Global Developmental Delay and Mental Retardation/Intellectual Disability

Published on 20/04/2015 by admin

Filed under Neurology

Last modified 20/04/2015

Print this page

rate 1 star rate 2 star rate 3 star rate 4 star rate 5 star
Your rating: none, Average: 1 (1 votes)

This article have been viewed 6966 times

Chapter 43 Global Developmental Delay and Mental Retardation/Intellectual Disability


Global developmental delay and mental retardation are related, complementary, nonsynonymous terms featuring both common and distinctive characteristics. In a manner analogous to cerebral palsy, each is a symptom complex highlighting a clinically recognizable entity that is etiologically heterogeneous, and which mandates a particular evaluation, management, and intervention approach. Both can be conceptualized as neurodevelopmental disabilities that can be defined as early-onset, chronic disorders that share the essential feature of a predominant disturbance in the acquisition of cognitive, motor, language, or social skills, which has a significant and continuing impact on the developmental progress of an individual [APA, 1994]. Diagnosis of these disorders occurs against a backdrop of a wide variation in “normality” of what can be highly individualized developmental trajectories that may not be smooth [Darrah et al., 2003]. Clear boundary lines may not be evident, mandating diagnosis over time rather than at a single point of clinical contact.

The consensus definition put forward in 2002 by the American Association on Mental Retardation (AAMR) defines mental retardation as “a disability characterized by a significant limitation both in intellectual functioning and in adaptive behavior as expressed in conceptual, social, practical, and adaptive skills.” This disability originates before the age of 18 [AAMR, 2002] and manifests with severe problems in the individual’s capacity to perform (i.e., impairment), ability to perform (i.e., activity limitations), and opportunity to function (i.e., participation restrictions) [WHO, 2001]. Correct and consistent application of this definition requires an awareness of five inherent contextual assumptions:

The present definition of mental retardation extends far beyond the traditional concept of a general subaverage level of intellectual function, as captured in a single numerical measurement such as an intelligence quotient (IQ). The recent emerging emphasis on functional behaviors and contextual factors is altering both the construct of disability and appropriate relevant terminology whereby the term “intellectual disability” is replacing “mental retardation.” The terms are essentially synonymous and both are used concurrently at present; however, it can be foreseen that “intellectual disability” will suplant “mental retardation” in the near future.

For the young child, the term global developmental delay has emerged to describe a disturbance across a variety of developmental domains [Batshaw and Shapiro, 1997; Fenichel, 2001; Kinsbourne and Graf, 2001; Majnemer and Shevell, 1995; Shevell et al., 2000; Simeonsson and Simeonsson, 2001]. Such a child has limitations or delay in the acquisition of developmental and functional skills that are both observable and measurable within the context of the natural progression of infants and young children [Shevell, 2008]. The latest consensus definition used by the American Academy of Neurology (AAN) practice parameter statement defines global developmental delay operationally as a significant delay in two or more developmental domains (e.g., gross/fine motor, cognitive, speech/language, personal/social, activities of daily living) [Shevell et al., 2003]. Typically, if there is delay in two domains, this implies delay across all domains.

Global developmental delay and intellectual disability are frequently diagnosed on the basis of experienced clinical judgment in advance of or as a substitute for detailed standardized assessments. Potential errors in measurements create a zone of uncertainty or range of confidence captured in the concept of the standard error of measurement (SEM) [Reschly, 1987]. This contributes to a conceptual hesitation regarding the application of strict numerical cutoffs. Accuracy is increased by repeated application of a particular measure over a longitudinal time interval [Batshaw, 1993]. The precise relationship between the diagnoses of mental retardation and global developmental delay must still be determined [Petersen et al., 1998; Yatchmink, 1996]; however, limitations and delay in the widespread acquisition of developmental skills, especially language, may be a harbinger of later intellectual disability.


Much information has been ascertained regarding the prevalence and causes of mental retardation; however, interpretation of these data necessitates an understanding of the assumptions inherent in these analyses. The operational parameters for mental retardation usually include an intelligence quotient (IQ) value that is 2 standard deviations below the mean (IQ below 70), identification before the age of 18, and difficulties in adaptive functioning. Although definitions are arbitrary, they allow for agreement on standardization of measurement. Assuming a normal distribution of IQ scores, approximately 2.25 percent of individuals will have an IQ below 70. Many of these individuals (given the high correlation between scores on IQ tests and standardized adaptive functioning tests) will also have adaptive difficulties, meeting the definition of mental retardation. Early population-based studies seem to have confirmed this point. Hagberg and Kyllerman [1983] documented a rate of mental retardation of 2 percent; 1.5 percent had mild mental retardation (IQ of 50–70), and 0.5 percent had moderate or severe mental retardation (IQ of less than 50). The population studied and the instruments used can influence the rate of mild mental retardation, but they seem to have little effect on the rate of severe mental retardation. The prevalence rates for mild mental retardation in 15 subsequent studies varied from 5 to 80 cases per 1000 people, with an average prevalence of 35 per 1000. In contrast, in 37 studies the prevalence of severe mental retardation varied only between 2.5 and 7 per 1000, with an average of 3.6 per 1000 [Leonard and Wen, 2002].

