CH A P T E R

7
Screening and Assessment Tools

and not absolute measures, and rating scales and test

7A. instruments are typically used to compare a child to
Measurement and a standardized, reference group of other children.
Approximately 5% of the general population obtains
Psychometric Considerations scores that fall outside the range of “normal.” However,
the range of normal is descriptive, not diagnostic: it
GLEN P. AYLWARD ■ TERRY STANCIN describes problem-free individuals, but does not
provide a diagnosis for them.3 No test is without error,
and scores may fall outside the range of normal simply
In a general pediatric population, practitioners can as a result of chance variation or issues such as refusal
expect 8% of their patients to experience significant to take a test. Three major sources of variation that
developmental or behavioral problems between the may affect test data include characteristics of a given
ages of 24 and 72 months, this rate increasing to 12% test, the range of variation among normal children,
to 25% during the fi rst 18 years.1,2 Therefore, consid- and the range of variation among children who have
eration and interpretation of tests and rating scales compromised functioning.
are part of the clinician’s day-to-day experience, Selection of which test to use depends on the refer-
regardless of whether the choice is made to adminis- ral questions posed, as well as time and cost con-
ter evaluations or review test or rating scale data straints. Testing results vary in terms of levels of
obtained by other professionals. detail, complexity, and defi nitiveness of fi ndings. The
This chapter is an introduction to the section on fi rst level of testing is screening, the results of which
assessment and tools. It contains topics such as: dis- are suggestive. The second level is administration of
cussion of descriptive statistics (e.g., mean, median, more formal tests designed to assess development,
mode), distributions of scores and standard devia- cognition, achievement, language, motor, adaptive, or
tions, transformation of scores (percentiles, z-scores, similar functions, the results being indicative. The
T-scores), psychometric concerns (sensitivity, speci- third tier involves administration of test batteries to
ficity, positive and negative predictive values), test assess various areas and abilities; these results are
characteristics (reliability, validity), and age and assumed to be defi nitive. This third tier typically
grade equivalents. Many of these topics are also includes a combination of formal tests or test batter-
elaborated in greater detail in subsequent chapters ies, history, interview, rating scales, and observations.
of this text. A more thorough discussion of psycho- The primary goal of more detailed testing is to delin-
logical assessment methods can be found in Sattler’s eate patterns of strengths and weaknesses so as to
text.3 provide a diagnosis and guidance for intervention and
Developmental and psychological evaluations placement purposes. Results gain meaning through
usually include measurement of a child’s develop- comparison with norms. A caveat is that tests differ
ment, behavior, cognitive abilities, or levels of achieve- markedly in their degree of accuracy.
ment. Comprehensive child assessments involve a In general, regardless of whether a measurement
multistage process that incorporates planning, col- tool is designed to be used as an assessment or a
lecting data, evaluating results, formulating hypoth- screening instrument, the normative sample on which
eses, developing recommendations, and conducting the test is based is critical. Test norms that are to be
follow-up evaluations.3 Test data provide samples of applied nationally should be representative of the
behavior, with scores representing measurements of general population. Demographics must proportion-
inferred attributes or skills. These scores are relative ately reflect characteristics of the population as a
123
124 DEVELOPMENTAL AND BEHAVIORAL PEDIATRICS: EVIDENCE AND PRACTICE

whole, taking into account factors such as region accurate comparison of the child’s data and those of
(e.g., West, Midwest, South, Northeast), ethnicity, the normative group.
socioeconomic status, and urban/rural setting. If a A major issue facing users of SNRAs is identifica-
test is developed with a nonrepresentative popula- tion of the question to be answered from the results
tion, characteristics of that specific sample may bias of testing. One of two contrasting questions is proba-
norms and preclude appropriate application to other bly the reason for testing: (1) How does this child
populations. Adequate numbers of children need to compare with his or her referent group? or (2) What
be included at each age across the age span evaluated are the limits of the child’s abilities, regardless of
by a given test so as to enhance stability of test scores. comparison to a referent group? SNRAs are suited to
Equal numbers of boys and girls should be included. answer the fi rst question. Examiners can subsequently
Clinical groups should also be included for compari- test limits or alter procedures to clarify clinical issues
son purposes. Convenience samples, or those obtained such as strengths and weaknesses after the standard
from one geographic location are not appropriate for administration is completed. However, these data,
development of test norms. although clinically useful, should not be incorporated
Tests generally need to be reduced and refi ned by into the determination of the test score because of the
eliminating psychometrically poor items during the reasons cited previously. Also, no single SNRA in
development phase. Conventional item analysis is one isolation can provide all the answers regarding a
such approach and involves evaluation of an item dif- child’s development or cognitive status; rather, it is a
ficulty statistic (percentage of correct responses) and component of the overall evaluation.
patterns of responses. The use of item discrimination Use of SNRAs is not universally endorsed, particu-
indexes (item-total correlations) and item validity larly with regard to infant assessment, because of
(discrimination between normative and special concerns regarding one-time testing in an unfamiliar
groups, by T-tests or chi square analyses) is routine. environment, different objectives for testing, and
More recent tests such as the Bayley Scales of Infant linkage to intervention, instead of diagnosis. There-
and Toddler Development–Third Edition (BSID-III)4 fore, emphasis is placed on alternative assessments
or the Stanford-Binet V5 employ inferential norming6 that rely on criterion-referenced and curriculum-based
or item response theory.7 Item response theory analy- approaches. In actuality, curriculum-based assess-
ses involve difficulty calibrations for dichotomous ment is a type of a criterion-referenced tool. These
items and step differences for polychotomous items, assessments can help to answer the second question
the goal being a smooth progression of difficulty posed previously and could also better delineate the
across each subtest (e.g., as in the Rasch probabilistic child’s strengths. Both provide an absolute criterion
model8). Item bias and fairness analysis are also com- against which a child’s performance can be evaluated.
ponents; this procedure is called differential item In criterion-referenced tests, the score a child obtains
functioning.9 See Roid5 or Bayley4 for a more detailed on a measurement of a specific area of development
description of these procedures. reflects the proportion of skills the child has mastered
in that particular area (e.g., colors, numbers, letters,
shapes). For example, in the Bracken Basic Concepts
STANDARDIZED ASSESSMENTS Scale—Revised,11 in addition to norm-referenced
scores, examiners can also determine the percentage
Standardized norm-referenced assessments (SNRAs) of mastery of skills in the six areas included in the
are the tests most typically administered to infants, School Readiness Composite. More specifically, in the
children, and adolescents. The most parsimonious colors subtest, the child is asked to point to colors
defi nition of SNRAs is that they compare an individ- named by the examiner. This raw score can be con-
ual child’s performance on a set of tasks presented in verted to a percentage of mastery, which is computed
a specific manner with the performance of children regardless of age. Similarly, other skills such as knowl-
in a reference group. This comparison is typically edge of numbers and counting or letters can be gauged.
made on some standard metric or scale (e.g., scaled In curriculum-based evaluations, the emphasis is on
score).10 Although there may be some allowance for specific objectives that are to be achieved, the poten-
flexibility in rate and order of administration proce- tial goal being intervention planning.12,13 The Assess-
dures (particularly in the case of infants), administra- ment, Evaluation, and Programming System for
tion rules are precisely defi ned. The basis for Infants and Children14 and the Carolina Curricula for
comparison of scores is that tasks are presented in the Infants and Toddlers with Special Needs15 are exam-
same manner across testings, and there are existing ples of curriculum-based assessments. Therefore,
data that represent how similar children have per- SNRAs, criterion-referenced tests, and curriculum-
formed on these tasks. However, if this format is based tests each have a role, depending on the intended
modified, additional variability is added, precluding purpose of the evaluation.
 CHAPTER 7 Screening and Assessment Tools 125

FIGURE 7A-1. The normal dis- Percentage of normal 0.1 2.5 13.5 34 34 13.5 2.5 0.1
tribution. distribution
Standard deviation –3 –2 –1 x +1 +2 +3
Deviation IQ (SD = 15) 55 70 85 100 115 130 145
T-score 20 30 40 50 60 70 80
Z-score –3 –2 –1 0 +1 +2 +3
Percentile 0.1 2 16 50 84 98 99.9
Stanine 1 2 5 8 9

PRIMER OF TERMINOLOGY tion, the interquartile range may be more useful: The
USED TO DETECT DYSFUNCTION distribution of scores is divided into four equal parts,
and the difference between the score that marks the
The normal range is a statistically defi ned range of 75th percentile (third quartile) and the score that
developmental characteristics or test scores measured marks the 25th percentile (fi rst quartile) is the inter-
by a specific method. Figure 7A-1 depicts a normal quartile range.16
distribution or bell-shaped curve. This concept is criti- The standard deviation (SD) is a measure of variabil-
cal in the development of test norms and provides a ity that indicates the extent to which scores deviate
basis for the following discussion. from the mean. The standard deviation is the average
of individual deviations from the mean in a specified
distribution of test scores. The greater the standard
deviation, the more variability is found in test scores.
Descriptive Statistics In Figure 7A-1, SD = 15 (the typical standard devia-
The mean (M) is a measure of central tendency and is tion in norm-referenced tests). In a normal distribu-
the average score in a distribution. Because it can be tion, the scores of 68% of the children taking a test
affected by variations caused by extreme scores, the will fall between +1 and −1 standard deviation (square
mean can be misleading in scores obtained from a root of the variance). In general, most intelligence
highly variable sample. In Figure 7A-1, the mean and developmental tests that employ deviation quo-
score is 100. tients have a mean of 100 and a standard deviation
The mode, also a measure of central tendency, is the of 15. Scaled scores, such as those found in the
most frequent or common score in a distribution. Wechsler tests, have a mean of 10 and a standard
The median is defi ned as the middle score that deviation of 3 (7 to 13 being the average range). If a
divides a distribution in half when all the scores have child’s score falls less than 2 standard deviations
been arranged in order of increasing magnitude. It is below average on an intelligence test (i.e., IQ < 70),
the point above and below which 50% of the scores he or she may be considered to have a cognitive-
fall. This measure is not affected by extreme scores adaptive disability (if adaptive behaviors are also
and therefore is useful in a highly variable sample. In impaired).
the case of an even number of data points in a dis- Skewness refers to test scores that are not normally
tribution, the median is considered to be halfway distributed. If, for example, an IQ test is administered
between two middle scores. Noteworthy is the fact to an indigent population, the likelihood that more
that in the normal distribution depicted in Figure children will score below average is increased. This is
7A-1, the mean, mode, and median are equal (all a positively skewed distribution (the tail of the distribu-
scores = 100), and the distribution is unimodal. tion approaches high or positive scores, i.e. the right
The range is a measure of dispersion that reflects portion of the x-axis). Here, the mode is a lower score
the difference between the lowest and highest scores than the median, which, in turn is lower than the
in a distribution (highest score − the lowest score +1). mean. Probabilities based on a normal distribution
However, the range does not provide information will yield an underestimate of the scores at the lower
about data found between two extreme values in the end and an overestimate of the scores at the higher
test distribution, and it can be misleading when the end. Conversely, if the test is administered to children
clinician is dealing with skewed data. In this situa- of high socioeconomic status, the distribution might
126 DEVELOPMENTAL AND BEHAVIORAL PEDIATRICS: EVIDENCE AND PRACTICE

be negatively skewed, which means that most children average is at the 84th percentile. Clinicians must be
will do well (the tail of the distribution trails toward aware that small differences in scores in the center of
lower scores or the left portion of the x-axis). In nega- the distribution produce substantial differences in
tively skewed distributions, the value of the median percentile ranks, whereas greater raw score differ-
< mean < mode scores at the lower end will be over- ences in outliers do not have as much of an effect on
estimated, and those at the upper end will be under- percentile scores. Oftentimes, the third percentile is
estimated. Skewness has significant ramifications in considered to be a clinical cutoff (e.g., in the case of
interpretation of test scores. In fact, the meaning of a the infant born small for gestational age). Deciles are
score in a distribution depends on the mean, standard bands of percentiles that are 10 percentile ranks in
deviation, and the shape of the distribution. width (each decile contains 10% of the normative
Kurtosis reflects the shape of the distribution in group). Quartiles are percentile bands that are 25 per-
terms of height or flatness. A flat distribution, in centile ranks in width; each quartile contains 25% of
which more scores are found at the ends of the dis- the normative group. Percentiles require the fewest
tribution and fewer in the middle, is platykurtic, in assumptions for accurate interpretation and can be
comparison with the normal distribution. Conversely, applied to virtually any shape of distribution. This
if the peak is higher than the normal distribution, metric is most readily understood by parents and pro-
scores do not spread out and instead are compressed fessionals and is recommended as the preferred way
and cluster around the mean. This is called a leptokur- to describe how a child’s score compares within a
tic distribution. group of scores. For example, a Wechsler Intelligence
Scale for Children–Fourth Edition (WISC-IV) Full
Transformations of Raw Scores Scale IQ score of 70 indicates that fewer than 3% of
children of a similar age score lower on that measure
LINEAR TRANSFORMATIONS of intelligence; conversely, more than 97% of chil-
Linear transformations provide information regard- dren taking the test have a higher score.
ing a child’s standing in comparison to group means. The stanine is short for standard nine, and this metric
The z-score is a standard score (standardization being divides a distribution into nine parts. The mean = 5,
the process of converting each raw score in a distribu- and the SD = 2, with the third to seventh stanine
tion into a z-score: raw score − the mean of the dis- being considered the average range. Approximately
tribution, divided by the standard deviation of the 20% of children score in the fi fth stanine, 17% each
distribution) that corresponds to a standard devia- in the fourth and sixth stanines, and 12% each in the
tion; that is, a z-score of +1 is 1 standard deviation third and seventh stanines (78% in total). Stanines
above average and a z-score of −1 is 1 standard devia- are frequently encountered with group administered
tion below average. The mean equals a z-score of 0; tests such as the Iowa Tests of Basic Skills, the Met-
therefore scores between z-scores of −1 and +1 are in ropolitan Achievement Tests, or the Stanford Achieve-
the average range. Stated differently, if a child receives ment Tests. The interrelatedness of these scores is
a z-score of +1, he or she obtained a score higher than depicted in Figure 7A-1.
those of 84% of the population (see Fig. 7A-1).
The T-score is another linear transformation and
can be considered a z-score × 10 + 50. The mean T- PSYCHOMETRIC CONCERNS
score is 50, and the standard deviation is 10. Therefore
a z-score of 1 equals a T-score of 60. T-scores are often Appropriate interpretation of test data necessitates
found in psychopathology-related test instruments consideration of other important test characteristics.
such as the Minnesota Multiphasic Personality Inven- As mentioned previously, when a child’s norm-
tory–A, the Conners rating scales, or the Child Behav- referenced test results are interpreted, the extent to
ior Checklist, on which T-scores of 70 or greater are which the child’s characteristics are represented in the
considered to be clinically relevant (approximately normative sample from which scores were derived is a
the 98th percentile); these cutoffs are depicted in critical concern. Moreover, caution is recommended
many scoring forms. when test results for children from cultural and ethnic
minorities drive academic or clinical decisions, unless
AREA TRANSFORMATIONS there is adequate representation of this diversity in
A percentile (the technical slang is “centile”) tells the standardization samples and validation studies.
practitioner how an individual child’s performance
compares to a specified norm group. If a percentile
score is 50, half of the children tested will score above
Sensitivity and Specificity
this, and half will score below. A score that is 1 stan- Frequently, interpretation of test results must take
dard deviation below average is at approximately the into account how well the instrument performs with
16th percentile; a score 1 standard deviation above set cutoff scores. Sensitivity is a measure of the propor-
 CHAPTER 7 Screening and Assessment Tools 127

tion of children with a specific problem who are posi-

tively identified by a test, with a specific cutoff score.
Children who have a disorder but are not identified
by the test are considered to have false-negative scores.
In developmental/behavioral pediatrics, the “gold
standard” (criterion used to determine the presence
of a given problem) often is not defi nitive but rather
is a reference standard. Comparison with an imper-
fect “gold standard” may lead to erroneous conclu-
sions that a screening test is inaccurate. As a result,
sensitivity may be better conceptualized as copositiv-
ity. Desired sensitivity rates are 70% to 80%, and
sensitivity is the true positive rate of a test.
Specificity is a measure of the proportion of children
who actually are normal and who also are correctly
determined by a given test to not have a problem.
Children who are normal but who are incorrectly
determined by a test cutoff score to be delayed or
learning disabled are considered to have false-positive
scores. Specificity is the true negative rate of a test.
Again, in cases such as developmental screening, the
presence of a reference (and not “gold”) standard
makes the term conegativity more appropriate. A speci-
FIGURE 7A-2. Example highlighting sensitivity and specificity.
ficity rate of 70% to 80% is desirable. However, in the
case of screening, it is better to have a higher sensitiv-
ity rate, perhaps at the cost of lowered specificity, so
as to enhance identification of infants and children
Frequency of a Disorder/Problem
who might be at risk. Prevalence rate refers to the number of children in the
Cutoff scores can be adjusted to enhance sensitiv- population with a disorder, in relation to the total
ity. By making criteria more inclusive, fewer children number of children in the population, measured at a
with true abnormalities will be missed; however, a given time. The incidence rate indicates the risk of
more restrictive cutoff will also increase the probabil- developing a disorder: namely, new cases of a problem
ity of false-positive fi ndings (overidentifying “normal” that develop over a period of time. The relationship
children as being abnormal). Conversely, if the cutoff between incidence and prevalence can best be illus-
score is made more exclusive to enhance specificity, trated by the following: prevalence rate = the inci-
the number of normal children inaccurately identi- dence rate × the duration of the disorder. In essence,
fied as abnormal is decreased, but some of those who the predictive value of screening takes into account
are truly abnormal will be erroneously called normal sensitivity and specificity of the screening procedure
(false-negative fi ndings). Sensitivity and specificity and the prevalence of the disorder.
are described in Figure 7A-2. Base rate is the naturally occurring rate of a given
Positive predictive value refers to the proportion of disorder. For example, the base rate of learning dis-
children with a positive test result who actually are abilities would be much higher in children referred
delayed or learning disabled. This reflects the proba- to a learning and attention disorders clinic than in
bility of having a problem when the test result is the general population. If a screening instrument
positive. The lower the prevalence of a disorder, the were used to detect learning disabilities for this group,
lower is the positive predictive value. Sensitivity may sensitivity and specificity values would differ from
be a better measure in low-prevalence problems. In those found in the general pediatric population. For
developmental screening, positive predictive values example, in the follow-up of low-birth-weight infants,
often are in the range of 30% to 50%. the base rate for major handicaps (moderate to severe
Negative predictive value refers to the proportion of mental retardation; cerebral palsy; epilepsy; deafness
children with a negative test result who indeed do not or blindness) is 15%; therefore, in 85% of this popula-
have developmental delays or learning problems. It tion, the fi ndings would be true negative. Low base
is the probability of not having the disorder when rates increase the possibility of false-positive results.
the test result is negative. This value is influenced by High base rates do not leave much room for improve-
the frequency or prevalence of a problem; in low- ment in terms of locating true-positive scores and
prevalence problems, specificity may be a better result in an increase in false-negative fi ndings. Tests
measure. can be most helpful in decision making when the base
128 DEVELOPMENTAL AND BEHAVIORAL PEDIATRICS: EVIDENCE AND PRACTICE

rate is in the vicinity of 0.50. Therefore, particularly in scores. Interrater reliability refers to how well inde-
in the case of screening, the relatively low base rates pendent examiners agree on results of a test. Alternate
of developmental problems in very young children forms involve use of parallel tests, so as to prevent
may increase the probability of false positive fi ndings. carryover (score inflation) if the parallel test is admin-
However, in such situations, this scenario is more istered soon after the fi rst. For example, the Peabody
desirable than the converse: false negative fi ndings. Picture Vocabulary Test–III has two forms, as does the
Relative risk provides an alternative strategy for Wide Range Achievement Test–4.
evaluating test accuracy.17,18 This approach involves Reliability is affected by test length (longer tests are
use of the likelihood ratio, which indicates the more reliable), test-retest interval (longer interval
increased probability that the child will display a lessens reliability), variability of scores (greater vari-
developmental problem, if the results of an earlier ance increases reliability estimate), guessing (increased
screening test were abnormal or suspect. This approach guessing decreases reliability), variations in test situ-
recognizes that not all children at early risk will later ation, and practice effects.3
manifest a developmental problem, but there is a
greater likelihood that they will. If a problem or dis- VALIDITY
order is rare, relative risk and odds ratios are nearly Validity refers to whether a test measures what it is
equal. supposed to measure for a specific purpose. A test
may be valid for some uses and not others. For
example, the Peabody Picture Vocabulary Test–III
Test Characteristics may be a valid measure of receptive vocabulary, but
it is not a valid measure of overall cognitive ability or
RELIABILITY even overall language ability. It is important to keep
Measurement is the ability to assign numbers to indi- in mind that test validation is context specific. In
viduals in systematic ways as a means of inferring order to determine whether an assessment method
properties of these individuals. Reliability refers to is “psychometrically sound” or “valid,” the clinician
consistency or accuracy in measurement. Reliability must consider how it is being used. For example, an
focuses on how much error is involved in measure- intelligence test may be a valid method for determin-
ment or how much an obtained score varies from the ing a child’s cognitive abilities but may have limited
“true score.” An observed test score = true score + validity for treatment design and planning (see previ-
measurement error. Internal consistency is a measure of ous discussion of SNRAs). Similarly, a test may have
whether all components of a test evaluate a cohesive demonstrated evidence as a valid measure of severity
construct or set of constructs (e.g., verbal ability or of general anxiety but not of phobias; a certain behav-
visual-motor skills). Stated differently, high internal ior rating scale may be valid as a measure of current
consistency means that all items are highly intercor- clinical symptoms but may not have validity for treat-
related. This is measured with Cronbach’s alpha, split- ment planning or for predicting outcomes. Thus, the
half reliability, or the Kuder-Richardson reliability purpose of the assessment needs to be considered in
estimate. Cronbach’s alpha is used to evaluate how order to properly evaluate the psychometric charac-
individual items relate to the test as a whole (intercor- teristics of an assessment method.
relation among items); split-half reliability relates half Content validity determines whether the items in the
of the test items to the remaining half, often by an test are representative of the domain the test purports
odd-even split; and the Kuder-Richardson reliability to measure: that is, whether the test does cover the
estimate is used for dichotomous (i.e., “yes”/“no”) material it is supposed to cover. Construct validity
items. Test-retest reliability is particularly pertinent in concerns whether the test measures a particular
developmental and psychological testing because it psychological construct or trait (e.g., intelligence).
takes into account the “true score” and error, address- Criterion-related validity involves the current relation-
ing whether the same score would be obtained if a ship between test scores and some criterion, such
specific test were readministered. The length of time as results of another test. Criterion-related validity
between the two administrations of the test is critical can be concurrent (convergent) or predictive. In both
in regard to this measurement; that is, the sooner the instances, the results of a test under consideration are
test is readministered, the greater the reliability esti- compared to an established reference standard to
mate is. In general, test-retest correlations of 0.70 are determine whether fi ndings are comparable. In con-
considered moderate, 0.80 moderate to high, and current validity, the two tests (e.g., a screen such as
0.90, high (scores >0.85 are desirable, although the Bayley Infant Neurodevelopmental Screener and
explicit, evidence-based criteria have not been defi ned a “reference standard” such as the BSID-II) are
yet). Tests with more items tend to have higher reli- administered at the same time, and the results are
ability, because of the likelihood of a greater variance correlated. With predictive validity, a screening test
 CHAPTER 7 Screening and Assessment Tools 129

might be given at one time, followed by administra- whether age or grade norms were used to obtain
tion of the reference standard at a later date (e.g., the standard scores. For example, if age norms are used
BSID-II is given to children aged 36 months, and the and the child had been retained in grade, he or she
Wechsler Preschool and Primary Scales of Intelli- would be at a significant disadvantage because he or
gence–III at age 41/2 years). Discriminant validity shows she would not have been exposed to the more
how well a screening test detects a specific type of advanced material. Conversely, if a child failed second
problem. For example, autism might be the condition grade and is being tested in early fall while repeating
of concern, and a screening test such as the Modi- second grade, he or she may receive inflated scores if
fied Checklist for Autism in Toddlers (M-CHAT) is grade norms are used.
used to distinguish children with this disorder from The IQ/DQ ratio (developmental quotient) is com-
those with mental retardation without autism. Face puted as mental age (obtained by the use of a test
validity involves whether the test appears to measure score) ÷ the child’s chronologic age and then multi-
what it is supposed to measure. Test-related factors plied by 100. Although developmental age refers to a
(examiner-examinee rapport, handicaps, motiva- level of functioning, DQ reflects the rate of develop-
tion), criterion-related factors, or intervening events ment.19 IQ/DQ ratio scores are not comparable at dif-
could affect validity. ferent age levels because the standard deviation
With regard to the interrelatedness among reliabil- (variance) of the ratio does not remain constant. As
ity and validity, reliability essentially sets the upper a result, interpretation is difficult, and these scores
limit of a test’s validity, and reliability is a necessary generally are not used very much in contemporary
but not sufficient condition for valid measurement. A standardized testing. Instead, the deviation IQ/DQ is
specific test can be reliable, but it may be invalid when employed. The deviation IQ is a method of estimation
used to evaluate a function that it was not designed that allows comparability of scores across ages and is
to measure. However, if a test is not reliable, it cannot used with most major psychological and developmen-
be valid. Stated differently, all valid tests are reliable, tal test instruments. The deviation IQ/DQ is norm
unreliable tests are not valid, and reliable tests may referenced and normally distributed, with the same
or may not be valid.19 standard deviation; typically, M = 100 and SD = 15.
Practitioners should also be cognizant of the fact Therefore, a deviation IQ of 85 obtained at age 6
that testing can involve a speed test, in which items should have the same meaning as a score of 85
are relatively easy but there is a specific time limit obtained at age 9.
and it is difficult to answer all of the items. The infa- The standard error of measurement (SEM) is an esti-
mous 2-minute math test is an example. A power test mate of the error factor in a test that is the result of
involves progressively more difficult items, this diffi- sampling or test characteristics, taking into account
culty being determined by the limits of a child’s the mean, standard deviation, and size of the sample.
knowledge base. The larger the standard error of measurement, the
greater the uncertainty associated with a given child’s
score. The SEM is produced by multiplying the stan-
Age and Grade Equivalents dard deviation of the test by the square root of (1− the
Age- and grade-equivalent scores are based on raw reliability coefficient of the test). In 95% of cases, the
scores and portray the average age or grade placement interval of approximately two times (1.96) the SEM
of children who obtained a particular raw score. above or below a child’s score would contain the
Although these metrics are useful in explaining “true” score: a 95% confidence interval. Stated differ-
results to parents and make conceptual sense, age- ently, a 95% confidence interval indicates that if a test
and grade-equivalent scores are uneven units of mea- is given 100 times with different samples, scores will
surement. For example, a six-month difference in fall in this interval 95% of the time. In a 90% confi-
performance at the age of 2 years is much more sig- dence interval, an interval of 1.64× the SEM above
nificant than a 6-month lag at age 8 years. Moreover, and below a child’s score would contain the “true”
a 9-year-old with an age equivalent of 7 years is quite score. Such estimates are important in test-retest
different from a 4-year-old functioning at a 7-year age situations or in the case of a child who does not
equivalent, or an average 7-year-old. These measures receive services because of missing a cutoff score by
lack precision, and in some test manuals, the same only a few points (e.g., a WISC-IV Full Scale IQ score
standard scores can produce somewhat different age/ of 72).
grade equivalents. Both metrics assume that growth A fi nal concern is the Flynn effect,20 in which test
is consistent throughout the school year and tend to norms increase approximately 0.3 to 0.5 points per
exaggerate small differences in performance. These year, which is equivalent to a 3- to 5-point increment
measures also vary from test to test. Furthermore, per decade. This fi nding has ramifications in compari-
with achievement testing, it is necessary to know sons of scores obtained on earlier versions of tests to
130 DEVELOPMENTAL AND BEHAVIORAL PEDIATRICS: EVIDENCE AND PRACTICE

more contemporary scores (e.g., WISC-Revised to the 12. Greenspan SI, Meisels SJ: Toward a new vision for the
WISC–Third Edition or WISC-IV; BSID to BSID-II; developmental assessment of infants and young chil-
Stanford-Binet form LM to the 5th edition). Caution is dren. In Meisels SJ, Fenichel E, eds: New Visions for
warranted when the practitioner attributes a decline the Developmental Assessment of Infants and Young
Children. Washington, DC: Zero to Three: National
in scores to a loss of cognitive ability, because in actu-
Center for Infants Toddlers and Families, 1996, pp
ality this decline may be attributable to the fact that a
11-26.
newer test has mean scores that are considerably lower 13. Meisels S: Charting the continuum of assessment and
than those of an earlier version of the test (e.g., 5-8 intervention. In Meisels SJ, Fenichel E, eds: New
points).20 This issue would also have ramifications for Visions for the Developmental Assessment of Infants
children whose IQ score on an older version of a test and Young Children. Washington, DC: Zero to Three:
is in the low 70s but decreases to below the cutoff for National Center for Infants Toddlers and Families,
mild mental retardation on a newer version. 1996, pp 27-52.
Although some practitioners may administer tests, 14. Bricker D: Assessment, Evaluation and Programming
all have occasion to respond to inquiries from parents System for Infants and Children, Volume 1: AEPS Mea-
about their child’s test performance or diagnosis surement for Birth to Three Years. Baltimore: Paul H.
Brookes, 1993.
derived from testing. The physician’s role includes
15. Johnson-Martin N, Jens K, Attermeir S, et al: The
explaining test results to parents, acknowledging
Carolina Curriculum, 2nd ed. Baltimore: Paul H.
parental concerns and advocating for the child, pro- Brookes, 1991.
viding additional evaluation, or referring to other 16. Urdan T: Statistics in Plain English. Mahwah, NJ:
professionals.21 Erlbaum, 2001.
17. Frankenburg WK, Chen J, Thornton SM: Common pit-
falls in the evaluation of developmental screening tests.
J Pediatr 113:1110-1113, 1988.
REFERENCES 18. Frankenburg WK: Preventing developmental delays: Is
1. Costello EJ, Edelbrock C, Costello AJ, et al: Psychopa- developmental screening sufficient? Pediatrics 93:586-
thology in pediatric primary care: The new hidden 593, 1994.
morbidity. Pediatrics 82:415-424, 1988. 19. Salvia J, Ysseldyke JE: Assessment, 8th ed. New York:
2. Lavigne JV, Binns HJ, Christoffel KK, et al: Behavioral Houghton Miffl in, 2001.
and emotional problems among preschool children in 20. Flynn JR: Searching for justice. The discovery of IQ
pediatric primary care: Prevalence and pediatricians’ gains over time. Am Psychol 54:5-20, 1999.
recognition. Pediatrics 91:649-657, 1993. 21. Aylward GP: Practitioner’s Guide to Developmental and
3. Sattler JM: Assessment of Children, 4th ed. San Diego: Psychological Testing. New York: Plenum Medical,
Jerome M. Sattler, 2001. 1994.
4. Bayley N: Bayley Scales of Infant and Toddler Develop-
ment, Third Edition: Technical Manual. San Antonio,
TX: PsychCorp, 2005.
5. Roid GH: Stanford-Binet Intelligence Scales for Early
Childhood, Fifth Edition: Manual. Itasca, IL: Riverside,
2005.
6. Wilkins C, Rolfhus E, Weiss L, et al: A Simulation
7B.
Study Comparing Inferential and Traditional Norming
with Small Sample Sizes. Paper presented at annual
Surveillance and Screening for
meeting of the American Educational Research Asso-
ciation, Montreal, Canada, 2005.
Development and Behavior
7. Wright BD, Linacre JM: WINSTEPS: Rasch Analysis for
All Two-Facet Models. Chicago: MESA, 1999. FRANCES P. GLASCOE ■
8. Rasch G: Probabilistic Models for Some Intelligence PAUL H. DWORKIN
and Attainment Tests. Chicago: University of Chicago
Press, 1980. More than three decades have elapsed since the iden-
9. Dorans NJ, Holland PW: DIF detection and description: tification of developmental, behavior, and psycho-
Mantel-Haenszel and standardization. In Holland PW, social problems as the so-called “new morbidity” of
Wainer H, eds: Differential Item Functioning. Mahwah,
pediatric practice.1 During the ensuing years, pro-
NJ: Erlbaum, 1993, pp 35-66.
10. Gyurke JS, Aylward GP: Issues in the use of norm-
found societal change, with public policy mandates
referenced assessments with at-risk infants. Child for deinstitutionalization and mainstreaming, has
Youth Fam Q 15:6-8, 1992. further influenced the composition of pediatric prac-
11. Bracken BA: Bracken Basic Concepts Scale—Revised. tice. Studies have documented the high prevalence of
San Antonio, TX: The Psychological Corporation, developmental and behavioral issues within the prac-
1998. tice setting, including disorders of high prevalence
 CHAPTER 7 Screening and Assessment Tools 131

and lower severity such as specific learning disability, physical handicaps, particularly when improved
attention-deficit/hyperactivity disorder, and speech family functioning is a measured outcome.6 More
and language impairment, as well as problems of recently, the benefits of early intervention for chil-
higher severity and lower prevalence such as mental dren at environmental risk has also been demon-
retardation, autism, cerebral palsy, hearing impair- strated. For example, enrollment and participation of
ment, and serious emotional disturbance.2 disadvantaged children in Head Start programs con-
The critical influence of the early childhood years tribute to a decreased likelihood of grade repetition,
on later school success and the well-documented ben- less need for special education services, and fewer
efits of early intervention provide a strong rational for school dropouts.7 Detection is also supported by the
the early detection of children at risk for adverse clearer delineation of adverse influences on children’s
developmental and behavioral outcomes. Neurobio- development. For example, the effect of such diverse
logical, behavioral, and social science research fi nd- factors as low-level lead exposure and adverse parent-
ings from the 1990s, the so-called decade of the brain, infant interaction on child development has implica-
have emphasized the importance of experience on tions for early identification.
early brain development and on subsequent develop- By virtue of their access to young children and
ment and behavior and the extent to which the less their families, child health providers are particularly
differentiated brain of the younger child is particu- well positioned to participate in early identification of
larly amenable to intervention.3 children at risk for adverse outcomes through ongoing
In this chapter, we highlight links between early monitoring of development and behavior. Clinicians’
detection and early intervention. Much has been knowledge of medical and genetic factors also facili-
written on this topic and the American Academy of tates early identification of conditions associated with
pediatrics has recently revised its policy statement on developmental problems. Furthermore, through their
developmental screening. The new statement includes relationships with children and their families, pedia-
expert opinion on how to provide quality develop- tricians and other child health providers are familiar
mental surveillance (the process of incorporating with the social and familial factors that place children
medical/developmental history, knowledge of the at environmental risk. Professional guidelines empha-
family, parents’ concerns, screening test results, and size the importance of early detection by child health
clinical observation) in order to make informed deci- providers. The American Academy of Pediatrics’ Com-
sions about any needed referrals. Thus, this chapter mittee on Children with Disabilities; Medicaid’s Early
offers a review of evidence and challenges in surveil- Periodic Screening, Diagnosis, and Treatment (EPSDT)
lance and screening, reconciles both approaches, program; and Bright Futures (guidelines for health
includes a list of quality screening measures, describes supervision of infants, children, and adolescents
effective early identification initiatives, and provides developed by the American Academy of Pediatrics
suggestions for enhancing the well-child visits to and the Maternal and Child Health Bureau) all
facilitate early detection of developmental and behav- encourage the effective monitoring of children’s
ioral problems. development and behavior and the prompt identifica-
tion of children at risk for adverse outcomes.8,9 The
emphasis on the primary care practice as a compre-
hensive medical home for all children also supports
BACKGROUND the office as the ideal medical setting for developmen-
tal and behavioral monitoring.10
Early identification and intervention affords the Despite this strong rationale, results of surveys of
opportunity to avert the inevitable secondary prob- parents and child health providers demonstrate that
lems with loss of self-esteem and self-confidence that current practices widely vary and suggest the need
result from years of struggle with developmental dys- to strengthen developmental monitoring and early
function. Federal legislation, the Individuals with detection. Only about half of parents of children aged
Disabilities Education Act (IDEA) of 2004, and related between 10 and 35 months recall their children’s ever
state legislation mandate early detection and inter- having received structured developmental assess-
vention for children with developmental and behav- ments from their child health providers.11 Parents also
ioral disabilities. Surveys indicate that parents have report gaps in the discussion of development and
strong interest in promoting children’s optimal related issues with pediatric providers.12 Most pedia-
development.4,5 tricians employ informal, nonvalidated approaches to
Perhaps the most compelling rationale for early developmental screening and assessment. The major-
detection is the effectiveness of early intervention. ity of pediatricians do not incorporate within their
Researchers have documented the benefits of early practice such tools as those recommended by Bright
intervention in children with mental retardation and Futures to aid in early detection.13
132 DEVELOPMENTAL AND BEHAVIORAL PEDIATRICS: EVIDENCE AND PRACTICE

