison, however, test results in and of themselves
would be little more than meaningless numbers. In general,
two standards of comparison are employed: 1) a normative
or extra-individual standard of comparison where an
individual’s performance is compared to that of others
with whom some important attributes are shared and 2)
a self-standard or intra-individual standard of comparison,
where an individual’s performance is compared to pre-injury
expectations regarding that individual’s level of ability.
Normative Standard
In employing the normative
standard of comparison, an individual’s test scores
are compared to the performance of a large group of
people of interest, permitting score adjustment for
such factors as age, sex or level of education. These
comparisons give the neuropsychologist information about
the individual’s current cognitive functioning relative
to a particular normative group. For example, knowing
that an individual performed at the 2nd percentile –
thus performing worse than 98% of the sample against
which the score is being compared – may be a strong
indicator of cerebral impairment. The adequacy of the
comparison between the individual being evaluated and
the normative group being employed is crucial and places
limitations on the type of conclusions that may be drawn
and the certainty one may have in them. For instance,
knowing that the above 2nd percentile performance occurred
on the entrance examination for Mensa and that the "normative"
comparison group was all current and past Mensa members,
might alter the interpretation of the score as a reflection
of brain impairment.
Self-Standard
The second standard of
comparison, the self-standard, compares an individual’s
post-injury performance with some measure of pre-injury
functioning. Ideally, one would directly compare post-injury
test results against results obtained from those very
same tests administered some time prior to the injury.
While this situation potentially does exist for military
personnel participating in the Defense and Veterans
Head Injury Program (DVHIP), it is a rare occurrence
for pre-injury testing to exist in nonmilitary settings.
Even for ex-military persons, access to the Armed Services
Vocational Aptitude Battery (ASVAB), the current version
of the military classification test taken by all noncommissioned
personnel upon entry into service, is restricted to
protect the test’s integrity, much like the various
civilian entrance examinations (e.g., SAT, MCAT, GRE).
While the DVHIP is able to administer the ASVAB, with
the permission of the military, six months post-injury
as part of its ongoing clinical and research program
(e.g., Grafman et al., 1988), in general clinical practice
it would not be available for re-administration.
Most commonly, the self-standard
of comparison involves estimating an individual’s pre-injury
level of ability. Unfortunately, this standard is frequently
applied incorrectly or neglected entirely. Lezak (1995)
considers it critical that the neuropsychologist obtains
or makes an educated estimate about the individual’s
cognitive functioning prior to his/her injury. Similarly,
Matarazzo (1990) wrote that too often the psychological
practitioner relies solely on data gathered from test
results and fails to consider and integrate other important
sources of data. Using only the normative standard of
comparison can result in erroneous conclusions.
For example, it is not
unusual to assess an individual with a superior pre-injury
level of cognitive ability whose post-injury scores
now fall within the average range. While such results
may be considered "within normal limits" from a statistical
standpoint, these average scores clearly represent a
loss in functioning for that individual. The impact
of this change in functioning will ultimately depend
on a variety of factors, including the interaction of
the person’s constellation of cognitive strengths and
weaknesses with the variety of occupational and environmental
demands present in his/her life.
To illustrate, "average"
ability in the area of information processing speed
may be adequate for a person employed in a position
that is repetitive or where rapid appraisal, integration
and utilization of new information is not required.
It may not be adequate, however, for someone working
as an air traffic controller or a trauma surgeon. Because
of the limitations evident in utilizing a strictly normative
approach, it is important that the neuropsychologist
also gather extra-test information in order to make
an estimation of the individual’s premorbid level of
cognitive functioning against which to compare current
test performance.
Methods of Premorbid
Estimation
Vanderploeg (1994) divides
the methods of estimating premorbid functioning into
four basic approaches: one based on historical data
reflecting past achievements, two based on an individual’s
post-injury test performance and one based upon demographic
information.
The first of these methods
involves information collected through clinical interview
and review of available records. Typically, the clinical
interview will cover such areas as educational achievements
and work accomplishments; marriage, family and social
functioning; and medical and psychological histories.