Because the highest percentage of children with mental retardation are in the mild range, any change between measured populations in mental retardation definitions or in the type or severity of deleterious environmental exposures (e.g., poor education, nutrition, environmental toxins, such as lead) can have a significant effect on the overall numbers of children with mild mental retardation. One of the more comprehensive investigations of mental retardation, the Metropolitan Atlanta Developmental Disabilities Study, identified a smaller than average rate for mild mental retardation (IQ of 50–70) of 8.4 per 1000, whereas that for severe mental retardation, at 3.6 per 1000, was consistent with the meta-analysis average [Boyle et al., 1996; Yeargin-Allsopp et al., 1997]. This study also documented two additional observations about the prevalence of mental retardation: race disparity and gender imbalance. The Atlanta study found that mental retardation was overrepresented by 50 percent among blacks relative to the white population [Yeargin-Allsopp et al., 1995], and that the male to female ratio was 1.4:1.0. A California study reported a similar rate of racial disparity for all categories of mental retardation and found that there was a substantial increased risk associated with low birth weight and maternal age [Croen et al., 2001]. Additional maternal influences observed in this study that increase the risk for mental retardation include poor nutritional status, tobacco smoking, and alcohol consumption. Paternal smoking also increases the risk of having affected children. In almost every study, gender disparity is a consistent finding, with males accounting for 20–40 percent more of the cases than females. A significant component of this difference may be related to X chromosome disorders, but studies have shown that maternal smoking or low birth weight has a more direct effect on IQ in male infants, suggesting that this disparity has many causes [Leonard and Wen, 2002].

Although there is considerable variation in the numbers of individuals diagnosed with mental retardation, there is less disagreement about the known causes of mental retardation and the high percentage of cases with unknown causes. For classification purposes, mental retardation cases are often grouped by prenatal, perinatal, and postnatal causes. The prenatal group includes known genetic syndromes and chromosomal disorders, central nervous system malformations, and toxic or infectious causes. Perinatal conditions include birth asphyxia, stroke, and infection. Postnatal conditions include infection, toxins (e.g., lead), and injury, such as nonaccidental trauma. Using this classification, the Metropolitan Atlanta Study found that 87 percent of children with mild mental retardation and 57 percent of moderately to severely affected children did not have identified causes. A similar breakdown – 77 percent for mild and 65 percent for severe cases – was found in the California study. Among the causes identified, chromosomal defects were the most common, accounting for 25 percent of all causes of mental retardation among more than 4.5 million live births in California; Down syndrome was the single most common known chromosomal cause. Other causes of retardation, including infection, central nervous system anomalies, and metabolic or endocrine causes, accounted for up to 8 percent of the total. In this analysis, low birth weight and other environmental exposures were not counted as biomedical causes, although they were associated with increased risk, as previously mentioned. Better focus on how these environmental influences affect ultimate IQ is essential for prevention and early intervention.

Many individuals with severe or mild mental retardation are unable to become productive members of society and require institutionalized or group-home care. The economic costs to society are substantial. In a Dutch study, mental retardation was the disease category with the largest health-care costs, almost equal to the economic impact of stroke, heart disease, and cancer combined [Meerding et al., 1998]. An analysis by the U.S. Centers for Disease Control and Prevention (CDC) estimates lifetime costs of more than $1 million dollars per person with mental retardation, which is more than that for cerebral palsy, hearing loss, or vision impairment [CDC, 2004].

History and Ethics

The first to draw a clear distinction between mental retardation and mental illness was the English philosopher John Locke, who wrote in his essay Concerning Human Understanding (1690), “Herein seems to lie the difference between idiots and madmen, that madmen put wrong ideas together and reason from them, but idiots make very few or no propositions and reason scarce at all.” The development of intelligence testing by Alfred Binet and Theodore Simon in 1905 ushered in an era of a more rigorous, scientific approach to defining mental retardation that became the basis for formulating an approach to its comprehensive management [Sherr and Ferriero, 2003].

Early 20th-century definitions of mental retardation focused on its incurability with permanent predetermined limitations and consistent prevention of participation in society (e.g., “mental deficiency is a state of incomplete mental development of such a kind and degree that the individual is incapable of adapting himself to the normal environment of his fellows in such a way to maintain existence independently of supervision, control or external support”; [Tredgold, 1937]). In the middle of the 20th century, adaptive behavior limitations were added as criteria for diagnosing mental retardation [Heber, 1961]. This addition was meant to implement a reduction in rigid reliance on IQ scores and to better reflect the social characteristics of the disability that is attached to mental retardation. Adaptive behavior deficiency was conceptualized as problems in maturation, learning, or social adjustment that were based on difficulties in adjusting to ordinary demands. By 1973, the AAMR definition of mental retardation required concurrent manifested deficits in general intellectual function and adaptive behavior [Grossman, 1973]. A 1992 AAMR definition specified the adaptive skill areas affected and emphasized limitations in current functioning across the individual’s life span [Luckasson et al., 1992]. As captured in the most recent effort of consensus definition, adaptive deficits lead to both activity limitations and participation restrictions that necessitate contextually driven systems of support to enable the individual’s fullest actualization of inherent potential and broadest societal integration [AAMR, 2002].