Not surprisingly, the early detection of children at such as “when your child becomes an adult, do you
risk for adverse developmental and behavioral out- think he or she will be above average, average, or
comes has proved elusive. Fewer than 30% of chil- below average?”) are also unhelpful in developmental
dren with such disabilities as mental retardation, monitoring, because parents are likely to expect
speech and language impairments, learning disabili- average functioning for children with delays and
ties, and serious emotional/behavioral disturbances predict overachievement for children developing at an
are identified before school entry.13 This lack of detec- average pace, a phenomenon dubbed the presidential
tion precludes the opportunity and benefits of timely, syndrome.15
early intervention. Although nearly half of parents During the physical examination, child health pro-
have some concerns for their child’s development or viders may interact with children by using an infor-
behavior, such concerns are infrequently elicited by mal collection of age-appropriate tasks. The lack of a
child health providers.14 standardized approach to measuring developmental
Multiple factors have been cited as barriers to effec- progress makes interpretation of children’s perfor-
tive developmental monitoring. Child health pro- mance on such tasks difficult. The reliance of child
viders report inadequate time during the office visit health providers on “clinical judgment,” based on
to deliver developmental services, including monitor- subjective impressions during the performance of the
ing and early detection. A professionally administered history and physical examination, are also fraught
developmental test (e.g., the Denver-II) cannot be with hazard. Such impressions are unduly influenced
adequately performed in a child health supervision by the extent to which a child is verbal and sociable
visit that lasts, on average, less than 20 minutes and in a setting that may be frightening, an effect likely
in which other content must be delivered. Other rec- to restrict affect and deter spontaneous demonstra-
ognized barriers include the inadequate training of tions of pragmatic language skills. Studies have docu-
child health providers and ineffective administrative mented the poor correlation between provider’s
and clinical practices, including staffi ng and record subjective impressions of children’s development and
keeping. Despite the assigning of a value to the billing the results of formal assessments. Clinical judgment
code for developmental screening (96110) by the identifies fewer than 30% of children with develop-
Centers for Medicare and Medicaid Services, reim- mental disabilities.15 The reliance on subjective
bursement for developmental services in general and impressions undoubtedly contributes to the late iden-
for developmental monitoring specifically by third- tification of children with such developmental issues
party payers remains inadequate. Health care organi- as mild mental retardation.
zations do not measure or prioritize the developmental According to research fi ndings and expert opinion,
content of child health supervision services. Further- surveillance and screening constitute the optimal
more, even if at-risk children are identified, the approach to developmental monitoring.16 As origi-
linkage of such children and their families to devel- nally described by British investigators, surveillance
opmentally enhancing programs and services is often encompasses all activities relating to the detection of
inefficient and challenging. developmental problems and the promotion of devel-
opment through anticipatory guidance during primary
care.17 Developmental surveillance is a flexible, longi-
tudinal, continuous process in which knowledgeable
DEVELOPMENTAL SURVEILLANCE professionals perform skilled observations during
child health care.17 Although surveillance is most
Currently, child health providers employ a variety of typically performed during health supervision visits,
techniques to monitor children’s development and clinicians may perform opportunistic surveillance during
behavior. History taking during a health supervision sick visits by exploring the child’s understanding of
visit typically includes a review of age-appropriate illness and treatment.18a
developmental milestones. Unfortunately, recall of The emphasis of developmental surveillance is on
such milestones is notoriously unreliable and typi- skillfully observing children and identifying parental
cally reflects parents’ prior conceptions of children’s concerns. Components include eliciting and attending
development.15 Although the accuracy in determin- to parents’ opinions and concerns, obtaining a rele-
ing the age of performing certain tasks is certainly vant developmental history, skillfully and accurately
improved by the use of diaries and records, the wide observing children’s development and parent-child
range of normal acquisition for such milestones limits interaction, and sharing opinions and soliciting input
their value in assessing children’s developmental from other professionals (e.g., visiting nurse, child
progress. Child health providers may also question care provider, preschool and school teacher), particu-
parents as to their predictions for their child’s devel- larly when concerns arise. Developmental history
opment. Predictions (typically elicited with questions should include an exploration of both risk and
 CHAPTER 7 Screening and Assessment Tools 133

protective factors, including environmental, genetic, ognition questions such as “Does your child use any
biological, social, and demographic influences, of the following words?” are more likely to yield
and observations of the child should include a care- helpful information than are such identification ques-
ful physical and neurological examination. Surveil- tions as “What words does your child say?” that rely
lance stresses the importance of viewing the child on parents’ spontaneous recall and report. Parental
within the context of overall well-being and report is likely to yield higher estimates of children’s
circumstance.17 functioning than is professional assessment. This dis-
The most critical component of surveillance is elic- crepancy is less likely to result from parental inaccu-
iting and attending to parents’ opinions and concerns. racy or exaggeration than from parents’ reports on
Research has elucidated the value of information newly emerging skills that are inconsistency de-
available from parents. Although there are several monstrated in the familiar and supportive home
ways to obtain quality information, research on environment.
parents’ concerns is voluminous. Concerns are par- Parents’ opinions and concerns must be considered
ticularly important indicators of developmental prob- within the context of cultural influences. Parents’
lems, particularly for speech and language function, appraisals and descriptions are influenced by expecta-
fi ne motor skills, and general functioning (e.g., “He’s tions for children’s normal development, and such
just slow”).15,18 Although concerns about self-help expectations vary among different ethnic groups. For
skills, gross motor skills, and behavior are less sensi- example, in a study of Latino (primarily Puerto
tive indicators of developmental functioning, such Rican), African American, and European American
opinions should serve as clinical “red flags,” mandat- mothers, Puerto Rican mothers expected personal
ing closer clinical assessment and developmental pro- and social milestones to be normally achieved at a
motion.15,18 The manner in which parental concerns later age than did the other groups, whereas fi rst steps
are elicited is important. Asking parents whether they and toilet training were expected at an older age by
have worries about their children’s development is European American mothers.23 Such differences were
unlikely to be useful, because they may be reluctant often explained by underlying cultural beliefs, values,
to acknowledge fears and interpret “development” as and childrearing practices. For example, the older age
merely reflecting physical growth. In contrast, asking for achievement of self-help skills is consistent with
parents whether they have any concerns about the the Puerto Rican concept of familismo and its emphasis
way their child is behaving, learning, and developing, on caring for children.
followed by more specific inquiry about functioning
in specific developmental domains, is more likely to
yield valid and clinically useful responses.18,19 Clini- USE OF SCREENING TOOLS
cians must be mindful of the complex relationship
between concerns and disability (some concerns are The effectiveness of developmental surveillance is
predictors of developmental status only at certain enhanced by incorporating valid measures of parents’
ages), the critical importance of eliciting concerns appraisals and descriptions of children’s development
rather than relying on parents to volunteer, and the and skilled professional observations. The process is
value of an evidence-based approach to interpreting enhanced by the periodic use of evidence-based
concerns.18,21 screening tools (meaning that measures are repeat-
Parents’ estimations are also accurate indicators of edly administered over time), including parent-
developmental status. For example, a study conducted completed questionnaires and professionally
in primary care demonstrated the extent to which administered tests. Screening tools that elicit infor-
parents’ estimates of cognitive, motor, self-help, and mation from parents may be used on a routine basis
academic skills correlate with fi ndings on develop- to supplement data gathering during health supervi-
mental assessments.22 Parental responses to the ques- sion visits, may be used periodically at select ages
tion, “Compared with other children, how old would (e.g., 9, 18, and 24 months), or may be used in a tar-
you say your child now acts?” are important indica- geted manner to further explore the significance of
tors of developmental delay, although such questions parental concerns. Similarly, professionally adminis-
are more challenging for parents than elicitations of tered screening tests may be administered periodi-
concerns.22 cally to help ensure that children do not elude early
In contrast to the limitations of parents’ recall of identification, or they may be used when concerns
developmental milestones, contemporaneous descrip- arise (so-called second-stage screening) or when
tions of children’s current skills and achievements are parents are not able to provide information.
useful indicators of developmental status. Similar to Table 7B-1 includes descriptions of screening tools
the solicitation of parental concerns, the format of that are highly accurate: that is, based on nation-
questions eliciting parental report is important. Rec- ally representative samples, fulfi lling psychometric
 134

TABLE 7B-1 ■ Developmental, Mental Health/Behavioral and Academic Screens

Developmental-Behavioral Screens Relying on Information from Parents

Age Range/
Screen Time Frame Description Scoring Accuracy Notes

Infant-Toddler Checklist for Language 6-24 months Parents complete the ITC’s 24 multiple-choice Cut-off scores at 1.25 Sensitivity: 78%
and Communication (ITC) (1998) questions in English. Reading level is 6th standard deviations Specificity: 84%
Paul H. Brookes Publishing, Inc., P.O. grade. Based on screening for delays in below in four Information about
Box 10624, Baltimore, MD 21285; language and social-emotional development domains. accuracy across
Phone: 1-800-638-3775; as the first evident symptom that a child is age ranges is not
http://www.pbrookes.com/ not developing typically. Does not screen available.
Part of Communication and Symbolic for motor milestones. Optional CD-ROM
Behavior Scales Developmental Profile ($99.95 facilitates factor scoring. The Checklist is
with CD-ROM) copyrighted but remains free for use at the
Brookes Web site.
Parents’ Evaluations of Birth to 8 years/ 10 questions eliciting parents’ concerns in Identifies children as Sensitivity ranges
Developmental Status (PEDS) 2-5 minutes English, Spanish, Vietnamese, Thai, being at low, from 74% to
(2007) Indonesian, Hmung, Somali, etc. Written moderate, or high 79% and
Ellsworth & Vandermeer Press, Ltd., at the 5th grade level. Determines when to risk for various specificity ranges
P.O. Box 68164, Nashville, TN refer, provide a second screen, educate kinds of disabilities from 70% to
37206; Phone: 615-226-4460; Fax: patient/parents, or carefully monitor and delays. 80% across age
615-227-0411; http://www.pedstest.com development, behavior/emotional, and levels.
$30.00 academic progress. Provides longitudinal
PEDS is also available online together surveillance and triage. Web site has
with the Modified Checklist of downloadable training materials.
Autism in Toddlers for electronic Computer-assisted telephone interview
records, individual parents, and versions are available.
computer-assisted telephone
DEVELOPMENTAL AND BEHAVIORAL PEDIATRICS: EVIDENCE AND PRACTICE

interviews.
PEDS: Developmental Milestones 0-95 months/ The PEDS:DM is a laminated book consisting Cutoffs tied to Sensitivity (0.75-
(PEDS:DM) (2007) 3-6 minutes of 6-8 items, one per domain (fine and performance above 0.87) and
Ellsworth & Vandermeer Press, Ltd., gross motor, receptive and expressive and below the 16th specificity (0.71-
P.O. Box 68164, Nashville, TN language, self-help, academics, and social- percentile for each 0.88) to
37206; Phone: 615-226-4460; Fax: emotional), across each age level (spanning item and its performance in
615-227-0411; http://www.pedstest.com the well-visit schedule). Can be domain. each domain.
$275.00 administered by parent report or by direct An assessment level Sensitivity (0.70-
Available electronically through elicitation. Helps comply with AAP policy version for NICU 0.94) and
[email protected] when used together with PEDS. follow-up and early specificity (0.77-
intervention, also 0.93) across age
provides age- levels
equivalent scores
and percentage of
delay.
Behavioral/Emotional Screens Relying on Information from Parents
Ages & Stages Questionnaires: 6-60 months/ Designed to supplement the ASQ, the Single cutoff scores Sensitivity ranged
Social-Emotional (ASQ:SE) (2004) 10-15 minutes ASQ:SE consists of 30 item forms (4-5 when a referral is from 71% -85%.
Paul H. Brookes, Publishers, P.O. Box pages long) for each of 8 visits between needed. Specificity from
10624, Baltimore, MD 21285; 6 and 60 months. Items focus on self- 90% to 98%.
Phone: 1-800-638-3775; regulation, compliance, communication,
http://www.pbrookes.com/ (likely to adaptive functioning, autonomy, affect,
be online soon) and interaction with people.
$125.00

Brief-Infant-Toddler Social- 12-36 months/ 42 item parent-report measure for identifying Cut-points based on Sensitivity (80% -
Emotional Assessment (BITSEA) 5-7 minutes social-emotional/behavioral problems and child age and sex 85%) in detecting
(2005) delays in competence. Items were drawn show presence/ children with
Harcourt Assessment, Inc., 19500 from the assessment level measure, the absence of social-emotional/
Bulverde Road, San Antonio, TX ITSEA. Written at the 4th-6th grade level. problems and behavioral
78259; Phone: 1-800-211-8378; Available in Spanish, French, Dutch, competence. problems and
www.harcourtassessment.com Hebrew. specificity 75%
$99.00 to 80%.

Eyberg Child Behavior Inventory/ 2-16 years of The ECBI/SESBI consists of 36-38 short Single refer/nonrefer Sensitivity 80%,
Sutter-Eyberg Student Behavior age/ statements of common behavior problems. score for specificity 86% to
Inventory. (ECBI/SESBI) (1999) 5-9 minutes More than 16 suggests referral for externalizing disruptive
Psychological Assessment Resources, behavioral interventions. Fewer than 16 problems—conduct, behavior
P.O. Box 998, Odessa, FL 33556; enables the measure to function as a attention, problems.
Phone: 1-800-331-8378; problems list for planning in-office aggression, etc.
http://www.parinc.com/ counseling and selecting handouts. The
$120.00 tools are helpful in monitoring behavioral
progress.
Ages and Stages Questionnaire 4-60 months/ Parents indicate children’s developmental Single pass/fail score Sensitivity ranged
(formerly Infant Monitoring System) 10-15 minutes skills on 25-35 items (4-5 pages) using a for developmental from 70% to 90%
(2004) different form for each well visit. Reading status. at all ages except
CHAPTER 7

Paul H. Brookes Publishing, Inc., PO level varies across items from 3rd to 12th the 4 month level.
Box 10624, Baltimore, MD 21285; grade. Can be used in mass mail-outs for Specificity ranged
Phone: 1-800-638-3775; child-find programs. In English, Spanish, from 76% to 91%.
http://www.pbrookes.com/ French.
$190.00

Developmental Screens (Relying on Directly Eliciting Skills from Children)

Bayley Infant Neurodevelopmental 3-24 months/ Uses 10-13 directly elicited items per 3-6 Categorizes via Specificity and
Screen (BINS) (1995) 10-15 minutes month age range to assess neurological cutscores, sensitivity are
The Psychological Corporation, 555 processes (reflexes and tone); performance into 75% to 86%
Academic Court, San Antonio, TX neurodevelopmental skills (movement and low, moderate, or across ages.
78204; Phone: 1-800-228-0752; symmetry) and developmental high risk in each
Screening and Assessment Tools

http://www.psychcorp.com accomplishments (object permanence, domain.

$265.00 imitation, and language).
135
 136

TABLE 7B-1 ■ Developmental, Mental Health/Behavioral and Academic Screens—cont’d

Developmental-Behavioral Screens Relying on Information from Parents

Age Range/
Screen Time Frame Description Scoring Accuracy Notes

Battelle Developmental Inventory 0-95 months/ Items (20 per domain) use a combination of Age equivalents and Sensitivity (72% to
Screening Test-II (BDIST)–2 (2006) 10-30 minutes direct assessment, observation, and cutoffs at 1.0, 1.5, 93%) to various
Riverside Publishing Company, 8420 parental interview. A high level of examiner and 2.0 SDs below disabilities;
Bryn Mawr Avenue, Chicago, IL skill is required. Well standardized and the mean in each specificity (79%
60631; Phone: 1-800-323-9540; validated. Scoring software including a PDA of 5 domains. to 88%). Accuracy
www.riversidepublishing.com application is available. English and information
$239.00 Spanish. across age ranges
is not available.
Brigance Screens-II (2005) 0-90 months/ Nine separate forms, one for each 12 month Cutoff, quotients, Sensitivity and
Curriculum Associates, Inc., 153 10-15 minutes age range. Taps speech-language, motor, percentiles, age specificity to
Rangeway Road, N. Billerica, MA readiness, and general knowledge at younger equivalent scores giftedness and to
01862; Phone: 1-800-225-0248; ages and also reading and math at older in various domains developmental
http://www.curriculumassociates.com/ ages. Uses direct elicitation and and overall. and academic
$501.00 observation. In the 0-2 year age range, can problems are
be administered by parent report. 70% to 82%
across ages.
Capute Scales: Cognitive Adaptive 0-36 months/ Measures visual-motor, expressive, and Developmental age Sensitivity: 0.21- Standardized
Test/Clinical Linguistic Auditory 6-20 minutes receptive language development. levels and 0.67 in low risk on a small,
Milestone Scale (CAT/CLAMS) Also available in Spanish and Russian. quotients. population. nonrepresentative
(2005) Sensitivity: 0.05- sample. Validated
Paul H. Brookes, Publishers, PO Box 0.88 in high risk against the Bayley
10624, Baltimore, MD 21285; populations. Scales of Infant
Phone: 1-800-638-3775; Specificity: 0.95- Development,
DEVELOPMENTAL AND BEHAVIORAL PEDIATRICS: EVIDENCE AND PRACTICE

http://www.pbrookes.com/ 1.00 in low risk largely on clinic-

$350.00 population. referred rather
Specificity: 0.82- than general
0.98 in high risk pediatric
populations). samples.
Denver-II (1992) Birth to 6 years Has four factors: personal-social, language, Produces four scores: With suspect scores Standardized in
Developmental Learning Materials, of age/ fine-motor-adaptive, and gross motor. Also abnormal, suspect, grouped with Colorado. Not
Inc., P.O. Box 371075, 10-20 minutes available in Spanish. A derivative measure, pass, untestable. passing scores: validated by the
Denver, CO 80237-5075; the Denver Prescreening Questionnaire Sensitivity: 56% authors. High
Phone: (800) 419-4729; (PDQ-II), is administered by parental Specificity: 80% reliability
www.denverii.com report and used to indicate need to With suspect scores suggests grouping
$90.00 administer the Denver-II. Training videos grouped with suspect with
are available for rent or purchase ($410). abnormal scores: abnormal scores
Publisher also offers a training institute Sensitivity: 83% is the better of
($395). Specificity: 43% the two accuracy
indices.
Developmental Indicators for the 3-0 to 6-11 Views performance in five domains: Offers five cutoff Sensitivity/ Very small sample
Assessment of Learning (DIAL-III) years/ motor, concepts, language, self-help, options (from specificity: in accuracy
(1998) 20-30 minutes and social. Scoring software is available 1.0 SDs to 2.0 SDs 50% -60% (no studies with no
American Guidance Service Inc., 4201 in both English and Spanish. Has below the mean) information on information on
Woodland Road, Circle Pines, MN training materials and administration for each domain. accuracy across which cutoffs
55014; Phone: (800) 627-7271; video available as well as parent-education Percentiles optional. age ranges). were used.
http://www.agsnet.com guides. SPEED-DIAL (3 of the 5 subtests)
$469.00 takes 10-20 minutes.
3-0 to 6-0 120 items, tapping performance in three Cutoffs indicating Sensitivity = 92% - Validation and
Early Screening Inventory-Revised years/ areas: visual-motor, language/cognitive, when to refer or 93% accuracy indices
(ESI-R) (1993) 15-20 minutes gross-motor. Onsite training and tapes are rescreen. Specificity = 80% computed against
Pearson Education, One Lake Street, available from the publisher. Computed on the 1974
Upper Saddle River, NJ 07458; performance McCarthy Scales
Phone: 201-236-7000; below/above the (normed on an
http://phcatalog.pearson.com 6th percentile. upper SES
$237.50 sample).

Family Screens
Family Psychosocial Screening (1996) Screens parents A two-page clinic intake form that identifies Refer/nonrefer scores All studies showed About 15 minutes
Kemper KJ, Kelleher KJ: Family for risk factors psychosocial risk factors including: (1) for each risk factor. sensitivity and (if interview
psychosocial screening: instruments a four item measure of parental history Also has guides to specificity to needed)
and techniques. Ambul Child Health of physical abuse as a child; (2) a six referring and larger inventories Materials ≈$.20
4:325-339, 1996 item measure of parental substance resource lists. greater than 90%. Admin. ≈$4.20
The measures are included in the abuse; and (3) a three item measure of Total ≈$4.40
article and downloadable at maternal depression.
http://www.pedstest.com

Academic Screens
Comprehensive Inventory of Basic 1st-6th grade Administration involves one or more of three Computerized or 70% to 80% Takes 10-15 minutes
Skills-Revised Screener (CIBS-R subtests (reading comprehension, math hand-scoring accuracy across Materials≈$.53
CHAPTER 7

Screener) (1985) computation, and sentence writing). Timing produces all grades. Admin. ≈$10.15
Curriculum Associates, Inc., 153 performance also enables an assessment of percentiles, Total ≈ $10.68
Rangeway Road, N. Billerica, MA information processing skills, especially rate. quotients, cutoffs.
01862; Phone: 1-800-225-0248;
http://www.curriculumassociates.com
$224.00/
Safety Word Inventory and Literacy 6-14 years Children are asked to read 29 common safety Single cutoff score 78% to 84% About 7 minutes
Screener (SWILS) (2002) words (e.g., High Voltage, Wait, Poison) indicating the sensitivity and (if interview
Glascoe FP: Clinical Pediatrics. Items aloud. The number of correctly read words need for a referral. specificity across needed)
courtesy of Curriculum Associates, is compared to a cutoff score. Results all ages. Materials ≈$.30
Inc. predict performance in math, written Admin. ≈$2.38
The SWILS can be freely downloaded language, and a range of reading skills. Test Total ≈$2.68
Screening and Assessment Tools

at: http://www.pedstest.com/ content may serve as a springboard to

injury prevention counseling.
137
 138

TABLE 7B-1 ■ Developmental, Mental Health/Behavioral and Academic Screens—cont’d

Developmental-Behavioral Screens Relying on Information from Parents

Age Range/
Screen Time Frame Description Scoring Accuracy Notes

Narrow-Band Screens for Autism and ADHD

Connors Rating Scale-Revised 3 to 17 years Although the CRS-R can screen for a range Cutoff tied to the Sensitivity: 78% to About 20 minutes
(CRS-R) (1997) of problems, several subscales specific to 93rd percentile for 92% Materials ≈$.2.25
Multi-Health Systems, Inc., P.O. Box ADHD are included: DSM-IV symptom each factor. Specificity: 84% to Admin. ≈$20.15
950, North Tonawanda, NY 14120- subscales (Inattentive, Hyperactive/ 94% Total ≈$22.40
0950; Phone: 1-800-456-3003 or Impulsive, and Total); Global Indices (GI)
1-416-492-2627; Fax: (Restless-Impulsive, Emotional Lability,
1-888-540-4484 or 1-416-492-3343; and Total), and an ADHD Index. The GI is
http://www.mhs.com/ useful for treatment monitoring. Also
$193.00 available in French.
Modified Checklist for Autism in 18-60 months Parent report of 23 questions modified for Cutoff based on Initial study shows About 5 minutes
Toddlers (M-CHAT) (1997) American usage at 4th-6th grade reading 2 of 3 critical items sensitivity at 90%; Print Materials
Free download at the First Signs Web level. Available in English and Spanish. Uses or any 3 from specificity at 99%. ≈$.10
site: http://www.firstsigns.org/ telephone follow-up for concerns. The checklist. Future studies are Admin. ≈$.88
downloads/m-chat.PDF M-CHAT is copyrighted but remains free needed for a full Total ≈$.98
($0.00) for use on the First Signs Web site. picture. Promising
tool.
Online for parents and EMRS at
DEVELOPMENTAL AND BEHAVIORAL PEDIATRICS: EVIDENCE AND PRACTICE

www.forepath.org ($1.00)

Compiled by Frances Page Glascoe, PhD, Adjunct Professor of Pediatrics ([email protected]). Copyright 2006, Glascoe FP: Collaborating with Parents. Nashville: Ellsworth & Vandermeer,
2006.
Numerous broad-band screening measures (meaning that multiple domains are measured) are listed. Several narrow-band tools essential for primary care (e.g., for ADHD and autism spectrum disorder)
are listed at the end. The left column provides publication information and the cost of purchasing a specimen set. The “Description” column offers information on what the instrument measures, what
factor or subtests are included, and administration methods. The “Scoring” column shows how scores were produced. The “Accuracy” column shows the percentage of patients with and without problems
identified correctly. Ideally, sensitivity should be at least 70%, meaning that the majority of children with disabilities are correctly detected in a one-time administration. Specificity, correct detection of
children without disabilities, should also be at least 70%. All measures, except where noted (see “Notes” column), were developed on nationally representative samples (meaning a group with geographic
and sociodemographic characteristics proportional to those found in the U.S. Census, including correct proportions of children with disabilities), have high levels of reliability (interrater, test-retest,
internal consistency), and have been validated against a range of criterion measures in general pediatric samples (because broadband screens must prove that they have validity across a range of
developmental domains and because calculation of sensitivity and specificity on referred populations is likely to be inflated).
AAP, American Academy of Pediatrics; ADHD, attention-defi cit/hyperactivity disorder; DSM-IV, Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (American Psychiatric Association, 1994);
NICU, neonatal intensive care unit; PDA, personal digital assistant; SD, standard deviation.
 CHAPTER 7 Screening and Assessment Tools 139

criteria (see Chapter 7A), and having both sensitivity identifying psychosocial risk factors and can be
and specificity of at least 70% to 80%. Two types of used as a standard intake form for new patients.
tools are presented: those relying on information 3. Elicit parents’ concerns and observations. Careful
from parents and those requiring direct elicitation of attention to wording is essential. Although facilitat-
children’s skills. The latter are useful in practices with ing conversations with parents can be informally
staff (e.g., nurses, pediatric nurse practitioners) who accomplished, several helpful sources suggest well-
have the time and skill to administer relatively worded questions. For example, Bright Futures con-
detailed screens. Such measures are also useful in tains useful trigger questions. A parent-completed
early intervention programs. Information is included measure, the Parents’ Evaluation of Developmental
on purchasing, cost, time to administer, scores pro- Status questionnaire (see Table 7B-1), has empiri-
duced, and age ranges of the children tested. cally tested wording and weighs the types of con-
cerns parents raise, assigns levels of risk, and
identifies optimal responses to concerns.
COMBINING SCREENING 4. Conduct a physical examination. Examination
should include attention to growth parameters,
AND SURVEILLANCE head shape and circumference, facial and other
body dysmorphology, eye findings (e.g., cataracts
We now present an algorithm for combining surveil-
in various inborn errors of metabolism), vascular
lance and screening into an effective, evidence-based
markings, and signs of neurocutaneous disorders
process for detecting and addressing developmental
(e.g., café-au-lait spots in neurofibromatosis,
and behavioral issues. The American Academy of
hypopigmented macules in tuberous sclerosis).
Pediatrics recently revised its policy statement on
Vision and hearing screening are essential.
early detection.8 We include the elements of the state-
5. Administer/score developmental screening tests.
ment, as follows.
Use of parent report measures, completed before
1. Review the patient’s chart for medical risk factors. the visit or in the waiting/examination room,
Take note of such potentially teratogenic exposures reduces the amount of time needed for screening.
as radiation or medications, infectious illnesses, Positive results may be followed by additional
fever, addictive substances, and trauma, and review screening of social-emotional functioning (e.g.,
results of neonatal screens, including phenylketon- Ages & Stages Questionnaires: Social-Emotional
uria, hypothyroidism, and other metabolic condi- and the Modified Checklist for Autism in Toddlers;
tions. Also consider the perinatal history, including see Table 7B-1) to better identify the areas of delay
birth weight, gestational age, Apgar scores, and and types of services needed. Note that the AAP’s
any medical complications. In addition, postnatal new statement recommends use of an autism spe-
medical factors to be considered include chronic cific screen like the M-CHAT at both 18 and 24
respiratory or allergic illness, recurrent otitis, head months, regardless of performance on broad-band
trauma, and sleep problems, including symptoms tools like PEDS or the ASQ.
of obstructive sleep apnea. 6. Provide additional medical screens when
2. Identify psychosocial risk factors. Common risk developmental-behavioral screens are positive.
factors for developmental and behavioral problems When indicated, common health-related causes for
include parents with less than a high school educa- delays and disorders should include screens for
tion, parental mental health or substance abuse iron deficiency and lead toxicity. Unless suggested
problems, four or more children in the home, by parental report (e.g., seizure activity) or clinical
single-parent family, poverty, frequent household findings (e.g., microcephaly, expanding head cir-
moves, limited social support, parental history of cumference), neurophysiological (e.g., electroen-
abuse as a child, and ethnic minority status. Four cephalogram) and neuroimaging (e.g., computed
or more risk factors are associated with develop- tomographic scan, magnetic resonance imaging)
mental performance that is well below average, studies are not routinely indicated. Developmental
which, in turn, has an adverse effect on future delay may suggest the need for metabolic screening
success in school.24 The presence of multiple risk for ammonia, organic, and amino acids (or referral
factors suggests the need for enrichment or reme- for such screens). A progressive loss of milestones
dial programs, regardless of screening results. suggests the possible need to screen for human
Examples include Head Start, after-school tutor- immunodeficiency virus (HIV).
ing, parenting training, social work services, men- 7. Explain results to parents. When parents’ concerns
toring, quality child care, and summer school. A have been elicited, the process of explaining find-
measure such as the Family Psychosocial Screen ings can begin with a simple affirmation of parents’
(available at www.pedstest.com) is often helpful for observations. It is important to present results in
140 DEVELOPMENTAL AND BEHAVIORAL PEDIATRICS: EVIDENCE AND PRACTICE

person and to maintain a positive outlook about domains associated with school success: language,
available services and their potential to improve academic/preacademic skills, and cognition. Clini-
outcome. Because screening/surveillance activities cians may provide patient education materials, lists
are not diagnostic in nature, the clinician should of informative and factual Web sites, lists of parent
avoid diagnostic labels in favor of euphemisms, training services, and contacts for social support
such as “developmental delay,” “behind other chil- programs. Group discussions for parents on devel-
dren,” and “having difficulties with. . . .” When a opmental topics are another potential strategy but
parent reports conflicting perceptions within the require careful planning and organization. Devel-
family about the possibility of problems, the clini- opmental promotion is assisted by a well-organized
cian should offer to explain findings to other family system for filing and retrieving parent-focused
members. Asking parents whether they know other materials (see www.dbpeds.org for materials and
families with children who have developmental- links).
behavioral differences may be helpful in clarifying 11. Establish a medical home. For children with
discussions. developmental-behavioral problems and/or
8. If indicated, make referrals for subspecialty medical complex health care needs, primary care contact
services. When medical factors are identified, and perspectives are of critical importance to
an appropriate response is referral for further promote optimal health and development. The
evaluation. American Academy of Pediatrics National Center
9. Seek nonmedical interventions. Nonmedical inter- of Medical Home Initiatives for Children with
ventions need not await a complete diagnosis. All Special Needs (www.medicalhomeinfo.org) provides an
children with apparent delays or disorders should essential guide for organizing practices to ensure
be referred promptly to appropriate programs and continuity of care, manage multiple referrals and
services. Public programs, including those man- comprehensive records, coordinate appointments,
dated by such legislation as the IDEA, should be and communicate with various providers.
available through community-based agencies or
the public schools without cost to the family and
generally provide a range of high-quality therapies SYSTEMWIDE APPROACHES TO
and evaluations, including speech-language, physi-
cal, and occupational therapy; assistive technology
SURVEILLANCE AND SCREENING
evaluations; and behavioral interventions. Most
State wide and countywide efforts to enhance col-
IDEA programs do not provide a detailed diagnosis
laboration among medical and nonmedical providers
but rather define functional skills and deficits. As
offer some of the most promising evidence for the
a consequence, a referral may also need to be made
effectiveness of surveillance and screening. Docu-
to a multidisciplinary diagnostic service. Because
mented outcomes include large increases in screening
such centers typically have long waiting lists and
rates during EPSDT visits;25 a fourfold increase in
because a final diagnosis is not necessary for initiat-
early intervention enrollment, resulting in a match
ing intervention, it is best to make such referrals
between the prevalence of disabilities and receipt of
concurrent with a referral to an IDEA program.
services26 ; a 75% increase in identification of children
Other services should be sought (e.g., Head Start,
from birth to age 3 with autism spectrum disorder27;
after school tutoring, quality daycare, parent train-
improvement in reimbursement for screening28 ; and,
ing) for children with psychosocial risk factors who
interestingly, increased attendance at well-child visits
do not fulfill specific eligibility requirements for
when parents’ concerns are elicited and addressed.25
early intervention or special education. Referral
Among the numerous initiatives—national, inter-
letters to programs and services should include
national, and regional—we selected a few to highlight
suggestions for the types of evaluations needed
because they employed varied models and gathered
(e.g., speech-language therapy, occupational and
outcome data to support their successes (and
physical therapy, social-emotional assessment,
challenges).
intelligence testing, academics). Programs offered
through IDEA often require documentation of
hearing and vision status. Some programs require The Assuring Better Child Health and
the completion of specific referral forms. Parental
consent should be obtained for sharing informa-
Development (ABCD) Program
tion, including copies of subsequent evaluations. Created by The Commonwealth Fund, the ABCD
10. Offer developmental promotion. Regardless of Program has identified policy strategies for state Med-
whether a child has developmental problems, icaid agencies to strengthen the delivery and fi nanc-
parents need advice and encouragement on ing of early childhood services for low-income
promoting optimal development, particularly in families. The emphasis is on assisting participating
 CHAPTER 7 Screening and Assessment Tools 141

states in developing care models that promote healthy district’s Childfi nd and Parents-as-Teachers programs,
development, including the mental development of and a parent-to-parent mentoring program for parents
young children. Models include developmental of children with special health care needs. The goal
screening, referral, service coordination, and educa- of PRIDE is earlier identification and intervention for
tional materials and resources for families and clinical children in Greenville County, South Carolina with
providers. The program has resulted in improvements developmental delays and improved support for their
in screening, surveillance, and assessment. Most parents.
notably, work in North Carolina facilitated a 75% The program has targeted key players in the lives
increase in screening, increased enrollment rates in of infants and toddlers as follows: Parents sign up
early intervention from 2.6% to 8% (in line with the around the time of their child’s birth to receive mile-
Centers for Disease Control and Prevention’s preva- stone cards every 3 to 6 months during the fi rst 3
lence projections), while simultaneously lowering years that describe the key developmental attain-
referral age26 (http://www.nashp.org; http://www.cdc. ments, activities to promote development at that age,
gov/ncbdd/child/interventions.htm). and red flags for potential developmental problems.
Parents are instructed to discuss any concerns with
their physician. Primary care physicians are provided
Help Me Grow with information and tools (the Parents Evaluation of
A program of the Connecticut Children’s Trust Fund, Developmental Status questionnaire) to improve their
Help Me Grow links children and families to com- system of developmental screening. A nurse practi-
munity programs and services by using a comprehen- tioner employed by PRIDE as the “physician office
sive statewide network. Components of the program liaison” works closely with practices, initially by
include the training of child health providers in setting up lunch meetings with physicians and staff
effective developmental surveillance; the creation of that are also attended by the PRIDE developmental-
a triage, referral, and case management system that behavioral pediatrician. With the agreement of the
facilitates access for children and families to services physicians, the liaison then assists the office staff in
through Child Development Infoline; the develop- implementing the system and provides a “Resource
ment and maintenance of a computerized inventory Guide” with information on local developmental
of regional services that address developmental and services and forms to facilitate referrals. Child care
behavioral needs of children and their families; and providers have the opportunity to attend educational
data gathering to systematically document capacity sessions (for credit hours) in which they learn about
issues and gaps in services. The program has increased child development, signs of developmental problems,
identification rates of at-risk children by child health and services that are available for these children. The
providers and increased referral rates of such children training sessions are provided in collaboration with
to programs and services. For example, chart reviews local programs that promote higher quality child care
conducted in participating practices noted an increase and early education (Success By 6 and First Steps),
in documented developmental or behavioral concerns and the attendees receive “toolkits” with information
from 9% before training to 18% after training. Fur- on the topics discussed. Initial results of the program
thermore, training resulted in significant differences indicate success; 16 of 17 local pediatric practices
in referral rates for certain conditions. Behavioral (which previously had no standardized system of
conditions were involved in 4% of referrals from developmental screening) now utilize the Parents
trained practices, in comparison with 1% from Evaluation of Developmental Status questionnaire.
untrained practices. Four percent of referrals Over the fi rst 18 months of the program, referrals to
from trained practices were for parental support and early intervention have increased almost 100% and
guidance, in comparison with fewer than 1% from un- referrals to the school’s Childfi nd program by 30%.
trained practices29 (http://www.infoline.org/Programs/ Other service providers have seen increases in new
helpmegrow.asp). referrals of up to 30%. The average age at referral to
early intervention has also dropped slightly.
Not surprisingly, increasing rates of referral raised
Promoting Resources in Developmental the likelihood of even longer waiting lists for tertiary-
level developmental-behavioral pediatric evaluations.
Education (PRIDE) To address this challenge, the PRIDE staff sought
This program is a 3-year project funded by the Duke funding from The Commonwealth Fund to study the
Endowment through a partnership of the Children’s feasibility and cost effectiveness of a model of “mid-
Hospital, the Center for Developmental Services (a level” developmental-behavioral pediatrics assess-
colocation of agencies serving children with develop- ment (as a step between telephone triage/record
mental disorders), the regional office of the state’s review and comprehensive diagnostic evaluation) for
early intervention system (BabyNet), the local school children younger than 6 years.30
142 DEVELOPMENTAL AND BEHAVIORAL PEDIATRICS: EVIDENCE AND PRACTICE