While this information is often based upon self-report,
which should be corroborated, it provides the neuropsychologist
with the individual’s perspective on how his/her life
has changed as a result of the injury. Given that it
is not unusual for persons with brain injury to lack
awareness of and insight into their cognitive and psychological
difficulties, it is extremely beneficial and frequently
crucial to speak with others who interact with the individual
on a regular basis. Family members, co-workers and personal
friends can provide valuable supplemental and corroborative
information on how the individual’s functioning has
changed since the TBI, as well as provide information
on the individual’s level of awareness and insight.
Historical records, such
as school transcripts, military C-file and employment
records, particularly job performance evaluations, are
helpful in establishing the capabilities the individual
had prior to the injury (Kay, 1992; Matarazzo, 1990;
McCaffrey et al., 1993; Sbordone, 1991). Academic records,
such as school transcripts and results from standardized
achievement tests, are another source of valuable information
in establishing an individual’s premorbid level of cognitive
ability. The patterns of performance present within
an individual’s grades may reflect prior cognitive strengths
or weaknesses. Caution should be exercised when reviewing
school records. While good grades and high scores on
academic tests are generally associated with average
to above average cognitive ability, poor performance
on these same measures does not necessarily reflect
lower premorbid functioning. Other factors such as motivation,
attitudes towards academic pursuits, the lack of a stable
supportive home environment and other life events can
result in actual academic achievement being lower than
intellectual capability.
In reviewing records
pertaining to work performance, the neuropsychologist
attempts to determine the cognitive abilities that were
required for the successful performance of a particular
person. If the records indicate a history of successful
work performance, then this information provides some
indication of the types of capabilities the individual
may have possessed. Military records often can provide
similar job skills related information. Ideally, all
of the collected historical data would converge to present
a unified picture of how the individual functioned prior
to their TBI. Some of the difficulties in estimating
premorbid intellectual ability from historical data
are the myriad of factors, other than intellectual ability,
that influence an individual’s life achievements and
the more subjective nature of the information being
collected. Standardized achievement test results can
assist greatly in making the above estimation process
more objectively based. Unfortunately, obtaining such
information is more the exception rather than the rule.
The remaining three methods
of estimation attempt to address the need for a more
objectively based procedure. Two of these methods make
use of current, post-injury neuropsychological test
results to estimate an individual’s prior level of cognitive
functioning (Ciplotti & Warrington, 1995; Lezak, 1995).
The first of these is referred to as the "hold" method
of estimation. This approach is named from the deterioration
quotient devised by Wechsler (1958). In this approach,
the neuropsychologist makes use of test scores which
are resistant to brain injury and frequently well preserved
(thus the name "hold" tests) in order to estimate prior
level of functioning. Abilities such as vocabulary (Yates,
1956) and reading (Nelson, 1978) are examples of cognitive
functions, shown to be more resilient to disrupting
effects of brain injury, that are typically used to
predict premorbid intellectual ability.
As Ciplotti and Warrington
(1995) point out, however, the use of these abilities
to predict prior functioning may be problematic in cases
where the individual has a long-standing learning problem
such as dyslexia or where there exists damage to the
parts of the brain directly involved in the performance
of those functions. For example, performance on tests
of reading or vocabulary would likely underestimate
the prior ability of an individual who sustained left
temporal lobe damage resulting in aphasia. In this instance,
the estimation of the individual’s premorbid abilities
would be skewed by a deficiency that was result of the
injury. It also should be pointed out that implicit
in the "hold" test method is the assumption that a single
level of performance accurately reflects an individual’s
pre-injury state. Research has shown that this assumption
may not necessarily be true, at least not for segments
of the population at the extremes of intellectual functioning
(Matarazzo et al., 1988). While normative data do indicate
significant variability in performance (i.e. strengths
and weaknesses) across a range of cognitive abilities,
research also has shown that a high degree of variability
in performance may reflect the presence of brain impairment
(Black, 1974; Simpson & Vega, 1971).