Evolution of the concept of mental retardation has been marked by an on-going debate regarding the nature of intelligence and its measurement. Despite the changing nuances of its definition, the primary criterion for diagnosing mental retardation over time has always been a deficit in intellectual ability [Gottfredson, 1997]. Factor analysis of data obtained from administering cognitive tests to large groups of individuals permitted the objectification of a unifactorial latent trait (i.e., general intelligence) to account for the variance between cognitive scores [Spearman, 1927]. Within this framework, intelligence was conceptualized as general mental capability that includes the ability to reason, solve problems, think abstractly, plan, and learn from experience [Carroll, 1997]. In essence, it is a reflection of a person’s ability to comprehend his or her surroundings. In contradistinction to the established unifactorial position of intelligence, several theoretical models of intelligence as a multidimensional construct have been put forward, with each intelligence component featuring distinctive developmental trajectories, problem-solving, and information-processing capacities [Gardner, 1993]. Validation by standardized and quantifiable measures of these intriguing and intuitively appealing constructs of intelligence remains elusive.

Historical controversy exists regarding the mechanisms by which intelligence can be measured and the social, cultural, and ethnic contexts of its measurement [Gould, 1981]. Establishing cutoff points for the labeling of “subaverage intellectual functioning” has also been problematic [MacMillan et al., 1995], especially with the single application of a particular measure and with scores that fall near the cutoff point within the range of the test SEM. IQ scores have also unfortunately been used at various historical points to further biologically determinist agendas and potentially to demonize minority groups [Gould, 1981].

A survey of the past century reveals a remarkable trajectory in the treatment of those affected by mental retardation or global developmental delay in Western society. With the prominence of eugenics (i.e., science of the improvement of the human race by better breeding) in the first part of the 20th century, those with mental retardation were targeted for involuntary eugenic sterilization in many jurisdictions [Kevles, 1985; Parent and Shevell, 1998; Proctor, 1988]. The global emphasis after World War II on individual civil rights resulted in national and international mandates that provided for legislative and judicial protection for the intellectually disabled from active discrimination (e.g., Americans with Disabilities Act of 1990 is but one national example). Early educational, rehabilitation, and school programs financed by public funds for those at risk of developmental disability or affected by mental handicap (e.g., Early Intervention Amendments to the Education of the Handicapped Act, Education for All Handicapped Children Act) have been implemented. These mandates for protection against discrimination and service provision exist with broad community and political support, providing a level of class protection not previously encountered. Legal standards have been upheld consistently by judicial authorities.

The medical care of individuals with global developmental delay and intellectual disability has been included in the thrust of broad medical principles. A common morality has emerged, framed by a generally accepted understanding of socially approved norms of human conduct [Shevell, 2009a]. Within this framework, ethical behavior is driven by mutually recognized duties and obligations. Within the medical sphere of human interactions, these duties and obligations focus on issues of autonomy, beneficence, non-maleficence, and justice [Bernat, 2002].

The absence of the capacity for competence (defined as “the ability to understand the context of the decision, the choices available, the likely outcomes of the varying choices, and to rationally process this information to make a decision”), together with minority age (typically less than 18 years), renders the pediatric patient with global developmental delay or intellectual disability doubly vulnerable with respect to ethical issues [Bernat, 2002]. The cornerstone of ethical modern medical practice is the respect for individual autonomy reflected in the primacy accorded to informed consent in all aspects of medical decision-making [Faden et al., 1986]. For those unable ever to provide informed consent, a “best interest” model for decision-making must be implemented that mandates careful selection of responsible proxy decision-makers and consideration of the risks and benefits for intervention from the unique perspective of the affected individual [Shevell, 1998]. Consensus exists that cognitive disability alone should never be the only reason to withhold or withdraw care. Developmentally appropriate models of assent that respect the cognitive capacity of the intellectually disabled offer an alternative to enhance our ethical efforts and are increasingly a feature of clinical practice. A determined effort must be made not to use the challenges faced in ethical practice with this population to abandon efforts through research to improve all aspects of care and outcome [Shevell, 2002].

Justice concerns itself with the fair distribution within society of what ultimately are limited resources in a broader socioeconomic context [Outka, 1974]. An appropriate standard would be objectively applied, equally valuing individual worth. Within the health-care sector, access frequently requires effective advocacy, which may favor better publicly organized and financially enabled disease advocacy groups.


Definitions and Testing

Accurate diagnosis of global developmental delay or intellectual disability is an essential precondition to proper management and service provision. Accurate diagnosis serves many functions, including understanding the specific associated medical and psychiatric complications, eligibility for service and support provision and its specific attributes, family counseling, and legal recognition of disability [Shevell, 2009b].