First Signs questionnaire (the standardized developmental-

behavioral surveillance and screening instrument
This national and international training effort is that elicits parents’ concerns about their children).
devoted to early detection of children with disabili- After training, screening rates increased from 0% to
ties, with a particular focus on autism spectrum dis- 43.5% during the pilot phase. At the same time, the
orders. This detection is accomplished through a mix practices experienced a 16% increase in attendance
of print materials and broadcast press, direct mail, at scheduled well-child visits, which suggests that
public service announcements, presentations (to focusing on parents’ concerns may increase their
medical and nonmedical professionals), a richly in- adherence to visit schedules. Blue Cross/Blue Shield
formative website (www.firstsigns.org), and detailed of Tennessee, together with the Tennessee Chapter of
program evaluation. Although First Signs initiatives the American Academy of Pediatrics, is now provid-
have been conducted in several states, including New ing training across the state.25 More information can
Jersey, Alabama, Delaware, and Pennsylvania, the be found through the Center for Health Care Strate-
Minnesota campaign is highlighted here because of gies, “Best Clinical and Administrative Practices for
that state’s assistance in program evaluation. Minne- State-wide” developmental and behavioral screening
sota is divided into discrete service regions. Central- initiatives as established by the Center for Health Care
ized train-the-trainers forums were conducted to Strategies [http://www.chcs.org/]
prepare 130 professionals as outreach trainers. These
individuals were from all regions of the state, and
most were early interventionists, family therapists, Healthy Steps for Young Children
and other nonmedical service providers. They then This a national initiative improves traditional pediat-
provided more than 165 workshops to 686 medical ric care with the assistance of an in-office child devel-
providers, to whom they offered individualized train- opment specialist, whose duties include expanded
ing tailored for health care clinics, as well as training discussions of preventive issues during well-child and
for more than 3000 early childhood specialists. First home visits, staffi ng a telephone information line,
Signs Screening Kits (which include video, informa- disseminating patient education materials, and net-
tion about and in some cases copies of appropriate working with community resources and parent
screening tools, wall charts and parent handouts on support groups. Now in its 12th year, Healthy Steps
warning signs) were distributed to more than 900 followed its original cohort of 3737 intervention and
practitioners and clinics. In addition, public service comparison families from 15 pediatric practices in
announcements were aired across the state in collab- varied settings. In comparison with controls, Healthy
oration with the Autism Society of Minnesota. Within Steps families received significantly more preventive
12 months, there was a 75% increase in the number and developmental services, were less likely to be dis-
of young children identified in the 0- to 2-year age satisfied with their pediatric primary care, and had
group and an overall increase of 23% in detection of improved parenting skills in many areas, including
autism spectrum disorders among all children aged 0 adherence to health visits, nutritional practices, devel-
to 21 years in that same period. The state has now opmental stimulation, appropriate disciplinary tech-
expanded the initiative to include childcare providers niques, and correct sleeping position. In practices
and is educating them about red flags and warning serving families with incomes below $20,000, use of
signs. In addition, physicians with the Minnesota telephone information lines increased from 37%
Chapter of the American Academy of Pediatrics Com- before the intervention to 87% after; office visits with
mittee for Children with Disabilities have begun someone who teaches parents about child develop-
incorporating First Signs information into physician ment increased from 39% to 88%; and home visits
training program at the University of Minnesota.27 increased from 30% to 92%. Low-income families
receiving Healthy Steps services were as likely as
high-income parents to adhere to age-appropriate
Blue Cross/Blue Shield of Tennessee well-child visits at 1, 2, 4, 12, 18, and 24 months.31,32
Blue Cross/Blue Shield of Tennessee requested that One program evaluation suggests that Healthy Steps
child health providers use standardized, validated offers a benefit comparable with that of Head Start at
screening at all EPSDT visits. To facilitate compliance, about one-tenth the cost,33 although this claim is
Blue Cross/Blue Shield of Tennessee piloted a program somewhat premature because Head Start data now
in 34 high-volume, Medicaid-managed care prac- extend to more than 35 years of follow-up research
tices. Outreach nurses, called regional clinical network with a proven return rate of $17.00 for each $1.00
analysts, trained providers on site how to admin- spent on early intervention, with savings realized
ister, score, interpret, and submit reimbursement for through reductions in teen pregnancy, increases in
the Parents’ Evaluation of Developmental Status high school graduation and employment rates, and
 CHAPTER 7 Screening and Assessment Tools 143

decreased adjudication and violent crime.7 Neverthe- measures both overrefer and underrefer to some
less, Healthy Steps is extremely promising and inex- extent). Other rich topics of inquiry include the fol-
pensive and includes a strong evaluation component lowing: How do surveillance methods enhance devel-
that will answer questions about its long-term opment and early detection, and which specific
effect. techniques most enhance decision making? Does
improved reimbursement have a positive effect on
provider behavior? How can surveillance and screen-
CONCLUSION ing be incorporated into electronic health records?
In the absence of regional and state initiatives, can
In summary, both expert opinion and research evi- primary care professionals engage in effective self-
dence support surveillance and screening as the study and thus positive practice change? What teach-
optimal clinical practice for monitoring children’s ing methods and content best help residents master
development and behavior, promoting optimal devel- efficient surveillance and screening techniques that
opment, and effectively identifying children at risk for work well in primary care? Perhaps the most critical
delays. The effectiveness of surveillance is enhanced area in need of further inquiry is determining the
by incorporating valid measures of parents’ appraisals longitudinal outcomes of families and children when
and descriptions of children’s development and behav- surveillance and screening are used together.
ior and skilled professional observations. Develop-
mental monitoring should combine surveillance at
REFERENCES
all health supervision visits with the periodic use of
evidence-based screening tools, including parent- 1. Haggerty RJ, Roughman KJ, Pless IB: Child Health and
completed questionnaires and professionally admin- the Community. New York: Wiley, 1975.
2. Dobos AE, Dworkin PH, Bernstein BA: Pediatricians’
istered tests.
approaches to developmental problems: Has the
To be effective, identification must lead to interven- gap been narrowed? J Dev Behav Pediatr 15:34-38,
tion through referral to appropriate programs and 1994.
services. Surveillance and screening activities must 3. Institute of Medicine: From Neurons to Neighborhoods:
ensure access to medical evaluations, developmental The Science of Early Childhood Development. Wash-
assessments, and intervention programs. Ultimately, ington, DC: National Academies Press, 2000.
collaboration among health care providers, parents, 4. Blumberg SJ, Halfon N, Olson LM: The national survey
and early intervention and other social service pro- of early childhood health. Pediatrics 113:1899-1906,
viders is crucial for effectively addressing the chal- 2004.
lenges of detection and timely enrollment in early 5. Young KT, Davis K, Schoen C, et al: Listening to
intervention programs and services. parents. A national survey of parents with young chil-
dren. Arch Pediatr Adolesc Med 152:255-262, 1998.
Establishing effective surveillance and screening in
6. Shonkoff JP, Hauser-Cram P: Early intervention for
primary care is nevertheless challenging.13 Effective disabled infants and their families: a quantitative anal-
initiatives consistently offer training to providers, ysis. Pediatrics 80:650-658, 1987.
office staff, and nonmedical professionals. Implemen- 7. Shonkoff JP, Meisels SJ: Handbook of Early Childhood
tation details are numerous (e.g., incorporation into Intervention, 2nd ed. New York: Cambridge University
existing office workflow, ordering and managing Press, 2000.
screening materials, gathering and organizing lists of 8. American Academy of Pediatrics, Council on Children
referral resources and patient education handouts, with Disabilities: Identifying infants and young chil-
identifying measures that work well with available dren with developmental disorders in the medical
personnel, and determining how best to communi- home: An algorithm for developmental surveillance
cate with nonmedical providers).18,18a,26,34 Ultimately, and screening. Pediatrics 118:403-420, 2006.
9. Green M, Palfrey JS, eds: Bright Futures: Guidelines
helping health care providers recognize the need to
for Health Supervision of Infants, Children, and Ado-
adopt effective detection methods is the critical fi rst lescents, 2nd ed. Arlington, VA: National Center for
step. Education in Maternal and Child Health, 2002.
10. American Academy of Pediatrics, Medical Home Initia-
tives for Children with Special Needs Project Advisory
DIRECTIONS FOR Committee: The medical home. Pediatrics 110:184-186,
2002.
FURTHER RESEARCH 11. Halfon N, Regalado M, Sareen H, et al: Assessing devel-
opment in the pediatric office. Pediatrics 113:1926-
Although much is known about the accuracy of 1933, 2004.
screening tools, studies are needed to determine their 12. Bethell C, Reuland CHP, Halfon N, et al: Measuring the
accuracy when used repeatedly (given that screening quality of preventive and developmental services for
144 DEVELOPMENTAL AND BEHAVIORAL PEDIATRICS: EVIDENCE AND PRACTICE

young children: National estimates and patterns of cli- 28. Inkelas M, Regalado H, Halfon N: Strategies for inte-
nicians’ performance. Pediatrics 113:1973-1983, 2004. grating developmental services and promoting medical
13. Silverstein M, Sand N, Glascoe FP, et al: Pediatricians’ homes. National Center for Infant and Early Childhood
reported practices regarding developmental screening: Health Policy, 2005. (Available at: http://www.healthy
Do guidelines work? And do they help? Pediatrics child.ucla.edu; accessed 10/13/06.)
116:174-179, 2005. 29. McKay K: Evaluating model programs to support dis-
14. Inkelas M, Glascoe FP, Regalado M, et al: National semination. An evaluation of strengthening the devel-
Patterns and Disparities in Parent Concerns about Child opmental surveillance and referral practices of child
Development. Paper presented at the annual meeting health providers. J Dev Behav Pediatr 27(1 Suppl):S26-
of the Pediatric Academic Societies, Baltimore, 2002. S29, 2006.
15. Glascoe FP, Dworkin PH. The role of parents in the 30. Kelly D: PRIDE. American Academy of Pediatrics’
detection of developmental and behavioral problems. Section on Developmental and Behavioral Pediatrics
Pediatrics 95:829-836, 1995. Newsletter. March, 2006 (Available at: www.dbpeds.org;
16. Regalado M, Halfon N: Primary care services promot- accessed 10/13/06.)
ing optimal child development from birth to age 3 31. McLearn KT, Strobino DM, Hughart N, et al: Narrow-
years: review of the literature. Arch Pediatr Adolesc ing the income gaps in preventive care for young chil-
Med 12:1311-1322, 2001. dren: Families in Healthy Steps. J Urban Health
17. Dworkin PH: British and American recommendations 81:206-221, 2004.
for developmental monitoring: The role of surveillance. 32. McLearn KT, Strobino DM, Minkovitz CS, et al: Devel-
Pediatrics 84:1000-1010, 1989. opmental services in primary care for low-income chil-
18. Glascoe FP: Collaborating with Parents: Using Parents’ dren: Clinicians’ perceptions of the Healthy Steps for
Evaluations of Developmental Status to Detect and Young Children Program. J Urban Health 81:556-567,
Address Developmental and Behavioral Problems. 2004.
Nashville: Ellsworth & Vandermeer, 1998. 33. Zuckerman B, Parker S, Kaplan-Sanoff M, et al: Healthy
18a. Houston HL, Davis RH: Opportunistic surveillance of Steps: A case study of innovation in pediatric practice.
child development in primary care: is it feasible? (Com- Pediatrics 114:820-826, 2004.
parative Study Journal Article) J R Coll Gen Pract 34. Hampshire A, Blair M, Crown N, et al: Assessing the
35(271):77-79, 1985. quality of child health surveillance in primary care. A
19. Glascoe FP: Toward a model for an evidenced-based pilot study in one health district. Child Health Care Dev
approach to developmental/behavioral surveillance, 28:239-249, 2002.
promotion and patient education. Ambul Child Health
5:197-208, 1999.
20. Rydz D, Shevell MI, Majnemer A, et al: Developmental
screening. Child Neurol 20:4-21, 2005.
21. Glascoe FP: Do parents’ discuss concerns about chil-
dren’s development with health care providers? Ambul
Child Health 2:349-356, 1997.
7C.
22. Glascoe FP, Sandler H: The value of parents’ age esti-
mates of children’s development. J Pediatr 127:831-835,
Assessment of Development
1995. and Behavior
23. Pachter LM, Dworkin PH: Maternal expectations about
normal child development in four cultural groups. Arch
Pediatr Adolesc Med 151:1144-1150, 1997.
TERRY STANCIN ■ GLEN P. AYLWARD
24. Glascoe FP: Are over-referrals on developmental screen-
ing tests really a problem? Arch Pediatr Adolesc Med “Assessment is a means to an end, not an end in itself.
155:54-59, 2001. —Jerome M. Sattler, 2001
25. Smith PK: BCAP Toolkit: Enhancing Child Devel-
opment Services in Medicaid Managed Care. Center Assessment of child development and behavior
for Health Care Strategies, 2005. (Available at: involves a process in which information is gathered
http://www.chcs.org/; accessed 10/13/06.) about a child so that judgments can be made. This
26. Pinto-Martin J, Dunkle M, Earls M, et al: Developmen- process generally includes a multistage approach,
tal stages of developmental screening: Steps to imple- designed to gain sufficient understanding of a child
mentation of a successful program . Am J Public Health
so that informed decisions can be made.1 In contrast
95:6-10, 2005.
to psychological testing (which includes the adminis-
27. Glascoe FP, Sievers P, Wiseman N: First Signs Model
Program makes great strides in early detection in Min- tration of tests), assessment is the process in which
nesota: Clinicians and educators play major role in data from clinical sources and tools (including history,
increased screenings. American Academy of Pediatrics’ interviews, observations, formal and informal tests),
Section on Developmental and Behavioral Pediatrics preferably obtained from multiple perspectives, are
Newsletter. August, 2004. (Available at: www.dbpeds. interpreted and integrated into relevant clinical
org; accessed 10/13/06.) decisions.
 CHAPTER 7 Screening and Assessment Tools 145

Developmental and behavioral assessments may be testing and measures of functional outcome. However,
conducted for several purposes.1,2 Screening involves we do not attempt to address the complex manner in
procedures to identify children who are at risk for a which information, obtained from different assess-
particular problem and for whom there are available ment data sources, is weighted and synthesized in the
effective interventions. Diagnosis and case formulation formulation of clinical judgments. The discussions of
procedures help determine the nature, severity, and assessment tools is not meant to be all-inclusive—
causes of presenting concerns and often result in clas- there are literally thousands of developmental and
sification or a label. Prognosis and prediction methods behavioral assessment measures in the literature—
result in generating recommendations for possible nor an endorsement of one instrument over others.
outcomes. Treatment design and planning assessment Rather, it is a sampling the array of instruments avail-
strategies aid in selecting and implementing inter- able to clinicians and researchers (Table 7C-1). We
ventions to address concerns. Treatment monitoring present implications and recommendations for future
methods track changes in symptoms and functioning research concerning measures of psychological assess-
targeted by interventions. Finally, treatment evaluation ment as they pertain to the field of developmental
procedures help investigators examine consumer sat- behavioral pediatrics.
isfaction and the effectiveness of interventions.
The purpose of this chapter is to describe methods
and tools for assessing children’s development and
behavior. In accordance with current discussions
CASE ILLUSTRATIONS
within the child psychology literature,2 we advocate
The following case examples are referred to through-
the development of integrated evidence-based assess-
out the discussion of assessment methods:
ment strategies for childhood problems with emphasis
placed on research concerning the reliability, validity, ■ Case 1: Jane is a 21-month-old (corrected age) girl
and clinical utility of commonly used measures in who was born at 27 weeks’ gestation, with a birth
assessment and treatment planning of developmental weight of 850 g, having a grade III intraventricu-
and behavioral problems (i.e., what methods have lar hemorrhage, bronchopulmonary dysplasia, and
been shown to be useful and valid for what purpose). hyperbilirubinemia. Her young, single mother
We describe general information about clinical inter- resides in low-income housing and may have used
viewing and observational methods required to cocaine during pregnancy. Her score on the revised
conduct comprehensive child assessments (for more Bayley Scales of Infant Development (BSID-II)
extensive discussions, see McConaughy3). To help Mental Developmental Index (MDI) was 90 at age
guide the pediatric practitioner’s and researcher’s 12 months (corrected age). Her developmental
appropriate use of assessment results, we provide status is being evaluated at a high-risk infant follow-
information on the range of methods used for assess- up clinic at this time to determine need for early
ing developmental abilities, intelligence and cognitive intervention services.
abilities, behavioral and emotional functioning, and ■ Case 2: Rachel is a 15-year-old girl with mild cere-
specialized testing, including neuropsychological bral palsy with no identified learning disorders who

TABLE 7C-1 ■ Illustrative Behavioral and Developmental Assessment Methods

Method Applications Illustrative Methods

Structured/semistructured Diagnostic assessments Diagnostic Interview for Children–IV (DISC-IV) 8

interviews Diagnostic Interview for Children and Adolescents (DICA-IV) 9
Assessment and treatment Comprehensive Assessment to Intervention System (CAIS)7
planning Child and Adolescent Psychiatric Assessment (CAPA)12
Semistructured Parent Interview (SPI) 3
Semistructured Clinical Interview for Children and Adolescents (SCICA)10
Standardized cognitive Developmental assessments Manual of Developmental Diagnosis 32
methods Cattell Infant Intelligence Scale31
Bayley Scales of Infant and Toddler Development–Third Edition
(BSID-III) 27
Battelle Developmental Inventory–Second Edition (BDI-2) 39
Mullen Scales of Early Learning (MSEL) 41
Differential Ability Scales (DAS) 43
McCarthy Scales of Children’s Abilities (MSCA) 44
146 DEVELOPMENTAL AND BEHAVIORAL PEDIATRICS: EVIDENCE AND PRACTICE

TABLE 7C-1 ■ Illustrative Behavioral and Developmental Assessment Methods—cont’d

Method Applications Illustrative Methods

Intelligence assessment Kaufman Brief Intelligence Test, Second Edition (KBIT-2) 45

Stanford-Binet Intelligence Scale–Fifth Edition26
Stanford-Binet Intelligence Scales for Early Childhood–5 (SB-5) 46
Kaufman Assessment Battery for Children–Second Edition (KABC-II) 47
Wechsler Preschool and Primary Scale of Intelligence–Third Edition
(WPPSI-III) 48
Wechsler Intelligence Scale for Children–Fourth Edition (WISC-IV) 49
Wechsler Abbreviated Scale of Intelligence (WASI) 50
Achievement Kaufman Test of Educational Achievement–II (KTEA-II) 52
Peabody Individual Achievement Test–Revised (PIAT-R) 53
Peabody Individual Achievement Test–Normative Update54
Wechsler Individual Achievement Test–II (WIAT-II) 55
Wide Range Achievement Test–3 (WRAT-3) 56
Woodcock-Johnson III Tests of Achievement (WJ III) 58
Neuropsychological Children’s Memory Scale (CMS) 60
assessments NEPSY—A Developmental Neuropsychological Assessment (NEPSY) 61
Behavior Rating Inventory of Executive Function (BRIEF) 62
Wide Range Assessment of Memory and Learning (WRAML; 2nd
edition: WRAML-2) 63,64
Global behavior rating Broad measures of Achenbach System of Empirically Based Assessment (ASEBA) 11,67-71
scales pathology Caregiver completed: Child Behavior Checklist (CBCL/11/2-5, CBCL/6-18)
Teacher Report Form
Youth Self-Report Form
Behavior Assessment System for Children–Second Edition (BASC-2)75
Parent Rating Scales
Teacher Rating Scales
Self-Report of Personality
Infant-Toddler Social-Emotional Assessment Scale (ITSEA)76
Minnesota Multiphasic Personality Inventory–Adolescent (MMPI-A)78
Peer reports Broad measure of Peer-Report Measure of Internalizing and Externalizing Behavior
pathology (PMIEB) 81
Observational Assessment of parent Dyadic Parent-Child Interaction Coding System (DPICS) 83
coding methods child interactions
Problem-specific Depression Children’s Depression Inventory (CDI) 87
questionnaires and Mood and Feeling Questionnaire (MFQ) 88
rating scales Reynolds Child Depression Scale (RCDS) 89
Reynolds Adolescent Depression Scale (RADS) 90
Children’s Depression Rating Scale Revised (CDRS-R) 91
Preschool Feelings Checklist92
Anxiety Multidimensional Anxiety Scale for Children (MASC) 94
Social Phobia and Anxiety Inventory for Children (SPAI-C) 95
Social Anxiety Scale for Children (SAS-C) 96 and Social Anxiety Scale for
Adolescents (SAS-A) 97
Revised Children’s Manifest Anxiety Scale (RCMAS) 98
Attention-deficit/ ADHD Rating Scale-IV99
hyperactivity disorder Vanderbilt ADHD Diagnostic Parent Rating Scales100
(ADHD)
Autism spectrum Autism Diagnostic Interview-Revised (ADI-R)106
disorders Autism Diagnostic Observation Schedule (ADOS)107
Social Communication Questionnaire (SCQ)108
Childhood Autism Rating Scale (CARS)110
Family assessment Parent and family Parenting Stress Index (PSI) 3rd ed113
methods assessment
Functional outcome Global functioning Children’s Global Assessment Scale (CGAS)115
methods Child and Adolescent Functional Assessment Scale (CAFAS)116
Adaptive behavior Vineland Adaptive Behavior Scales (Vineland-II)117
Health-related quality of PedsQL 4.0118
life
 CHAPTER 7 Screening and Assessment Tools 147

presents with depressed mood and falling grades INTERVIEWS

in her ninth grade placement. Academic strengths
have been language arts, but she has always been
weak in math. Historically she has been a B and C Clinical assessment interviews are face-to-face inter-
student, but in her freshman year, she is in danger actions with bidirectional influence for the purpose
of failing math. Rachel complains about trouble of planning, implementing, or evaluating treatment.3
getting work done this year, especially in algebra. The interview is a fundamental technique for gather-
Her mother is puzzled that Rachel has requested ing assessment data for clinical purposes and is
counseling. considered by many clinicians to be an essential com-
■ Case 3: Jose is a 9-year-old third grader referred ponent. Interviews provide respondents the oppor-
to the Developmental-Behavioral Pediatrics Clinic tunity to offer personal reflections of concerns and
because of the following problems: poor academic historical events. Thoughts, feelings, and other private
performance, disruptive behavior, and trouble get- experiences are conveyed in conversation that is not
ting along with peers. His fi rst language is Spanish, readily obtainable in any other format. The interview
but he is considered fluent in English. He was born often serves a dual purpose. Not only does a clinical
in Puerto Rico, and his parents do not speak or interview provide valuable assessment data, but it also
read English. A note from his teacher indicates con- is probably the fi rst opportunity for a clinician to
cerns that Jose has a short attention span and fails begin to build a positive therapeutic relationship that
to complete many assignments. is the foundation for effective behavioral change. In
practice, most clinical assessment interviews use
unstructured or semistructured formats in order to
“WHAT MEASURE SHOULD I USE?” obtain detailed information about a particular pre-
senting problem. Greater flexibility in interview
Kazdin4 noted that in clinical situations, this question formats is often desirable when the clinical goals
suggests a misunderstanding of the assessment include not just reaching a diagnosis but also estab-
process, because it is unlikely that any one measure lishing a therapeutic relationship with a family and
or method can suitably capture child functioning. developing a treatment plan.
Although some measures have been shown to be An effective clinical interview needs to establish a
perform better than others, a single “gold standard” condition of trust and rapport so that the interviewee
tool does not exist for assessing most aspects of chil- can feel comfortable in divulging personal informa-
dren’s functioning. Valid child assessment often tion.6 It is important to outline the purpose and nature
requires data from multiple sources, including inter- of the interview at the outset and to discuss issues
views, direct observations, standardized parents’ and and limits of confidentiality. Effective interviewing
teachers’ rating scales, self-reports, background ques- requires listening skills, strategic use of open-ended
tionnaires, and standardized tests. Multiple methods and direct questions, and verbal and nonverbal
are needed not only to evaluate different facets of empathic communications. The clinician needs to
problems but also because of the high rate of comor- offer careful statements that reflect, paraphrase,
bidity in children with developmental and behavioral reframe, summarize, and restate to verify accurate
conditions. In clinical settings, methods should be interpretation of client statements.6 At the same time,
tailored to address the specific referral questions and the clinician is gathering verbal and nonverbal infor-
assessment goals; therefore, preordained “assessment mation conveyed by the client. Most interviewers
batteries” should be avoided. Moreover, clinical assess- take notes during interviews.
ments often have multiple goals, such both diagnosis Most clinical assessments of children begin with
and treatment planning. Diagnostic methods, shown a parent interview, the content of which depends
to be evidenced-based (e.g., structured diagnostic on its purpose. Interviewing in the context of a
interviews or rating scales), are often not helpful in developmental-behavioral problem usually focuses
treatment planning, whereas a functional analysis of on identification and analysis of parental concerns so
impairment (i.e., identification of environmental con- that an intervention plan can be developed and imple-
texts and socially valid target behaviors) are more mented. Psychosocial interviews typically elicit parent
useful.5 Different methods of data collection yield dif- perceptions about the specific nature of the problem
ferent information, and one is not inherently better (including antecedents and consequences of the
than the other; each method contributes unique ele- problem), family relations and home situation, social
ments. Moreover, assessments must adopt a frame- and school functioning, developmental history, and
work that maintains a correct developmental medical history. A practical interview format that is
perspective, including use of methods and procedures well suited for primary care settings is the Compre-
that fit a child’s developmental stage. hensive Assessment to Intervention System, devel-
148 DEVELOPMENTAL AND BEHAVIORAL PEDIATRICS: EVIDENCE AND PRACTICE

oped by Schroeder and Gordon.7 This behaviorally a diagnosis or a specific judgment with high inter-
oriented format clusters information in six areas for assessor reliability, as would be desired in research
quick response: referral question, social context of studies on specific psychiatric diagnoses, standard-
question, general information about the child’s devel- ized, structured psychiatric interviews are often
opment and family, specifics of the concern and func- preferable. Structured interviews contain specific,
tional analysis of behavior, effects of the problem, and predetermined questions with a format designed to
areas for intervention. Schroeder and Gordon used elicit information efficiently and thoroughly. Key
this system both in their telephone call-in service and questions are followed by specified branch ques-
in their pediatric psychology office practices. tions with restricted, closed (“yes”/“no”) or brief
Child interviews are generally viewed as an essen- responses.
tial component of clinical assessments and can be An example of a structured interview is the
conducted with children as young as age 3 years.3 National Institute of Mental Health Diagnostic Inter-
Child clinical interviews are useful for establishing view for Children–IV.8 This instrument is a highly
rapport, learning the child’s perspective of function- structured interview with nearly 3000 questions
ing, selecting targets for interventions, identification designed to assess more than 30, psychiatric disorders
of the child’s strengths and competencies, and assess- and symptoms listed in the American Psychiatric
ing the child’s view of intervention options. More- Association’s Diagnostic and Statistical Manual of Mental
over, child interviews offer an opportunity to observe Disorders, Fourth Edition (DSM-IV) 8a in children and
the child’s behavior, affect, and interaction style adolescents aged 9 to 17 years. Parent and child ver-
directly. However, competent interviewing of chil- sions in English and Spanish are available, and lay
dren and adolescents interviews requires considerable interviewers can administer it for epidemiological
skills and knowledge of development. For example, research. The Diagnostic Interview for Children and
preschool children often respond better in interviews Adolescents9 is another structured diagnostic inter-
that the interviewer conducts while sitting at the view for children ages 6 to 17. This instrument con-
child’s level on the floor or at a small table and with sists of nearly 1600 questions that address 28 DSM-IV
toys, puppets, and manipulative items. School-age diagnoses relevant to children. Interrater reliability
children may end communication if they feel bar- estimates of individual diagnoses range from poor to
raged by too many direct questions, especially if asked good, and diagnoses are moderately correlated with
“why” about motives, or if questions are abstract or clinicians’ diagnoses and self-rated measures.
rhetorical. Adolescent interviews may require addi- Structured interviews result in higher interrater
tional attention to matters of confidentiality, trust, (or interobserver) reliability because there is little
and respect. opportunity for the interviewer to influence the
Interviews of children and adolescents may include content of data collected. Although sometimes con-
a brief observational, descriptive report of clinician sidered to be the “gold standard” for psychiatric diag-
impressions, summarized as a behavioral observa- nostic and epidemiological research, standardized
tions or a mental status examination. Key areas of interviews are not impervious to reporter bias. In
psychological functioning are examined, including addition, structured diagnostic interviews tend to rely
general appearance and behavior (physical appear- on DSM-IV symptoms which may not be developmen-
ance, nonverbal behaviors, attitudes), emotional tally appropriate, particularly for very young chil-
expression (mood and affect), characteristics of speech dren. Moreover, structured diagnostic interviews may
and language, form (how thoughts are organized) and take 1 to 3 hours to complete, which renders them
content (e.g., delusions, obsessions, suicidal/homi- impractical for most clinical settings, especially
cidal ideation) of thought, perceptual disturbances because they typically do not assess background and
(e.g., hallucinations, dissociation), cognition (orien- family factors that are necessary for developing and
tations, attention, memory), and judgment and insight implementing an intervention plan.
(developmentally appropriate). Semistructured interviews combine aspects of
traditional and behavioral interviewing techniques.
Specific topic areas and questions are presented, but,
Structured and Semistructured in contrast to structured interviews, more detailed
responses are encouraged. Semistructured formats
Diagnostic Interviews also support use of empathic communication described
Assessment data obtained from unstructured clinical previously (e.g., reflecting, paraphrasing). For exam-
interviews tend to vary considerably and are largely ple, the Semistructured Parent Interview3 contains
interviewer dependent. As a result, unstructured sample questions organized around six topic areas:
interviews have particularly poor reliability and valid- concerns about the child (open ended), behavioral or
ity. When the primary assessment goal is to provide emotional problems (eliciting elaboration to begin a
 CHAPTER 7 Screening and Assessment Tools 149

functional analysis of behavior), social functioning, ment from a motivational interviewing perspective
school functioning, medical and developmental involves addressing the patient’s ambivalence about
history, and family relations and home situations. making a change in behavior, exploring the negative
Like other semistructured formats, the Semistruc- and positive aspects of this choice, and discussing the
tured Parent Interview encourages parent interviews relationship between the proposed behavior change
built around a series of open-ended questions to (e.g., compliance with mediations) and personal
introduce a topic, followed by more focused questions values (e.g., health). This information is elicited in an
about specific areas of concern. empathic, accepting, and nonjudgmental manner and
The Semistructured Clinical Interview for Children is used by the patient to select goals and create a col-
and Adolescents (SCICA)10 is an interview designed laborative plan for change with the provider.
for children aged 6 to 16. It is part of the Achenbach The effectiveness of motivational interviewing with
System of Empirically Based Assessment (ASEBA)11 children and young adolescents has not been estab-
and was designed to be used separately or in conjunc- lished. However, there is emerging evidence of its
tion with other ASEBA instruments (e.g., Child utility with adolescents and young adults, particularly
Behavior Checklist [CBCL], Teacher Report Form). in the areas of risk behavior, program retention, and
The SCICA contains a protocol of questions and pro- substance abuse.15,16
cedures assessing children’s functioning across six
broad areas: (1) activities, school, and job; (2) friends;
(3) family relations; (4) fantasies; (5) self-perception
and feelings; and (6) problems with parent/teacher. TESTING METHODS:
There are additional optional sections pertaining to DEVELOPMENTAL AND
achievement tests, screening for motor problem, and COGNITIVE
adolescent topics (e.g., somatic complaints, alcohol
and drug abuse, trouble with the law). Interview Infancy and Early Childhood
information (observations and self-report) are scored Since the 1980s, there has been increased interest in
on standardized rating forms and aggregated into the developmental evaluation of infants and young
quantitative syndrome scales and DSM-IV–oriented children.17,18 This began with the 1986 Education of
scales. Test-retest, interrater, and internal consistency the Handicapped Act Amendments (Public Law 99-
evaluations indicate excellent to moderate estimates 457) and continues with the Individuals with Dis-
of reliability. Accumulating evidence for validity of abilities Education Improvement Act of 2004 (Public
the SCICA includes content validity, as well as crite- Law 108-446), a revision of the Individuals with Dis-
rion-related validity (ability to differentiate matched abilities Education Act (IDEA). These laws involve
samples of referred and nonreferred children). provision of early intervention services and early
The Child and Adolescent Psychiatric Assessment12 childhood education programs for children from birth
is another semistructured diagnostic interview for through 5 years of age. Developmental evaluation is
children and adolescents aged 9 to 17. One interesting necessary to determine whether children qualify
feature of this instrument is the inclusion of sections for such intervention services. Part C of the IDEA
assessing functional impairment in a number of areas revision (Section 632) delineates five major areas of
(e.g., family, peers, school, and leisure activities), development: cognitive, communication, physical,
family factors, and life events. social-emotional, and adaptive. However, defi nitions
of delay vary, criteria being set on a state by state
basis. These can included a 25% delay in functioning
Motivational Interviewing in comparison with same-aged peers, 1.5 to 2.0 stan-
Motivational interviewing is an empirically supported dard deviations below average in one or more areas
interviewing approach gaining considerable attention of development, or performance on a level that is a
in medical and mental health settings. More than an specific number of months below a given child’s
assessment strategy, motivational interviewing is a chronological age. However, pressure to quantify
brief, client-centered directive intervention designed development has caused professionals working with
to enhance intrinsic motivation for behavior change infants and young children to attribute a degree of
through the exploration and reduction of patient preciseness to developmental screening and assess-
ambivalence.13 Based on a number of social and ment that is neither realistic nor attainable. Addi-
behavioral principles, including decisional balance, tional problems include test administration by
self-perception theory, and the transtheoretical model examiners who are not adequately trained and use of
of change,14 motivational interviewing combines rog- instruments that have varying degrees of psychomet-
erian and strategic techniques into a directive and yet ric rigor.19 Nonetheless, developmental evaluation is
patient-centered and collaborative encounter. Assess- critical, because timely identification of children with
150 DEVELOPMENTAL AND BEHAVIORAL PEDIATRICS: EVIDENCE AND PRACTICE

developmental problems affords the opportunity for memory. In contrast, the purpose of early develop-
early intervention, which enhances skill acquisition mental measures such as the Bayley Scales of Infant
or prevents additional deterioration. Development (BSID)28 or the Gesell Developmental
Again, choice of the type of developmental assess- Schedules29 was to be diagnostic of developmental
ment that is administered is driven by the purposes delays, providing a benchmark of developmental
of the evaluation: for example, determination of acquisitions (or lack thereof) in comparison to same-
eligibility for early intervention or early childhood aged peers. Nonetheless, this distinction is often
education services, documentation of developmental blurred, perhaps because there is no specific age at
change after provision of intervention, evaluation of which a child shifts from “development” to “intelli-
children who are at risk for developmental problems gence” (although the culmination of the infancy
because of established biomedical or environmental period is often indicated), nor is there a clear-cut
issues, documentation of recovery of function, or pre- transformation from a delay to a deficit. Developmen-
diction of later outcome. Assessment of infants and tal tests also tend to include motor and social-adaptive
young children is in many ways unique, because it skills. Both tests of development and intelligence are
occurs against a backdrop of qualitative and quanti- driven by the theoretical model of the test developer
tative developmental, behavioral, and structural and the constructs measured by the test. Those that
changes, the velocity of change being greater during assess the former are considered more dynamic or
infancy and early childhood than at any other time. fluid; those that assess intelligence are more consis-
The rapidly expanding behavioral repertoire of the tent and predictive. Herein, we discuss both develop-
infant and young child and the corresponding diver- mental and intelligence tests that are used in children
gence of cognitive, motor, and neurological functions in this age level.
pose distinct evaluation challenges.18,19
Another significant testing concern in this age
range is test refusal.20 Test refusal, where a child Developmental Assessment Instruments
either declines to respond to any items, or eventually GESELL DEVELOPMENTAL SCHEDULES/
stops responding when items become increasingly CATTELL INFANT INTELLIGENCE TEST
difficult, occurs in 15% to 18% of preschoolers.21-24
Occasional refusals occur in 41% of young children. The Gesell Developmental Schedules29,30 and the
In addition to the immediate ramifications problem- Cattell Infant Intelligence Test31 are the oldest devel-
atic test-taking behaviors have on actual test scores, opmental test instruments and exemplify the blurring
there is evidence that early high rates of refusals are of developmental and intelligence testing boundaries.
associated with similar behaviors at later ages, and The most recent version of the former is Knobloch and
with lower intelligence, visual perceptual, neuropsy- associates’ Manual of Developmental Diagnosis (for chil-
chological, or behavioral scores in middle child- dren aged 1 week to 36 months).32 Gesell specified
hood.22-25 Non-compliance has been reported to occur key ages at which major developmental acquisitions
in verbal production tasks, gross motor activities, or occur: 4, 16, 28, and 40 weeks and 12, 18, 36, and 48
toward the end of the testing session, and it occurs months. Gross motor, fi ne motor, adaptive, language,
more in children born at biologic risk or those from and personal-social areas are assessed, with 1 to 12
lower socioeconomic households. Children who refuse items at each age. A developmental quotient is com-
any aspect of testing differ from those who refuse puted for each area with the formula maturity age
some items, or who are compliant and cooperative to level/chronological age ×100. The Cattell test is essen-
a certain point and then refuse more difficult items. tially an upward extension of the Gesell schedule over
This situation prompted inclusion of the Test Observa- the fi rst 21 months and a downward extension of
tion Checklist (TOC) in the Stanford-Binet Scales for early versions of the Stanford-Binet tests from age 22
Early Childhood, 5th Edition (SB5).26 months and older (the Cattell age range is 2 to 36
A distinction is often made between developmental months). A major drawback of both instruments is
tests and intelligence tests,27 and both are used in the the limited standardization sample size (e.g., 107 for
age range under discussion. The assessment of intel- the Gesell schedule, 274 for the Cattell test). As a
ligence originated from the need to determine which result, neither is used frequently at this time, although
children would be able to learn in a classroom and the Cattell test does yield so-called IQ scores below 50
which would be mentally deficient. In fact, this was (the floor of the BSID).
the original purpose of the Binet test. Intelligence
tests have become more psychometrically sophisti- BAYLEY SCALES OF INFANT DEVELOPMENT27,28,33
cated but still assess different facets of primary cogni- The original BSID28 evolved from versions adminis-
tive abilities such as reasoning, knowledge, quantitative tered to infants enrolled in the National Collaborative
reasoning, visual-spatial processing, and working Perinatal Project. It was the reference standard for the
 CHAPTER 7 Screening and Assessment Tools 151

assessment of infant development, administered to Scaled scores (M = 10, SD = 3), composite scores
infants 2-30 months of age. The BSID was theoreti- (M = 100, SD = 15), percentile ranks, and growth
cally eclectic and borrowed from different areas of scores are provided, as are confidence intervals for the
research and test instruments. It contained three scales and age-equivalent scores for subtests. Growth
components—the MDI, the Psychomotor Develop- scores are new and, with caution, are used to plot the
mental Index (PDI), and the Infant Behavior Record child’s growth over time for each subtest in a longi-
(M = 100, SD = 16)—and was applicable for children tudinal manner. This metric is calculated on the basis
aged 2 to 30 months. The BSID subsequently was of the subtest total raw score and ranges from 200 to
revised as the BSID-II,33 this partly because of the 800 (M = 500, SD = 100). As in the original BSID,
upward drift of approximately 11 points on the MDI there are basal rules (passing the fi rst three items at
and 10 points on the PDI, reflecting the Flynn effect34 the appropriate age starting point) and a ceiling or
(M = 100, SD = 15). As a result, the BSID-II scores discontinue rules (a score of 0 for five consecutive
were 12 points lower on the MDI and 10 points lower items).
on the PDI in comparison with the original BSID.35 The correlation between the BSID-III Language
The Behavior Rating Scale was developed to enable Composite and the BSID-II MDI is 0.71; that between
assessment of state, reactions to the environment, the Motor Composite and the BSID-II PDI is 0.60; and
motivation, and interaction with people. The age that between the Cognitive Composite and the BSID-
range for the BSID-II was expanded to 1 to 42 months. II MDI is 0.60. The moderate correlation between the
Unfortunately, this instrument had 22 item sets and older PDI and MDI and their BSID-III counterparts
basal and ceiling rules that differed from the original underscores the significant differences between the
BSID. These rules were controversial in that if correc- old and new BSIDs. However, in contrast to the
tion is used to determine the item set to begin admin- expected Flynn effect (see Chapter 7A and Flynn34),
istration, or if an earlier item set is employed because the BSID-III Cognitive and Motor composite scores
of developmental problems, scores tend to be some- are approximately 7 points higher than the corre-
what lower, because the child is not automatically sponding BSID-II MDI and PDI. This phenomenon
given credit for passing the lower item set. It was also has also been reported with the Peabody Picture
criticized because it did not provide area scores com- Vocabulary Test–Third Edition,38 and the Battelle
patible with IDEA requirements such as cognitive, Developmental Inventory–Second Edition39 (Box
motor communication, and social and adaptive 7C-1).
function.35
For the newest version of the BSID, the Bayley BATTELLE DEVELOPMENTAL
Scales of Infant and Toddler Development–Third INVENTORY–SECOND EDITION (BDI-2) 39
Edition (BSID-III),27 norms were based on responses The norms of the BDI-2 were based on the perfor-
of 1700 children. The BSID-III assesses development mances of 2500 children, and this instrument is
(at ages 1 to 42 months) across five domains: cogni- applicable to children from birth through age 7 years
tive, language, motor, social-emotional, and adaptive. 11 months. Data are collected through a structured
Like its predecessors, the BSID-III is a power test. test format, parent interviews, and observations of the
Assessment of the fi rst three domains is accomplished child. The scoring system is based on a 3-point scale:
by item administration, whereas the latter two are 2 if the response met a specified criteria, 1 if the child
evaluated by means of caregiver’s responses to a ques- attempted a task but it was incomplete (emerging
tionnaire. A Behavior Observation Inventory is com- skill), and 0 if the response was incorrect or absent.
pleted by both the examiner and the caregiver. The The original Battelle Developmental Inventory40 and
Language scale includes a Receptive Communication the BDI-2 were developed on the basis of milestones:
and an Expressive Communication scaled score; the that is, development reflects the child’s attainment of
Motor Scales includes a Fine Motor and a Gross Motor critical skills or behaviors. Five domains are assessed:
score. The BSID-III Social-Emotional Scale is an adap- (1) The Adaptive Domain, which contains the Self-Care
tation of the Greenspan Social-Emotional Growth (e.g., eating, dressing toileting) and Personal Respon-
Chart: A Screening Questionnaire for Infants and sibility subdomains (initiate play, carry out tasks,
Young Children.36 The Adaptive Behavior Scale is avoid dangers); (2) the Personal-Social Domain, which
composed of items from the Parent/Primary Caregiver contains the Adult Interaction (e.g., identifies famil-
Form of the Adaptive Behavior Assessment System– iar people), Peer Interaction (shares toys, plays coop-
Second Edition;37 it measures areas such as communi- eratively) and Self-Concept and Social Role subdomains
cation, community use, health and safety, leisure, (express emotions, aware of gender differences); (3)
self-care, self-direction, functional preacademic per- the Communication Domain, which contains the Recep-
formance, home living, and social and motor skills tive Communication and Expressive Communication
and yields a General Adaptive Composite score. subdomains; (4) the Motor Domain, which contains
152 DEVELOPMENTAL AND BEHAVIORAL PEDIATRICS: EVIDENCE AND PRACTICE