The second of these methods
utilizing current test performance to estimate premorbid
cognitive ability, the "best performance method" is
an offshoot of the "hold" test approach. Rather than
using tests which are pre-determined to be resistant
to brain injury, the entire testing record is reviewed
and the highest scores, in combination with historical
data, are used to make predictions concerning the individual’s
prior level of functioning (Lezak, 1995). While at times
providing useful information, caution should be exercised
using this approach in that I often results in an overestimation
of pre-injury level of cognitive ability (Vanderploeg,
1994).
The last of the four
methods employs an actuarial approach to the estimation
of pre-injury ability, utilizing known relationships
between demographic variables and performance on intelligence
testing. Variables such as age, education and occupation
are put into one of a variety of regression formulas
that exist in the research literature to yield a predicted
"IQ" score. This score is then compared to the individual’s
actual test performance. In the event that current performance
levels are significantly below predicted performance,
such discrepancies may be suggestive of impaired cognitive
functioning. There are numerous studies utilizing the
actuarial approach to predict IQ scores (Barona, Reynolds
& Chastain, 1984; Eppinger et al., 1987; Karzmark et
al., 1985; Wilson, Rosenbaum & Brown, 1979) and performance
on other neuropsychological measures (Karzmark et al.,
1984). The actuarial approach has its limitations, particularly
when the individual’s premorbid functioning lies at
the extremes of ability where there exists much less
accuracy in prediction (Ciplotti & Warrington, 1995;
Sweet, Moberg & Tovian, 1990).
Discussion
Clearly, the preferred
method of determining alteration in functioning would
be to reassess the individual on a battery of standardized
tests that encompass a sufficient range of cognitive
abilities administered prior to the injury and then
examine any differences in performance. Unfortunately,
this possibility rarely exists. As a result, a number
of approaches have been developed over the years which
neuropsychologists use to estimate a premorbid level
of functioning against which to compare test results.
It should be emphasized that none of these methods are
without criticism. Consequently, it is ill advised to
rely too heavily on any single method. While the above
methods have their limitations, when used in combination
they can provide fairly accurate estimations of premorbid
functioning that allow the neuropsychologist to better
determine what changes in cognitive function may have
occurred as a result of the individual’s brain injury.
In conclusion, the following
example is given to illustrate the combined use of the
above methods. An individual who was involved in a motor
vehicle crash is referred for a neuropsychological evaluation
to determine the nature and extent of any cognitive
impairment. Prior to the injury, this person had graduated
college with excellent grades and had established a
successful work history as a financial broker. Since
the accident, she reports experiencing difficulties
with forgetfulness, inattention, distractibility and
irritability that are detrimentally affecting her work
performance. The results of all the neuropsychological
testing reveal that the vast majority of scores fall
around the upper end of the average to high average
range (75th percentile). None of the scores fall into
what is commonly considered the impaired range. If a
strictly normative interpretive approach were applied
to the data, the results would be considered reflective
of normal cognitive functioning. Upon closer inspection,
however, the neuropsychologist notes that tests of attention
and vigilance, speed of information processing, new
verbal learning and executive functioning generally
fall in the low average to the low end of the average
range (15th to 30th percentiles). Further, school grades,
standardized test scores such as the Scholastic Aptitude
Test (SAT) and several actuarial methods of premorbid
IQ estimation all suggest premorbid functioning falling
within the high average to superior range of ability.
Utilizing this information as the standard of comparison
suggests that this individual is experiencing diminished
functioning in attention, memory, processing speed and
executive functioning relative to her other cognitive
abilities – a pattern consistent with the known sequelae
of TBI.
Estimating an individual’s
pre-injury level of cognitive functioning is a complex
but essential part of the neuropsychological evaluation.
As in the above example, going from a high average/superior
level of ability to an average level of ability reflects
a loss of cognitive function which can have dire effects
for an individual’s wage earning potential and his/her
overall quality of life. Proper estimation of premorbid
functioning ensures more accurate diagnosis of cognitive
changes, which ultimately increases the chances of individuals
with brain injury getting the services they need.
Richard A. Lanham, Jr.,
PhD, is the clinical and research neuropsychologist
for the Traumatic Brain Injury Clinic at the Minneapolis
VA Medical Center and a co-investigator in the defense
and Veterans Head Injury Program (DVHIP). He holds a
clinical assistant professor appointment with the Department
of Psychology at the University of Minnesota and is
in private practice with Minnesota Clinical and Neuropsychological
Associates, P.A., Bloomington, MN.