The diagnosis of intellectual disability, as presently defined, requires demonstration of significant limitations in intellectual functioning and adaptive behavior [AAMR, 2002]. Intelligence, conceptualized as general mental capabilities, is represented “objectively” by an IQ score obtained through proper application of an appropriate assessment measure [Hernstein and Murray, 1994]. Adaptive behavior refers to skills (i.e., conceptual, social, and practical) that a person learns in order to function within the context of the expectations and challenges of everyday life. Limitations in these skills affect the ability to respond to changes and demands encountered, affecting performance in daily life and participation in available opportunities. Multiple standardized, age-appropriate measures have been normed and validated on normal populations to assess adaptive behavior skills [Spreat, 1999]. For intellectual function and adaptive behavior, a significant limitation thought sufficient to trigger possible inclusion under the rubric of intellectual disability is performance at least 2 standard deviations below the mean for an appropriate test [AAMR, 2002]. Some of the standardized measures for the evaluation of intellectual, neurodevelopmental, and behavioral testing are summarized in Table 43-1, and a qualitative description of IQ and index scores on Wechsler tests that categorize intelligence levels is provided in Table 43-2.

Table 43-1 Measures for Evaluation of Intellectual, Neurodevelopmental, and Behavioral Progress

Test Name, Age Range Of Subjects, And Test Publication Data Test Description Administration and Scoring Information
Bayley Scales of Infant Development, Second Edition (Bayley II)
16 days to 3 years 6 months 15 days
Nancy Bayley, The Psychological Corporation, 1993, (Bayley III, 2005)
Standardized assessment of cognitive, motor, and behavioral development for children aged 1–42 months
Mental Scale has 178 items, Psychomotor Scale has 111 items, and Behavior Rating Scale has 30 items
Mental Scale yields a normalized standard score called the Mental Development Index, evaluating a variety of abilities, including sensory/perceptual acuities, learning, and problem-solving; verbal communication; abstract thinking; and mathematical concept formation
Motor Scale assesses skills of degree of body control, large muscle coordination, finer manipulatory skills of the hands and fingers, dynamic movement, dynamic praxis, postural imitation, and stereognosis
Behavior Rating Scale assesses the child’s relevant test-taking behaviors and measures the following factors: attention/arousal, orientation/ engagement, emotional regulation, and motor quality
Yields: Mental Development Index (MDI) and Psychomotor Development Index (PDI); and five behavior factors: Attention/Arousal, Orientation/Engagement, Emotional Regulation, Motor Quality, and Total Behavior Rating
MDI and PDI means = 100, SD = 15
Behavior ratings as percentile ranks:

Administration time: younger, about 30 minutes; older, about 60 minutes

Bayley Infant Neurodevelopmental Screener (BINS) 3 months to 24 months
Glen P Aylward, The Psychological Corporation, 1995, Screening test to identify infants who are developmentally delayed or have neurologic impairments
Four domains: Neurological Functions/Intactness (N), Receptive Functions (R), Expressive Functions (E), and Cognitive Processes (C)
The infant is administered 11 or 13 items and scored as optimal or non-optimal performance
Total scores can range from 0 to 11 or 13, with a higher score indicating better functioning
This screen can be used to prompt a more comprehensive evaluation Yields: A Total Score summing performance across the four domains
Chart shows age-appropriate cut-off values for high, moderate, or low risk for developmental delay or neurologic impairments
Administration time: 10 minutes Denver Developmental Screening Test II (Denver II) Birth to 6 years
WK Frankenburg and JB Dodds et al., Denver Developmental Materials, 1992, Surveillance and monitoring instrument to determine if a child’s development is within the normal range. Acquired skills are checked off on a chart, and results demonstrate in a graphic manner the child’s pattern of developmental skills compared with age-mates
It is not intended as a diagnostic tool and may be more appropriate as a developmental chart or inventory rather than as a screener
Four domains: Personal-Social, Language, Fine Motor-Adaptive, and Gross Motor; 125 items If a child fails items that are successfully completed by 90 percent of younger children, delay may be suspected, and further evaluation may be needed INTELLIGENCE OR COGNITIVE TESTS Wechsler Intelligence Scale for Children, Fourth Edition (WISC-IV) 6 years 0 months 0 days to 16 years 11 months 30 days
David Wechsler, The Psychological Corporation, 2003 (Spanish version, 2005), This test of intellectual or cognitive ability for school-age children is part of a series that includes the WPPSI-III (see below) for preschool and primary ages and the WAIS-III for adolescents and adults
It is the most commonly used intelligence test for school-age children
Ten core subtests and five optional subtests Yields: Verbal Comprehension Index (VCI), Perceptual Reasoning Index (PRI), Working Memory Index (WMI), Processing Speed Index (PSI), and Full Scale Intelligence Quotient (FSIQ)
Subtest means = 10, SD = 3
Index and IQ means = 100, SD = 15
Administration time: 65–80 minutes for core subtests Wechsler Preschool and Primary Scale of Intelligence, Third Edition (WPPSI-III) 2 years 6 months 0 days to 7 years 3 months 30 days