BOX 7C-1 ing). The BDI-2 full assessment incorporates all five
domains, whereas the screening test includes two items
CASE 1: DEVELOPMENTAL ASSESSMENT
at each of 10 age levels for each of the five domains.
DISCUSSION
A developmental quotient is produced for each domain
The toddler in Case 1 was given a developmental and for a total BDI-2 Composite score (M = 100, SD =
assessment that included the BSID-III. Results are 15); scaled scores are applied to the subdomains (M =
shown in the table below. 10, SD = 3). Noteworthy is the fact that these are
normalized standard scores and not ratio scores. Per-
95% centiles, age-equivalent scores, and confidence inter-
Standard Confidence
vals are provided; the domain developmental quotients
BSID-III Scale Score Percentile Interval
are the most reliable scores. The correlation between
Cognitive 9 the original Battelle Developmental Inventory and
Cognitive composite 95 37 87-103 the BDI-2 total developmental quotient is 0.78; the
Receptive 9 total BDI-2 score is 1.1 points higher than that of
communication
Expressive 8 the original Battelle Developmental Inventory, with
communication domain differences ranging from 1.4 to 2.8 points.
Language composite 91 27 84-99 Again, this is in contrast to the Flynn effect.
Fine motor 7
Gross motor 6 MULLEN SCALES OF EARLY LEARNING (MSEL) 41
Motor composite 79 16 73-88
Social-emotional 8 The MSEL assess the learning abilities and patterns
Social-emotional 90 25 83-99 in various developmental domains in children 2 to 51/2
composite years of age. Particular emphasis is placed on differ-
General Adaptive 81 18 76-86 entiation of visual and auditory learning, thereby
Composite (GAC) enabling measurement of unevenness in learning.
BSID, Bayley Scales of Infant Development. The MSEL differentiates receptive or expressive prob-
lems in the visual or auditory domain through four
BSID-III results indicate that the child had average scales: Visual Receptive Organization, Visual Expres-
cognitive abilities, low-average language skills, bor- sive Organization, Language Receptive Organization,
derline motor abilities (Gross Motor worse than Fine and Language Expressive Organization. At the recep-
Motor scores), low-average social-emotional func- tive level, processing that involves one modality
tioning, and borderline adaptive skills. Her low (visual or auditory) is defi ned as intrasensory reception;
average language may be influenced by the nonopti- processing that involves two modalities (auditory and
mal environment; the motor deficits are most likely visual) is termed intersensory reception. This design pro-
attributable to the grade III bleed. The Cognitive vides assessment of visual, auditory, and auditory/
composite score is 5 points higher than the previous visual reception and of visual-motor and verbal
BSID-II MDI score that the child had received at age expression. The MSEL AGS Edition42 combines the
1; this is in contrast to the Flynn effect (whereby Infant MSEL and Preschool MSEL and is applicable
scores generally increase 0.5 points per year) but is to children from birth to age 68 months. A gross
within the 7-point increment that is found when the motor scale is also included (T-scores, Early Learning
BSID-II and BSID-III scores are compared (BSID-III Composite [M = 100, SD = 15]). The Early Learning
scores are somewhat higher than BSID-II scores). On Composite has a correlation of 0.70 with the BSID
the basis of these data, early intervention services MDI.
geared toward language and adaptive skills are rec-
ommended. Moreover, the motor deficits will require DIFFERENTIAL ABILITY SCALES43
occupational and physical therapy services. The Differential Ability Scales is applicable to children
aged 21/2 to 17 years but is most useful in the range
from age 21/2 to 7 years. Many clinicians consider the
Differential Ability Scales an intelligence test, although
it yields a range of scores for developed abilities and
the Gross Motor, Fine Motor, and Perceptual Motor not an IQ score; it is rich in developmental informa-
subdomains (stacks cubes puts small object in bottle); tion of a cognitive nature. On the basis of reasoning
and (5) the Cognitive Domain, which contains Atten- and conceptual abilities, a composite score, the
tion and Memory (follows auditory and visual General Conceptual Ability score (M = 100, SD = 15;
stimuli), Reasoning and Academic Skills (names range, 45 to 165), is derived. Subtest ability scores
colors, uses simple logic), and Perception and Con- have a mean of 50 and a standard deviation of 10 (T-
cepts subdomains (compares objects, puzzles, group- scores). In addition, verbal ability and nonverbal
 CHAPTER 7 Screening and Assessment Tools 153

ability cluster scores are produced for upper preschool- Composite IQ scores (M = 100, SD = 15), as well as
age children (31/2 years and older). For ages 2 years 6 90% confidence intervals, age-equivalent scores, and
months to 3 years 5 months, four core tests constitute scaled scores for two of the three subtests. The Verbal
the General Conceptual Ability composite (block scale consists of two subtests: Verbal Knowledge (60
building, picture similarities, naming vocabulary, items measuring both receptive vocabulary and range
and verbal comprehension), and there are two sup- of general information; child points to the picture
plementary tests (recall of digits, recognition of pic- matching the word or question) and Riddles (48 items
tures). For ages 3 years 6 months to 5 years 11 months, measuring verbal comprehension, reasoning, vocabu-
six core tests are included in the General Conceptual lary knowledge, and deductive reasoning, based on
Ability composite (copying, pattern construction, and two or three clues). The Riddles subtest replaces the
early number concepts in addition to verbal compre- Defi nitions from the original Kaufman Brief Intelli-
hension, picture completion, and naming vocabulary; gence Test, thereby circumventing reading. Matrices
block building is now optional). The test is unique in is the nonverbal scale (46 items with meaningful
that it incorporates a developmental and an educa- stimuli [people, objects] and abstract stimuli [designs,
tional perspective, and each subtest is homogeneous symbols]). Discrepancies between Verbal and Non-
and can be interpreted in terms of content. verbal scores are of interest. The KBIT-2 Verbal score
is approximately 1 point lower than that of the ori-
MCCARTHY SCALES OF CHILDREN’S ginal Kaufman Brief Intelligence Test, the KBIT-2
ABILITIES (MSCA) 44 Non-verbal score is 3 points lower, and the KBIT-2
The MSCA essentially bridges developmental and IQ Composite is, on average, 2 points lower. The
tests.17 It is most useful in the 3- to 5-year age range KBIT-2 composite score is typically within 2 points of
(age range, 21/2 to 81/2 years). Some clinicians would the Wechsler Intelligence Scale for Children–Fourth
question viewing the MSCA as a developmental test; Edition (WISC-IV), composite score, and correlations
however the term IQ was avoided initially, with the with the Verbal Comprehension Index, Perceptual
test considered to measure the child’s ability to inte- Reasoning Index, and the Full Scale IQ (FSIQ) are
grate accumulated knowledge and adapt it to the tasks 0.79, 0.56, and 0.77, respectively.
of the scales. Eighteen tests in total are divided into STANFORD-BINET INTELLIGENCE SCALES,
Verbal (five tests), Perceptual-Performance (seven FIFTH EDITION/STANFORD-BINET
tests), Quantitative (three), Memory (four tests), and INTELLIGENCE SCALES FOR EARLY
Motor (five) categories. Several tests are found on CHILDHOOD–5 (EARLY SB5) 26,46
two scales. The Verbal, Perceptual-Performance, and
Quantitative scales are combined to yield the General The 10 subtests of the Early SB5 are drawn from the
Cognitive Index (M = 100, SD = 16; 50 is the lowest SB5, and the norms are derived from approximately
score). The mean scale standard score (T-score) for 1660 children aged 7 years 3 months or younger. The
each of the five scales is 50 (SD = 10). The MSCA is test is applicable from age 2 to 71/4 years (the SB5
attractive because in enables production of a profi le extends to adulthood). The 10 subtests constitute the
of functioning (with age-equivalent scores) and it FSIQ, and various combinations of these subtests con-
includes motor abilities; conversely, the test was stitute other scales. An Abbreviated Battery IQ scale
devised in 1972, and hence there is inflation of scores consists of two routing subtests: Object Series/
vis-à-vis the Flynn effect (i.e., increments in test Matrices and Vocabulary. Routing subtests enable the
norms over time result in lower scores on newer tests examiner to know the level at which to begin subse-
than those obtained on measures with older norms; quent subtests. The Nonverbal IQ scale consists of five
see Chapter 7A for a discussion of the Flynn effect34). subtests measuring the factors of nonverbal fluid rea-
Short forms of the MSCA are available, but these are soning, knowledge, quantitative reasoning, visual-
not useful in the younger age ranges.17 spatial processing, and working memory. The Verbal
IQ scale is composed of five subtests measuring verbal
ability domains in the same five factor areas as for the
Intelligence Assessment Instruments Nonverbal IQ scale. The Early SB5 also includes the
Test Observation Checklist. The test differs markedly
KAUFMAN BRIEF INTELLIGENCE TEST, from the fourth edition of the Stanford-Binet Intelli-
SECOND EDITION (KBIT-2) 45 gence Tests. Nonverbal IQ, Verbal IQ, and FSIQ scores
The KBIT-2 was released 14 years after the original are obtained (M = 100, SD = 15), as are total factor
Kaufman Brief Intelligence Test and is applicable for index scores (sum of verbal and nonverbal scaled
ages 4 to 90 years. It is particularly useful as an esti- scores) for fluid reasoning, knowledge, quantitative
mate of IQ, for screening, and in time-limited situa- reasoning, visual-spatial processing, and working
tions. The test produces Verbal, Non-verbal, and memory; scaled scores (M = 10, SD = 3) can be com-
154 DEVELOPMENTAL AND BEHAVIORAL PEDIATRICS: EVIDENCE AND PRACTICE

puted for each of the nonverbal and verbal domains. not the case with the WPPSI-III. The current version,
Optional change-sensitive scores and age-equivalent with norms based on scores of 1700 children, contains
scores are also computed. The SB5 FSIQ is approxi- 14 subtests (7 new, 7 revised) and has two age ranges:
mately 3.5 points lower than the that of the fourth from 2 years 6 months to 3 years 11 months and from
edition. The SB5 FSIQ is approximately 5 points lower 4 years 0 months to 7 years 3 months. In the first age
than the FSIQ for the Wechsler Intelligence Scale for range, FSIQ, Verbal IQ, and Performance IQ scores are
Children–Third Edition (WISC-III). obtained, through four core subtests. Seven core sub-
tests are applicable to the second age range. Supple-
KAUFMAN ASSESSMENT BATTERY FOR mental and optional subtests are used to obtain a
CHILDREN–SECOND EDITION47 General Language Composite in the younger children
This battery, with norms based on scores from 3025 and a Processing Speed Quotient in the older children.
children, is applicable in children aged 3 to 18 years Inclusion of the Picture Concepts, Matrix Reasoning,
(the original Kaufman Assessment Battery for Chil- and Word Reasoning subtests allows for better assess-
dren ceiling was 12) and contains 18 core and sup- ment of fluid reasoning. For IQ and composite scores,
plementary subtests (the number of core and M = 100 and SD = 15; for scaled scores, M = 10, SD = 3.
supplementary tests administered varies, depending Children tested with the WISC-III and the WPPSI-III
on age). It is similar to the original battery in that at overlapping ages had a WISC-III FSIQ score that
there is a simultaneous and sequential processing was, on average, 4.9 points higher than the WPPSI-III
approach, vis-à-vis the Luria neuropsychological FSIQ score; correlations with the BSID-II MDI score
model. However, the test also uses the Cattell-Horn- were 0.80; those with the Differential Ability Scales
Carroll abilities model that includes fluid crystallized General Conceptual Ability composite were 0.87. As
intelligence. As a result, interpretation is based on the in many of the newer IQ tests, various composite
model that is selected; the number of scales produced scores allow for testing of more specific cognitive abili-
is also model-dependent. The five areas assessed ties and better interpretation of fi ndings.
include (1) simultaneous processing (eight subtests;
e.g., triangles, face recognition, pattern reasoning, WECHSLER INTELLIGENCE SCALE FOR
block counting, gestalt closure), (2) sequential pro- CHILDREN–FOURTH EDITION49
cessing (word order, number recall, hand move- The WISC-IV, with norms based on responses from
ments), (3) planning (a new scale applicable for ages 2200 children, is applicable to ages 6 years 0 months
7 to 18; includes pattern reasoning, story comple- to 16 years 11 months, and contains 15 subtests (10
tion), (4) learning (four subtests, e.g., Atlantis, Rebus), core, 5 supplementary). The Verbal IQ and Perfor-
and (5) knowledge (optional and only for the Cattell- mance IQ scores of the WISC-III are no longer used.
Horn-Carroll model; includes riddles, verbal knowl- Gone also are the Picture Arrangement, Object
edge, and expressive vocabulary, some of which were Assembly, and Mazes subtests from the WISC-III, to
previously achievement tests). decrease the emphasis on performance time. Instead,
For subjects at age 3 years, a Mental Processing the WISC-IV contains a Verbal Comprehension Index
Index (from the Luria model) and a Fluid Crystallized (Similarities, Vocabulary, Comprehension, Informa-
Index (FCI-from the Cattell-Horn-Carroll model) are tion,* and Word Reasoning*), a Perceptual Reasoning
derived. For children by age 7 years, the full array of Index (Block Design, Picture Concepts, Matrix Rea-
scores can be derived; this includes the Mental Pro- soning, Picture Completion*), a Working Memory
cessing Index, a Global Score, a Fluid-Crystallized Index (Digit Span, Letter-Number Sequencing, Arith-
Index, and a Nonverbal Index (four or five subtests, metic*), and a Processing Speed Index (Coding,
depending on age, and including language-reduced Symbol Search, Cancellation*). In addition to these
instructions and nonverbal responses). The number four index scales, a measure of general intellectual
of core subtests for the Cattell-Horn-Carroll mode is function (FSIQ) is produced. The more narrow
7 to 10, depending on age, and the number of core domains and emphasis on fluid reasoning reflect con-
subtests for the Luria approach is 5 to 8. Subtest scale temporary thinking with regard to intelligence per se.
scores have a mean of 10 (SD = 3); the index score For index and FSIQ scores, M = 100 and SD = 15; the
mean is 100 (SD = 15). As with the SB5 and WISC-IV, mean scaled score is 10 (SD = 3). The WISC-IV is
intraindividual differences can be computed. highly correlated with WISC-III indexes (rs = 0.72 to
0.89). The FSIQ score is approximately 2.5 points less
WECHSLER PRESCHOOL AND PRIMARY SCALE than that of its predecessor; the Verbal Comprehen-
OF INTELLIGENCE–THIRD EDITION (WPPSI-III) 48 sion Index score is 2.4 points less than the WISC-III
Whereas the Wechsler Preschool and Primary Scale of Verbal IQ score; the Perceptual Reasoning Index score
Intelligence–Revised was a downward extension of
the Wechsler Intelligence Scale for Children, this is *Supplementary tests.
 CHAPTER 7 Screening and Assessment Tools 155

is 3.4 points less than the Performance IQ score; the deficiencies, thereby clarifying the nature of the
Working Memory Index score is 1.5 points lower than learning problem; and assist in planning, instruction,
the Freedom from Distractibility Index score; and the and intervention. Unfortunately, achievement tests
Processing Speed Index score is 5.5 points lower than do not adequately meet these needs. In general, stan-
its WISC-III counterpart. In comparison with the dard scores (with percentiles) are the most precise
Wechsler Abbreviated Scale of Intelligence (WASI) metric; age- and grade-equivalent scores are least
(described next), the WISC-IV FSIQ score is 3.4 points useful. With regard to the Wechsler tests, the Verbal
lower, the Verbal Comprehension Index score is 3.5 IQ (or Verbal Comprehension Index) and FSIQ are
points lower than the WASI Verbal IQ, and the Per- most highly correlated with achievement, particu-
ceptual Reasoning Index score is 2.6 points lower. A larly reading; the Performance IQ (Perceptual Rea-
General Ability Index (containing three verbal com- soning Index), with mathematics.51 Achievement tests
prehension and three perceptual reasoning subtests), differ in terms of content and type of response required
can be computed; this is less sensitive to the influence (e.g., multiple choice vs. recall of information), and
of working memory and processing speed and there- these differences sometimes cause one test to produce
fore is useful with children who have learning dis- lower scores than another.
abilities or attention-deficit/hyperactivity disorder
(ADHD) (Box 7C-2). KAUFMAN TEST OF EDUCATIONAL
ACHIEVEMENT–II52
WECHSLER ABBREVIATED SCALE This test is available in two formats: the Comprehen-
OF INTELLIGENCE50 sive form (with parallel forms A and B) and the Brief
The WASI is applicable to ages 6 years 0 months form. The mean score is 100 (SD = 15). Noteworthy
through 89 years. Verbal IQ, Performance IQ, and is the fact that this test’s norms were based on the
either FSIQ-4 (with four subtests) or FSIQ-2 (two sub- scores of the same population as for the Kaufman
tests) scores are obtained. Although subtests are Assessment Battery for Children–Second Edition. The
similar to those found in other Wechsler scales, the Comprehensive form, applicable from ages 4 years 6
actual items differ. Subtests include Vocabulary, Matri- months to 25, assesses reading (letter/word recogni-
ces, Block Design, and Similarities (the first two are tion, comprehension), math (computation, concepts
used to compute the FSIQ-2). T-scores are used for and application), written language (spelling, written
subtests (M = 10, SD = 5). The WASI is very useful in expression), and oral language (listening comprehen-
both clinical and research settings, because of its sion and written expression). Several reading-related
reduced administration time. The downside is a reduc- skill areas are also assessed (e.g., phonological aware-
tion in the amount of information obtained, particu- ness). The Brief form (for ages 4 years 6 months to 90
larly in terms of more specific indexes of cognitive years) measures reading (word recognition and com-
abilities. The scores are generally a few points higher prehension), math computation and application prob-
than those of more detailed tests, but they still are lems, and written expression (written language and
comparable; the correlation between the FSIQ-2 and spelling) and yields a battery composite score as well.
WISC-IV FSIQ scores is 0.86; between the FSIQ-4 and Age- and grade-equivalent scores are provided. The
WISC-IV scores, 0.83 (comparable with the correla- test differs significantly from the original Kaufman
tion among the WISC-III and WISC-IV FSIQ scores). Test of Educational Achievement and from the version
Very small differences are noted on the subtest level. with normative data update.

PEABODY INDIVIDUAL ACHIEVEMENT

Achievement Testing TEST–REVISED–NORMATIVE UPDATE53,54
Use of individually-administered achievement tests This test is applicable for kindergarten through grade
has increased dramatically since the introduction of 12 (ages 5 to 19). It differs from others in that spelling
Public Law 94-142 (Education of All Handicapped and math are presented in a multiple-choice format
Children Act), and these tests continue to be a critical and in other subtests such as reading comprehension,
component in the evaluation of children with aca- the student selects a picture that best illustrates the
demic difficulties under the IDEA revision of 2004. sentence that was read. The test includes scores for
The major reason is that achievement tests enable the general information, reading recognition, reading
delineation of aptitude-achievement discrepancies, a comprehension, total reading, math, spelling, written
hotly debated requirement for establishment of a expression, written language, and the total test. The
learning disability (versus response to treatment normative update version is the same test, with
intervention). It is assumed that such tests identify updated norms. Some clinicians argue that the multi-
children who need special instructional assistance; ple-choice format may yield higher test scores because
help recognize the nature of a child’s difficulties/ of the recognition, as opposed to recall, format.
156 DEVELOPMENTAL AND BEHAVIORAL PEDIATRICS: EVIDENCE AND PRACTICE

BOX 7C-2
CASE 2: COGNITIVE ASSESSMENT DISCUSSION
Because of concerns related to academic ability and required abstract perceptual reasoning were particularly
performance, Rachel was administered the WISC-IV. difficult for her. Despite cognitive weaknesses, Rachel’s
These results revealed that Rachel’s cognitive abilities cluster scores on the Woodcock-Johnson III Tests of
have developed very unevenly (probably in relation to Achievement were all in the average range or better. This
underlying cerebral palsy). Her verbal comprehension suggests that she has been able to use her verbal abili-
abilities are within the high average range and represent ties to compensate for weaknesses in other areas.
a significant strength for her. Significant weaknesses However, she has struggled in some academic subject
are perceptual reasoning and processing speed, which areas, especially algebra, as the content has become
are in the borderline range of functioning. Tasks that more abstract.

Cognitive Assessment Results

WISC-IV Subtest Index Subtest Index Percentile Description

Verbal Comprehension 114 82 High average

Similarities 12
Vocabulary 13
Comprehension 13
Perceptual Reasoning 75 5 Borderline
Block Design 4
Picture Concepts 9
Matrix Reasoning 5
Picture Completion 7
Working Memory 97 42 Average
Digit Span 8
Letter-Numbering
Sequencing 11
Arithmetic 10
Processing Speed 70 2 Borderline
Coding 4
Symbol Search 5
Cancellation 2
Full Scale IQ 88 21 Low average

WISC-IV, Wechsler Intelligence Scale for Children–Fourth Edition.

Academic Achievement Assessment Results on the Woodcock-Johnson III Tests of Achievement,

Form B (Actual Grade: 9)

Cluster Standard Score Grade Equivalent

Oral Language 113 13.3

Total Achievement 103 10.4
Broad Reading 106 11.0
Broad Math 93 7.9
Broad Written Language 108 12.9
Math Calculation Skills 97 8.9
Written Expression 101 9.9
Academic Skills 106 11.5
Academic Fluency 104 10.9
Academic Applications 94 8.0
Academic Knowledge 110 13.9
 CHAPTER 7 Screening and Assessment Tools 157

WECHSLER INDIVIDUAL ACHIEVEMENT TEST–II55 school systems. Of note is the fact that the WJ III Tests
This test is applicable for prekindergarten through of Achievement norms were based on the scores of
college (ages 4 to 85). This is an updated form of the the same population as those of the WJ III Tests of
original Wechsler Individual Achievement Test. There Cognitive Abilities and are designed to be used in
are four composite scores: (1) Reading (word reading, combination. Standard scores (M = 100, SD = 15),
pseudoword decoding, reading comprehension); (2) percentile scores, and age- and grade-equivalent
Mathematics (numerical operations, math reason- scores are the most helpful metrics. Computer scoring
ing); (3) Written Language (spelling, written expres- is necessary.
sion); and (4) Oral Language (listening comprehension,
oral expression). Standard scores (M = 100, SD = 15), Neuropsychological Testing
age- or grade-equivalent scores, and quartile scores
are reported. Reading rate can also be assessed, and There are three approaches to neuropsychological
the test form includes qualitative observational testing of children, and all involve the assessment
descriptions for various subtests. The test is linked to of brain-behavior relationships. The fi rst approach
Wechsler IQ tests, and aptitude/achievement discrep- entails modification of traditional neuropsychological
ancy tables are included. batteries such as the Halstead-Reitan Neuropsycho-
logical Battery or the Luria-Nebraska Neuropsycho-
WIDE RANGE ACHIEVEMENT TEST–356 logical Battery, to form corresponding children’s
batteries.59 The second approach involves interpre-
This is the seventh edition of the Wide Range Achieve- tation of standard tests such as those measuring
ment Test and is applicable for ages 5 to 75 years. intelligence, with the use of a neuropsychological
There are two equivalent forms (Blue, Tan) and each “mind-set.” In this case, results from standardized
contains reading (read letters, pronounce words), tests are tied into neuropsychological constructs and
spelling (write letters, words from dictation) and functions (e.g., the Kaufman Assessment Battery for
arithmetic (40 computation problems) tests. The test Children–Second Edition). The third approach
is based on norms by age and not grade. Critics of this includes tests or rating scales designed to assess spe-
test argue that it is outdated and provides very gross cific areas of neuropsychological function. Neuropsy-
estimates of academic achievement because it con- chological testing generally is more specific in terms
tains few items within each content area; conversely, of pinpointing strengths and deficits, and the results
it is easy and quick to administer. An Expanded more precisely describe brain-behavior relationships.
Version is also available57 that contains a group (G) Neuropsychological testing may elucidate more subtle
form with reading/reading comprehension, math, problems that contribute to cognitive, academic, or
and nonverbal reasoning (some tests are multiple social difficulties; these problems may not be appar-
choice), and an individual (I) form that assesses ent from results of more routine measures used to
reading, mathematics, listening comprehension, oral detect learning disabilities. Noteworthy is the fact
expression, and written language. The Expanded that standard intellectual assessment is typically part
Version group form is applicable to grades 2 to 12; the of a neuropsychological workup. Selected tests from
Individual form, to ages 5 to 24. this third approach are discussed as follows.

WOODCOCK-JOHNSON III TESTS OF CHILDREN’S MEMORY SCALE60

ACHIEVEMENT (WJ III) 58 The Children’s Memory Scale assesses learning and
The WJ III has two parallel forms (A and B) that are memory function with nine subtests. There are two
divided into a standard battery (12 subtests) and an levels: one for ages 5 to 8 years and one for ages 9 to
extended battery (10 tests); therefore, there are 22 16 years. The Children’s Memory Scale includes three
subtests in all. The latter provides the opportunity for domains: Auditory/Verbal, Visual/Nonverbal, and
more in-depth diagnostic evaluation of specific aca- Attention/Concentration, each with two core and
demic functions (e.g., word attack, oral comprehen- one supplemental test. The first two domains have
sion). The WJ III contains a reading cluster, an oral an immediate-memory component and a delayed-
language cluster, a math cluster, a written language memory component (tested 30 minutes later). Eight
cluster, and an academic knowledge cluster. Clusters index scores are produced: verbal immediate, verbal
are designed to correspond with IDEA areas. The delayed, delayed recognition, learning visual imme-
standard battery provides 10 cluster scores, and the diate, visual delayed, attention/concentration, and
extended battery provides an additional 9 cluster general memory (global memory function). Core sub-
scores. Broad reading, broad math, and broad written tests include: Stories, Word Pairs, Dot Locations,
language are often used to provide an overview of the Faces, Numbers, and Sequences. Word Lists, Family
child’s achievement. The WJ III is used by many Pictures, and Picture Locations are the supplemen-
158 DEVELOPMENTAL AND BEHAVIORAL PEDIATRICS: EVIDENCE AND PRACTICE

tary tests. The general memory score is moderately Regulation Index and Metacognition Index. There are
correlated with IQ scores. also two validity scales, the Inconsistency and Nega-
tivity scales, that assist in detecting response biases.
NEPSY–A DEVELOPMENTAL T-scores and percentiles are computed from raw scores
NEUROPSYCHOLOGICAL and can be graphed on the reverse side of the scoring
ASSESSMENT (NEPSY) 61 summary sheet. T-scores higher than 65 (1.5 standard
The NEPSY is based on Luria’s theoretical model,59 is deviations above average) are considered to have
applicable for ages 3 to 12 years, and consists of 27 reached a clinical threshold. There are different norms
subtests that encompass five domains: (1) Attention for boys and girls. The BRIEF is particularly useful in
and Executive Functions (e.g., Tower test, Auditory evaluating children with ADHD, traumatic brain
Attention and Response Set, Visual Attention); (2) injury, autism spectrum disorders (ASDs), and learn-
Language (Speeded Naming, Comprehension of ing disorders and those who experience cognitive,
Instructions, Phonological Processing); (3) Sensori- behavioral, or academic problems and whose initial
motor Functions (e.g., Fingertip Tapping, Visuomotor test results are inconclusive.
Precision); (4) Visuospatial Functions (Design Copying,
Arrows, Block Construction); and (5) Learning and WIDE RANGE ASSESSMENT OF MEMORY
Memory (e.g., Memory for Faces, Names, Sentence AND LEARNING (WRAML)/WIDE RANGE
Repetition). There is an 18-subtest core assessment. In ASSESSMENT OF MEMORY AND
general, each domain contains five to six subtests. LEARNING–2 (WRAML-2) 63,64
Subtest scaled scores are obtained (M = 10, SD = 3), and The WRAML (ages 5-17) and WRAML-2 (ages 5-90)
these can be combined into summary domain scores are designed to test visual and verbal memory. The
(M = 100, SD = 15). Correlations with the Children’s WRAML-2 contains six core subtests (the WRAML
Memory Scale range from 0.36 to 0.60. has nine): Story Memory, Verbal Learning, Design
BEHAVIOR RATING INVENTORY OF EXECUTIVE Memory, Picture Memory, Finger Windows, and
FUNCTION (BRIEF) 62 Number/Letter Memory. Verbal Memory Index (Story
Memory, Verbal Learning), Visual Memory Index
Executive function is an umbrella construct that refers (Design Memory, Picture Memory) and Attention/
to interrelated neuropsychological functions that are Concentration (Finger Windows, Number/Letter
responsible for purposeful, problem-solving, goal- Memory) summary scores are obtained (M = 100, SD
directed behavior. Executive function is involved in = 15). There are optional Sentence Memory, Sound-
guiding, directing, regulating, and managing cogni- Symbol, Verbal Working Memory, and Symbolic
tive, behavioral, and emotional functions. The BRIEF Memory subtests. Delayed recall and recognition
measures executive function in an ecological manner: memory can also be assessed. A General Memory
namely, it is a questionnaire given to parents and/or Index is computed from the core subtests. Scores on
teachers, thereby assessing executive function in Memory Screening, consisting of the fi rst four core
home and school environments. The BRIEF is appli- subtests (taking 20 minutes), correlate highly with
cable for school-aged children (5 to 18 years), although those of the General Memory Index (r = 0.91). In
a preschool version is also available (BRIEF-P). In contrast to the WRAML, there is no Learning Index
addition, a BRIEF-SR (self-report) version has become in the WRAML-2. The WRAML-2 also allows assess-
available for ages 11 to 18 years, requiring a fi fth ment of primary/recency effects, immediate/delayed
grade reading level. Each version consists of 86 items recall, rote versus meaningful information, visual/
scored “never” (1), “sometimes” (2), or “often” (3). verbal differences, working memory, short-term
There are eight clinical scales: Inhibit (controlling memory, sustained attention, and recognition versus
impulses, modifying behavior), Shift (cognitive flex- retrieval memory. This test is useful in evaluation of
ibility, transitioning), Emotional Control (emotional children with learning disorders, those suspected of
modulation), Initiate (beginning a task/activity, in- having verbal processing problems, and those sus-
dependently generating ideas), Working Memory pected of having ADHD.
(holding information in mind, persistence), Plan/
Organize (anticipating future events, setting goals),
Organization of Materials (workspace, play areas,
orderliness), and Monitor (work checking, keeping TESTING METHODS: BEHAVIORAL
track of how behaviors affect others). The fi rst three AND EMOTIONAL
scales combine to form the Behavioral Regulation
Index; the remaining five constitute the Metacog- Assessment of social, emotional, and behavioral
nition Index. The Global Executive Composite is adjustment of children typically begins with a parent
computed from the combination of the Behavioral or caregiver interview regarding the nature, severity,
 CHAPTER 7 Screening and Assessment Tools 159

and frequency of concerns. Most child assessment ing. Most rating scales use a standard questionnaire,
techniques rely on caregiver reports because it is pre- checklist, or Likert-response format for surveying
sumed that adults who interact daily with a child are areas of interest and usually are completed by care-
the most knowledgeable informants about a child’s givers without much assistance. Rating scales include
functioning. School-aged children and adolescents brief screening measures that assess global, broad-
should also have the opportunity to provide their own based measures, and problem-specific scales.
perceptions and information about their symptoms. Broad-based behavioral assessment instruments
Younger children (younger than 10 years) can provide assess multiple dimensions of behavior in children.
assessment information, but their self-descriptions Most are empirically developed taxonomies that are
tend to be less reliable; therefore, direct and multiple symptom driven and do not necessarily correspond to
observations and interviews may be necessary. specific diagnostic schemas. On rating scales, infor-
A criticism of reliance on caregiver reports in child mants rate the child on a broad range of social com-
assessments is that they are subject to reporter bias. petencies and problematic behaviors. Results produce
However, all reports are subject to “bias,” including empirically derived factor scores on broad dimensions
those from the child, parents, clinicians, teachers, and (e.g., internalizing and externalizing problems) and
other observers. All reports are to some extent limited specific symptom areas (e.g., depression or aggressive-
(or “biased”) by the perspectives, knowledge, recall, ness) based on age and gender norms. Parent, teacher,
and candor of the informants. Because there is no and self-report forms are available for cross-
unbiased “gold standard” source of data about chil- informant comparisons. Rating scales yield very
dren’s problems, data from multiple sources are always useful information about a child’s functioning in
needed. Regardless of the child’s age, behavioral and comparison with children of the same age and gender,
emotional assessment strategies almost always should and generally are viewed as necessary components of
include information obtained from multiple sources, most child assessments.
including parents, teachers, and the child, as well as by
direct observation of the child. Data from multiple ACHENBACH SYSTEM OF EMPIRICALLY
informants with different perspectives provide critical BASED ASSESSMENT/CHILD
information about how the child functions in different BEHAVIOR CHECKLIST11,67-71
settings such as at home, at school, and with friends. The CBCL was one of the fi rst broad-based rating
Even when there is discrepant information obtained scales of behavior in children to be developed, and it
from caregivers (as is often true), multiple vantage continues to be the most widely used method for
points are useful in determining the scope and func- behavioral assessments in children. Achenbach began
tional effect of behavior problems.65 work on what would become the CBCL in the 1960s
Assessment of child and adolescent emotional and in an effort to differentiate child and adolescent psy-
behavioral problems is further complicated because chopathology.68 At that time, the DSM provided just
of the high rate of comorbidity, heterogeneity, and two categories for childhood disorders: Adjustment
severity of concerns. Children referred for assess- Reaction of Childhood and Schizophrenic Reaction,
ments often meet diagnostic criteria for multiple dis- Childhood Type. Achenbach and collaborators applied
orders or display symptoms associated with multiple an empirically based approach to child psychopathol-
disorders. Thus, it is often important to assess not only ogy much like what was used in the development of
a referred problem but also a broad range of social, the Minnesota Multiphasic Personality Inventory.
emotional, and behavioral domains. For example, in This approach involved recording problems for large
their review of evidence-based assessment of conduct samples of children and adolescents, performing mul-
problems, McMahon and Frick66 concluded that tivariate statistical analyses to identify syndromes of
because of the high rate of comorbid disorders (e.g., problems that co-occur, using reports to assess com-
ADHD, depressive and anxiety disorders, substance petencies and adaptive functioning, and constructing
use problems, language impairment, and learning age- and gender-specific profi les of scales on which to
difficulties), initial assessments of youth with conduct display individuals’ scores.11 These taxonomic proce-
problems should include broadband measures to dures revealed that most behavior problems in chil-
screen for all conditions, followed by disorder-specific dren could be broadly divided into “internalizing”
scales, interview strategies, and standardized testing and “externalizing” conditions. This pioneering work
of conduct and comorbid disorders. had enormous influence on clinical and research
assessment practices and established the empirical
foundation for contemporary conceptualizations of
Behavioral Rating Scales child psychopathology.
Behavior rating scales are an extremely useful and The CBCL was published fi rst in 1983 as a measure
efficient method for obtaining data on child function- of behavior problems in children aged 4 to 18 years.
160 DEVELOPMENTAL AND BEHAVIORAL PEDIATRICS: EVIDENCE AND PRACTICE

Currently, there are ASEBA materials for ages 11/2 to tional problems. The competency scale includes 20
older than 90 years. There are forms for preschoolers items about a child’s activities, social relations, and
(11/2 to 5 years, parent and teacher/daycare versions) 69 school performance. Specific behavioral and emo-
and school-aged children (parent, teacher versions for tional problems are described in 118 items that are
children aged 6 to 18 years and youth self-report for rated along the 0-to-2 scale described previously,
ages 11 to 18 years),67 as well as for adults (18 to 59 along with two open-ended items for reporting addi-
years)70 and older adults (60 to older than 90 years)71 tional problems. A scoring profi le provides raw scores,
(both with caregiver and self-report formats). For T-scores, and percentiles for three competence scales
each problem listed, informants provide ratings on (Activities, Social, and School); Total Competence;
the following scale: 0 = “not true,” 1 = “somewhat or eight cross-informant (clinical scale) syndromes; and
sometimes true,” and 2 = “very true or often true.” Internalizing, Externalizing, and Total Problems
Hand-scored and computer-scored profi les are avail- (broad scales). The eight clinical scales scored from
able, as are Spanish-language forms. the CBCL/6-18 Teacher Report Form and Youth Self-
The Child Behavior Checklist for Ages 11/2-5 Report are Aggressive Behavior; Anxious/Depressed;
(CBCL/11/2-5) obtains parents’ ratings of 99 problem Attention Problems; Rule-Breaking Behavior; Social
items along with descriptions of concerns and com- Problems; Somatic Complaints; Thought Problems;
petencies. Scales are based on parent ratings of 1728 and Withdrawn/Depressed. Now available are also six
preschool children; norms are based on a national DSM-oriented scales associated with affective prob-
sample of 700 children. Raw scores can be translated lems, anxiety problems, somatic problems, attention-
into standard T-scores, yielding interpretative infor- deficit/hyperactivity problems, oppositional defiant
mation on three summary scales (Internalizing, problems, and conduct problems. The school-age
Externalizing, and Total Problems), as well as on scales are based on new factor analyses of parents’
clinical syndromes scales (Emotionally Reactive, ratings of nearly 5000 clinically referred children,
Anxious/Depressed, Somatic Complaints, With- and norms are based on results from a nationally
drawn, Attention Problems & Aggressive Behavior, representative sample of 1753 children aged 6 to 18
and Sleep Problems). A Language Development years11 (Box 7C-3).
Survey is included to screen for language delays. ASEBA materials are backed by extensive research
DSM-oriented scales pertaining to affective problems, in their development and have been used in more
anxiety problems, pervasive developmental problems, than 6000 studies pertaining to a broad range of
attention-deficit/hyperactivity problems, and opposi- behavioral health topics. There is strong support for
tional defiant problems are now available. its use with multidimensional child assessments in
The Child Behavior Checklist for Ages 6-18 pediatric settings, (e.g., Mash and Hunsley2 ; Riekert
(CBCL/6-18) similarly obtains reports from parents, et al,72 Stancin and Palermo73), although criticisms
close relatives, and/or guardians regarding school- have been raised about the validity of the CBCL for
aged children’s competencies and behavioral/emo- populations of chronically ill children.74

BOX 7C-3
CASE 2: BEHAVIORAL AND EMOTIONAL ASSESSMENT DISCUSSION
The behavior problem profiles obtained on the CBCL/6- sic Personality Inventory–Adolescent indicated that she
18 and the Youth Self-Report for Rachel are shown in was experiencing high levels of general distress. Eleva-
the following two illustrations. On the CBCL problem tions on clinical scales 2,3,7,8,0 suggested that she may
scales (completed by her mother), Rachel’s Total Prob- have felt anxious, lonely, and pessimistic much of the
lems, Internalizing, and Externalizing scores and syn- time and may have felt isolated from others and inferior.
drome scales were all in the normal ranges for girls aged In other words, Rachel reported having high levels of
12 to 18. Similarly, a teacher completed a Teacher Report internalizing symptoms, as well as difficulties managing
Form, and results were all within the normal range. social relationships and aggression. Cross-informant
However, on the Youth Self-Report problem scales, comparisons indicate that adults in Rachel’s life were
Rachel reported more problems than are typically not aware of the level of her internal distress. Discrepan-
reported by teenage girls, particularly withdrawn behav- cies between Rachel’s self-report of symptoms and the
ior, somatic complaints, problems of anxiety or depres- ratings by her mother became a springboard for validat-
sion, problems in social relationships, thought problems, ing Rachel’s need for mental health attention and led to
attention problems, and problems of an aggressive better communication within the family.
nature. Rachel’s responses on the Minnesota Multipha-
 BOX 7C-3
CASE 2: BEHAVIORAL AND EMOTIONAL ASSESSMENT DISCUSSION—cont’d
CBCL/6-18—Syndrome Scale Scores for Girls 12-18

ID: Gender: Female Date Filled: 01/03/2006 Clinician: Informant: Jane Doe
Name: Rachel (none) Age: 15 Birth Date: 10/01/1990 Agency: Relationship: Biological Mother
Verified: Yes

100 Internalizing Externalizing

95 C
L
90 I
N
85 I
T
C
S 80 A
C L
O 75
R 70
E
65
N
60 O
R
55 M
50 A
L
Anxious/ Withdrawn/ Somatic Social Thought Attention Rule-Breaking Aggressive
Depressed Depressed Complaints Problems Problems Problems Behavior Behavior
Total Score 0 5 1 2 0 5 2 2
T Score 50 63 53 54 50 59 54 50
Percentile 50 90 62 65  50 81 65 50

0 14.Cries 1 5.EnjoysLittle 0 47.Nightmares 0 11.Dependent 0 9.MindOff 0 1.ActsYoung 0 2.Alcohol 0 3.Argues

0 29.Fears 1 42.PreferAlone 0 49.Constipate 0 12.Lonely 0 18.HarmSelf 1 4.FailsToFinish 0 26.NoGuilt 0 16.Mean
0 30.FearSchool 0 65.Won’tTalk 0 51.Dizzy 0 25.NotGetAlong 0 40.HearsThings 1 8.Concentrate 1 28.BreaksRules 0 19.DemAtten
CHAPTER 7

0 31.FearDoBad 1 69.Secretive 0 54.Tired 0 27.Jealous 0 46.Twitch 0 10.SitStill 0 39.BadFriends 0 20.DestroyOwn

0 32.Perfect 0 75.Shy 0 56a.Aches 0 34.OutToGet 0 58.PicksSkin 1 13.Confused 1 43.LieCheat 0 21.DestroyOther
0 33.Unloved 0 102.LacksEnergy 0 56b.Headaches 0 36.Accidents 0 59.SexPartsP 1 17.Daydream 0 63.PreferOlder 1 22.DisbHome
0 35.Worthless 1 103.Sad 0 56c.Nausea 0 38.Teased 0 60.SexPartsM 0 41.Impulsive 0 67.RunAway 0 23.DisbSchool
0 45.Nervous 1 111.Withdrawn 0 56d.EyeProb 0 48.NotLiked 0 66.RepeatsActs 1 61.PoorSchool 0 72.SetsFires 0 37.Fights
0 50.Fearful 1 56e.SkinProb 1 62.Clumsy 0 70.SeesThings 0 78.Inattentive 0 73.SexProbs 0 57.Attacks
0 52.Guilty 0 56f.Stomach 0 64.PreferYoung 0 76.SleepsLess 0 80.Stares 0 81.StealsHome 0 68.Screams
0 71.SelfConsc 0 56g.Vomit 1 79.SpeechProb 0 83.StoresUp 0 82.StealsOut 1 86.Stubborn
0 91.TalkSuicide 0 84.StrangeBehv 0 90.Swears 0 87.MoodChang
0 112.Worries 0 85.StrangeIdeas 0 96.ThinksSex 0 88.Sulks
0 92.SleepWalk 0 99.Tobacco 0 89.Suspicious
Screening and Assessment Tools

0 100.SleepProblem 0 101.Truant 0 94.Teases

0 105.UsesDrugs 0 95.Temper
0 106.Vandalism 0 97.Threaten
0 104.Loud
161

Copyright 2001 T.M Achenbach B  Borderline clinical range; C  Clinical range Broken lines  Borderline clinical range
 BOX 7C-3
162