Thomas Misukanis, PhD,
is a clinical neuropsychologist in private practice
with Minnesota Clinical and Neuropsychological Associates,
P.A., Bloomington, MN. He also serves as an adjunct
professor in the Department of Psychology at the University
of Saint Thomas in St. Paul, MN.
References
Barona A, Reynolds CR
& Chastain R: A Demographically based index of premorbid
intelligence for the WAIS-R. Journal of Consulting and
Clinical Psychology. 52(5):885-887, 1984.
Black FW: Cognitive effects
of unilateral brain lesions secondary to penetrating
missile wounds. Perceptual and Motor Skills. 38:387-391,
1974.
Ciplotti L & Warrington
E: Neuropsychological assessment. Journal of Neurology,
Neurosurgery and Psychiatry. 58:655-664, 1995.
Eppinger MG, Craig PL,
Adams RL et al.: The WAIS-R index for estimating premorbid
intelligence: Cross validation and clinical utility.
Journal of Consulting and Clinical Psychology. 55(1):86-90,
1987.
Grafman J, Jonas BS,
Martin A et al.: Intellectual function following penetrating
head injury in Vietnam veterans. Brain. III:169-184,
1988.
Karzmark P, Heaton RK,
Grant I et al.: Use of demographic variables to predict
overall performance on the Halstead-Reitan battery.
Journal of Consulting and Clinical Psychology. 52:663-665,
1984.
Karzmark P, Heaton RK,
Grant I et al.: Use of demographic variables to predict
full scale IQ: A replication and extension. Journal
of Clinical and Experimental Neuropsychology. 7(4):
412-420, 1985.
Kay T: Neuropsychological
diagnosis: Disentangling the multiple determinants of
functional disability after mild traumatic brain injury.
Physical Medicine and Rehabilitation. 6:109-127, 1992.
Lezak M: Neuropsychological
Assessment, (3rd ed.). New York: Oxford University Press,
1995.
Matarazzo JD, Daniel
MH, Prifitera A et al.: Inter-subtest scatter in the
WAIS-R standardization sample. Journal of Clinical Psychology.
44:940-950, 1988.
Matarazzo JD: Psychological
assessment versus psychological testing: Validation
from Binet to the school, clinic and courtroom. American
Psychologist. 45(9):999-1017, 1990.
McCaffrey RJ, Williams
AD, Fisher JM et al.: Forensic issues in mild head injury.
Journal of Head Trauma Rehabilitation. 8(3):38-47, 1993.
Nelson HE & O’Connell
A: Dementia: The estimation of premorbid intelligence
levels using the new adult reading test. Cortex. 14:234-244,
1978.
Sbordone R: Neuropsychology
for the Attorney. Orlando, FL: Deutsch, 1991.
Simpson CD & Vega A:
Unilateral brain damage and patterns of age-corrected
WAIS subtest scores. Journal of Clinical Psychology.
27:204-208, 1971.
Sweet JJ, Moberg PJ and
Tovian SM: Evaluation of the Wechsler Adult Intelligence
Scale-Revised premorbid IQ formulas in clinical populations.
Psychological Assessment: A Journal of Consulting and
Clinical Psychology. (2(1):41-44, 1990.
Vanderploeg RD: Estimating
premorbid level of functioning. In: Clinician’s Guide
to Neuropsychological Assessment. RD Vanderploeg (Ed.).
New Jersey: Lawrence Erlbaum Associates, 1994.
Wechsler D: The Measurement
and Appraisal of Adult Intelligence (4th ed.). Baltimore:
Williams and Wilkins, 1958.
Wilson RS, Rosenbaum
G & Brown G: The problem of premorbid intelligence in
neuropsychological assessment. Journal of Clinical Neuropsychology.
1(1):49-53, 1979.
Yates AJ: The use of
vocabulary in the measurement of intellectual deterioration.
Journal of Mental Science. 102:409-440, 1956.