David Wechsler, The Psychological Corporation, 2002, Test of young children’s intellectual ability


Subtest means = 10, SD = 3
IQ, PSQ, GLC means = 100, SD = 15
Administration times:

Stanford-Binet Intelligence Scales, Fifth Edition (SB5) 2 years 0 months to 85 years+
Gale H Roid, Riverside Publishing, 2003, Assessment of intelligence and cognitive abilities
Two domains: Verbal (V) and Nonverbal (NV)
Each has five factors of cognitive ability: Fluid Reasoning (FR), Knowledge (KN), Quantitative Reasoning (QR), Visual-Spatial Processing (VS), and Working Memory (WM)
Use of Change-Sensitive Score (CSS) makes it possible to compare changes in an individual’s scores over time if retested Yields: ten subtest scores and four types of composite scores – five factor indices, two domain scales (Verbal IQ and Nonverbal IQ), Abbreviated Battery IQ, and the Full Scale IQ of all 10 subtests
Change-Sensitive Scores (CSS) are criterion-referenced (rather than norm-referenced) and therefore reference absolute levels of ability from the 2-year-old level to the adult level
Subtest score means = 10, SD = 3
Index and IQ means = 100, SD = 15
Change-Sensitive Scores range = 376-592, from 2-year-old level to adult (most fall in the range of 420–530); 500 is an average score for an individual 10 years 0 months old Administration time: 50–60 minutes for all ten subtests Differential Ability Scales (DAS) 2 years 6 months to 17 years 11 months

Colin D. Elliott, The Psychological Corporation, 1990 (DAS-II, 2006), Test of cognitive ability and of basic academic skills; includes nine cognitive and three achievement subtests
For language-impaired and non-English-speaking children, the examiner can obtain Special Nonverbal Composite Yields: Cognitive:

Academic Achievement: Basic Number Skills, Spelling, and Word Reading scores Subtest T-score means = 50, SD = 10
General Conceptual Ability (GCA), Cluster, and Academic Achievement means = 100, SD = 15
Administration time:

Leiter International Performance Scale, Revised (Leiter-R) 2 years 0 months to 20 years 11 months
Gale H Roid and Lucy J Miller, Stoelting Company, 1997, Totally nonverbal test of intelligence and cognitive abilities
Two domains: Visualization and Reasoning (VR) for measuring IQ, and Attention and Memory (AM)
There are 10 subtests in each domain
Includes four social-emotional rating scales (by Examiner, Parent, Self-Rating, and Teacher)
Nonverbal tasks, especially suited for children and adolescents who are nonverbal, do not speak English, or use English as Second Language (ESL); speech-hearing-motor impaired; attention-deficit hyperactivity disorder (ADHD), autistic; delayed; disadvantaged; or traumatic brain injury (TBI) Yields: 20 subtests
Five composite scores: Fluid Reasoning, Fundamental Visualization, Spatial Visualization, Attention, and Memory; and Brief IQ (based on four VR subtests) and Full Scale IQ scores
Growth-scale scores for subtests and IQ in a metric that can track growth over time if retested
Subtest score means = 10, SD = 3
Composite and IQ score means = 100, SD = 15
Administration time: VR, about 40 minutes; AM, about 35 minutes Comprehensive Test of Nonverbal Intelligence (CTONI) 6 years to 90 years
Donald D Hammill, Nils A Pearson, and J Lee Wiederholt, The Psychological Corporation, 1997, Test measures nonverbal reasoning abilities and estimates intelligence of individuals who experience undue difficulty in language (e.g., bilingual, language other than English, deaf) or fine motor skills. It can be administered orally in English or in pantomime. No oral responses, reading, writing, or object manipulation are required; responses are made by pointing to alternative choices
Analogical Reasoning, Categorical Classifications, and Sequential Reasoning in two different contexts: pictures of familiar objects and geometric designs; and solving problems
Six subtests: Pictorial Analogies, Pictorial Categories, Pictorial Sequences, Geometric Analogies, Geometric Categories, Geometric Sequences Yields: three composite scores – Nonverbal Intelligence Quotient, Pictorial Nonverbal Intelligence Quotient, and Geometric Intelligence Quotient – and an overall score
Subtest means = 10, SD = 3
Composite and overall means = 100, SD = 15
Administration time: 1 hour NEUROPSYCHOLOGICAL TESTS NEPSY (NE for neuro and PSY for psychology) 3 years 0 months to 12 years 11 months

Marit Korkman, Ursula Kirk, and Sally Kemp, The Psychological Corporation, 1997, (NEPSY-II, 2007) Test of neuropsychological development in preschool and early school-age children
Subtests can be used in various combinations, according to the needs of the child
Five domains: Attention and Executive Functions, Language, Sensorimotor Functions, Visuospatial Processing, and Memory and Learning