CASE 2: BEHAVIORAL AND EMOTIONAL ASSESSMENT DISCUSSION—cont’d

CBCL/6-18—Internalizing, Externalizing, Total Problems, Other Problems for Girls 12-18

ID: Gender: Female Date Filled: 01/03/2006 Clinician: Informant: Jane Doe
Name: Rachel (none) Age: 15 Birth Date: 10/01/1990 Agency: Relationship: Biological Mother
100
Other Problems

95
0 6.BMOut
0 7.Brags
90 0 15.CruelAnimal
0 24.NotEat
85 C 0 44.BiteNail
L
0 53.Overeat
I
N 0 55.Overweight
80
I
0 56h.OtherPhys
T C
A 0 74.ShowOff
75 L
S 0 77.SleepsMore
C
O 0 93.TalkMuch
R 70 0 98.ThumbSuck
E
0 107.WetsSelf
65 0 108.WetsBed
0 109.Whining
DEVELOPMENTAL AND BEHAVIORAL PEDIATRICS: EVIDENCE AND PRACTICE

0 110.WishOppSex
60
N 0 113.OtherProb
O
R
55
M
A
L
50

Internalizing Problems Externalizing Problems Total Problems

Total Score 6 4 17

T Score 52 49 50

Percentile 58 46 50

B Borderline clinical range; C  Clinical range Broken lines  Borderline clinical range
 CBCL/6-18—DSM-Oriented Scales for Girls 12-18

ID: Gender: Female Date Filled: 01/03/2006 Clinician: Informant: Jane Doe
Name: Rachel (none) Age: 15 Birth Date: 10/01/1990 Agency: Relationship: Biological Mother
100

95

90 C
L
I
85 N
I
80 C
T A
75 L
S
C
O 70
R
E 65
N
60 O
R
55 M
A
50 L
Attention
Deficit/ Oppositional
Affective Anxiety Somatic Hyperactivity Defiant Conduct
Problems Problems Problems Problems Problems Problems
Total Score 2 0 1 2 2 2
T Score 54 50 54 52 52 55
Percentile 65 50 65 58 58 69
1 5.EnjoysLittle 0 11.Dependent 0 56a.Aches 1 4.FailsToFinish 0 3.Argues 0 15.CruelAnimal
0 14.Cries 0 29.Fears 0 56b.Headaches 1 8.Concentrate 1 22.DisbHome 0 16.Mean
CHAPTER 7

0 18.HarmSelf 0 30.FearSchool 0 56c.Nausea 0 10.SitStill 0 23.DisbSchool 0 21.DestroyOther

0 24.NotEat 0 45.Nervous 0 56d.EyeProb 0 41.Impulsive 1 86.Stubborn 0 26.NoGuilt
0 35.Worthless 0 50.Fearful 1 56e.SkinProb 0 78.Inattentive 0 95.Temper 1 28.BreaksRules
0 52.Guilty 0 112.Worries 0 56f.Stomach 0 93.TalkMuch 0 37.Fights
0 54.Tired 0 56g.Vomit 0 104.Loud 0 39.BadFriends
0 76.SleepsLess 1 43.LieCheat
0 77.SleepsMore 0 57.Attacks
0 91.TalkSuicide 0 67.RunAway
0 100.SleepProb 0 72.SetsFires
0 102.Underactiv 0 81.StealsHome
1 103.Sad 0 82.StealsOut
0 90.Swears
Screening and Assessment Tools

0 97.Threaten
0 101.Truant
0 106.Vandalism
B  Bord rline clinical range; C  Clinical range Broken lines  Borderline clinical range
163
 BOX 7C-3
CASE 2: BEHAVIORAL AND EMOTIONAL ASSESSMENT DISCUSSION—cont’d
164

YSR/11-18—Syndrome Scale Scores for Girls

ID: Gender: Female Date Filled: 01/03/2006 Clinician: Informant: Self

Name: Rachel (none) Age: 15 Birth Date: 10/01/1990 Agency: Relationship: Self
Verified: Yes

100 Internalizing Externalizing

95 C
90 L
I
85 N
T I
S 80 C
C A
O 75 L
R 70
E
65
60 N
O
55 R
M
50 A
L
Somatic Anxious/ Social Thought Attention Delinquent Aggressive
Withdrawn Complaints Depressed Problems Problems Problems Behavior Behavior
Total Score 12 13 27 12 8 16 5 17
T Score 85-C 79-C 88-C 85-C 69-B 90-C 62 67-B
Percentile 98 98 98 98 97 98 89 96

2 42.PreferAlone 1 51.Dizzy 2 12.Lonely 0 1.ActsYoung 2 9.MindOff 0 1.ActsYoung 2 26.NoGuilt 2 3.Argues

1 65.Won’tTalk 2 54.Tired 2 14.Cries 2 11.Dependent 1 40.HearsThings 2 8.Concentrate 0 39.BadCompany 0 7.Brags
2 69.Secretive 2 56a.Aches 0 18.HarmSelf* 2 25.NotGetAlong 0 66.RepeatsActs 2 10.SitStill 0 43.LieCheat 2 16.Mean
DEVELOPMENTAL AND BEHAVIORAL PEDIATRICS: EVIDENCE AND PRACTICE

1 75.Shy 2 56b.Headaches 2 31.FearDoBad 1 38.Teased 1 70.SeesThings 2 13.Confused 0 63.PreferOlder 0 19.DemAtten

2 102.Underactive 2 56c.Nausea 2 32.Perfect 2 48.NotLiked 0 83.StoresUp* 2 17.Daydream 1 67.RunAway 0 20.DestroyOwn
2 103.Sad 0 56d.EyeProb 2 33.Unloved 2 62.Clumsy 2 84.StrangeBehav 2 41.Impulsive 1 72.SetsFires 0 21.DestroyOther
2 111.Withdrawn 2 56e.SkinProb 2 34.OutToGet 1 64.PreferYoung 2 85.StrangeIdeas 2 45.Nervous 0 81.StealsHome 0 23.DisobeySchl
2 56f.Stomach 2 35.Worthless 2 111.Withdrawn* 2 61.PoorSchool 0 82.StealsOther 1 27.Jealous
0 56g.Vomit 2 45.Nervous 2 62.Clumsy 1 90.Swears 2 37.Fights
2 50.Fearful 0 101.Truant 1 57.Attacks
0 52.Guilty 0 105.AlcDrugs 1 68.Screams
2 71.SelfConsc 0 74.ShowOff
2 89.Suspicious 1 86.Stubborn
1 91.ThinkSuic* 2 87.MoodChang
2 103.Sad 1 93.TalkMuch
2 112.Worries 0 94.Teases
2 95.Temper
1 97.Threaten
1 104.Loud
B  Borderline clinical range; C  Clinical range Broken lines  Borderline clinical range *Not on CBCL or TRF construct
 YSR/11-18—Internalizing, Externalizing, Total Problems, Other Problems, Profile ICCs, Clinical T Scores for Girls

ID: Gender: Female Date Filled: 01/03/2006 Clinician: Informant: Self

Name: Rachel (none) Age:15 Birth Date: 10/01/1990 Agency: Relationship: Self
100
Other Problems
95
0 5.ActOppSex 2 55.Overweight
90 C 1 22.DisobeyParent 2 56h.OtherPhys
L 1 24.NotEat 2 58.PicksSkin
85 I
N 2 29.Fears 2 76.SleepsLess
T I
80 2 30.FearsSchool 0 77.SleepsMore
C
S A
C 75 L
2 36.GetsHurt 2 79.SpeechProb
O 2 44.BiteNail 0 96.ThinksSex
R
E 70
2 46.Twitch 1 99.TooNeat
65 2 47.Nightmares 2 100.SleepProb
1 53.Overeat 0 110.WishOpSx
60 N
O
55 R
M
A
50 L
Internalizing Problems Externalizing Problems Total Problems
Total Score 50 22 130
T Score 87-C 66-C 85-C Not In Total Problem Score
Percentile 98 95 98 0 2.Allergy 0 4.Asthma

B  Borderline clinical range; C  Clinical range Broken lines  Borderline clinical range Copyright 1999 T.M. Achenbach
ADM Version 4

Cross-Informant Profile Types Profile Types Specific to YSR

Intraclass Significant Intraclass Significant
Corr.(ICC) Similarity* Corr.(ICC) Similarity*
CHAPTER 7

Withdrawn 0.474 No YSR Social 0.588 No

Somatic Complaints 0.663 No Delinquent 0.869 No
Social Problems 0.036 No
Delinquent-Aggressive 0.878 No
*Intraclass correlations 0.444 indicate statistically significant similarity between a child’s profile and previously identified profile types.

Somatic Anxious/ Social Thought Attention Delinquent Aggressive

T scores based
Screening and Assessment Tools

Withdrawn Complaints Depressed Problems Problems Problems Behavior Behavior

on clinical
sample 73 71 71 75 62 74 48 56

Copyright by T. M. Achenbach. Reproduced with permission.

165
166 DEVELOPMENTAL AND BEHAVIORAL PEDIATRICS: EVIDENCE AND PRACTICE

BEHAVIOR ASSESSMENT SYSTEM FOR This measure contains 42 items, is completed by a

CHILDREN–SECOND EDITION (BASC-2) 75 parent or caregiver, and can be used fi rst to screen for
The BASC-2 is another broad, multidimensional rating possible concerns and then followed with the ITSEA
scale system designed to measure behavior and emo- for more comprehensive evaluation.
tions of children and adolescents. It includes a Parent
MINNESOTA MULTIPHASIC PERSONALITY
Rating Scale, a Teacher Rating Scale, and a Self-Report
INVENTORY–ADOLESCENT (MMPI-A) 78
of Personality. Norms are provided for ages 2 years 0
months through 21 years 11 months (Teacher Rating Although it is not a behavior rating scale per se, the
Scale and Parent Rating Scale) and 8 years 0 months MMPI-A is a self-report questionnaire that yields
through college age (Self-Report of Personality). T- indices pertaining to the nature and severity of symp-
scores and percentiles for a general population and toms in relation to peers with psychiatric disorders.
clinical populations are available for interpretation. Norms are based on a nationally representative sample
Computer scoring and Spanish language forms are of more than 1600 male and female adolescents in the
available. The Parent Rating Scale requires approxi- United States. The MMPI-A scoring yields T-scores for
mately a fourth grade reading level; forms pertaining 7 validity scales, 10 clinical scales, 15 content scales,
to three age levels—preschool (ages 2 to 5), child (ages and other supplementary scales and indices. The
6 to 11), and adolescent (ages 12 to 21)—measure MMPI-A is a lengthy measure (478 true/false items)
adaptive and problem behaviors in the community that requires at least a sixth grade reading level;
and home setting. The Parent Rating Scale contains therefore, some adolescents fi nd it to be difficult to
134 to 160 items and entails use of a four-choice complete. However, a shortened 350-item version
response format. Clinical scales include Aggression, yielding basic results can be administered to save
Anxiety, Attention Problems, Atypicality, Conduct administration time.
Problems, Depression, Hyperactivity, Somatization,
and Withdrawal. Adaptive scales include Activities of Projective Techniques
Daily Living, Adaptability, Functional Communica-
tion, Leadership, and Social Skills. The Teacher Rating Projective assessment techniques encourage a respon-
Scale similarly measures adaptive and problem behav- dent to “project” issues, concerns, and perceptions
iors in the preschool or school setting. An additional onto ambiguous stimuli such as an inkblot or a picture.
clinical domain in the Teacher Rating Scale is Learn- The basic premise is that when the child is faced with
ing Problems; Study Skills are measured on the Adap- an ambiguous stimulus or one requiring perceptual
tive Scales. The Self-Report of Personality provides organization, underlying psychological issues affect-
insight into a child’s or adolescent’s thoughts and feel- ing the child will influence interpretation of these
ings, including scales such as Anxiety, Attention Prob- stimuli. The most commonly used projective tech-
lems, Sense of Inadequacy, Social Stress, Interpersonal niques with children include use of child human
Relations, and Self Esteem (among others). One strong figure or family drawings, storytelling responses to
advantage of the BASC-2 over other rating scales is the pictures or photographs, and reactions to Rorschach
inclusion of validity and response set indexes that may inkblots. Once the mainstay of personality assess-
be used to judge the quality of responses. ment, projective assessment techniques have fallen
out of favor in the era of evidence-based assessment
techniques. However, some techniques continue to
INFANT-TODDLER SOCIAL-EMOTIONAL have clinical utility and validity with specific assess-
ASSESSMENT SCALE (ITSEA) 76 ment purposes. They can provide clues that subse-
The ITSEA provides a comprehensive analysis of quently can be pursued with interviews and other
emerging social-emotional development of infants techniques. For example, family drawings can be a
and toddlers aged 12 to 36 months. It includes parallel helpful source of qualitative information about a
parent and child care provider forms that contains child’s view of family relations, especially with
166 items focusing on behavioral and emotional prob- younger children with more limited verbal expres-
lems and competencies. A national normative sample sions. Responses to incomplete sentences, story cards,
consisted of 600 children, with clinical groups that and “3 wishes” (“if you could have 3 wishes, what
included children with autism, language delays, pre- would they be?”) can reveal insights into a child’s
maturity, and other disorders. English and Spanish internal representations of relationships. In addition,
forms are available with computer or hand scoring the Rorschach has been shown to be a valid method
that yield T-scores for 4 broad domains, 17 specific for examining perceptual accuracy in youth with
subscales, and 3 index scores. An interesting feature possible thought disorders when used with validated
of the ITSEA is its companion measure, the Brief scoring systems such as John E. Exner’s system for
Infant-Toddler Social-Emotional Assessment Scale.77 scoring the Rorschach test.79
 CHAPTER 7 Screening and Assessment Tools 167

Assessing Peer Relationships on a range of child social behaviors. However, for

clinical purposes, it may difficult (and impractical) to
Peer perspectives contain unique and important in- obtain peer ratings of an individual child. For this
formation about children but are usually missing in reason, parent, teacher, and rating scales of behavior
multi-informant clinical assessments. Peers play criti- in children can be used to as a more practical alterna-
cal social roles in children’s lives and have access to tive for multi-informant assessments of peer relation-
information that adults may not have and that chil- ships and functioning. For example, the ASEBA scales
dren may be reluctant to self-report. For example, (e.g., the CBCL Teacher Report Form) include positive
social acceptance within a peer group is an important peer relationship items on the competence scales and
aspect of a child’s functional status, but it can be dif- a social problems scale highlighting peer difficulties
ficult to assess accurately by interview or parent on the problem scales.
report. Sociometric assessments that use peer nomi-
nation methods have been developed as a systematic
way of gathering information about the extent to Testing for Specific Problems
which a child is accepted or rejected within a peer
group.80 Strategies may involve asking children by PARENT-CHILD INTERACTIONS
interview or on paper to nominate three classmates Parent-child interaction problems contribute signifi-
with whom they most like to play (positive nomina- cantly to the origin and maintenance of a wide range
tions/peer acceptance) and three classmates with of behavior problems in children. Therefore, treat-
whom they would least like to play (negative nomina- ment of children in mental health settings, especially
tions/peer rejection). An alternative method is for children with negative, externalizing behaviors, often
children to rate how much they like to play with each focuses on promoting optimal parenting styles and
classmate, for example, on a scale from 1 (“I don’t like parent-child interactions. For these reasons, assess-
to”) to 5 (“I like to a lot”). Using various statistical ment of parent-child interactions is essential when
classification schemes, children can be considered treatment interventions are planned for children with
to be popular, accepted, rejected, neglected, or a wide range of behavioral problems.82
controversial. Parent-child interactions may be assessed through
Peer nomination assessment instruments have observation, Q-sorts (cards with descriptive labels
been used to measure specific domains of child func- are “sorted” into piles as to how well they pertain
tioning besides peer acceptance. Techniques often to a child), or rating scales. Qualitative assessments
involve presenting children in a classroom with a list through observations may be conducted in vivo or by
of behavioral descriptions and asking them to select using videotape recordings of parent-child interac-
which of their peers best match each descriptor. tions. The Dyadic Parent-Child Interaction Coding
Peer nomination approaches with acceptable reliabil- System83 is widely used in clinical and research set-
ity and validity have been developed to obtain tings to code direct observations in a standardized
peer ratings for a number of specific behavioral or laboratory setting. Observations (through a one-way
emotional problem domains in children, such as mirror or videotape) are made during three standard
ADHD symptoms, aggression and withdrawal, and parent-child interaction settings: child-led play,
depression.81 parent-led play, and cleaning up. Parent and child
The Peer-report Measure of Internalizing and verbalizations and physical behaviors are coded along
Externalizing Behavior81 was developed to assess a 25 categories. Reliability and validity studies provide
broad range of peer-reported externalizing and inter- good support for use of the Dyadic Parent-Child Inter-
nalizing child psychopathology. As with other peer- action Coding System to evaluate baseline and post
nomination inventories, students are provided with treatment behaviors, as well as to measure ongoing
classroom roster sheets that contain listings of all of treatment progress.83 In addition to this structured
the children in the classroom. Then, they are asked method of observation, it is sometimes useful to
to select up to three classmates (either gender) who observe parent-child interactions in more naturalistic
best fit the description read to them (e.g., “worry settings.84
about things a lot” or “get mad and lose their temper”). The classic research method for assessing the
Preliminary reports suggest that this measure dem- quality of parent-child relationships is the laboratory
onstrates adequate reliability and validity as a broad Strange Situation Paradigm developed by Ainsworth
measure of psychopathology. (described by Shaddy and Colombo85). Strength and
Peer nomination procedures may be useful for psy- quality of an infant’s attachment to a caregiver are
chosocial screening in the classroom, evaluating the assessed by placing the child in situations in which
effectiveness of mental health interventions on social he or she is alone with the caregiver, separated from
behaviors in school settings, and conducting research caregiver and introduced to a stranger, and then
168 DEVELOPMENTAL AND BEHAVIORAL PEDIATRICS: EVIDENCE AND PRACTICE

reunited with the caregiver. The infant can be classi- depression, it does not distinguish between depres-
fied as securely attached, ambivalent/resistant, avoid- sion and anxiety very well.86
ant, or disorganized on basis of reactions in those Assessment of depression in infants and preschool
situations. children is very challenging because of the difficulty
Information about parent-child interactions in of eliciting self-report information in a reliable or
clinical settings can be obtained from sorting tech- valid manner. Caregiver reports obtained with broad-
niques and rating scales. The Attachment Q-Set (as band measures (such as the CBCL/11/2-5 or Teacher
described by Querido and Eyberg) 82 is a measure of a Report Form 1-5) may be a useful alternative or
child’s attachment related behaviors. Parents sort 90 adjunctive tool. A new parent report screening
behavioral dimensions of security, dependency, and measure of preschool depression is the Preschool
sociability into piles according to the extent to which Feelings Checklist.92 This 20-item checklist of depres-
they describe the child. Results of the Q-set are related sive symptoms in young children was shown to have
to results obtained by exposing infants to the Strange high internal consistency and to be correlated highly
Situation Paradigm. In addition, there are a variety of with the Diagnostic Interview for Children–IV and
measures by which to assess various dimensions of the CBCL on a sample of 174 preschool children from
parent-child relationships and interactions through a primary care setting. Moreover, preliminary study
the use of rating scales and checklists.82 suggested that it had acceptable sensitivity and speci-
ficity when a cutoff score of 3 was used.92
DEPRESSION
Self-report questionnaires and rating scales are usu- ANXIETY
ally preferred over parent or teacher rating scales for Screening for anxiety disorders is most often done
screening depression in children and teens and for with rating scales, although data supporting their use
monitoring symptoms during treatment. However, are sparse, and several scales have been shown to
they tend to have limited sensitivity and specificity measure different anxiety constructs.93 The Multi-
and therefore should be used cautiously.86 Moreover, dimensional Anxiety Scale for Children94 is a youth
they can be influenced by respondent bias if the child self-report rating scale that assesses anxiety in four
does not want to divulge information. The most domains: physical symptoms, social anxiety, harm
widely used depression rating scale for children and avoidance, and separation/panic. Children aged 8 to
adolescents is the Children’s Depression Inventory.87 19 are asked to rate how true 39 items are for them.
This instrument includes 27 items covering a range of Internal consistency reliability coefficients of sub-
depressive symptoms and associated features and it scales and total scores range from 0.74 to 0.90, although
can be used in youth ages 7-17. Research on the Chil- interrater reliability is lower (0.34 to 0.93). The Mul-
dren’s Depression Inventory has generally shown it to tidimensional Anxiety Scale for Children has some
have good internal consistency, test-retest reliability, support for use as a screener for anxiety disorders, as
and sensitivity to change, but the evidence for dis- does the Social Phobia and Anxiety Inventory for
criminant validity is more limited.86 Children,95 the Social Anxiety Scale for Children96
The Mood and Feeling Questionnaire88 is a 32-item and the Social Anxiety Scale for Adolescents.97 The
measure of depression (and there is an even briefer Revised Children’s Manifest Anxiety Scale,98 although
13-item version) that has been shown to have good widely used, does not appear to discriminate between
estimates of reliability, discriminant validity, and sen- children with anxiety disorders and those with other
sitivity to change for children aged 8 to 18 years.86 psychiatric conditions and therefore should be used
The Reynolds Child Depression Scale89 and the Reyn- cautiously as a screening or diagnostic tool.93 However,
olds Adolescent Depression Scale90 are 30-item scales it does appear to be sensitive to change and therefore
for youth aged 8 to 12 and 13 to 18. These scales have may be a useful tool for monitoring treatment effects.
also been shown to be internally consistent and stable,
although there is more limited evidence of discrimi- ATTENTION-DEFICIT/
nant validity and sensitivity to change.86 HYPERACTIVITY DISORDER
The Children’s Depression Rating Scale91 is an ADHD is one of the most common childhood mental
interesting hybrid measure that combines separately health disorders and a frequent diagnostic consider-
obtained responses from a child and an informant ation in developmental-behavioral pediatric settings.
along with the clinician’s behavioral observations. Despite the vast literature on ADHD psychopathology
Seventeen items assess cognitive, somatic, affective, and treatment, considerably less research has been
and psychomotor symptoms; cutoff scores provide directed toward determining best assessment prac-
estimates of level of depression. Moderate reliability, tices.5 The most efficient empirically based assessment
convergent validity, and sensitivity to treatment have methods for diagnosing ADHD are parent and teacher
been demonstrated, but, as with most measures of symptom rating scales based on DSM-IV criteria (e.g.,
 CHAPTER 7 Screening and Assessment Tools 169

the ADHD Rating Scale99 or the Vanderbilt ADHD BOX 7C-4

Diagnostic Scales100 ) or derived from a rational or CASE 3: BEHAVIORAL ASSESSMENT RESULTS
empirical basis (e.g., BASC or CBCL).101 Broadband
rating scales (such as the BASC or CBCL) were not Jose is reported to have a short attention span and
recommended for diagnosing ADHD in the American to display social and academic impairment. Parent
Academy of Pediatrics Diagnostic Guidelines102 because and teacher CBCL measures were obtained to broadly
broad domain factors (e.g., externalizing) do not examine the nature and severity of behavior problems
discriminate children referred for ADHD from non- (the Spanish version was administered to parents).
referred peers.103,104 However, a more recent review5 Clinically significant scores were on the following
challenged this recommendation, concluding that the parent and teacher subscales: Social Problems, Atten-
Attention Problems subscales within the CBCL and tion Problems, and Aggressive Behavior. Scores on
BASC do accurately identify children with ADHD. the Teacher Report Form Attention Problems sub-
Because of their ability to identify other comorbid scales were further clinically significant for Inatten-
conditions and impairments, broadband measures tion (98th percentile) and Hyperactivity-Impulsivity
(which also have advantages of extensive normative (97th percentile). Scores on the Vanderbilt ADHD
information across gender and developmental ages) Diagnostic Scales, used to collect information about
are probably more efficient than DSM-IV–based rating the presence of DSM-IV symptoms, showed that Jose’s
scales for diagnosing ADHD.5 mother and teacher considered him to display symp-
As with any disorder, ADHD should not be diag- toms associated with ADHD, combined type. Mater-
nosed with symptom rating scales alone. Clinical nal reports on the Vanderbilt scales were considered
interviews and other sources of data are needed to cautiously because of possible language and cultural
establish pertinent history, to rule out other disorders differences from the normative sample. Clinical inter-
that may better account for symptoms (e.g., autism, views and academic screening to rule out communi-
low intellectual functioning, post-traumatic stress cation problems and learning disorders led to
disorder, adjustment problems), and to assess comor- diagnoses of ADHD and Oppositional Defiant Disor-
bid conditions. Interestingly, DSM-based structured der. Idiographic measures of daily behavior problems
interviews have not been shown to add incremental targeted out-of-seat behaviors during instruction,
validity to parent and teacher rating scales.5 Behav- schoolwork completion, and peer aggression for
ioral observation assessment procedures have been behavioral interventions. A treatment plan was devel-
shown to be empirically valid in numerous studies oped to include a trial of stimulant medication, home
but practically impossible in most clinical settings, and classroom behavioral interventions, parent train-
although parent and teacher proxy observational ing in behavior management skills, and a social skills
measures have been developed.5 Measures of child group.
functioning and impairment in key domains includ-
ing peer relationships, family relationships, and aca-
demic settings, should be included in an ADHD diagnostic parent interview that elicits current
assessment and are likely to be more useful for treat- behavior and developmental history. It yields three
ment purposes than are global ratings of impairment. algorithm scores measuring social difficulties, com-
Moreover, assessment of ADHD needs to emphasize munication deficits, and repetitive behaviors; these
situational contexts and socially valid target behav- scores have been shown to distinguish children with
iors (i.e., functional analysis of behavior) necessary autism from children with other developmental
for treatment planning (Box 7C-4). delays. I is very labor intensive in terms of training
(3 days) and administration time (3 hours) and there-
AUTISM SPECTRUM DISORDERS fore has been used more in research than in clinical
Empirically based procedures for assessing ASDs have settings.105
emerged since the 1990s, greatly improving the accu- The Autism Diagnostic Observation Schedule
racy and validity of the diagnoses and the ability to (ADOS)107 is a widely used semistructured, interac-
plan and evaluate interventions. Ozonoff and associ- tive assessment of ASD symptoms. It includes four
ates105 summarized the current state of the art with graded modules and can be used with a broad range
regard to assessment of ASDs and recommended a of patients from the very young and nonverbal to
core assessment battery that includes collecting diag- high-functioning, verbal adults. Modules 1 and 2,
nostic information from parents and by direct geared toward developmentally younger children,
observation along with standardized measures of assess social interest, joint attention, communication
intelligence, language, and adaptive behavior. One behaviors, symbolic play, and atypical behaviors.
ASD-specific measure is the Autism Diagnostic Inter- Modules 3 and 4 assess higher level functioning indi-
view–Revised,106 a comprehensive, semistructured viduals, with a focus on conversational reciprocity,
170 DEVELOPMENTAL AND BEHAVIORAL PEDIATRICS: EVIDENCE AND PRACTICE

empathy, insight into social relationships, and special There are many family self-report questionnaires tar-
interests. Administration time is typically less than geting different aspects of functioning that may be
an hour. For either pair of modules there are empiri- useful in family assessments, especially in research
cally derived cutoff scores for autistic disorder and for settings.112 Although questionnaires have psycho-
broader ASDs (such as Asperger syndrome). Studies metric appeal, they carry biases of the individual
on the psychometric properties of the Autism Diag- completing them, which is counter to the spirit of
nostic Observation Schedule indicate excellent reli- family assessment. Moreover, questionnaires may
ability (interrater, internal consistency, and test-retest have limited utility when specific treatment recom-
reliability) for each module, as well as excellent diag- mendations are developed in clinical settings for a
nostic validity.105 particular family’s set of concerns.111 A popular
A parent-report alternative to the Autism Diagnos- example of a parent report family questionnaire with
tic Interview–Revised for children older than 4 years research and clinical applications is the Parenting
is the Social Communication Questionnaire.108 This Stress Index.113 This index consists of 120 items about
instrument has a lifetime-behavior version helpful for child characteristics, parent personality, and situa-
diagnostic purposes, as well as a current-behavior tional variables, and it yields a Total Stress Score, as
version that can be used for evaluating a person’s well as scale scores for child and parent characteris-
change over time.105 Currently, the widely popular tics. It has been translated and validated for use with
Gilliam Autism Rating Scale109 has not been subjected a variety of international populations and has been
to sufficient psychometric study to recommend its shown to be useful in a clinical contexts.
use.105 Several parent report measures have been
developed to help diagnose other ASD disorders (e.g.,
Asperger syndrome), but at present, there is not suf-
Functional Outcomes
ficient empirical study to recommend their use. A Measures of global functioning are typically ratings
clinically practical method of direct observation for of a clinician’s judgment about a child or adolescent’s
children older than 24 months is the Childhood overall functioning in day-to-day activities at school,
Autism Rating Scale.110 Little training is necessary to at home, and in the community.114 Measures of global
rate 15 items on a 7-point scale (from “typical” to functioning are useful for identifying need for treat-
“severely deviant”); the results yield a composite score ment, as well as for monitoring treatment effects and
that is correlated highly with that of the Autism predicting treatment outcome. The importance of
Diagnostic Interview–Revised (although it may over- global functioning is reflected in the placement of the
identify children with mental retardation as having Global Assessment of Functioning—which stipulates
ASD). that impairment in one of more areas of functioning
is necessary in order to meet criteria for a diagno-
sis—as Axis V on the DSM-IV. The Global Assessment
Family Assessment of Functioning is a scale of a mental health contin-
Evaluations in developmental and behavioral pediat- uum from 1 to 100 with 10 anchor descriptions;
rics often include a family assessment in order to higher scores reflect better functioning. For example,
understand the interpersonal dynamics of the family a score between 31 and 40 would be given for a child
system.111 Using an unstructured interview format, a cli- with major functional impairment in several areas
nician may inquire about family structure, roles, and (frequently beats up younger children, is unruly at
functioning and explore each family member’s per- home, and is failing in school); a score between 61
ception of a presenting issue or problem. This assess- and 70 is given to a child with mild symptoms (mild
ment approach is often useful in family therapy depressed mood) or some difficulties in functioning
sessions. Structured interviews may be employed to (disruptive in school) but who generally functions
ensure that specific areas or topics are covered. Geno- fairly well and who has good social relationships.
grams are graphic representations of families that Shaffer and colleagues modified the anchors of the
begin with a family tree and may include additional Global Assessment of Functioning to pertain better
details about family structure, cohesiveness or con- to youth, creating the Children’s Global Assessment
fl icts, timelines of events, and family patterns (e.g., Scale (CGAS).115 This instrument yields one score
domestic violence, substance abuse, divorce, suicides, and has been used in a large number of psychia-
health conditions, presence of behavioral disorder). tric outcome studies, especially medication-related
Formal, validated observational approaches to family research.111
assessment typically involved trained observers who A widely used measure of functioning is the Child
coded ratings during live or videotaped observations and Adolescent Functional Assessment Scale.116 This
of family interactions and are mostly confi ned to measure is a clinician-rated instrument consisting of
research settings. behavioral descriptions (e.g., is expelled from school,
 CHAPTER 7 Screening and Assessment Tools 171

bullies peers) grouped into levels of impairment for referred for developmental and behavioral services.
each of five domains: role performance (school/work, As a result of the comprehensive evaluation, the teen-
home, community), behavior toward others, moods/ ager in Case 2 (Rachel) received a diagnosis of Major
self-harm, substance use, and thinking. The Child Depression, single episode, along with Cognitive Dis-
and Adolescent Functional Assessment Scale has been order not otherwise specified. Treatment recommen-
shown to have considerable criterion-related and pre- dations included individual cognitive behavior therapy
dictive validity and is widely used to evaluate outcome to focus on adaptive coping, a trial of antidepressive
in clinical settings and in clinical research.111 medication, family education, and educational adjust-
Adaptive functioning measures such as the Vine- ments to allow her to have more time to complete
land Adaptive Behavior Scales117 are used to assess school work. She opted to continue to take advanced
personal and social skills needed for everyday living language courses but enrolled in slower paced math
and are especially useful for identifying children with courses. Interventions were very successful; subse-
mental retardation, developmental delays, and per- quent assessments were used to verify treatment
vasive developmental disorders. The Vineland scales effects.
include survey interview and parent/caregiver rating A psychological evaluation is complete when assess-
forms that yield domain and adaptive behavior com- ment data have been organized, synthesized, inte-
posite standard scores (M = 100, SD = 15), percentile grated, and presented, usually in the form of a written
ranks, adaptive levels, and age-equivalent scores for report.1,17 Reports are usually independent documents
individuals from birth to age 90 years. Domains written with an intended audience in mind. They
assessed include Communication, Daily Living Skills, should include assessment fi ndings, such as relevant
Socialization, Motor Skills, and an optional Maladap- history, current problems, assets, and limitations, as
tive Behavior Index. well as behavioral observations and test interpreta-
Health-related quality-of-life (HRQOL) measures tions. A typical report includes the following sections
have been developed to evaluate functional outcomes or elements: identifying information, reason for refer-
in clinical and health services research. HRQOL mea- ral, sources of assessment information (including tests
sures differ from more traditional measures of health administered if any), behavioral observations, results
status and physical functioning by also assessing and impressions, recommendations, and summary.
broader psychosocial dimensions such as emotional, A major concern in developmental and behavioral
behavioral, and social functioning. The Pediatric assessment has been the misuse of test data.1 For
Quality of Life Inventory (PedsQL 4.0)118 is an example example, deviations from standardized procedures in
of an HRQOL measure that has been developed and test administration, disrespect for copyrights, use of
validated for use in pediatric settings. The PedsQL 4.0 tests for purposes without adequate research support,
Generic Core Scales assess physical, emotional, social, interpretation of results without taking into account
and school functioning with child self-report (ages 5 appropriate norms or reference groups, and use of a
to 18) and parallel parent proxy-report formats (for single test score for making decisions about a child
children aged 2 to 18 years). Physical Health and are among more common problems with test use. Led
Psychosocial Health summary scores are transformed by a consortium of professional associations (includ-
to a scale of 0 to 100 in which higher scores reflect ing the American Educational Research Association,
better health-related quality of life. The PedsQL 4.0 the American Psychological Association, and the
had excellent internal consistency reliability in a large National Council on Measurement in Education), the
pediatric sample, distinguished healthy children from Joint Committee on Testing Practices has ongoing
those with chronic health conditions, and was related workgroups charged with improving quality of test
to other indicators of health status.118 use. Several documents have been created to guide
professionals who might develop or use educational
or psychological tests, including Standards for Educa-
tional and Psychological Testing119 and the Code of Fair
SUMMARY AND IMPLICATIONS Testing Practices in Education (Revised).120
FOR CLINICAL CARE Another important clinical issue pertains to what
qualifications are necessary for psychological test
Interviews, psychological tests, rating scales, and administrators. Although a thorough review of these
other measurement strategies are central in the com- issues is beyond the scope of this chapter, the Joint
prehensive assessment of behavior and development Committee on Testing Practices has developed guide-
of children. Use of assessment techniques in the lines that address this issue.121,122 Most discussions
cases featured in this chapter highlight the contribu- about user qualifications emphasize knowledge and
tions of multi-informant, multimethod evidence- skills necessary to administer and interpret tests in
based approaches to the clinical care of children the context in which a particular measure is being
172 DEVELOPMENTAL AND BEHAVIORAL PEDIATRICS: EVIDENCE AND PRACTICE

used, as opposed to a particular professional degree 1980s. Critics have argued that intelligence and
or license. Some instruments can be administered achievement tests used to allocate limited educational
with relatively little training in psychometric issues resources penalize children whose family, cultural,
(e.g., clinical rating scales such as the Vanderbilt and socioeconomic status are different from middle-
ADHD Diagnostic Scales), whereas other instruments class European American children.1 Specifically, it
require extensive training and supervised experience has been argued that intelligence and achievement
(e.g., individually administered ability tests such as tests are culturally biased and thus harmful to African
the BSID or Wechsler tests). To be qualified to admin- American children and other ethnic minorities. Other
ister most of the instruments discussed in this chapter, experts have been critical of test use to label children
a test user should have extensive knowledge and skills or have argued that norm-referenced tests are imper-
related to psychometrics and measurement, selection fect in what they measure and therefore have little or
of appropriate tests, test administration, and other no utility in the classroom. Dialog on these criticisms
variables that influence test data. Such knowledge has led to improved test practices, including more
and skills generally require advanced graduate level representative normative groups, increased availabil-
coursework in psychology and supervised clinical ity of tests in languages other than English, increased
experience. Psychologists (among others) are gener- awareness of cultural factors among clinicians admin-
ally those who are qualified to use psychological tests istering and interpreting tests, and use of criterion- or
properly. curriculum-based assessments.
Proper use of tests in clinical assessments require Computers are playing more of a role in clinical
high level skills and professional judgments in order assessments. They can facilitate administration and
to make valid interpretation of scores and data col- scoring of some tests and interview methods, record-
lected from multiple sources, with the use of proper ing of observational data, preparation of reports, and
test selection, administration, and scoring proce- transmittal of assessment information.1 For example,
dures.122 When selecting methods, the clinician the CBCL’s computer scoring program yields several
evaluates whether the construction, administration score profi les, including useful cross-informant com-
procedures, scoring, and interpretation of the methods parisons along with a narrative report.67 Computer-
under consideration match the current assessment administered assessment methods have several
need, knowing that mismatches may invalidate test advantages, including eliminating human clinicians’
interpretation. Instrument selection also is influenced biases, calculation errors, and memory difficulties.
by practical considerations such as training, fami- Computers will probably be used more extensively in
liarity, personal preference, and availability of test the future to assist in selecting assessment instru-
materials. Cost considerations may also factor into ments, making diagnoses, designing interventions,
instrument selection. Test development can be very and monitoring treatment effects. However, it unlikely
costly, especially if normative samples are broadly that computers will supplant the clinician, who will
developed. Therefore, it may not be fi nancially fea- still be needed to integrate computer-generated results
sible to purchase test materials for all clinical into meaningful recommendations. In fact, there are
assessments. potential dangers of using computer-generated reports,
We wish to emphasize the importance of adhering and knowledgeable professionals understand that
to standardized administration procedures in using these reports should be used cautiously when being
psychological tests. Valid interpretation of measure- incorporated into assessment reports.
ment results cannot be made if there are deviations
in administration or scoring procedures. For example,
interpretations based on test procedures that have
been altered or shortened for convenience or other SUMMARY AND IMPLICATIONS
reasons without accompanying psychometric study FOR RESEARCH
are not valid or clinically sound. Likewise, interpreta-
tion of assessment results should never rely solely on Selecting the right measure for a specific research or
test scores.1 Clinical judgments should be made by clinical purpose can be a daunting prospect. It is
integrating assessment and observational data, taking important to recognize that developmental and
into consideration whether results are congruent with behavioral measures are not limited to published tests
other pieces of information, discrepancies from dif- and that literally thousands of unpublished, non-
ferent sources, and factors affecting the reliability and commercial inventories, checklists, scales, and other
validity of results (e.g., motivation of child, language instruments exist in the behavioral sciences litera-
barriers). ture. To avoid the time-consuming task re-creating
Use of standardized ability, achievement, and instruments, researchers are urged to investigate what
behavioral tests has come under attack since the existing measures are available to suit a particular
 CHAPTER 7 Screening and Assessment Tools 173

need. The American Psychological Association Web mendations for intervention. Thus, although knowl-
site (http://www.apa.org/science/faq-findtests.html) pro- edge about tests is important, ultimately it is the
vides helpful information about locating both pub- clinician who is the most important component of the
lished and unpublished test instruments. For example, evaluation process.
the PsycINFO database (usually available at a local
library) is an excellent source of information on the
very latest behavioral science research, including REFERENCES
testing. In addition, the Buros Mental Measurements 1. Sattler JM: Assessment of Children: Cognitive Appli-
Yearbooks123 have provided consumer-oriented, critical cations, 4th ed. San Diego: Jerome M. Sattler, 2001.
test reviews since 1938 and can provide evaluative 2. Mash EJ, Hunsley J: Evidence-based assessment of
information for informed test selection. The Buros child and adolescent disorders: Issues and challenges.
Center for Testing also offers online reviews and J Clin Child Adolesc Psychol 34:362-379, 2005.
information about nearly 4000 measures at www.unl. 3. McConaughy SH: Clinical Interviews for Children
edu/buros. Fortunately, most commercially available and Adolescents: Assessment to Intervention. New
tests can be located and purchased easily by accessing York: Guilford, 2005.
Web sites on the Internet. 4. Kazdin AE: Evidence-based assessment for child and
adolescents: Issues in measurement development and
Practitioners may be tempted to use measures
clinical applications. J Clin Child Adolesc Psychol 34:
developed as research tools for clinical purposes. This 548-558, 2005.
is unwise and may represent a misuse of an instru- 5. Pelham WE, Fabiano GA, Massetti GM: Evidence-
ment. Measures shown to be “reliable and valid” for based assessment of attention deficit hyperactivity
a research study application may have little evidence disorder in children and adolescents. J Clin Child
of validity for a particular child in a clinical context. Adolesc Psychol 34:449-476, 2005.
For example, a measure of family dysfunction that 6. Spence SH: Interviewing. In Ollendick TH, Schroeder
predicts symptom improvement in group compari- CS, eds: Encyclopedia of Clinical Child and Pediatric
sons may have little applicability in clinical settings. Psychology. New York: Kluwer Academic/Plenum,
Often, the length of the research instrument and/or 2003, pp 324-326.
the scoring procedures precludes clinical use. 7. Schroeder CS, Gordon BN: Assessment and Treatment
of Childhood Problems: A Clinician’s Guide, 2nd ed.
Unfortunately, there are as yet no clear guidelines
New York: Guilford, 2002.
or criteria with which to evaluate measures or to 8. Shaffer D, Fisher P, Lucas CP, et al: NIMH Diagnostic
decide what measures are better than others.4 How- Interview for Children–IV (NIMH DISC-IV): Descrip-
ever, with psychometric study and refi nement, many tion, differences from previous versions, and reliabil-
research tools can become important clinical mea- ity of some common diagnoses. J Am Acad Child
sures with evidence to support their use. Adolesc Psychiatry 39:28-38, 2000.
Assessment in developmental-behavioral pediat- 8a. American Psychiatric Association: Diagnostic and
rics is continually evolving in response to new research Statistical Manual of Mental Disorders, 4th ed.
and clinical problems. This chapter highlights some Washington, DC: American Psychiatric Association,
of the emerging assessment trends being studied such 1994.
as development of empirically based assessment pro- 9. Reich W: Diagnostic Interview for Children and Ado-
lescents (DICA). J Am Acad Child Adolesc Psychiatry
cedures, expansion of measures appropriate for ethnic
39:59-66, 2000.
minorities and culturally diverse populations (espe- 10. McConaughy SH, Achenbach TM: Manual for the
cially children with limited English proficiency), and Semistructured Clinical Interview for Children and
use of computer-assisted technologies. Internet and Adolescents, 2nd ed. Burlington: University of
Web-based assessment applications are of particular Vermont, Research Center for Children, Youth, &
interest, but they also raise concerns about threatened Families, 2001.
test security, psychometric integrity, and ethical and 11. Achenbach TM: ASEBA: Achenbach System of
legal ramifications.124 Empirically Based Assessment, 2005. (Available at:
In conclusion, tests and other assessment instru- ht t p : //w w w. a seba.org /about us /about us.ht m l ;
ments provide valuable data. Clinicians must consider accessed 10/18/06.)
the levels of assessment involved in obtaining these 12. Angold A, Prendergast M, Cox A, et al: The Child and
Adolescent Psychiatric Assessment (CAPA). Psychol
data: (1) refi nement of questions to be answered, (2)
Med 25:739-753, 1995.
selection of appropriate tests to answer questions 13. Miller WR, Rollnick S: Motivational Interviewing:
posed, (3) proper administration and scoring, (4) Preparing People to Change, 2nd ed. New York: Guil-
interpretation, and (5) synthesis with other informa- ford, 2002.
tion and observations. The art of assessment resides 14. DiClemente CC, Prochaska JO: Toward a compre-
in how these fi ndings are integrated and interpreted, hensive, transtheoretical model of change: Stages of
so as to develop diagnostic hypotheses and recom- change and addictive behaviors. In Miller WR, Heather
174 DEVELOPMENTAL AND BEHAVIORAL PEDIATRICS: EVIDENCE AND PRACTICE