Yields: subtest and five domain scores
Subtest Scaled Score means = 10, SD = 3
Core Domain Score means = 100, SD = 15
Percentile ranks:

Administration times:

Delis–Kaplan Executive Function System (D-KEFS) 8 years to 89 years
Dean C Delis, Edith Kaplan, and Joel H Kramer, The Psychological Corporation, 2001, Test of executive function in older school-age children and adults evaluates component processes of tasks thought to be especially sensitive to frontal lobe dysfunction, such as flexibility of thinking, inhibition, problem-solving, planning, impulse control, concept formation, abstract thinking, and creativity in verbal and spatial modalities
Nine stand-alone tests Yields: nine test scores
Scaled score means = 10, SD = 3
Administration time: 90 minutes for all nine tests INDIRECT FUNCTIONAL RATINGS BY PARENT OR CAREGIVER Vineland Adaptive Behavior Scales, Second Edition (Vineland-II) 0 to 89 years (parents/caregivers)
3 to 21 years 11 months (teachers)
Sara S Sparrow, Domenic V Cicchetti, and David A Balla, American Guidance Service (AGS), 2004, A measure of personal and social skills as reported by the parent or caregiver
Four domains: Communication, Daily Living Skills, Socialization, and Motor Skills, with 11 subdomains
The Survey Interview, Expanded Interview, and Parent/Caregiver Rating Forms contain an optional maladaptive behavior domain for pinpointing undesirable behaviors that may interfere with adaptive functioning Yields: 11 subdomain scores, 4 domain scores, and an overall Adaptive Behavior Composite Subdomain score means = 10, SD = 3
Domain and Adaptive Behavior Composite score means = 100, SD = 15
Administration Time: Survey Interview and Parent/Caregiver Rating Forms, 20–60 minutes Infant Development Inventory Birth to 18 months
Harold Ireton, Behavior Science Systems, 1994, A measure of skills as reported by parent/caregiver
Five areas: Social, Self-Help, Gross Motor, Fine Motor, and Language
The child’s level of development in each area is determined by asking the parent, “What’s your baby doing?” and by observing the infant. The questionnaire can be completed by the parent or done as an interview
The child’s level of development in each area is compared with the child’s actual age Yields: a profile of development in the five areas compared with age norms of children from birth to 21 months
Administration time: 10 minutes Child Development Inventory 15 months to 6 years
Harold Ireton and Edward J. Thwing, Behavior Science Systems, 2005, A measure of skills as reported by the parent or caregiver
Nine scales: Social, Self-Help, Gross Motor, Fine Motor, Expressive Language, Language Comprehension, Letters, Numbers, and an overall General Development
On an inventory of the child’s observed developmental skills, parents answer “yes” (present or already acquired) or “no” (not yet)
There are 270 statements about the child’s behavior and 30 problem items describing various symptoms and behavior problems Yields: profile of development in the eight behavior domains and of the overall level of development compared with age norms of children between the ages of 1 year and 6.5 years
The age level assigned to each behavior item was defined as the age at which at least 75% of parents answered “yes” to that statement
Administration time: 30–50 minutes

(Courtesy of Rita J Jeremy, Ph.D., Developmental Psychologist, Department of Pediatrics, University of California, San Francisco, CA.)

Table 43-2 Qualitative Description of IQ and Index Scores on Wechsler Tests

Score Classification Percentage Included in Theoretical Normal Curve
130 and above Very superior 2.2
120–129 Superior 6.7
110–119 High average 16.1
90–109 Average 50.0
80–89 Low average 16.1
70–79 Borderline 6.7
69 and below Extremely low 2.2

Widely used measures for intelligence testing for children (5 to 16 years old) include the Wechsler Intelligence Scale for Children III (WISC-III) [Wechsler, 1991], the Cognitive Assessment System (CAS) [Naglieri and Das, 1997], and the Kaufman Assessment Battery for Children (K-ABC) [Kaufman and Kaufman, 1983]. For adults, the Wechsler Adult Intelligence Scale III (WAIS-III) is widely used [Wechsler, 1997]. The Stanford–Binet IV has applicability for children and adults [Thorndike et al., 1986]. The Wechsler Preschool and Primary Scale of Intelligence – Revised (WPPSI-R) [Wechsler, 1967] has been standardized for children as young as 3 years and has recognized limitations in interpretability [Sattler, 1982]. The standard for a systematic measurement of adaptive behavior from birth to adulthood has been the Vineland Adaptive Behavior Scale (VABS) [Sparrow et al., 1984], although acceptable alternatives exist, including the AAMR Adaptive Behavior Scale (ABS) [Lembert et al., 1993; Nihira et al., 1983], the Scales of Independent Behavior – Revised (SIB-R) [Bruininks et al., 1996], the Comprehensive Test of Adaptive Behavior – Revised (CTAB-R) [Adams, 1999], and the Adaptive Behavior Assessment System (ABAS) [Harrison and Oakland, 2000].

For an accurate diagnosis of global developmental delay, careful attention must be paid to its underlying concept and operational definition [Shevell, 1998, 2002, 2003, 2006]. A significant delay (i.e., greater than 2 standard deviations below the mean) needs to be demonstrated in two or more developmental domains exclusive of the qualitative impairment in language and social interaction that has been used to define an autistic spectrum disorder. Practically, delay in the child with global developmental delay is typically evident across all developmental domains. The practitioner should be aware of the general psychometric properties of testing instruments, their intended domains of evaluation and potential sources of error, and inherent standard error of measurement.

Standardized tests for assessment of infant, toddler, or preschool child development exist and include the Bayley Scales of Infant Development (2nd edition) [Bayley, 1993], the Battelle Developmental Inventory [Newborg et al., 1984], and the Denver Developmental Screening Test (2nd edition) [Frankenburg et al., 1992]. Frequently, rather than using a broad developmental instrument, domain-specific measures are individually applied and an overall assessment arrived at. Examples of domain-specific developmental measures include the following groups:

1. Motor profile: Alberta Infant Motor Scale (AIMS) [Piper and Darrah, 1994], the Peabody Developmental Motor Scales (PDMS) [Folio and Dubose, 1974], the Bruininks–Oseretsky Test of Motor Proficiency (BOTMP) [Bruininks, 1978].
2. Language skills: Peabody Picture Vocabulary Test – Revised (PPVT-R) [Dunn and Dunn, 1997], Expressive One Word Picture Vocabulary Test – Revised (EOWPVT-R) [Gardner, 1990], the Clinical Linguistic and Auditory Milestone Scales (CLAMS) [Capute and Accardo, 1991], the Clinical Evaluation of Language Function (CELF, 4th edition) [Semel et al., 2003], and the Slossen Intelligence Test (SIT) [Slossen, 1983].
3. Behavior and activities of daily living: Vineland Adaptive Behavior Scale (VABS) [Sparrow et al., 1984], Pediatric Evaluation of Disability Inventory (PEDI) [Haley et al., 1992], and the Pediatric Functional Independent Measure (WeeFIM) [Msall et al., 1994].

Often, diagnosis may be initially formulated or, less frequently, entirely based on clinical judgment [AAMR, 2002]. To be valid, such clinical judgment must be based on extensive direct experience with individuals with global developmental delay or intellectual disability. Typically, clinical judgment may be necessary because of various social, cultural, and linguistic contexts or because of unavailability, inappropriateness, or delay in the administration of standardized assessment procedures. Validation of clinical judgment is increased by direct observation, reliance on reliable third-party informants, input from an interdisciplinary team skilled in multidimensional standardized assessments, and repeated observations of an individual over time.

Advances in Diagnostic Testing

How does the clinician approach the diagnostic evaluation of a child with global developmental delay or intellectual disability? The AAN practice parameter and evidence report for global developmental delay [Shevell et al., 2003; Michelson et al., 2011] provide a framework for such an evaluation (Figure 43-1). The recommendations incorporate a combination of broad screening tools and disease-specific testing based on a heightened pretest probability, given identifying clinical features. Correctly applied, each has a reasonable pretest probability (>1 percent) of diagnosis. The current algorithm begins with a complete clinical assessment. For those patients in whom a specific diagnosis is considered, targeted testing is recommended, whether this be an MRI for an asymmetric physical examination or methylation testing for Angelman’s syndrome. For the remaining patients a step-wise approach is recommended that begins with array comparative genomic hybridization (CGH), followed by chromosomal karyotype if that is negative. This recommendation is an outgrowth of many recent studies showing the enhanced utility of array CGH to detect clinically relevant chromosomal changes, and is also recommended by a consortium of clinical genetics laboratories and clinicians [Miller et al., 2010; Michelson et al., 2011]. Fragile X is recommended to evaluate mildly affected children of both genders and MeCP2 the gene that causes Rett syndrome testing is considered important in severely affected females. If these tests are unrevealing, the algorithm recommends conducting head magnetic resonance imaging (MRI; with single proton spectroscopy, where that is available). If this approach is not diagnostic, comprehensive metabolic testing is then recommended (see Figure 43-1). This then diverges based on an a priori consideration of a specific diagnosis. As all these diagnostic tools advance (a discussion of some recent advances are expanded upon below), these algorithms will continue to change to reflect these technical improvements. Regardless, clinical judgment will still be tantamount.