N, eds: Treating Addictive Behaviors, 2nd ed. New 33. Bayley N: Bayley Scales of Infant Development–II.
York: Plenum Press, 1998, pp 3-24. San Antonio, TX: The Psychological Corporation,
15. Baer JS, Peterson PL: Motivational interviewing with 1993.
adolescents and young adults. In Miller WR, Rollnick 34. Flynn JR: Searching for justice. The discovery of IQ
SR, eds: Motivational Interviewing: Preparing People gains over time. Am Psychol 54:5-20, 1999.
to Change, 2nd ed. New York: Guilford, 2002, pp 35. Black M, Matula K: Essentials of Bayley Scales of
320-332. Infant Development–II Assessment. New York: Wiley,
16. Sindelar HA, Abrantes AM, Hart C, et al: Motivational 2000.
interviewing in pediatric practice. Curr Prob Pediatr 36. Greenspan SI: Greenspan Social-Emotional Growth
Adolesc Health Care 34:322-339, 2004. Chart: A Screening Questionnaire for Infants and
17. Aylward GP: Practitioner’s Guide to Developmental Young Children. San Antonio, TX: Harcourt Assess-
and Psychological Testing. New York: Plenum Medical, ment, 2004.
1994. 37. Harrison PL, Oakland T: Adaptive Behavior Assess-
18. Aylward GP: Infant and Early Childhood Neuropsy- ment System, 2nd ed. San Antonio, TX: The Psycho-
chology. New York: Plenum Press, 1997. logical Corporation, 2003.
19. Aylward GP: Measures of infant and early childhood 38. Dunn LM, Dunn LM: The Peabody Picture Vocabu-
development. In Goldstein G, Beers SR, eds: Compre- lary Test–III. Circle Pines, MN: American Guidance
hensive Handbook of Psychological Assessment, vol 1. Service, 1997.
Hoboken, NJ: Wiley, 2004, pp 87-97. 39. Newborg J: Battelle Developmental Inventory–Second
20. Aylward GP, Carson AD: Use of the Test Observation Edition. Itasca, IL: Riverside, 2005.
Checklist with the Stanford-Binet Intelligence Scales 40. Newborg J, Stock JR, Wnek L, et al: The Battelle
for Early Childhood, Fifth Edition (Early SB5). Pre- Developmental Inventory. Itasca, IL: Riverside,
sented at the National meeting of the National Asso- 1994.
ciation of School Psychologists, Atlanta, GA, April 1, 41. Mullen EM: Mullen Scales of Early Learning. Circle
2005. Pines, MN: American Guidance Service, 1984.
21. Bishop D, Butterworth GE: A longitudinal study using 42. Mullen EM: Mullen Scales of Early Learning: AGS
the WPPSI and WISC-R with an English sample. Br J Edition. Circle Pines, MN: American Guidance
Educ Psychol 49:156-168, 1979. Service, 1995.
22. Mantynen H, Poikkeus AM, Ahonen T, et al: Clinical 43. Elliott CD: Differential Ability Scales. San Antonio,
significance of test refusal among young children. TX: The Psychological Corporation, 1990.
Child Neuropsychol 7:241-250, 2001. 44. McCarthy DA: McCarthy Scales of Children’s
23. Ounsted M, Cockburn J, Moar VA: Developmental Abilities. New York: The Psychological Corporation,
assessment at four years: Are there any differences 1972.
between children who do, or do not, cooperate? Arch 45. Kaufman AS, Kaufman NL: Kaufman Brief Intel-
Dis Child 58:286-289, 1983. ligence Test, Second Edition. Circle Pines, MN: Amer-
24. Wolcaldo C, Rieger I: Very preterm children who do ican Guidance Service, 2004.
not cooperate with assessments at three years of age: 46. Roid G: The Stanford-Binet Intelligence Scale–Fifth
Skill differences at five years. J Dev Behav Pediatr Edition. Itasca, IL: Riverside, 2003.
21:107-113, 2000. 47. Kaufman AS, Kaufman NL: Kaufman Assessment
25. Langkamp DL, Brazy JE: Risk for later school prob- Battery for Children–Second Edition. Circle Pines,
lems in preterm children who do not cooperate for MN: American Guidance Service, 2004.
preschool developmental testing. J Pediatr 135:756- 48. Wechsler D: Wechsler Preschool and Primary Scale of
760, 1999. Intelligence–Third Edition. San Antonio, TX: The Psy-
26. Roid G: Stanford-Binet Intelligence Scales for Early chological Corporation, 2002.
Childhood. Itasca, IL: Riverside, 2005. 49. Wechsler D: Wechsler Intelligence Scale for Children–
27. Bayley N: Bayley Scales of Infant and Toddler Devel- Fourth Edition. San Antonio, TX: The Psychological
opment. San Antonio, TX: The Psychological Corpora- Corporation, 2003.
tion, 2006. 50. Wechsler D: The WASI: Wechsler Abbreviated Scale
28. Bayley N: Bayley Scales of Infant Development. of Intelligence. San Antonio, TX: The Psychological
San Antonio, TX: The Psychological Corporation, Corporation, 1999.
1969. 51. Ramsay MC, Reynolds CR: Relations between intelli-
29. Gesell AL, Halverson HM, Amatruda CS: The First gence and achievement tests. In Goldstein G, Beers
Five Years of Life: A Guide to the Study of the Pre- SR, eds: Comprehensive Handbook of Psychological
school Child, From the Yale Clinic of Child Develop- Assessment, vol 1. Hoboken, NJ: Wiley, 2004, pp
ment. New York: Harper, 1940. 25-50.
30. Gesell A: The Mental Growth of the Preschool Child. 52. Kaufman AS, Kaufman NL: Kaufman Test of
New York: Macmillan, 1925. Educational Achievement, 2nd ed. Circle Pines, MN:
31. Cattell P: Cattell Infant Intelligence Scale. New York: American Guidance Service, 2004.
The Psychological Corporation, 1940. 53. Markwardt EC: Peabody Individual Achievement
32. Knobloch H, Stevens F, Malone AE: Manual of Devel- Test–Revised. Circle Pines, MN: American Guidance
opmental Diagnosis. New York: Harper & Row, 1980. Service, 1989.
 CHAPTER 7 Screening and Assessment Tools 175

54. Markwardt EC: Peabody Individual Achievement and impact in a primary care setting. J Pediatr Psychol
Test–Normative Update. Circle Pines, MN: American 24:405-414, 1999.
Guidance Service, 1998. 73. Stancin T, Palermo TM: A review of behavioral
55. Wechsler D: The Wechsler Individual Achievement screening practices in pediatric settings: Do they
Test, 2nd ed. San Antonio, TX: The Psychological Cor- pass the test? J Dev Behav Pediatr 18:183-194,
poration, 2001. 1997.
56. Wilkerson G: Wide Range Achievement Test, 3rd ed. 74. Perrin EC, Stein REK, Drotar D: Cautions in using the
Wilmington, DE: Wide Range, Inc., 1993. Child Behavior Checklist: Observations based on
57. Robertson GJ: Wide Range Achievement Test– research about children with a chronic illness. J
Expanded Version. Odessa, FL: Psychological Assess- Pediatr Psychol 16:411-421, 1991.
ment Resources, 2002. 75. Reynolds CR, Kamphaus RW: Behavior Assessment
58. Woodcock RW, McGrew KS, Mather N: Woodcock- System for Children–Second Edition (BASC-2)
Johnson III. Tests of Achievement. Itasca, IL: River- Manual. Circle Pines, MN: AGS Publishing, 2006.
side, 2001. 76. Carter A, Briggs-Gowan M: Infant Toddler Social
59. Leark RA: The Luria-Nebraska Neuropsychological Emotional Assessment (ITSEA). San Antonio, TX:
Battery–Children’s Revision. In Goldstein G, Beers Harcourt Assessment, 2006.
SR, eds: Comprehensive Handbook of Psychological 77. Carter A, Briggs-Gowan M: Brief Infant Toddler Social
Assessment, vol 1. Hoboken, NJ: Wiley, 2004, pp Emotional Assessment (BITSEA). San Antonio, TX:
147-156. Harcourt Assessment, 2006.
60. Cohen M: Children’s Memory Scale. San Antonio, TX: 78. Butcher JN, Williams CL: Minnesota Multiphasic Per-
The Psychological Corporation, 1997. sonality Inventory–Adolescent (MMPI-A) Manual.
61. Korkman M, Kirk U, Kemp SL: NEPSY—A Develop- Minneapolis, MN: University of Minnesota Press,
mental Neuropsychological Assessment. San Antonio, 1992.
TX: The Psychological Corporation, 1998. 79. Society for Personality Assessment: The status of the
62. Gioia GA, Isquith PK, Guy SC, et al: Behavior Rating Rorschach in clinical and forensic practice: An official
Inventory of Executive Function (BRIEF). Odessa, statement by the Board of Trustees of the Society for
FL: Psychological Assessment Resources, 2000. Personality Assessment. J Pers Assess 85:219-237,
63. Sheslow D, Adams W: The Wide Range Assessment of 2005.
Memory and Learning. Wilmington, DE: Jastak Asso- 80. Morris TL: Sociometric assessment. In Ollendick TH,
ciates, 1990. Schroeder CS, eds: Encyclopedia of Clinical Child and
64. Sheslow D, Adams W: Wide Range Assessment of Pediatric Psychology. New York: Kluwer Academic/
Memory and Learning, 2nd ed. Odessa, FL: Psycho- Plenum, 2003, pp 632-634.
logical Assessment Resources, 2003. 81. Weiss B, Harris V, Catron B: Development and initial
65. Achenbach TM, McConaughy SH, Howell CT: Child/ validation of the Peer-Report Measure of Internaliz-
adolescent behavioral and emotional problems: Impli- ing and Externalizing Behavior. J Abnorm Child
cations of cross-informant correlations for situational Psychol 30:285-294, 2002.
specificity. Psychol Bull 101:213-232, 1987. 82. Querido JG, Eyberg SH: Assessment of parent-child
66. McMahon RJ, Frick PJ: Evidence-based assessment of interactions. In Ollendick TH, Schroeder CS, eds:
conduct problems in children and adolescents. J Clin Encyclopedia of Clinical Child and Pediatric Psychol-
Child Adolesc Psychol 34:477-505, 2005. ogy. New York: Kluwer Academic/Plenum, 2003, pp
67. Achenbach TM, Rescorla LA: Manual for ASEBA 40-41.
School-Age Forms & Profi les. Burlington: University 83. Eyberg SM, Nelson MMcD, Duke M, et al: Manual for
of Vermont, Research Center for Children, Youth, & the Dyadic Parent-Child Interaction Coding System
Families, 2001. Third Edition, 2005. (Available at: http://www.phhp.
68. Achenbach TM: The classification of children’s psy- ufl .edu/~seyberg/PCITWEB2004/Measures/DPICS%20III
chiatric symptoms: A factor-analytic study. Psychol %20final%20draft.pdf; accessed 10/18/06.)
Monogr 80(No. 615), 1966. 84. White S: Parent training: Use of videotaped interac-
69. Achenbach TM, Rescorla LA: Manual for ASEBA tions. Presented at the Great Lakes Society of Pediatric
Preschool Forms & Profi les. Burlington: University Psychology Conference, Cleveland, OH, March 20,
of Vermont, Research Center for Children, Youth, & 2000.
Families, 2000. 85. Shaddy DJ, Colombo J: Attachment. In Ollendick TH,
70. Achenbach TM, Rescorla LA: Manual for ASEBA Schroeder CS, eds: Encyclopedia of Clinical Child and
Adult Forms & Profi les. Burlington: University of Pediatric Psychology. New York: Kluwer Academic/
Vermont, Research Center for Children, Youth, & Plenum, 2003, pp 43-44.
Families, 2003. 86. Klein DN, Dougherty LR, Olino TM: Toward guide-
71. Achenbach TM, Newhouse PA, Rescorla LA: Manual lines for evidence-based assessment of depression in
for ASEBA Older Adult Forms & Profi les. Burlington: children and adolescents. J Clin Child Adolesc Psychol
University of Vermont, Research Center for Children, 34:412-432, 2005.
Youth, & Families, 2004. 87. Kovacs M: Children’s Depression Inventory (CDI)
72. Riekert KA, Stancin T, Palermo TM, et al: A psycho- Manual. North Tonawanda, NY: Multi-Health Systems,
logical behavioral screening service: Use, feasibility, 1992.
176 DEVELOPMENTAL AND BEHAVIORAL PEDIATRICS: EVIDENCE AND PRACTICE

88. Angold A, Costello EJ, Messer SC, et al: Development Acad Child Adolesc Psychiatry 36(10 Suppl):85S-
of a short questionnaire for use in epidemiological 121S, 1997.
studies of depression in children and adolescents. Int 105. Ozonoff S, Goodlin-Jones BL, Solomon M: Evidence-
J Methods Psychiatric Res 25:237-249, 1995. based assessment of autism spectrum disorders in
89. Reynolds WM: Reynolds Child Depression Scale: Pro- children and adolescents. J Clin Child Adolesc Psychol
fessional Manual. Odessa, FL: Psychological Assess- 34:523-540, 2005.
ment Resources, 1989. 106. Rutter M, LeCouteur A, Lord C: Autism Diagnostic
90. Reynolds WM: Reynolds Adolescent Depression Scale: Interview–Revised Manual. Los Angeles: Western
Professional Manual. Odessa, FL: Psychological Assess- Psychological Services, 2003.
ment Resources, 1987. 107. Lord C, Rutter M, DiLavore PC, et al: Autism Diag-
91. Poznanski EO, Mokros HB: Children’s Depression nostic Observation Schedule Manual. Los Angeles:
Rating Scale–Revised (CDRS-R). Los Angeles: Western Western Psychological Services, 2002.
Psychological Services, 1999. 108. Rutter M, Bailey A, Berument SK, et al: Social
92. Luby JL, Heffelfi nger A, Koenig-McNaught AL, et al: Communication Questionnaire (SCQ) Manual. Los
The Preschool Feelings Checklist: A brief and sensitive Angeles: Western Psychological Services, 2003.
measure for depression in young children. J Am Acad 109. Gilliam JE: Gilliam Autism Rating Scale. Austin, TX:
Child Adolesc Psychiatry 43:708-717, 2004. PRO-ED, 1995.
93. Silverman WK, Ollendick TH: Evidence-based assess- 110. Schopler E, Reichler R, Renner B: Childhood Autism
ment of anxiety and its disorders in children and Rating Scale (CARS). Los Angeles: Western Psycho-
adolescents. J Clin Child Adolesc Psychol 34:380-411, logical Services, 1988.
2005. 111. Kazak A: Family assessment. In Ollendick TH,
94. March JS, Parker JDA, Sullivan K, et al: The Multi- Schroeder CS, eds: Encyclopedia of Clinical Child and
dimensional Anxiety Scale for Children (MASC): Pediatric Psychology. New York: Kluwer Academic/
Factor structure, reliability, and validity. J Am Acad Plenum, 2003, pp 231-232.
Child Adolesc Psychiatry 36:554-565, 1997. 112. Touliatos J, Straus M, Perlmutter B: Handbook of
95. Beidel DC, Turner SM, Morris TL: A new inventory family measurement techniques. Thousand Oaks, CA:
to assess childhood social anxiety and phobia: The Sage Publications, 2000.
Social Phobia and Anxiety Inventory for Children. 113. Abidin RR: Parenting Stress Index, 3rd Edition
Psychol Assess 7:73-79, 1995. Manual. Lutz, FL: Psychological Assessment Resources,
96. La Greca AM, Stone WL: Social Anxiety Scale for 1995.
Children–Revised: Factor structure and concurrent 114. Hodges K: Assessment of global functioning. In
validity. J Clin Child Psychol 22:7-27, 1993. Ollendick TH, Schroeder CS, eds: Encyclopedia of
97. La Greca AM, Lopez N: Social anxiety among adoles- Clinical Child and Pediatric Psychology. New York:
cents: Linkages with peer relations and friendships. Kluwer Academic/Plenum, 2003, pp 38-40.
J Abnorm Child Psychol 26:83-94, 1998. 115. Shaffer DM, Gould S, Brasic J, et al: A Children’s
98. Reynolds CR, Richmond BO: Revised Children’s Global Assessment Scale (CGAS). Arch Gen Psychia-
Manifest Anxiety Scale: Manual. Los Angeles: try 40:1228-1231, 1983.
Western Psychological Services, 1985. 116. Hodges K: Child and Adolescent Functional Assess-
99. DuPaul GJ, Power TJ, Anastopoulos AD, et al: ADHD ment Scale (CAFAS). In Marnish ME, ed: The Use of
Rating Scale-IV—Checklists, Norms, and Clinical Psychological Testing for Treatment Planning and
Interpretations. New York; Guildford, 1998. Outcomes Assessment, 2nd ed. Mahwah, NJ: Erlbaum,
100. Wolraich ML, Lambert W, Doffi ng MA, et al: Psycho- 1999, pp 631-664.
metric properties of the Vanderbilt ADHD Diagnostic 117. Sparrow SS, Cicchetti DV, Balla DA: Vineland Adap-
Parent Rating Scale in a referred population. J Pediatr tive Behavior Scales, 2nd ed. Circle Pines, MN: AGS
Psychol 28:559-567, 2003. Publishing, 2006.
101. Collett BR, Ohan JL, Myers KM: Ten-Year Review of 118. Varni JW, Burwinkle TM, Seid M, et al: The PedsQLTM
Rating Scales. V: Scales Assessing Attention-Deficit/ 4.0 as a pediatric population health measure: feasibil-
Hyperactivity Disorder. J Am Acad Child Adolesc ity, reliability, and validity. Ambul Pediatr 3:329-341,
Psychiatry 42:1015-1037, 2003. 2003.
102. American Academy of Pediatrics: Diagnosis and eval- 119. American Educational Research Association, Ameri-
uation of the child with attention-deficit/hyperactiv- can Psychological Association, National Council on
ity disorder. Pediatrics 105:1158-1170, 2000. Measurement in Education: The Standards for Edu-
103. Brown RT, Freeman WS, Perrin JM, et al: Prevalence cational and Psychological Testing. Washington, DC:
and assessment of attention-deficit/hyperactivity dis- AERA Publications, 1999. (Available at: http://www.
order in primary care settings. Pediatrics 107:e43, apa.org/science/standards.html; accessed 10/18/06.)
2001. (Available at: http://pediatrics.org/cgi/content/ 120. Joint Committee on Testing Practices: Code of
full/107/3/e43; accessed 10/18/06). Fair Testing Practices in Education (Revised). Educa-
104. Dulcan M: Practice parameters for the assessment tional Measurement: Issues and Practice 2:23-26,
and treatment of children, adolescents, adults with 2005.
attention-deficit/hyperactivity disorder. American 121. Eyde LD, Moreland KL, Robertson GJ: Test User Qual-
Academy of Child and Adolescent Psychiatry. J Am ifications: A Data-Based Approach to Promoting Good
 CHAPTER 7 Screening and Assessment Tools 177

Test Use. Washington, DC: American Psychological for language and speech in children are also based
Association, 1988. on the normal progression of milestones throughout
122. Turner SM, DeMers ST, Fox HR, et al: APA’s guide- early childhood and on evidence of substantial delay
lines for test user qualifications: An executive or difference. Accordingly, this chapter begins with
summary. American Psychologist 56:1099-1113,
defi nitions of language, speech, and the subcompo-
2001.
nents. It proceeds to a review of the course of lan-
123. Spies RA, Plake BS, eds: The Sixteenth Mental Mea-
surements Yearbook. Lincoln: University of Nebraska guage development in children from birth to school
Press, 2005. age and a description of individual variations within
124. Naglieri J, Drasgow F, Schmit M, et al: Psychological the normal range. Next follows a discussion of the
testing on the Internet: new problems, old issues . Am approaches to assessing language in infants, toddlers,
Psychol 59:150-162, 2004. and young preschoolers. The approaches for young
children are contrasted with the approaches for
school-aged children and adolescents. Finally, tables
of measures that can be used across the age range for
assessments of language and speech are included.

7D. BASIC DEFINITIONS

Assessment of Language
Language is the main medium through which humans
and Speech share ideas, thoughts, emotions, and beliefs. Unlike
other methods of communication, language is sym-
HEIDI M. FELDMAN ■ CHERYL MESSICK bolic; meaning is conveyed by arbitrary signs. Cries
and giggles are signs that arise as reflexive responses
The development of language represents an important or as emissions from the emotional or motivational
accomplishment for young children, allowing them state that they represent. Therefore, cries and giggles
to participate fully in the human community. Lan- are communicative but not language. In contrast,
guage learning progresses rapidly in the toddler and words and sentences are arbitrary and therefore can
preschool era. At age 1 year, typically developing chil- vary from language to language. A dog can be labeled
dren are just beginning to understand and produce perro in Spanish, chien in French, and so forth. Lan-
words. By age 4 to 5 years, they can participate guage is also rule-governed. In English, for example,
actively in conversations and construct long and the order of words in a sentence cannot be signifi-
complex discussions. The process of language learn- cantly altered without changing meaning or render-
ing proceeds in a predictable and orderly manner for ing the sequence ungrammatical. For example, “Bill
the majority of children. However, the pace is slow kissed Sue” has a different meaning than “Sue kissed
and the pattern disordered for many children. The Bill.” Moreover, “See the dog” is a grammatical sen-
overall prevalence of language disorders at school tence, but “dog the see” is not. These features of lan-
entry has been estimated at approximately 7%,1 and guage—the use of symbols in a systematic manner to
the overall prevalence of speech disorders, at nearly convey meanings—provide people with the ability to
4%.2 In view of the pivotal role of language and create and understand an infi nite number of
speech in learning, communication, and social rela- messages.
tionships, and because of the high prevalence of dis- Language is distinct from speech. Speech refers
orders, screening for language delays and disorders is specifically to articulation of sounds and syllables
appropriate for all children and comprehensive assess- created by the complex interaction of the respiratory
ments of language and speech is appropriate for those system, the larynx, the pharynx, the mouth struc-
at high risk for delays or disorders. tures, and the nose. Sign languages also meet the
Language is conceptualized as being composed of defi nition of language but entail the configuration
receptive and expressive domains and as having and movement of hands, arms, facial muscles, and
multiple components within those domains. These body to articulate meaning. Similarly, written lan-
subcomponents are usually coordinated in normal guages convey meaning through the use of arbitrary
functioning. However, in disorders, one or more sub- symbols on a page. It is possible to have a speech dis-
components may become deficient or abnormal. Sim- order without a language disorder and the converse.
ilarly, speech is composed of multiple, independent However, children may exhibit disorders in speech
features. Assessment strategies and tools for language and language concurrently. In this chapter, discussion
and speech at any age are designed to assay skills in focuses on the assessment of verbal language and
several components. Assessment strategies and tools speech.
178 DEVELOPMENTAL AND BEHAVIORAL PEDIATRICS: EVIDENCE AND PRACTICE

TABLE 7D-1 ■ Domains and Components of Language and Speech

Term Definition Example

Language
Receptive language Ability to understand another’s language A father says, “Where are my shoes?” and the child points under
the chair to the father’s sneakers.
Expressive language Ability to produce language A father says, “Where are my shoes?” and the child responds,
“Under the chair.”
Phoneme The smallest units of the sound system /b/ and /p/ are different phonemes, and their use results in
that change meaning of a word different words, as in bat and pat.
Morpheme The smallest unit of meaning in language The plural /s/ is a morpheme; when added to the word book, it
conveys a different meaning: books.
Syntax The set of rules for combining “The boy ate his supper” follows English syntax. “Ate the boy
morphemes and words into sentences supper the” does not follow English syntax.
Semantics The meaning of words and sentences Vocabulary, categorization of meanings, and sentence structures
all contribute to semantics.
Pragmatics Social aspects or actual use of language Pragmatic behaviors focus on discourse rules, presuppositional
behavior, and communicative functions.

Speech
Intelligibility The ability of speech to be understood Speech sound errors, rate of speech, familiarity with the speaker
and message, and background noise are some of the factors
that may decrease intelligibility.
Fluency The forward flow of speech Dysfluency may involve pausing or repetition of sounds, words,
and phrases
Stuttering Repetitions of consonant sounds, An example of stuttering is the following: “W-w-would you give
prolongation of vowel sounds, or m-m-m-eeeee the m-m-milk.” Stuttering is often accompanied
other forms of fragmentation, by secondary behaviors, such as head movements or facial
blockage, or dyscoordination of the expressions that appear designed to permit forward flow.
forward flow of speech
Voice and resonance Qualities of speech based on the passage Hoarseness may be caused by laryngeal inflammation or
of air through the larynx, mouth, and nodules. Hyporesonance may be caused by adenoidal
nose hypertrophy. Hypernasality may be due to velopharyngeal
insufficiency.

For purposes of understanding mature language Language is also subdivided into subsystems or
use, language is subdivided into several components. components, in large part on the basis of the size of
Table 7D-1 lists some of the terms used to describe units. Comprehensive assessments evaluate multiple
components of language and their defi nitions. In subsystems of language in terms of both comprehen-
terms of language, an important division is between sion and production.
receptive and expressive language. Receptive language
refers to the ability to understand or comprehend ■ Phonemes are the smallest units in the sound system
another person’s language. Expressive language refers of a language that serve to change the meaning of
to the ability to produce language. Receptive language a word. For example, in English bat, pat, bit, and bid
typically begins to develop before expressive lan- are all recognized as different words. Therefore, the
guage. The two components typically progress in rela- single sounds that differentiates among them—/b/,
tive synchrony. In some toddlers, however, the ability /p/, /a/, /i/, /t/, and /d/—all represent different
to produce language lags significantly behind the phonemes in English. The phonological system of a
ability to understand language. Older children may language is composed of the inventory of phonemes
show uneven skills in their abilities to understand and the rules by which phonemes can interact with
and produce, with either domain more advanced than each other. For example, if a new word in English
the other. Therefore, comprehensive assessments of were needed, the sounds represented by /i/ and /b/
language usually include separate evaluations of could be combined to create the word ib, but the
receptive language or comprehension and expressive sounds /b/ and /d/ could not be combined because
language or production. Some standardized measures that combination violates the phonological rules of
include separate subtests for comprehension and pro- English.
duction. Some measures focus on one or the other ■ Morphemes are considered the smallest unit of
component. meaning in oral and written language. Words are
 CHAPTER 7 Screening and Assessment Tools 179

free-standing morphemes that are the meaningful background information. For example, once a
building blocks of larger units, such as sentences. speaker realizes that his or her listeners do not
Meaningful parts of words, such as the plural “-s” know that “Bob” is his or her cousin, the speaker
or past tense “-ed” markers are bound morphemes, needs to tell listeners who he is, to increase their
which, when attached to another morpheme, alter understanding of the message.
the meaning of the word. In English, there are rela-
Several aspects of verbal production are considered
tively few bound morphemes, but in other lan-
parts of speech. Table 7D-1 includes defi nitions and
guages, such as Hebrew and Italian, there are many
examples of these components. Speech includes the
morphemes that can be attached to other mor-
accuracy of speech sound production. Assessments of
phemes and change the meaning of words.
speech typically include analysis of the types of speech
■ Syntax comprises the rules for combining mor-
sound errors. Estimates of intelligibility are used to
phemes and words into organized and meaningful
describe the functional consequences of speech sound
sentences. In English, most sentences begin with a
errors. Another component of speech production is
noun phrase, such as “The boy,” followed by a verb
fluency, defi ned as the forward flow of speech. Stut-
phrase, such as “gave the girl a red book.” In addi-
tering is a type of dysfluency, characterized by repeti-
tion, the adjective red should come before the noun
tion or prolongation of sounds and other fragmentation
book, but that arrangement is reversed in some lan-
of the sounds, often accompanied by a sense of effort
guages. In other languages, such as Italian and
and by secondary behavioral characteristics that the
German, the syntactic rules require a different
speaker uses to attempt to reinitiate forward flow of
arrangement of words: for example, the adjective
speech. Voice and resonance also affect speech. The
occurring after the noun it describes, and the verb
flow of air through the vocal cords into the nose and
appearing at the end of the sentence rather than in
mouth affect the quality of speech. Voice disorders
the middle.
include hoarseness, which may be caused by tempo-
■ Semantics refers to the meaning of words and sen-
rary inflammation of the larynx or by nodules from
tences. The number of words that a child produces
vocal abuse. Resonance disorders include hyponasal-
and understands can be considered one element of
ity, which is a reduction in the usual amount of air
the child’s semantic knowledge. The meaning of
through the nose and may be caused by adenoidal
sentences is described in such terms as agents and
hypertrophy, and hypernasality, which results from
actions, as distinct from syntax in which sentences
excessive air through the nose and may be secondary
may be described in terms of noun and verb phrases.
to a cleft palate.
In the sentence “The boy gave the girl a red book,”
the boy is the agent, gave is the action, and the girl
is the recipient or dative. Semantics also includes
meaning at concrete and abstract levels, word defi- TYPICAL LANGUAGE
nitions, and word categories such as synonyms
and antonyms. During school age, semantic skills
DEVELOPMENT
that are learned include knowledge of metaphorical
language as in idioms, proverbs, and similes.
Infancy
■ Pragmatics refers to social aspects or actual use of Newborns demonstrate the basic building blocks for
language. Pragmatic skills address three broad areas language development through social interactions
of using language: discourse rules, communicative with adults. They show preferences for looking at the
functions, and presuppositional skills. Examples of human face over other visual stimuli and for looking
discourse rules include features such as appropriate at the eyes and the mouth over other body parts. They
use of intonation and tone of voice, as well as the also show preferences for listening to the human
inclusion of politeness markers in communication. voice over other auditory stimuli. As newborns, they
Discourse guidelines also consider the ability to ini- prefer to listen to their mother’s voice over the voice
tiate, respond, maintain topics, and appropriately of an unfamiliar woman.3,4 Some evaluations of new-
take turns. Discourse rules also cover aspects such borns include these prerequisites for language in their
as varying the language used in relation to different assessment.
environments and social interactions. Children Experimental studies show that infants have fun-
vary the style and tone of voice when asking an damental abilities for perceiving, discriminating, and
adult for a favor in comparison with asking a peer. learning speech sounds. At very young ages, they are
Communicative functions examine the purpose able to differentiate, for example, between two similar
behind a communication act (e.g., requesting, com- sounds (e.g., /ba/ vs. /pa/).5 During infancy, they can
menting, protesting). Presuppositional skills address also detect and use the statistical properties of sound
the ability to provide a listener with appropriate co-occurrences in continuous streams of speech to
180 DEVELOPMENTAL AND BEHAVIORAL PEDIATRICS: EVIDENCE AND PRACTICE

group syllables into wordlike units.6 These nonspe- infants demonstrate the ability to take turns, by
cific perceptual mechanisms are probably extremely vocalizing and cooing responsively with other people
important in helping children to parse the sound in their environment while maintaining eye contact.
stream and begin to comprehend language.7,8 At the These patterns constitute the initial phases of com-
time of this writing, these abilities have been demon- munication and establish the patterns for later con-
strated in research studies but have not yet been inte- versational exchanges. At approximately 6 months of
grated into language assessments. age, children produce more differentiated vocal pro-
The language that babies hear from the adults in ductions with the addition of consonant sounds.
their environment refi nes these innate mechanisms. Sounds are produced in syllable chains referred to as
By the time that they are about 9 months of age, babbling. Initially the babble is simple repetition of a
children show greater precision in differentiating the single syllable, such as “bababa,” and as the child gets
phonemes of the native language than phonemes older, it becomes a chain of different syllables, such
from other languages.7,9 For example, all infants less as “dabigu.” By 12 months of age, some children add
than 6 months of age can differentiate between the sentence-like intonation patterns to the babbling. At
sounds /r/ and /l/. However, by the time they reach this point, the output is referred to as jargon. Around
9 months of age, Japanese infants no longer make that the same time, children also begin to produce their
distinction because /r/ and /l/ are not separate pho- fi rst words.
nemes in Japanese7 while English or American infants
still can make the distinction. Although infants as a
group show these early abilities in speech perception,
Second Year of Life
it is not clear whether individual differences in the In the fi rst half of the second year, receptive language
nature or timing of how children process the speech skills progress from understanding single words to
stream are predictive of later functioning. Therefore, simple commands. Toddlers demonstrate the ability
clinical assessments of infants currently do not include to follow common routines, such as “Let’s go bye-bye”
these types of measures of speech perception. or “Time for bath” and then simple commands that
Table 7D-2 summarizes key milestones in language make arbitrary connections, such as “kiss the pencil.”
and speech. All of these milestones should be consid- In the second half of the second year, they are able
ered approximations, in view of the wide range of to identify body parts, and as they approach age 2
normal. Delays in one or more specific behaviors may years, they begin to follow two-part instructions
or may not prove clinically significant, depending on (e.g., “Get the ball and give it to Daddy”).
such variables as risk factors, patterns of develop- The pace of expressive language skills is initially
ment, severity, and the rate of progress in other devel- slow. After the fi rst words appear, at approximately
opmental areas. 12 months of life, vocabulary initially grows at a rate
In terms of receptive language skills, a few mile- of about 5 to 10 words each month, with some words
stones are worth highlighting because either they play entering and then disappearing from the repertoire.
a key role in screening tests and assessments of early The early vocabulary generally includes more nouns
communication or because they may be demonstrable than verbs.10 Early words may be immature in terms
in a health supervision clinical visit with a child and of their sound patterns, restricted to simple combina-
parent. In terms of receptive language, by about 6 tions of consonants and vowels, such as “baba” for
months of age, babies often demonstrate recognition bottle or “wawa” for water. The meaning of early
of their own names, either by pausing in their activity vocabulary words may be quite different from mature
when they hear their names or even by looking toward meanings. Children may apply a word very selec-
a speaker. By 9 months of age, they typically partici- tively, such as the word dog only for the family’s pet
pate in some social routines, linking appropriate or relatively indiscriminately, such as the word dog for
actions with commands, such as “Wave bye-bye,” or any four-legged animal, including cows and cats.
responding with arms raised to “Would you like me to In the second half of the fi rst year, many children
pick you up?” At around 12 months of age, they show developing typically undergo a rapid change in the
understanding of simple words, responding appropri- rate of word learning.11,12 This spurt usually occurs
ately to questions, such as “Where’s mama?” or com- after a child has at least 35 to 50 different words in
mands, such as “Show me the ball.” his or her vocabulary. The vocabulary grows at a rate
In terms of expressive language skills, children of 4 or 5 words a day. At about that time, two-word
begin to produce voluntary vocalizations by about 2 phrases emerge. Thus, by 2 years of age, children typi-
to 3 months of age. The fi rst form of sound production cally can say about 100 words and some two-word
is called cooing, which is composed of musical vowel- phrases. Children use their language to talk about
like sounds and occasional /k/- and /g/-like conso- things in the here and now. From that point on, lan-
nants. Shortly after producing such coos in isolation, guage development proceeds rapidly.
 CHAPTER 7 Screening and Assessment Tools 181

TABLE 7D-2 ■ Developmental Milestones in Receptive and Expressive Language

Age Range Receptive Expressive

0-1 months Startles or widens eyes to sound Cries

2-4 months Quiets to voice, blinks eyes to sound Makes musical sounds, called cooing
Coos in reciprocal exchanges
5-7 months Turns head toward sound Babbles
Looks and responds to own name Repeats self-initiated sounds
8-10 months Links actions to words, responding with raised arms when a Says mama or dada indiscriminately
parent says “up” or waves when person says “wave bye-bye” Points at interesting objects or events
Looks at the direction of a point
Stops action when hears “no”
11-14 months Listens selectively to familiar words Uses symbolic gestures
Indicates understanding of single words Repeats parent-initiated sounds
Respond to simple questions such as “Where’s mama?” Babbles with intonation patterns of sentences,
Follows simple commands, such as “Show me the ball” called jargon
Uses a few words, such as names, mama or dada
specifically, or animal noises
15-18 months Points to three body parts (eyes, nose, mouth) Uses words to express needs
Understands up to 50 words Learns at least 20 words
Recognizes common objects by name (dog, cat, bottle, ball, Uses words inconsistently and mixed with jargon,
book) echolalia, or both
Follows one-step commands accompanied by gesture
(“Give me the doll,” “Hug your bear,” “Open your mouth”)
19 months- Points to pictures when asked, “Show me” Says 50 words
2 years Understands in, on, and under Uses telegraphic two-word sentences (“Go
Begins to distinguish you from me bye-bye,” “Up daddy,” “Want cookie”)
Can formulate negative judgments (a pear is not a cookie)
25-30 months Follows two-step commands Uses jargon and echolalia infrequently
Can identify objects by use Generates average sentence of 21/2 words
Learns adjectives and adverbs
Begins to ask questions with what or who
Asks adults to repeat actions (“Do it again”)
3 years Knows several colors Uses pronouns and plurals
Knows what people do when they are hungry, thirsty, or sleepy Can tell stories that begin to be understood
Is aware of past and future Uses negative (“I can’t,” “I won’t”)
Understands today and not today Verbalizes toilet needs
Can state full name, age, and gender
Creates sentences of 3 to 4 words
Asks why
31/2 years Can answer such questions as “Do you have a doggie?”; Can relate experiences in sequential order
“Which is the boy?”; and “What toys do you have?” Says a nursery rhyme
Understands little, funny, and secret Asks permission
4 years Understands same versus different Tells a story
Follows three-step commands Uses past tense
Completes opposite analogies (“A brother is a boy, a sister Counts to 3
is a. . . .”) Names primary colors
Understands why people have houses, stoves, and umbrellas Enjoys rhyming nonsense words
Enjoys exaggerations
5 years Understands what people do with eyes and ears Indicates “I don’t know”
Understands differences in texture (hard, soft, smooth) Indicates funny and surprise
Understands if, when, and why Can define in terms of use
Identifies words in terms of use Asks definition of specific words
Begins to understand left and right Makes serious inquiries (“How does this work?”
and “What does it mean?”)
Uses mature sentence structure and form,
including complex sentences
182 DEVELOPMENTAL AND BEHAVIORAL PEDIATRICS: EVIDENCE AND PRACTICE

Unfamiliar listeners may have some difficulty in adults is that this developmental dysfluency repre-
understanding children younger than age 2 years. sents poor coordination of language, speech, and
The children often show phonetic variability in the thought. For most children, the dysfluency gradually
production of consonants and multiple processes that disappears between ages 4 and 5 years. Characteris-
simplify speech sounds. It is generally stated that only tics of clinically significant stuttering are as follows:
about half of what 2-year-olds say is intelligible to repetition of initial sounds, prolongation of sounds,
strangers, although accurate estimation during con- the need for effort to speak, appearance of secondary
versation is quite challenging. By age 2 years, chil- behavioral characteristics such as grimacing or repeti-
dren typically master consonants created at the front tive movements, or the child’s feelings of inadequacy
of the mouth, including /b/, /p/, /m/, and /w/, and or embarrassment.
sometimes the sounds produced when the tongue is By ages 3 to 4 years, preschoolers begin to partici-
placed behind the teeth, including /t/, /d/, and /n/. At pate in conversations, with gradual mastery of prag-
this age, children use /w/ for many sounds they matic skills. The also begin to talk about past events
cannot produce accurately. Their speech is also ren- and tell short stories, although their initial efforts may
dered less intelligible because they reduce consonant be marked by considerable disorganization. However,
clusters to a single sound, such as top for stop, and they by ages 4 to 5 years, they connect sentences to describe
leave off the ends of words or other sounds (“da” for sequences or scenes or to tell stories in a logical or
dog; “nana” for banana). chronological way. These multiple sentence produc-
tions are called narratives. They also improve conver-
sational abilities, which allows for longer dialogs. By
Preschool Period this age, children are able to converse easily on a
By age 3 years, children understand much of what is variety of topics with familiar as well as unfamiliar
said to them, commensurate with their cognitive listeners. They also show expertise in concepts and
abilities. For example, they learn to recognize colors vocabulary according to their individual interests
and can respond appropriately to questions, such as (e.g., names of different types of dinosaurs).
“What do we do when we are hungry?” They can also
appreciate what a parent means when he or she says,
“We will go to the park today” versus when he or she
School Age through Adulthood
answers, “Not today” to a question about going to the The fundamentals of language are established by
park. Preschoolers gradually begin to answer differ- school age. At kindergarten entry, typically develop-
ent types of questions, including which, what, and ing children can understand and produce complex
when questions. sentences. Language sophistication is typically com-
In expressive language, they gradually include pro- mensurate with cognitive abilities. Because children
nouns, increase the number of verbs and adjectives, have facility with understanding and creating at least
and introduce abstract vocabulary items, such as simple sentences, individual differences in language
color, quantity, and size terms. Children acquire a and speech abilities may be difficult to detect in
variety of grammatical morphemes, including plural routine conversation. It may not be apparent, even to
marker “-s,” possessive marker “-’s,” and the “-ing” parents and teachers, that children’s lack of compli-
attached to verbs to convey an ongoing action (e.g., ance is related to poor comprehension of what they
jumping; eating). Their syntactic skills expand to have been told. Thus, in the late preschool period and
include the ability to ask questions and create nega- at school age, systematic evaluation, with standard-
tive sentences. They build sentences of increasing ized assessment measures, should form the basis of
length and, although still immature in the mastery of evaluation, rather than informal observational
syntax, begin to produce sentences with increasing techniques.
grammatical complexity, including compound sen- By school age, most speech sounds are mature,
tences with independent and dependent clauses (e.g., although some sounds may still be underdeveloped.
“that the one that jump”). These include /sh/, /th/, /s/, /z/, /l/, and /r/ and con-
A child’s phonological system develops as well sonant blends such as /sp/, /tr/, and /bl/. Such errors
during the preschool period, and so an increasing however should not affect intelligibility significantly.
proportion of his or her sentences become fully intel- By 8 years of age, children should articulate all sounds
ligible. On the third birthday, a child’s speech is typi- of the English language correctly in spontaneous
cally intelligible to unfamiliar adults approximately conversation.
75% of the time. However, at this age, it is common
for children to experience dysfluency in their speech.
Often they repeat entire words or phrases, such as, “I
want, I want, I want an apple.” The impression on
 CHAPTER 7 Screening and Assessment Tools 183

INDIVIDUAL VARIATIONS IN purposes, evaluation of parental input along with the

SPEECH AND LANGUAGE child’s language may provide insights into the causes
of developmental deficits.
DEVELOPMENT It is difficult to predict which “late talkers”—that
is, children who show initial delays—are destined to
Patterns of early language development vary among develop language disorders. Careful analysis of their
children.13 Understanding these variations are impor- language production has not yielded valid prediction.
tant for interpretation of assessment data. Some chil- Two features of the child’s development that have
dren early on build a vocabulary of object names and been associated with resolution of the delays are good
use language to talk about the items around them. receptive language skills and mature symbolic play
These children typically pass through the stages skills.19 For this reason, assessment of language in
described previously, with a clear one-word phase fol- young children should include an evaluation of their
lowed by telegraphic two-word sentences (e.g., play. In addition, the range of communicative func-
“Mommy sock” or “baby eat”). Other children learn tions and social skills that children show may be very
social expressions, such as “Thank you” or “Give it to indicative of the nature of a language disorder. Mature
me,” and use language to express needs or to interact communication entails using language for varied pur-
with others. It appears that these children may not poses, including expressing wants and needs, greeting
understand the components within these rehearsed others, describing objects or actions, and answering
forms. Their vocabulary of single nouns may be questions. Communicative and social functioning
limited, and their speech often includes a lot of jargon. may be assessed in some evaluation protocols. Finally,
Their progression through the stages described previ- the more components of language affected, and the
ously may also vary. These early stylistic differences do greater the severity of deficit, the higher the likeli-
not seem predictive of major differences in later devel- hood that the child will have long-term communica-
opment and may reflect individual strategies or style, tion deficits. A thorough speech/language evaluation
although children who use language to refer to objects therefore requires the assessment of multiple com-
come from families with higher levels of education.13 ponents of language. It includes standardized mea-
The rate of language learning also varies among sures and analysis of spontaneous communication
children. One important factor that has been shown behaviors.
to be predictive of the rate of vocabulary and syntactic
development is the amount of child-directed language
in the environment.14,15 Descriptive studies have dem-
ASSESSMENTS
onstrated that, on average, parents in the lower socio-
economic strata provide less child-directed language
Purposes of Assessment
than do parents in the middle socioeconomic class.16 There is no single procedure or scheme for the
Children from the lower socioeconomic classes are assessment of language and speech in all children.
more likely to show slower language development, Assessment procedures vary as a function of the
reduced vocabulary, and a higher prevalence of lan- age, cognitive level, and social characteristics of the
guage delays than are children from the middle socio- children. In addition, the purpose of evaluation must
economic classes.17 It is important to consider the be considered in the selection of assessment
environmental input of children when the results of procedures.
language assessments are interpreted. There is no pro- Screening is appropriate to establish whether
fessional consensus about whether and when these asymptomatic children are at high risk for language
early delays constitute an actual language disorder; or speech impairment. Screening is appropriate for
however, prognosis for spontaneous improvement is language and speech disorders because the conditions
affected by the severity of the gap between a child’s are prevalent and early treatment can reduce the
development and that of peers and by delays across severity of the resulting condition. Screening instru-
multiple components of language. ments must be inexpensive to administer because
Another factor associated with delayed and disor- they are designed to be used with large populations.
dered development is a family history of speech/lan- They must also have good sensitivity and specificity,
guage or learning difficulties in a fi rst-degree relative. or accuracy in prediction, so that only children at the
These fi ndings suggest that there are both environ- highest risk of disorder are required to go through a
mental and biological contributors to the rate and full assessment. In most cases, screening of infants
pattern of language development.18 It is useful to and toddlers relies heavily on parent report because
know the family history in interpreting the assess- direct assessment of children at young ages is time
ment results of children who have minor delays in consuming and nonetheless may not produce repre-
language or speech. In addition, for some assessment sentative samples of what the child can do.
184 DEVELOPMENTAL AND BEHAVIORAL PEDIATRICS: EVIDENCE AND PRACTICE

Diagnostic testing establishes the clinical status of development in young children.22 The advantage of
the child in terms of language abilities and perfor- the diary method, particularly in the hands of parents
mance. Comprehensive diagnostic testing requires who are linguists, is that it can be a comprehensive
characterization of multiple aspects of language and report of verbal output. Creative strategies can be
speech skills. In infants, toddlers, and young pre- employed to assess the level of comprehension, as well
schoolers, the purpose of assessment is often to iden- as production. Of course, it is impractical for average
tify children with delays and disorders who could parents to keep a comprehensive diary for clinical
benefit from early intervention services. The earlier assessments. Each alternative method requires some
such children can be identified, the more likely they degree of sampling from the rich array of child
are to benefit from treatment. These diagnostic assess- capacities.
ments may also be useful for designing intervention An authentic informal approach to language assess-
strategies and targets, as well as to monitor the effec- ment is language sampling. For toddlers, this method
tiveness of treatment. typically involves the analysis of parent-child (or
In older preschool- and school-aged children, the clinician-child) conversations. Children and their
purpose of assessment is often to explain academic, communicative partners are typically observed as
social, or communication difficulties and to identify they play with a set of toys, either the child’s own or
children in need of therapeutic and support services. a standardized collection. A sample of at least 50 to
Early speech and language delays are often associated 100 utterances is obtained. Experienced speech and
with later reading and spelling problems.20,21 Speech language pathologists are able to transcribe and then
and language assessment are also important for chil- analyze the conversation with young children in real
dren who have behavior difficulties, because compre- time, identifying patterns that are used frequently.
hension deficits may be one factor contributing to One major advantage of language sampling is that it
behavior disorders. Again, such evaluations may assesses functional communication in a naturalistic
establish the nature of intervention or specific target setting. Another advantage is that multiple compo-
outcomes. As children get older, assessments are more nents of language, such as vocabulary, syntax, and
likely to provide insights into the prognosis for future pragmatics, can be assessed concurrently. Many
functioning. At all ages, language and speech assess- formal assessment tools do not have strategies for
ment are prerequisites for planning treatments and assessing pragmatics, and therefore conversational
monitoring progress. analysis or language sampling is often a secondary
procedure in a comprehensive assessment. Speech
sounds in context can also be assessed simultane-
Assessing Language in Infants, ously. Formal tests often assess speech sounds in
Toddlers, and Preschoolers single words rather than in continuous speech. A
third advantage is that parental language can be
Accurate assessment of infants and toddlers is very
assessed at the same time as child language; this pro-
challenging. First, the behaviors of interest occur
vides the clinician with insight into the quality of the
infrequently and unpredictably in young children
language environment. Often for more detail or for
who are just learning language. Second, young chil-
research purposes, the conversation is videotaped
dren may have difficulty cooperating for formal
and/or audiotaped for later transcription and analysis.
assessment procedures. Infants and toddlers are more
If the transcript is prepared as a computer fi le and
likely to demonstrate their emerging skills in interac-
transcribed according to a few basic conventions, two
tions with parents and other familiar adults rather
prominent programs now available can analyze mul-
than with strangers. Third, the attention span of
tiple features of the child’s language, as well as the
young children is short. Finally, infants and toddlers
parental language input. Child Language Data
are not used to remaining seated and following the
Exchange System is publicly available (http://childes.
adult lead in interactions. For all of these reasons,
psy.cmu.edu).23 Systematic Analysis of Language
informal observational studies, parent interview tools,
Transcripts (SALT) software is commercially available
and/or natural assessments play an important role in
(http://www.languageanalysislab.com/).24 In addition,
the evaluation of young children. Formal assessments
an automated method to analyze speech sounds
become more central to evaluation as the child reaches
is called Programs to Examine Phonetic and Phono-
preschool age and beyond.
logical Evaluation Records (PEPPER) (http://www.
waisman.wisc.edu/phonology/project/project.htm) is also
OBSERVATIONAL AND INTERVIEW available.24
ASSESSMENT STRATEGIES Analysis of parent-child conversation has several
Parent diaries have made an important contribution limitations in clinical practice. It may require sub-
to the initial understanding of the course of language stantial time to transcribe and analyze the conversa-
 CHAPTER 7 Screening and Assessment Tools 185

tion. Except for a few measures, such as the mean language abilities in infants and toddlers because the
length of utterances (average sentence length), gram- relevant data can be collected efficiently. An example
matical morpheme usage, and syntactic complexity,25 is the Pediatric Evaluation of Developmental Status,
there are no norms for the child’s output. Interpreta- which asks parents questions and scores their
tions about the child’s level of functioning are based responses in reference to the child’s age.28 This parent
on the types of sentence patterns used, in comparison report screening tool has comparable sensitivity and
to the expectations for the child’s age. Parents who specificity as screening tests that assess the child
themselves are not highly communicative may not directly.
elicit a representative sample of the child’s full capaci- Direct child observation is another strategy of eval-
ties. Finally, the method does not directly assess com- uation for young children. For example, with the
prehension. In many situations, the advantages of this Communication and Symbolic Behavior Scales, clini-
approach outweigh the limitations. This approach is cians evaluate the communication skills of children
very useful to monitor progress over time in individ- by observing their play in structured and unstruc-
ual children. tured situations and their interactions with adults.29
Parent report inventories circumvent some of the It is recommended as a tool to use for children with
challenges of conversational analysis. The inventories disorders in the autism spectrum (see Chapter 13).
tap into a parent’s extensive knowledge and frequent On the basis of these observations, the professional
observations of their child’s abilities. To improve reli- administering the test rates the child on multiple
ability and validity, these inventories concern current scales organized into clusters, such as communicative
abilities rather than past skills or age at acquisition functions and social-affective signaling. These ratings
and rely heavily on a recognition rather than recall are converted to standard norm-referenced scores.
format. The MacArthur-Bates Communicative De- The Autism Diagnostic Observation Schedule com-
velopment Inventory (CDI)11 (http://www.brookes bines parent interview with professional observations
publishing.com/store/books/fenson-cdi/index.htm) is de- of social and communicative behaviors under struc-
signed for children 8 to 30 months of age. The version tured situations.30 A cutoff score distinguishes chil-
for children 8 to 16 months of age prompts parents to dren who meet the criteria for autism from children
indicate, from a list of more than 400 words, the with normal development or other disorders. Finally,
number of words understood and produced and also the combination of parent interview and direct obser-
asks parents about specific symbolic gestures and vation constitutes other language and communication
actions that the child performs. The version for chil- screening tests, such as the Early Language Milestone
dren aged 16 to 30 months asks parents to indicate, test,31 a test that is used as a screening instrument in
from a list of almost 700, the words the child produces some pediatric practices.
and also to assess early grammatical development. A
shorter version of this inventory is now available. The FORMAL ASSESSMENTS
Language Development Survey (LDS) (http://www. Individually administered formal assessments of
aseba.org/research/language.html) assesses vocabulary young children can be subdivided into two categories:
development in children 18 to 35 months of age.26 It norm-referenced and criterion-referenced tests. Norm-
uses a similar format, with a vocabulary list of 310 referenced tests have standardized procedures for
words selected on the basis of diary studies. It includes administering and scoring the items. Raw scores
questions about the average length of the child’s are converted to age-adjusted standard scores that
phrases. Both of these parent report inventories have allow a designated child’s results to be compared with
been shown to have good to excellent reliability and those children of the same age. Norm-referenced tests
concurrent validity in relation to direct assessments of language development that are used to assess
and analysis of conversational samples. Predictive infants, toddlers, and preschoolers are listed in Table
validity is only fair.27 Specificity of the inventory is 7D-3.
higher than sensitivity, which indicates that many Such tools are often used to qualify a child for early
children with early delays catch up to within the intervention services. In addition, norm-referenced
normal range at older ages. tests are good for comparing the level of language
Parent interview tools provide an alternative skills to the level of cognitive or motor abilities.
method of language assessment. The Receptive- Criterion-referenced tests measure what skills the
Expressive Emergent Language Test–Third Edition, child has mastered from a set of skills in the usual
published by PRO-ED, is designed for infants to sequence of development or from a curriculum used
3-year-olds.27a The test has two subtests, Receptive to treat children who are delayed in development.
Language and Expressive Language, and a new sup- Administration of items is flexible and can be adjusted
plementary subtest, Inventory of Vocabulary Words. if, for example, children have sensory or motor
Parent interviews are also useful in screening tests of impairment. Multiple sources of data, including indi-
186 DEVELOPMENTAL AND BEHAVIORAL PEDIATRICS: EVIDENCE AND PRACTICE

TABLE 7D-3 ■ Norm-Referenced Tests for the Assessment for Infants, Toddlers, and Preschoolers

Age Rang
Assessment Tool Publisher, Date (Years) Features

Clinical Evaluation of Language The Psychological 3.0-6.11 Assesses expressive and receptive skills
Fundamentals: Preschool (CELF-P) Corporation, 1992 Total language and subscale scores
Kaufman Survey of Early Academic American Guidance 3.0-6.11 Screens speech, language, and preacademic
and Language Skills (K-SEALS) Service, 1993 skills
Generates scaled scores
Mullen Scales of Early Learning American Guidance 0-5.8 Includes broad array of abilities
Service, 1993
Preschool Language Scale: Fourth The Psychological 0-6.11 Assesses auditory comprehension and
Edition (PLS-4) Corporation, 2002 expressive communication
Three supplemental tests
Generates total score plus separate subscale
scores
Receptive-Expressive Emergent PRO-ED, 2003 0-3.0 Assesses comprehension and expressive
Language Test (REEL-3) communication through parent interview
and observation format
Sequenced Inventory of University of Washington 0.4-4.0 Assesses expressive and receptive skills and
Communication Development: Press, 1984 areas in need of further assessment
Revised (SICD-R) Generates age-equivalent score
Test of Early Language Development: PRO-ED, 1999 2.0-7.11 Assesses receptive and expressive language,
Third Edition (TELD-3) syntax, and semantics
Generates scaled scores plus subtest scores

TABLE 7D-4 ■ Criterion-Referenced Assessments for Infants, Toddlers, and Preschoolers

Assessment Tool Publisher, Date Age Range (Years) Features

Battelle Developmental Riverside Publishing, 0-7.11 Allows multiple assessment methods

Inventory–2 2005 Provides scaled scores and age-equivalent scores
Brigance Diagnostic Inventory Curriculum Associates, 0-6.11 Generates a developmental quotient and
of Early Development: 1991 developmental age
Revised Edition
Developmental Assessment of PRO-ED, 1998 0-5.0 Allows multiple assessment methods
Young Children Provides scaled scores and age-equivalent scores
Diagnostic Evaluation of Harcourt, 2003 4.0-9.0 Appropriate for children who speak nonstandard
Language Variance (DELV) English
Integrates assessment of phonology, semantics,
syntax, and pragmatics
Hawaii Early Learning VORT Corporation, 0-3.0 Allows multiple assessment methods
Profile: Birth to Three 1994 Generates age-equivalent score
Rossetti Infant-Toddler Lingui Systems, Inc., 0-3.0 Assesses interaction-attachment, gesture, play,
Language Scale 1990 comprehension and expression

vidually administered test items, parent reports, and An advantage of criterion-referenced tests, particu-
casual professional observations, can all be used to larly for young children, is that they can be used to
determine whether a child should be given credit for simultaneously assess children and plan educational
any given item. Criterion-referenced measurement or therapeutic interventions. For this reason, these
emphasizes the specific behaviors that have been measures are often used in federally funded early
mastered, rather than the relative standing of the intervention programs to qualify children for ser-
child in reference to the group. A listing of represen- vices and generate the Individualized Family Service
tative criterion-referenced tests can be found in Table Plans. Most of these tests are comprehensive and
7D-4. include one or more sections on communication or
 CHAPTER 7 Screening and Assessment Tools 187

language. Many criterion-referenced tests generate choice of an instrument from this list is often related
age-equivalent scores or developmental quotients, to the purpose of the evaluation.
rather than or in addition to scaled scores.
FORMAL MEASURES OF LANGUAGE
COMPONENTS AND SPEECH
Assessing Language in Older
In many situations, a single comprehensive measure
Preschool- and School-Aged Children fails to provide the necessary information for under-
As children grow older, their language and speech standing a child’s profi le of strengths and weaknesses
skills become increasingly differentiated. Assessment in language and speech. In these circumstances,
of language and speech skills often requires either a speech and language pathologists design an assess-
comprehensive test or multiple measures to survey ment protocol, often choosing one or more formal
the full array of language components. Formal norm- measures for specific language components to supple-
referenced measures play an increasingly prominent ment the comprehensive tests or to test specific
role, although informal assessment continues to hypotheses about a child’s profi le. For example, a
provide interesting insights into functional commu- child’s scores on a receptive vocabulary test may be
nication, as well as speech patterns. depressed because the child impulsively pointed
at a picture after presentation of the stimulus
COMPREHENSIVE FORMAL MEASURES before carefully considering all options. In such a
In assessing speech and language skills in children in case, the comparison of receptive and expressive
the late preschool- or school-age period, speech and vocabulary may be informative. If the child’s perfor-
language pathologists frequently choose a compre- mance on one domain of language is particularly
hensive formal measure that surveys a variety of lan- weak on a comprehensive test, validation with a
guage components. These tests usually assess receptive second measure might be advisable. Table 7D-6 lists
and expressive skills in separate subtests. They typi- some of the measures that assess specific components
cally generate subscale scores, as well as a composite of language.
score. It is essential to evaluate the pattern of subtest In addition to measures of oral language, the speech
scores, as well as the composite, to determine whether and language pathologist may also be called upon to
a child’s disorder is general or specific. A representa- provide measures of preliteracy and literacy skills.
tive list of comprehensive language measures is found Children with deficits in language or speech are at
in Table 7D-5. Note that the age ranges for the tests high risk for deficits in reading and writing skills. In
and the types of subtests vary across instruments. The preschool children, formal measures of prereading

TABLE 7D-5 ■ Comprehensive Norm-Referenced Tests of Language Abilities for School-Aged Children and Adolescents

Assessment Tool Publisher, Date Age Range (Years) Features

Clinical Evaluation of Psychological 5.0-21.0 Assesses language content, structure, and

Language Fundamentals: Corporation, 2003 memory
Fourth Edition Generates total language and subtest scores
Comprehensive Assessment of American Guidance 3.0-21.0 Measures comprehension, expression, and
Spoken Language Service, 1999 retrieval
Generates scaled scores for total and
supplemental indices
Language Processing Test: Lingui Systems, Inc., 5.0-11.11 Assesses ability to process, organize, and attach
Revised 1995 meaning to sound
Oral and Written Language American Guidance 3.0-21.0 Assesses listening comprehension, oral
Scales (OWLS) Service, 1995 expression, and written expression
Composite and Listening comprehension scores
Test of Adolescent Language: PRO-ED, 1994 12.0-24.11 Assesses verbal language, reading and written
Third Edition (TOAL-3) language
Generates composite and subscale scores
Test of Language PRO-ED, 1997 8.0-12.11 Assesses spoken language, semantics, syntax,
Development–Intermediate: listening, and speaking
Third Edition (TOLD-I:3) Generates quotients for each section
Test of Language PRO-ED, 1997 4.0-8.11 Assesses listening, organizing, speaking,
Development–Primary: semantics, syntax, and spoken language
Third Edition (TOLD-P:3) Generates scaled score for each section
188 DEVELOPMENTAL AND BEHAVIORAL PEDIATRICS: EVIDENCE AND PRACTICE

TABLE 7D-6 ■ Tests to Assess Specific Subcomponents of Language

Assessment Tool Publisher, Date Age Range (Years) Features

Expressive One Word Academic Therapy 2.0-12.1 Primarily assesses words but may provide
Vocabulary Test (EOWVT) Publications, 2000 information on other components
Provides scaled scores
Listening Skills Test (LIST) The Psychological 3.0-11.0 Assesses ability to make decisions about verbal
Corporation, 2001 language
Provides total score
Peabody Picture Vocabulary American Guidance 2.5-adulthood Primarily assesses single-word vocabulary by
Test: Third Edition Service, 1997 pointing, but may provide information on
(PPVT-III) attention
Receptive One-Word Academic Therapy 2.0-18.11 Single-word vocabulary comprehension
Vocabulary Test: 2000 Publications, 2000 Provides scaled score
Structured Photographic Janelle Publications, 4.0-9.11 Assesses morphology and syntax
Expressive Language Test: 2003 Generates scaled score
Third Edition (SPELT-3)
Test for Auditory PRO-ED, 1999 3.0-9.11 Assesses understanding of vocabulary, grammar,
Comprehension of Language: and sentence structure
Third Edition (TACL-3) Generates total composite score
Test of Pragmatic Language PRO-ED, 1992 5.0-13.11 Assesses appropriateness of pragmatic and
(TOPL) social skills
Generates composite score

TABLE 7D-7 ■ Individually Administered Speech Assessment Tools

Assessment Tool Publisher, Date Age Range (Years) Features

Arizona Articulation Proficiency Western Psychological 1.5-18.0 Generates age-equivalent, intelligibility, and
Scale: Third Edition Services, 2000 severity ratings
Goldman-Fristoe Test of American Guidance 2.0-20.0 Three subtests: Sounds-in-Words, Sounds-
Articulation: Second Edition Service, 2000 in-Sentences, Stimulability
Generates age- and grade-equivalent scores,
gender-specific norms
Khan-Lewis Phonological American Guidance 2.0-21.0 To analyze speech sound errors, this is used with
Analysis: Second Edition Service, 2002 the Goldman Fristoe test
Generates standard score, age- and grade-
equivalent scores, and percentage of
occurrence
Photo Articulation Test: Third PRO-ED, 1997 3.0-8.11 Generates standard, age-, and grade-equivalent
Edition scores
Stuttering Severity Instrument PRO-ED, 1994 2.10-adulthood Provides frequency and duration scores,
for Children and Adults: physical concomitants, total score
Third Edition Mean scaled score and descriptive severity level
Voice Assessment Protocol PRO-ED, 1997 4.0-18.0 Evaluates pitch, loudness, quality, breath
for Children and Adults features, rate, and rhythm
Generates scaled score for pitch

and prewriting skills, tools such as the Test of Early Because a child’s skills in language and speech may
Reading Abilities provide measures of whether the be completely different, it is usually appropriate to
child is acquiring the foundation for beginning to include specific procedures to assess speech in a com-
acquire reading skills. In school-aged children, prehensive evaluation of a young child. Several tests
reading and writing skills can be assessed by stan- are available to assess speech sound development. A
dardized tools such as the Woodcock-Johnson battery representative list of such tests is included in Table
or the Test of Written Language. 7D-7. In school-aged students who may exhibit artic-
 CHAPTER 7 Screening and Assessment Tools 189

ulation difficulties on a small subset of sounds, mea- ments must address multiple components of commu-
sures from conversational samples may be used, rather nication, including comprehension and production of
than testing from single-word articulation tests. language and speech. In assessments of language, cli-
nicians should consider strengths and weaknesses in
INFORMAL ASSESSMENTS the various subcomponents, including vocabulary,
syntax, and pragmatics. Assessments of speech should
Informal assessment strategies continue to play an
include evaluation of sounds in single words and in
important role in the evaluation of school-aged chil-
connected discourse and should also address the
dren. Informal assessments are sometimes the best
issues of fluency, voice, and resonance, when appro-
strategy for evaluating pragmatic skills, such as topic
priate. In addition, level of speech intelligibility in
maintenance, speech acts, and sensitivity to the needs
conversation should be addressed. When these ele-
of a listener. They may also serve to demonstrate how
ments are included in the assessment, the nature of
a child integrates knowledge and skills at the level of
the child’s communication deficits can be understood,
words and sentences into connected discourse, such
appropriate diagnostic workups conducted, and suit-
as in telling or retelling stories, relating the sequence
able interventions initiated.
of a day or daily activity, or describing a complex
Research to evaluate and demonstrate the reliabil-
picture. Finally, direct observations or parent-child,
ity and validity of measures to assess speech and
clinician-child, or peer-peer interaction may be used
language functioning and disorders in children is still
to generate a speech/language sample. The advantage
needed. Some measures in common use have limited
of using observations is that formal tests tend to eval-
reliability and validity. Many of the measures do not
uate only speech sounds in individual words. Obser-
have norms for subgroups within the populations,
vations of speech in conversation and narratives allow
such as children from low socioeconomic status, chil-
the clinician to determine whether sounds that are
dren from racial and ethnic minorities, and bilingual
intelligible in individual words remain interpretable
children. There is a stunning lack of appropriate
in connected discourse.
instruments for assessing speech and language in
many subgroups defi ned in terms of language, dialect,
or cultural characteristics.
CONCLUSIONS AND Research is also needed to determine whether
EMERGING ISSUES speech and language assessment instruments in
current use are appropriate for assessment of children
Evaluation of language and speech in young children with different disorders, such as hearing impairment,
is an essential component of developmental assess- cognitive impairments, and autism. Such research
ments because language and speech play a vital role would require testing the measures on large and rep-
in multiple functional domains, including learning, resentative samples of these subgroups. It would also
communication, controlling behavior, and interacting require establishing the reliability and validity of
with others; because developmental delays and dis- instruments for the subpopulations and not just for
orders in these domains are highly prevalent; and the total normative sample.
because early treatment is effective at reducing long- A major issue in language and speech assessments
term adverse outcomes. Screening assessments for at present is that current evaluations of children, par-
language and speech should be part of routine health ticularly at the youngest ages, have limited predictive
supervision in children up to school age. Screening validity with regard to language or speech skills at
measures tend to rely on or incorporate parental older ages.27 Specificity of these early assessments is
reports of the child’s communication. These reports considerably higher than sensitivity. Determining the
tend to be comparable to observational measures. developmental skills in which early delays confer the
Comprehensive assessments of speech and language highest risk for language or speech disorder would
by a speech and language pathologist should be com- allow early intervention to be appropriately targeted
pleted when parents, physicians, or educators have to the neediest children. Similarly, identifying the
concerns about a child or when a child does not pass aspects of language or speech most predictive of later
a screening test. Observational or interview proce- reading disorders is a prerequisite for early interven-
dures are important in the assessment of young chil- tion for reading.
dren, because children may have difficulty in Finally, research has uncovered at least some
cooperating with formal procedures. Comprehensive mechanisms with which infants detect and analyze
assessments of school-aged children must include the speech stream. However, individual differences in
formal techniques because children may have ade- the adequacy or use of these mechanisms have not
quate conversational skills that mask difficulties in yet been described. Assessment of differences in these
comprehension or production. Comprehensive assess- very basic mechanisms may identify children at risk
190 DEVELOPMENTAL AND BEHAVIORAL PEDIATRICS: EVIDENCE AND PRACTICE

for language and speech disorders at much younger language impairment among second-grade children.
ages than is currently possible. Further research into J Child Psychol Psychiatry 41:473-482, 2000.
the nature of these mechanisms may also reveal strat- 21. Catts HW, Fey ME, Tomblin J, et al: A longitudinal
egies for early treatment. investigation of reading outcomes in children with lan-
guage impairments. J Speech Lang Hear Res 45:1142-
1157, 2002.
22. Brown RW: A First Language: The Early Stages. Cam-
REFERENCES bridge, MA: Harvard University Press, 1973.
1. Tomblin JB, Records NL, Buckwalter P, et al: Preva- 23. Sagae K, MacWhinney B, Lavie A: Automatic parsing
lence of specific language impairment in kindergarten of parental verbal input. Behav Res Methods Instrum
children. J Speech Lang Hear Res 40:1245-1260, Comput 36:113-126, 2004.
1997. 24. Weston A, Shriberg L, Miller J: Analysis of language-
2. Shriberg LD, Tomblin J, McSweeny JL: Prevalence of speech samples with SALT and PEPPER. J Speech Hear
speech delay in 6-year-old children and comorbidity Res 32:755-766, 1989.
with language impairment. J Speech Lang Hear Res 25. Miller J, Chapman R: The relation between age and
42:1461-1481, 1999. mean length of utterance in morphemes. J Speech Hear
3. DeCasper A, Fifer W: Of human bonding: Newborns Res 24:154-161, 1981.
prefer their mothers’ voices. Science 208:1174-1176, 26. Rescorla L, Achenbach TM: Use of the language devel-
1980. opment survey (LDS) in a national probability sample
4. Eisenberg R: Auditory competence in early life: The of children 18 to 35 months old. J Speech Lang Hear
roots of communicative-behavior. Baltimore, MD: Uni- Res 45:733-743, 2002.
versity Park Press, 1976. 27. Feldman HM, Dale PS, Campbell TF, et al: Concurrent
5. Eimas PD, Siqueland ER, Jusczyk P, et al: Speech per- and predictive validity of parent reports of child lan-
ception in infants. Science 171:303-306, 1971. guage at ages 2 and 3 years. Child Dev 76:856-868,
6. Saffran JR, Aslin RN, Newport EL: Statistical learning 2005.
by 8-month-old infants. Science 274:1926-1928, 1996. 27a. Bzoch KR, League R, Brown VL: 2003 Receptive-
7. Kuhl PK: Early language acquisition: Cracking the Expressive Emergent Language Test. Austin, TX:
speech code. Nat Rev Neurosci 5:831-843, 2004. Pro-Ed.
8. Saffran JR: Musical learning and language develop- 28. Glascoe F: Parents’ Evaluation of Developmental Status.
ment. Ann N Y Acad Sci 999:397-401, 2003. Nashville, TN: Ellsworth & Vandermeer, 2005.
9. Werker JF, Tees RC: Cross-language speech perception: 29. Wetherby AM, Prizant BM: Communication and Sym-
Evidence for perceptual reorganization during the fi rst bolic Behavior Scales: Developmental Profi le, 1st
year of life. Infant Behav Dev 7:49-63, 1984. normed ed. Baltimore: Paul H. Brookes, 2002.
10. Gentner D: On relational meaning: The acquisition of 30. Lord C, Risi S, Lambrecht L, et al: The Autism Diag-
verb meaning. Child Dev 49:988-998, 1978. nostic Observation Schedule–Generic: A standard
11. Fenson L, Dale PS, Reznick J, et al: Variability in early measure of social and communication deficits associ-
communicative development. Monogr Soc Res Child ated with the spectrum of autism. J Autism Dev Disord
Dev 59:1-173, 1994. 30:205-223, 2000.
12. Houston-Price C, Plunkett K, Harris P: “Word-learning 31. Coplan J, Gleason JR, Ryan R, et al: Validation of an
wizardry” at 1;6. J Child Lang 32:175-189, 2005. early language milestone scale in a high-risk popula-
13. Bates E, Marchman VA, Thal D, et al: Developmental tion. Pediatrics 70:677-683, 1982.
and stylistic variation in the composition of early
vocabulary. J Child Lang 21:85-123, 1994.
14. Huttenlocher J: Language input and language growth.
Prev Med 27:195-199, 1998.
15. Huttenlocher J, Vasilyeva M, Cymerman E, et al: Lan-
7E.
guage input and child syntax. Cognit Psychol 45:337- Assessment of Motor Skills
374, 2002.
16. Hart B, Risley TR: Meaningful Differences in the Every-
day Experience of Young American Children. Balti-
MARIA A. JONES ■
more: Paul H. Brookes, 1995. IRENE R. MCEWEN ■
17. Dollaghan CA, Campbell TF, Paradise JL, et al: Mater- LYNN M. JEFFRIES
nal education and measures of early speech and lan-
guage. J Speech Lang Hear Res 42:1432-1443, 1999.
Motor disorders in children are associated with a
18. Campbell TF, Dollaghan CA, Rockette HE, et al: Risk
factors for speech delay of unknown origin in 3-year-
number of conditions that vary widely in age at mani-
old children. Child Dev 74:346-357, 2003. festation, the type and severity of the motor deficits,
19. Thal DJ: Language and cognition in normal and late- and prognosis.1 Arthrogryposis, developmental coor-
talking toddlers. Top Lang Disord 11(4):33-42, 1991. dination disorder, Down syndrome, cerebral palsy,
20. Tomblin J, Zhang X, Buckwalter P, et al: The associa- meningomyelocele, muscular dystrophy, osteogenesis
tion of reading disability, behavioral disorders, and imperfecta, spinal muscular atrophy, and traumatic
 CHAPTER 7 Screening and Assessment Tools 191

brain injury are some of the diagnoses in children as the result of the interaction between a health con-
who have motor disorders. dition (disease, disorder, or injury) and contextual
Assessment of motor skills is a process of gathering factors, including those related to the person and
and synthesizing information to describe and under- those related to the environment. A health condition
stand motor skills, through such means as interviews, can be classified by three interrelated domains: body
observations, questionnaires, and formal assessment structures and functions, activities, and participation.
tools.1 Information obtained through assessment can Examples of body structures are brain formation,
be useful for such purposes as diagnosing conditions bone density, and muscle composition. Strength,
associated with disordered movement, documenting balance, and coordination are examples of body func-
eligibility for services available for children with tions. Activities are specific tasks or actions, such as
developmental delays and disabilities, planning inter- walking, running, and climbing, which when com-
vention to remediate or compensate for motor deficits, bined contribute to participation in home, school,
and evaluating change in motor skills over time. community, and other situations of life. Figure 7E-1
Many professionals have interest and expertise in shows the ICF classification system and the defi ni-
motor skill assessment. Child development specialists, tions and interrelations of the components. Figure
educators, neuropsychologists, occupational thera- 7E-2 illustrates an application of the ICF framework
pists, pediatricians, and physical therapists are some for a child with Down syndrome.
of the potential team members in assessment and The ICF framework is helpful for deciding what to
intervention for children with conditions that affect assess to answer specific questions related to a child’s
their motor skills. To promote a shared understanding motor skills.3 If parents were concerned, for example,
of motor skill assessment, the purposes of this chapter about why their young child is not yet sitting, use of
are (1) to describe a common framework for assess- a tool that assessed the child’s body functions and
ment of motor skills in children, with emphasis on structures, such as strength and postural reactions
the focus of the assessment and its purpose; (2) to would be appropriate. (Table 7E-1 contains defi ni-
review general considerations for measurement of tions of some terms used in motor assessment.) A test
motor skills; and (3) to summarize formal tools that designed to measure the child’s ability to sit (activity)
are commonly used for assessment of motor skills of would elicit results that confi rm or contradict the
neonates and infants, preschool-aged children, school- parent’s observations but would not provide informa-
aged children, and adolescents. tion about possible limitations in body functions and
structures that prevented the child from sitting. If
parents were seeking early intervention services
FOCUS OF MOTOR under the Individuals with Disabilities Education
SKILLS ASSESSMENT Improvement Act,4 then a tool that included mea-
sured activities would be most helpful for determin-
The World Health Organization’s International Clas- ing the child’s eligibility for services on the basis of
sification of Functioning, Disability and Health (ICF)2 delayed motor skills in comparison to typical peers.
provides a useful framework for deciding which aspect Measurement of activity and participation would be
of motor skills to measure when planning an assess- required if a child with motor deficits entered fi rst
ment (also see Chapter 6). The ICF is a biopsychoso- grade in a new school and the parents and school
cial model in which health and disability are viewed team members questioned the child’s ability to func-

Health condition
(disease/disorder)

FIGURE 7E-1 Flowchart depicting the International Body function and structure Activities Participation
Classification of Functioning, Disability and Health (Physiological functions of body (Execution of a (Involvement in
(ICF). (From World Health Organization: International systems; anatomical parts of the task or action) a life situation)
Classification of Functioning, Disability and Health. body such as organs, limbs, and
Geneva: World Health Organization, 2001.) their components)

Environmental factors Personal factors

192 DEVELOPMENTAL AND BEHAVIORAL PEDIATRICS: EVIDENCE AND PRACTICE

Condition
Down syndrome

Impairments related to body Activity limitations Participation restrictions

structures and function Walking on uneven Playing at recess
Decreased muscle strength surfaces Physical education
Low muscle tone Climbing on/off
Lax joints various surface heights
Slow reaction time Jumping
Throwing ball

Environmental factors Personal factors

Teachers’ concerns about safety Attitude toward exercise
Distance to playground Cognitive abilities
Children crowded on equipment Motivation

FIGURE 7E-2 Flowchart illustrating the Application of the International Classifi cation of Functioning, Dis-
ability and Health to a child with Down syndrome who wants to participate in recess and physical
education.

TABLE 7E-1 ■ Motor Assessment Terms

Term Definition

Automatic reactions Coordinated patterns of movement that occur in response to a stimulus, such as reactions that maintain
or movements balance (equilibrium reactions) or align the head and body (righting reactions); may include primitive
reflexes, described at end of table.76
Developmental Observable milestones of typical children that represent progress toward achieving upright posture, mobility,
motor skills and manipulation.73
Functional motor Self-chosen, self-directed motor skills that are meaningful for the child and family.77
skills
Fine motor skills Skills that involve the small muscles of the body, especially in eye-hand coordination tasks, to make small,
precise movements.
Gross motor skills Skills or movement, such as jumping, that involve the large muscles of the body.
Muscle tone Tension or stiffness of muscles at rest; resistance to quick passive movement. Stiffness may be abnormally
high (hypertonia), low (hypotonia), or fluctuating. Muscle tone varies with position and activity. The
relationship of passive stiffness at rest to active movement is unclear.76
Postural control or Regulation of the body’s position in space for stability and orientation. Stability (or balance) maintains or
reactions regains the position of the body over the base of support. Orientation aligns the body parts, in relation to
one another, so that they are appropriate for the movement or task being accomplished.78
Primary, early, or Coordinated patterns of movement demonstrated spontaneously by normally developing infants that may
primitive reflexes also be elicited by external stimuli. Examples include the rooting, Moro, and asymmetrical tonic neck
reflexes.79

tion in the school environment and wanted to iden- disease. The overall purpose of the ICF is to provide
tify supports that the child might need or goals for a common language and framework for describing
intervention. health and health-related status.2,3
The ICF is similar to the older frameworks that the Because many factors are related to children’s per-
World Health Organization and other groups have formance of motor skills, a number of tests and mea-
developed, but more positive terminology is used to sures exist for assessment of the dimensions of the
focus on “components of health” (World Health Orga- ICF, particularly body structure and function and
nization,2 p 4) rather than on the consequences of activity. When clinicians decide on a tool to use, the
 CHAPTER 7 Screening and Assessment Tools 193

purpose for the assessment is another important mative data obtained from measurement research are
consideration. not available.8
Predictive measures are used for screening and
diagnostic purposes to identify which children have
PURPOSES OF MOTOR or are likely to have a particular condition or status
ASSESSMENT in the future.5 Testing of infants who are at risk for
abnormal motor development, for example, is an
Kirshner and Guyatt5 described three purposes for attempt to predict which infants will be later receive
clinical measurement: discrimination, prediction, diagnoses of conditions such as cerebral palsy. Early
and evaluation. These purposes provide a framework identification can lead to intervention aimed at pre-
to use in conjunction with ICF domain for identifying venting or ameliorating the effects of the condition.
appropriate tools for measurement of children’s motor Evaluative measures are used to assess change over
skills. Discriminative measures identify children with time or as a result of intervention.5,6 Good evaluative
and without a particular characteristic or with varying measures are responsive to change that occurs,
degrees of a characteristic,6 such as delayed gross whether in body structures and functions, activity, or
motor skills, impaired balance, or superior manual participation. Although measuring change in body
dexterity. Discriminative measures can be norm ref- structures and function can be appropriate, evalua-
erenced or criterion referenced. In norm-referenced tion of change also should include measurement of
tests, a child’s motor skills are compared with those activities and participation that are meaningful to the
of typical children of the same age, and scores indi- child and family.6
cate how the child’s skills compare within the normal Most tests and measures for assessment of motor
distribution of scores of typical children.7 Criterion- skills are useful for only one or two purposes. Norm-
referenced tests can be used to assess such body struc- referenced developmental tests, for example, often are
tures and functions as postural control and reactions, not good predictors of infant’s motor performance at
but they also are used to measure performance of later ages.9 Other assessments effectively identify chil-
activities such as the ability to kneel, drink from a dren with delayed motor development but are not
cup, or open a locker. Although criterion-referenced useful for evaluation of change as a result of interven-
tests often are constructed to allow comparison of a tion. Selection of tools that match the purpose of the
child’s development or performance with estimates of assessment is key to obtaining useful test results.
development or performance of typical children, nor- Table 7E-2 lists commonly used motor assessment

TABLE 7E-2 ■ Common Motor Assessment Tools

Type of Test, Most

Test Name Useful Purpose Age for Testing ICF Dimension

Alberta Infant Motor Scale Discriminative, norm-referenced 0-18 months Body structure and function: postural
(AIMS) 9 Evaluative for infants with control
delayed, but not abnormal Activity: motor performance
movement
Predictive
Assessment of Preterm Discriminative, norm-referenced Preterm to full- Body structure and function: physiological/
Infant Behavior (APIB) 20 term infants autonomic system reactions, attention,
state and motor organization,
self-regulation
Battelle Developmental Discriminative, norm-referenced 0-7.11 years Activity: cognitive, communication, social-
Inventory–2nd ed. emotional, motor, and adaptive skills
(BDI-2) 31
Bayley Scales of Infant and Discriminative, norm-referenced 1-42 months Activity: cognitive, motor, language, social-
Toddler Development– emotional, adaptive behavior
Third Edition (BSID-III) 32
Birth to Three Assessment 33 Discriminative, criterion- 0-36 months Activity: gross motor, fine motor, language,
referenced personal-social skills; nonverbal thinking
Bleck’s Locomotor Prognosis Predictive of ambulation in 12 months and Body structure and function: postural
in Cerebral Palsy48 children with cerebral palsy older reactions and reflexes
at 7 years of age
194 DEVELOPMENTAL AND BEHAVIORAL PEDIATRICS: EVIDENCE AND PRACTICE

TABLE 7E-2 ■ Common Motor Assessment Tools—cont’d

Type of Test, Most

Test Name Useful Purpose Age for Testing ICF Dimension

Bruininks-Oseretsky Test of Discriminative, norm-referenced 4-21 years Body structure and function: balance and
Motor Proficiency, 2nd ed. coordination
(BOT-2) 42 Activity: gross and fine motor performance
Canadian Occupational Discriminative Any Any
Performance Measure, Evaluative
4th ed67
Children’s Handwriting Discriminative Grades 1-2 Activity: manuscript handwriting
Evaluation Scale for performance
Manuscript Writing
(CHES) 59
Diagnosis and Remediation Discriminative, criterion- Grade 3+ Activity: handwriting performance
of Handwriting referenced
Problems62
Early Learning Discriminative, criterion- 0-36 months Activity: gross motor, fine motor, cognitive,
Accomplishment Profile referenced language, self-help, social-emotional
(E-LAP) 34 development
Evaluation Tool of Children’s Discriminative, criterion- Grades 1-2 Activity: handwriting performance
Handwriting (ETCH) 63 referenced
General Movements29 Discriminative, criterion- Preterm to 4 Body structure and function: general
referenced months movement of trunk and extremities
Gross Motor Function Discriminative, criterion- Any age Activity: lying and rolling, sitting, crawling,
Measure (GMFM) 45 referenced kneeling, standing, walking, running, and
Evaluative jumping
Hawaii Early Learning Discriminative, curriculum- 0-6 years Activity: cognition, language, motor, fine
Profile35 referenced motor, social, self-help skills
Infant Neurobiological Discriminative, criterion- 0-18 months Body structure and function: postural
International Battery referenced control, muscle tone, vestibular function
(INFANIB) 21 Predictive
Minnesota Handwriting Discriminative, norm-referenced Grades 1-2 Activity: manuscript handwriting
Test61 performance
Movement Assessment Discriminative, norm-referenced 4-12 years Activity: Manual dexterity, balance, ball
Battery for Children handling, visual-motor skills
(Movement ABC) 55
Neonatal Behavioral Discriminative, criterion- Full-term Body structure and function: oral-motor,
Assessment Scale18 referenced infants muscle tone, vestibular function
Neonatal Individualized Discriminative, criterion- Preterm–4 weeks Body structure and function: autonomic
Developmental Care and referenced and motor organization, attention
Assessment Program
(NIDCAP)17
Newborn Behavioral Discriminative, criterion- Birth–2 months Body structure and function: physiological,
Observation system19 referenced motor, state organization
Peabody Developmental Discriminative, norm- and 0-5 years Activity: gross motor and find motor skills
Motor Scales, 2nd ed. criterion-referenced
(PDMS-2) 30
Pediatric Evaluation of Discriminative, norm- and 6 months–71/2 Activity and participation: self-care,
Disability Inventory criterion-referenced years functional mobility, social function
(PEDI) 44 Evaluative level
School Function Assessment Discriminative, criterion- Kindergarten– Participation, task supports, activity
(SFA) 66 referenced sixth grade performance (physical tasks, cognitive/
Evaluative behavioral tasks)
Test of Infant Motor Discriminative, norm-referenced 32 weeks’ Body function/structure and activity:
Performance (TIMP) 23 Predictive gestation to 4 postural control; ability to orient and
months post stabilize head in space and in response
term to stimulation; selective control of distal
movements; antigravity control of trunk
and extremities
Test of Legible Handwriting60 Discriminative 7-18.5 years Activity: handwriting performance

ICF, International Classification of Functioning, Disability, and Health.

 CHAPTER 7 Screening and Assessment Tools 195

tools, their most useful purposes, and the dimensions for the delays are important, as are measures that are
of the ICF that they measure. predictive of future diagnoses, such as cerebral palsy.11
Motor assessment of infants often focuses on body
structures and function, such as muscle tone and
GENERAL CONSIDERATIONS reflexes,12 as well as neuromotor development, pos-
tural reactions, and fi ne and gross motor skills (see
FOR ASSESSMENT OF
Table 7E-2).13 For infants born prematurely or with
MOTOR SKILLS other risk factors, periodic monitoring beyond the
neonatal period is important; the assessment of motor
When deciding among tools to assess children’s motor skills is a component of the monitoring process.l4
skills, examiners should identify the ICF dimension Repeated assessment is recommended for early iden-
of interest, the purpose of the assessment, the psycho- tification of infants with motor dysfunction or delay
metric properties of the tests, and the age group for and to predict which infants may later receive diag-
which the tests were developed. Psychometric proper- noses of conditions not evident at birth or shortly
ties, such as reliability and validity, are important for thereafter.15,16
controlling measurement error and ensuring that the
measurements will be useful. Because reliability of
measurements is population specific, reporting of reli-
ability coefficients in a text such as this could be Tool That Focus Specifically on
misleading without a full description of the charac- Motor-Related Function
teristics of the study participants and the examiners.8
The Neonatal Individualized Developmental Care and
For diagnostic tests, sensitivity and specificity are Assessment Program17 is a comprehensive criterion-
important, and responsiveness is important for evalu- referenced assessment for preterm or full-term infants
ative tests. Psychometric properties of tests often are up to 4 weeks’ post-term age. It involves a systematic
provided in test manuals and in reports of research observation of the infant’s autonomic, motor, and
conducted after the tests were published. attention responses during caregiving routines and
Examiners also need to be consistent with test discriminates infants with difficulty in the three
administration and knowledgeable in interpreting the areas. It is a total program that encompasses both the
results. The results of repeated testing and observa- assessment and related caregiving recommendations.
tion of motor skills in the environment where they Other neonatal assessments that focus on the motor,
will be used will provide information to guide and behavioral, and physiological function of infants born
modify intervention. However, intervention should
at term or before term include the Neonatal Behav-
not be driven by items failed on motor assessments; ioral Assessment Scale,18 the Newborn Behavioral
rather, they should focus on activities that children
Observations system,19 and the Assessment of Preterm
and their families identify as meaningful and that
Infants’ Behavior.20
children perform in everyday environments.10 The
The Infant Neurobiological International Battery
questions to be answered often are different for chil- (INFANIB)21 is a criterion-referenced tool used to
dren of different ages, and many tools are age specific. assess neuromotor status of infants from birth to age
In the following sections, we describe considerations 18 months who were born prematurely. In addition
for motor skills assessment of children in four age to the test’s discriminating between normal and
groups and describe some of the tools available for abnormal development, an infant’s scores on spastic-
assessment of each group. ity and head and trunk subscales at 6 months of age
are highly predictive of cerebral palsy at 12 months
(86.8% for spasticity and 87.1% for head and trunk
MOTOR ASSESSMENT OF subscales).22
INFANTS AND TODDLERS The Test of Infant Motor Performance23 assesses
motor development of infants from 32 weeks after
Motor development in typical infants and toddlers is conception through 4 months’ post-term age. The test
predictable and shifts from reflexive to purposeful discriminates infants at risk for motor dysfunction
movement, leading to the ability to move against from typically developing infants24 and had 0.92 sen-
gravity, transition in and out of different body posi- sitivity for identifying infants at age 3 months with
tions, and explore the environment by crawling, delayed performance on the Alberta Infant Motor
walking, and climbing.9 For infants born prematurely Scale (AIMS)1 at 12 months of age.25 The test also is
and for infants and toddlers who are not achieving one of the best predictors of a later diagnosis of cere-
typical motor milestones, discriminative measures to bral palsy. Of infants whose motor performance was
identify delays in motor development and the reasons delayed at 3 months according to their Test of Infant
196 DEVELOPMENTAL AND BEHAVIORAL PEDIATRICS: EVIDENCE AND PRACTICE

Motor Performance scores, 75% received a diagnosis ingful, measurable goals identified by the family and
of cerebral palsy by preschool age.11 other team members.10
The AIMS is a norm-referenced discriminative
measure of the infant’s gross motor development from
40 weeks after conception to independent walking. MOTOR ASSESSMENT OF
The AIMS differentiates infants’ motor development PRESCHOOL-AGED CHILDREN
as normal, at risk, and abnormal. A score at the 10th
percentile or below at age 4 months (sensitivity, 0.77; Motor development of preschool-aged children is
specificity, 0.82) or at the 5th percentile or below at characterized by active movement throughout their
age 8 months (sensitivity, 0.86; specificity, 0.93) is environments and continued refi nement of skills pre-
predictive of motor delay at 18 months of age.26 viously acquired.36 In the preschool years, motor skills
The general movements assessment 27 also is a dis- become particularly important for social interaction
criminative measure with predictive validity. Exam- and play. Children with motor impairments that limit
iners observe the quality of infants’ gross movements their ability to explore and interact with their envi-
at variable speeds and amplitudes and then classify ronment are at risk for delayed development in cogni-
the movements as normal or abnormal. When infants tive, communication, and social domains.37-39 Because
show abnormal general movements at 2 and 4 months most children with moderate to severe motor impair-
after term, the test is predictive of cerebral palsy, with ments have been identified as having delayed motor
accuracy of 0.85 to 0.98.28,29 development or have a medical diagnosis by this age,
As infants develop and motor delays are suspected, norm-referenced discriminative measures are rarely
motor assessment often focuses on early identification useful, but other types of discriminative measures
to determine whether infants meet eligibility criteria can be helpful for measurement of motor-related
for early intervention services under the Individuals dimensions within ICF, such as range-of-motion
with Disabilities Education Improvement Act.4 Crite- (body functions/structures); mobility, and self-help
ria vary from state to state; however, most are based skills (activities); and the child’s ability to participate
on presence of a qualifying condition, such as Down within family routines and community settings
syndrome, or on a documented development delay. A (participation).
tool that is widely used to identify and document For all preschool-aged children with delayed motor
motor delays is the Peabody Developmental Motor development, measurement of the effect of motor
Scales,30 a norm-referenced discriminative test of skills on functioning is more important than simply
gross and fi ne motor development for children from documenting a motor delay.10 The ability of a child
birth to age 72 months. to function in age-appropriate daily activities of the
family and community40 also needs to be assessed,
and interventions must be provided to remediate
Comprehensive Developmental when possible or to compensate when children are
Assessment Tools That Include unlikely to achieve necessary motor skills. Observa-
tion of motor skills in the environments in which
Motor Development children use them usually yields the most useful
Most other assessments used for early identification information for intervention planning.41
are comprehensive criterion-referenced or norm-
referenced tools that assess infants’ and toddlers’
development in several areas, such as motor, cogni-
Tools That Measure Motor Skills
tive, social-emotional, communication, and adaptive In children with mild motor disorders or with acquired
development. Frequently used norm-referenced tests or progressive conditions, such as Duchene muscular
include the Battelle Developmental Inventory31 and dystrophy, a motor delay might fi rst be identified
the Bayley Scales of Infant and Toddler Develop- during the preschool years. The Peabody Develop-
ment.32 Examples of criterion-referenced tests include mental Motor Scales,30 a norm-referenced tool com-
the Birth to Three Assessment,33 the Early Learning monly used to assess infants’ fi ne and gross motor
Accomplishment Profi le,34 and the Hawaii Early development, also is widely used for children of pre-
Learning Profi le.35 Although discriminative tools for school age. The Bayley Scales of Infant and Toddler
infants and toddlers measure skills that might be Development32 is appropriate for children up to 42
meaningful for evaluation of change in some indi- months of age, and motor skills of children as young
vidual children, they are not useful as evaluative as 4 years can be assessed with the Bruininks-
measures for most children with motor impairments. Oseretsky Test of Motor Proficiency (BOT-2).42
The most useful individual evaluative measure is Few tools have been developed to assess the motor
often to determine whether a child achieves mean- capabilities of children with disabilities or change in
198 DEVELOPMENTAL AND BEHAVIORAL PEDIATRICS: EVIDENCE AND PRACTICE

of GMFM scores and the children’s severity of motor assess manual dexterity, balance, ball handling, and
impairment.52 visual motor skills. The M-ABC can be used with
children aged 4 to 12 years. Although these tests are
often used, authors have expressed concern about the
MOTOR ASSESSMENT OF potential lack of agreement between the tests in iden-
tifying children with developmental coordination
SCHOOL-AGED CHILDREN disorder.53,56
Both the BOT-2 and the M-ABC have components
As the demands of the environment increase and
that address fi ne motor skills, but neither specifically
children are required to perform more complex motor
addresses the development of handwriting, which
tasks, such as writing and physical education activi-
is a common reason that school-aged children are
ties, parents and teachers may become concerned
referred for motor assessments.57,58 Frequently used
about uncoordinated movements in children not pre-
tools for assessing handwriting include the Children’s
viously identified as having motor deficits. Uncoordi-
Handwriting Evaluation Scale for Manuscript
nated movements are characterized by inconsistency
Writing,59 the Test of Legible Handwriting,60 the Min-
in performance, asymmetry, loss of balance, falling,
nesota Handwriting Test,61 Diagnosis and Remedia-
slow reaction and movement timing, decreased muscle
tion of Handwriting Problems,62 and the Evaluation
force, and poor motor planning.53 Children who con-
Tool of Children’s Handwriting.63 Although these and
sistently show uncoordinated movements may have a
other tools for assessing handwriting are used, the
developmental coordination disorder or other devel-
identification of the reason for handwriting problems
opmental disability that was not apparent until the
can be difficult because a child’s motor and visual
child reached school age.
perceptual abilities, as well as orthographic, spelling,
If children have not previously been identified as
and written language processing, all contribute to
having difficulty with motor skills, norm-referenced
writing success.64,65
discriminative tools can be useful for comparing a
child’s performance with typical children of the same
age and for identifying strengths and deficits in com-
ponents of motor skills, such as balance, coordina- Tools That Measure Motor-Related
tion, and visual-motor skills. Norm-referenced tools Functional Skills
usually are not helpful for school-aged children whose
Another frequent purpose of motor assessment of
motor disorders were previously identified, but tools
school-aged children is to measure the effects of motor
that help identify functional deficits and that evaluate
skills on children’s ability to function within the
change over time can be useful. Observation in chil-
school environment. This is similar to assessing
dren’s own environments often is the most valuable
preschool-aged children with motor delays, in which
method for identifying the effects of motor disorders
the focus of assessment shifts from identifying the
on activities and participation and for identifying
presence of motor delays or evaluating the effects of
potential goals of intervention.10 The previously
intervention aimed at improving developmental
described methods for predicting motor development
motor skills in isolation to measuring functional
of preschool-aged children with cerebral palsy and
changes within the activities and participation
Down syndrome continue to be useful for predicting
levels of the ICF framework. The use of assess-
motor skills in some school-aged children.
ment tools designed to measure functional changes
over time is important for identifying and measuring
Tools That Measure Developmental individually meaningful goals and planning
intervention.
Motor Skills The PEDI44 continues to be appropriate for assess-
A motor-related condition that often is identified ing change in function and caregiver assistance over
during the school years is developmental coordina- time for school-aged children. Tools such as the School
tion disorder, which affects movement in the absence Function Assessment (SFA) 66 and the Canadian Occu-
of identified neurological dysfunction.54 Two norm- pational Performance Measure (COPM) 67 also can be
referenced tests that commonly are used to help iden- useful for these purposes. The SFA includes items that
tify children with developmental coordination address the activity and participation levels of the ICF
disorder are the BOT-242 and the Movement Assess- framework. The SFA is intended to determine the
ment Battery for Children (M-ABC).55 The BOT-2 child’s current level of participation and performance
assesses fi ne motor skills, gross motor skills, balance, in elementary school activities and to document the
and coordination and can be used with persons aged supports a child needs to participate and perform in
4 to 21 years. The M-ABC also includes items that those activities. The SFA can be completed by one or
 CHAPTER 7 Screening and Assessment Tools 199

more school professionals who have observed the tools are available to measure motor-related participa-
child during typical school activities and routines. tion, and, except for the individual-specific COPM,67
One weakness of the SFA is that it takes about 1.5 those that do exist are most appropriate for children
hours to complete the assessment. of elementary school age or younger. Development of
The COPM was designed to identify goals of inter- new tools to measure broader aspects of motor-related
vention and to measure outcomes. It has been widely participation for children of all ages would not only
used in adult rehabilitation to detect change in a enable researchers to identify participation limita-
person’s self-perception of performance over time68 ; tions that might be ameliorable but, if developed as
however, its use for children with disabilities and discriminative and evaluative measures, would also
their families is increasing.69 Authors of the COPM allow measurement of change in participation over
recognize the limitations with using the COPM with time or with intervention.72 Psychometrically sound
children younger than 8 years because of the diffi- measures of participation also would be useful for
culty with the self-assessment necessary to complete aggregating data for program evaluation purposes.
the COPM, but they reported that research is under Research to determine minimal clinically impor-
way to develop a different method for accessing young tant change in scores on evaluative tools is important
children’s goals and priorities. for understanding the relevance of change in indi-
vidual children and for program evaluation.73,74 Devel-
opment of computer-adapted testing to minimize the
MOTOR ASSESSMENT number of test items that need to be administered
would reduce the time required for assessment.75 New
OF ADOLESCENTS tools also are needed to help predict motor disorders
other than cerebral palsy in young infants and to help
Adolescence is a time of increased independence,
predict the likely limits of development in children
beginning separation from family, and physical body
with a variety of conditions associated with motor
changes. More complex motor skills develop and often
disorders.
are practiced through participation in sports and
involvement in community activities.36 Adolescence
also is a time when risk of injury increases, and the
incidence of traumatic brain injury, which can have REFERENCES
a range of motor and behavioral consequences,70 is 1. Tupper DE, Sondell SK: Motor disorders and neuropsy-
particularly high between the ages of 15 to 19 chological development. In Dewey D, Tuper DE, eds:
years.71 Developmental Motor Disorders: A Neuropsychological
For adolescents with relatively high levels of motor Perspective. New York: Guilford, 2004, pp 3-25.
2. World Health Organization: International Classifica-
function, the BOT-2,42 for which normative data up
tion of Functioning, Disability and Health. Geneva:
to age 21 are available, could be a useful discrimina- World Health Organization, 2001.
tive measure. Depending on the purpose for the 3. Lollar DJ, Simeonsson RJ: Diagnosis to function: clas-
assessment, tools used with school-aged children also sification for children and youths. J Dev Behav Pediatr
may be appropriate for adolescents with motor disor- 26:323-330, 2005.
ders. For those functioning at less than a 7-year age 4. Individuals with Disabilities Education Improvement
level, the PEDI44 could be useful for identifying mobil- Act of 2004 (Public Law 108-446), 20 U.S.C. 1400.
ity and self-care skills that are lacking and for mea- 5. Kirshner B, Guyatt B: A methodological framework for
suring change in those skills over time if intervention assessing health and disease. J Chron Dis 38:27-36,
is determined to be appropriate. The COPM is a par- 1985.
ticularly valuable measure for helping adolescents 6. Rosenbaum PL: Clinically based outcomes for children
with cerebral palsy: Issues in the measurement of func-
identify meaningful motor-related goals and to rate
tion. In Sussmann M, ed: The Diplegic Child. Parkridge,
their perceptions of changes and satisfaction with the IL: American Academy of Orthopaedic Surgeons, 1992,
goals over time.64 pp 125-132.
7. Long T, Toscano K: Handbook of Pediatric Physical
Therapy, 2nd ed. Philadelphia: Lippincott Williams &
RECOMMENDATIONS FOR Wilkins, 2002.
8. Portney LG, Watkins MP: Foundations of Clinical
THE FUTURE Research: Applications to Practice. 2nd ed. Upper
Saddle River, NJ: Prentice-Hall Health, 2000.
Although many tools have been developed to assess 9. Piper MC, Darrah J: Motor Assessment of the Develop-
children’s motor skills and their underlying compo- ing Infant. Philadelphia: WB Saunders, 1994.
nents, most are discriminative tools that assess the 10. Harris SR, McEwen IR: Assessing motor skills. In
body function and activities domains of the ICF. Few McLean M, Bailey DB Jr, Wolery M, eds: Assessing
200 DEVELOPMENTAL AND BEHAVIORAL PEDIATRICS: EVIDENCE AND PRACTICE

Infants and Preschoolers with Special Needs, 2nd ed. 27. Prechtl HFR: Qualitative changes of spontaneous move-
Englewood Cliffs, NJ: Prentice-Hall, 1996, pp 305- ments in fetus and preterm infant are a marker of
333. neurological dysfunction. Early Hum Dev 23:151-158,
11. Kolobe THA, Bulanda M, Susman L: Predictive validity 1990.
of the Test of Infant Motor Performance at pre-school 28. Hadders-Algra M, Groothuis AMC: Quality of general
age. Phys Ther 84:1144-1156, 2004. movements in infancy is related to the development of
12. Andre-Thomas CY, Saint-Anne Dargassies S: The Neu- neurological dysfunction, attention deficit hyperactiv-
rological Examination of Infants. Little Club Clinics ity disorder and aggressive behavior. Dev Med Child
in Developmental Medicine, No. 1. London: National Neurol 41:381-391, 1999.
Spastics Society, 1960. 29. Prechtl HFR, Einspieler C, Cioni G, et al: An early
13. Case-Smith J, Bigsby R: Motor assessment. In Singer marker of neurological deficits after prenatal brain
LT, Aeskind PS, eds: Biobehavioral Assessment of lesions. Lancet 349:1361-1363, 1997.
Infants. New York: Guilford, 2001, pp 423-442. 30. Folio MR, Fewell RR: Peabody Developmental Motor
14. Allen MC, Alexander GR: Using motor milestones Scales, 2nd ed. Austin, TX: PRO-ED, 2000.
as a multistep process to screen preterm infants 31. Newborg J: Battelle Developmental Inventory–2nd ed.
for cerebral palsy. Dev Med Child Neurol 39:12-16, Itasca, IL: Riverside, 2005.
1997. 32. Bayley N: Bayley Scales of Infant and Toddler Develop-
15. Escobar GJ, Littenberg B, Pettiti DB: Outcome among ment–Third Edition. San Antonio, TX: The Psychologi-
surviving very low birthweight infants: A meta- cal Corporation, 2005.
analysis. Arch Dis Child 66:204-211, 1991. 33. Ammer JJ, Bangs TE: Birth to Three Assessment and
16. Ferrari F, Cioni G, Prechtl HFR: Qualitative changes of Intervention System. Austin, TX: PRO-ED, 2000.
general movements in preterm infants with brain 34. Glover ME, Preminger JL, Sanford AR: Early Learning
lesions. Early Hum Dev 23:193-231, 1990. Accomplishment Profi le. Lewisville, NC: Kaplan Early
17. Als H: Program Guide: Newborn Individualized Devel- Learning Company, 2002.
opmental Care and Assessment Program (NIDCAP): 35. Parks S: Hawaii Early Learning Profi le. Palo Alta, CA:
An Education and Training Program for Health Care Vort Corporation, 2004.
Professionals. Boston: Children’s Medical Center Cor- 36. Cech DJ, Martin ST: Functional Movement Develop-
poration, 1986. ment across the Lifespan, 2nd ed. Philadelphia: WB
18. Brazelton TB: The Neonatal Behavioral Assessment Saunders, 2002.
Scale. Philadelphia: JB Lippincott, 1984. 37. Hays RM: Childhood motor impairments: Clinical
19. Nugent JK, Keefer C, O’Brien S, et al: The Newborn overview and scope of the problem. In Jaffee KM, ed:
Behavioral Observations system. Boston: Brazelton Childhood Powered Mobility: Developmental, Techni-
Institute, 2005. cal and Clinical Perspectives. Washington, DC: Reha-
20. Als H, Lester BM, Tronick EZ, et al: Toward a research bilitation Engineering and Assistive Technology Society
instrument for the Assessment of Preterm Infants’ of North America, 1987, pp 1-10.
Behavior (APIB). In Fitzgerald H, Lester B, Yogman M, 38. Tefft D, Guerette P, Furumasu J: Cognitive predictors
eds: Theory and Research in Behavioral Pediatrics, vol of young children’s readiness for powered mobility. Dev
1. New York: Plenum Press, 1982, pp 35-63. Med Child Neurol 41:665-670, 1999.
21. Ellison PH: INFANIB. A Reliable Method for the Neu- 39. Kermoian, R: Locomotion experience and psychologi-
romotor Assessment of Infants. Tucson, AZ: Therapy cal development in infancy. In Furumasu J, ed: Pedi-
Skill Builders, 1994. atric Powered Mobility: Developmental Perspectives,
22. Stavrakas PA, Kemmer-Gacura GE, Engelke SC, et al: Technical Issues, Clinical Approaches. Arlington, VA:
Predictive validity of the Infant Neurological Interna- Rehabilitation Engineering and Assistive Technology
tional Battery. Dev Med Child Neurol Suppl 64:35-36, Society of North America, 1997, pp 7-21.
1991. 40. Miller LJ, Robinson C: Strategies for meaningful assess-
23. Campbell SK, Kolobe THA, Osten ET, et al: Construct ment of infants and toddlers with significant physical
validity of the Test of Infant Motor Performance. Phys and sensory disabilities. In Meisels SJ, Fenichel E, eds:
Ther 75:585-596, 1995. New Visions for the Developmental Assessment of
24. Campbell SK, Hecker D: Validity of the Test of Infant Infants and Young Children. Washington, DC: Zero to
Motor Performance for discriminating among infants Three, 1996, pp 313-329.
with varying risk for poor motor outcome. J Pediatr 41. McEwen IR, Hansen LH: Children with cognitive
139:546-551, 2001. impairments. In Campbell SK, Vander Linden DW,
25. Campbell SK, Kolobe THA, Wright BD, et al: Validity Palisano RJ, eds: Physical Therapy for Children, 3rd ed.
of the Test of Infant Motor Performance for prediction Philadelphia: WB Saunders, 2006, pp 591-624.
of 6-, 9-, and 12-month scores on the Alberta Infant 42. Bruininks RH, Bruininks BD: Bruininks-Oserestsky
Motor Scales. Dev Med Child Neurol, 44:263-272, Test of Motor Proficiency, 2nd ed. Circle Pines, MN:
2002. American Guidance Service Publishing, 2005.
26. Darrah J, Piper MC, Watt MJ: Assessment of gross 43. Ketelaar M, Vermeer A, Helders PJ: Functional motor
motor skills of at-risk infants: Predictive validity of the abilities of children with cerebral palsy: A systematic
Alberta Infant Motor Scale. Dev Med Child Neurol literature review of assessment measures. Clin Rehabil
40:485-491, 1998. 12:369-380, 1998.
 CHAPTER 7 Screening and Assessment Tools 201

44. Hayley SM, Coster WJ, Ludlow LH, et al: Pediatric 63. Amundson SJ: Evaluation Tool of Children’s Handwrit-
Evaluation of Disability Inventory. Development, Stan- ing. Homer, AK: OT Kids, 1995.
dardization and Administration Manual. Boston: 64. Adams MJ: Beginning to Read: Thinking and Learning
Health and Disability Research Institute, 1992. about Print. Cambridge, MA: MIT Press, 1990.
45. Russell DJ, Rosenbaum PL, Avery LM, et al: Gross 65. Feder KP, Majnemer A: Children’s handwriting evalu-
Motor Function Measure (GMFM-66 & GMFM-88) ation tools and their psychometric properties. Phys
User’s Manual. London: Mac Keith Press, 2002. Occup Ther Pediatr 23(3):65-84, 2003.
46. Russell DJ, Russell DJ, Avery LM, et al: Improved 66. Coster W, Deeney T, Haltiwanger J, et al: School Func-
scaling of the Gross Motor Function Measure for chil- tion Assessment Materials. San Antonio, TX: The Psy-
dren with cerebral palsy: Evidence of reliability and chological Corporation, 1998.
validity. Phys Ther 80:873-885, 2000. 67. Law M, Baptiste S, Carswell A, et al: Canadian Occu-
47. Russell D, Palisano R, Walter S, et al: Evaluating motor pational Performance Measure. 4th ed. Ottawa, ON:
function in children with Down syndrome: Validity of CAOT Publications ACE, 2005.
the GMFM. Dev Med Child Neurol 40:693-701, 1998. 68. Atwal A, Owen S, Davies R: Struggling for occupa-
48. Bleck EE: Locomotor prognosis in cerebral palsy. Dev tional satisfaction: Older people in care homes. Br J
Med Child Neurol 17:18-25, 1975. Occup Ther 66:118-124, 2003.
49. Campos du Paz A, Burnett S, Braga L: Walking prog- 69. Law M, Darrah J, Pollock N, et al: Family-centered
nosis in cerebral palsy. Dev Med Child Neurol 36:130- functional therapy for children with cerebral palsy: An
134, 1994. emerging practice model. Phys Occup Ther Pediatr
50. Rosenbaum PL, Walter SD, Hanna SE, et al: Prognosis 18(1):83-102, 1998.
for gross motor function in cerebral palsy: Creation of 70. Dewey D, Bottos S, Tupper DE: Acquired childhood
motor development curves. JAMA 288:1357-1363, conditions with associated motor impairments. In
2002. Dewey D, Tuper DE, eds: Developmental Motor Disor-
51. Palisano RJ, Rosenbaum P, Walter S, et al: Develop- ders: A Neuropsychological Perspective. New York:
ment and reliability of a system to classify gross motor Guilford, 2004, pp 169-196.
function in children with cerebral palsy. Dev Med 71. Langlois JA, Rutland-Brown W, Thomas KE: Traumatic
Child Neurol 39:214-223, 1997. brain injury in the United States: Emergency depart-
52. Palisano RJ, Walter SD, Russell DJ, et al: Gross motor ment visits, hospitalizations, and deaths. Atlanta, GA:
function of children with Down syndrome: Creation of Centers for Disease Control and Prevention, National
motor growth curves. Arch Phys Med Rehabil 82:494- Center for Injury Prevention and Control, 2004.
500, 2001. 72. Wright FV, Boschen K, Jutai J: Exploring the compara-
53. Barnhart RC, Davenport MJ, Epps SB, et al: Develop- tive responsiveness of a core set of outcome measures
mental coordination disorder. Phys Ther 83:722-731, in a school-based conductive education programme.
2003. Child Care Health Dev 31:291-302, 2005.
54. Dewey D, Wilson BN: Developmental coordination dis- 73. Beaton DE, Boers M, Wells GA: Many faces of The
order: What is it? Phys Occup Ther Pediatr 20(2-3): Minimal Clinically Important Difference (MCID): A
5-27, 2001. literature review and directions for future research.
55. Henderson SE, Sugden DA: Movement Assessment Curr Opin Rheumatol 14:109-114, 2002.
Battery for Children. San Antonio, TX: The Psychologi- 74. Iyer LV, Haley SM, Watkins MP, et al: Establishing
cal Corporation, 1992. minimal clinically important differences for scores on
56. Crawford SG, Wilson BN, Dewey D: Identifying devel- the Pediatric Evaluation Of Disability Inventory for
opmental coordination disorder: Consistency between inpatient rehabilitation. Phys Ther 83:888-898, 2003.
tests. Phys Occup Ther Pediatr 20(2-3):29-50, 2001. 75. Haley SM, Raczek AE, Coster WJ, et al: Assessing
57. Cermak SA: Fine motor functions and handwriting. In mobility in children using a computer adaptive testing
Fisher AG, Murray EA, Bundy A, eds: Sensory Integra- version of the Pediatric Evaluation Of Disability Inven-
tion: Theory and Practice. Philadelphia: FA Davis, 1991, tory. Arch Phys Med Rehabil 86:932-939, 2005.
pp 166-170. 76. Palisano RJ: Neuromotor and developmental assess-
58. Missiuna C: Strategies for success: Working with chil- ment. In Wilhelm IJ, ed: Physical Therapy Assessment
dren with developmental coordination disorder. Phys in Early Infancy. New York: Churchill Livingstone,
Occup Ther Pediatr 20(2-3):1-4, 2001. 1993, pp 173-224.
59. Phelps J, Stempel L: Children’s Handwriting Evaluation 77. Campbell SK: The child’s development of functional
Scale for Manuscript Writing. Dallas: Texas Scottish movement. In Campbell SK, Vander Linden DW,
Rite Hospital for Crippled Children, 1987. Palisano RJ, eds: Physical Therapy for Children, 3rd ed.
60. Larsen S, Hammill D: Test of Legible Handwriting. Philadelphia: WB Saunders, 2006, pp 33-76.
Austin, TX: PRO-ED, 1989. 78. Shumway-Cook A, Woollacott M: Theoretical issues in
61. Reisman JE: Development and reliability of the research assessing postural control. In Wilhelm IJ, ed: Physical
version of the Minnesota Handwriting Test. Phys Occup Therapy Assessment in Early Infancy, New York:
Ther Pediatr 13(2):41-55, 1993. Churchill Livingstone, 1993, pp 161-171.
62. Stott DH, Moyes FA, Henderson SE: Diagnosis and 79. Crutchfield CA, Barnes MR: Motor Control and Motor
Remediation of Handwriting Problems. Cardiff, UK: Learning in Rehabilitation. Atlanta, GA: Stokesville,
Drake Educational Associates, 1985. 1993.