Fig. 43-1 Algorithm for the evaluation of the child with unexplained global developmental delay or intellectual disability. A detailed history, a complete physical examination, psychoeducational testing, and screening tests for visual and hearing deficits are recommended for all children with GDD/ID. EEG is recommended when there is concern about seizures or an epileptic encephalopathy. In children with features suggesting a specific etiology, genetic testing, neuroimaging, and metabolic testing may be useful for confirmation. For children without features suggesting a specific etiology, testing can be done in a stepwise or parallel manner for genetic abnormalities, structural brain abnormalities, and metabolic abnormalities. Although an extensive list of metabolic tests is provided in this algorithm, there is insufficient evidence to make specific recommendations as to which testing sequence would have the greatest diagnostic yield. The algorithm is explained in greater detail in the Clinical Context section of this guideline. CGD = congenital disorder of glycosylation, CSF = cerebrospinal fluid, EEG = electroencephalogram, RBC = red blood cell, MRI = magnetic resonance imaging, MRS = magnetic resonance spectroscopy, VLCFA = very long chain fatty acids, XLID = X-linked intellectual disability. This algorithm is based on data contained in an evidence-based review on this topic [Michelson et al, 2011].

(Report of the Quality Standards Subcommittee of the American Academy of Neurology and the Practice Committee of the Child Neurology Society. Neurology 2011, in press.)

Genomic Microarray

Genomic microarray technology is an evolving platform that arrays a representation of the human chromosome on a glass slide and hybridizes fluorescently labeled control and patient DNA to detect copy number changes in the genome [Bejjani et al., 2005]. The first-generation chips had approximately 3000 elements spaced approximately every 1.0 Mb [Pinkel et al., 1998]. Newer-generation chips have been developed that completely cover the genome (i.e., tiling arrays) using more than 100,000-oligonucleotide spots [Ishkanian et al., 2004]. All of these platforms aim to identify interstitial copy number changes or copy number variants. Initially, this approach has been used very successfully to identify chromosomal regions (and genes) involved in cancer progression. Studies done thus far suggest that this approach will likely find many additional regions that are implicated in neurodevelopmental disabilities, even in the setting of prior normal high-resolution karyotyping [Harada et al., 2004; Shaw-Smith et al., 2004; Engels et al., 2007; Shevell et al., 2008]. Many studies in autism, mental retardation, and cohorts with multiple congenital anomalies (who also have neurodevelopmental disabilities) show that a few loci occur repetitively and at a much higher frequency in affected individuals than in controls [Berkel et al., 2010; Koolen et al., 2006; Sharp et al., 2008]. This type of large-scale analysis can provide clinicians with the necessary information to advise families about the significance of these genetic findings.

However, one challenge that arises from this whole-genome analysis is distinguishing normal variation from disease-causing changes, particularly in cases in which the genetic variation has never been reported previously in the literature, or has been reported but its incidence in cases and controls is not well documented. The first step is to establish whether a documented copy number change is de novo or familial; this is essential with respect to determining the possible pathogenicity of such changes [Speicher and Carter, 2005], although there is an increasing number of reported cases now in which inherited loci have been shown to correlate with disease presentation [Girirajan et al., 2010; Kumar et al., 2008; Shinawi et al., 2009; Weiss et al., 2008]. As we continue to make strides in genetics that move away from simple mendelian disorders, this concept of “partial penetrance” will grow in clinical relevance. The establishment of comprehensive, computerized, publicly available genotype–phenotype databases will aid in the process of establishing pathogenicity of these less common copy number changes in the future [Speicher and Higgins, 2007].

Advances in Imaging

High-quality MRI has significantly advanced the ability to detect many brain malformations. Certain studies suggest that MRI is useful for detecting abnormalities in up to 50 percent of children with developmental delay. In some cases, such as bifrontal polymicrogyria, lissencephaly, and bilateral periventricular nodular heterotopia, certain genes are known to cause these inherited syndromes [Gaitanis and Walsh, 2004]. For most cases, the genetic or other cause of the malformation has yet to be determined. Proton MR spectroscopy measures the resonance of molecules in the brain. Given the unique resonance frequencies of many molecules, their abundance can be noninvasively measured. This approach has been useful for detecting changes in cerebral lactate in mitochondrial disorders, and for observing the absence of creatine in the three disorders of creatine deficiency [Stromberger et al., 2003], another cause of potentially treatable nonsyndromic mental retardation [Schulze, 2003]. This technique also has been useful for the detection of abnormalities in succinic semialdehyde dehydrogenase deficiency [Ethofer et al., 2004], and for those changes that precede the changes on conventional T2-weighted MRI for X-linked adrenoleukodystrophy patients [Eichler et al., 2002].

Another mode of MRI that holds promise for refinement of diagnostics is diffusion tensor imaging (DTI). Predicated on the assumption that water molecules most likely diffuse along the trajectory of axons within the white matter, DTI can measure this water diffusion and, by employing certain algorithms, can approximate the position and direction of the major white matter tracts in the cerebrum. This approach should refine the imaging results for mental retardation or autism in which no changes can be identified by conventional MRI [Barnea-Goraly et al., 2004].

Proteomics in Disease Analysis

Practical and theoretical concerns limit exclusive reliance on genetic information to understand the causes of mental retardation (even including only those that are genetically based). Proteomics potentially offers a rapid means to screen individuals for many specific diseases within a general category, particularly when the proteins are expressed in the blood or other tissue that is readily available for analysis. For example, the true prevalence of mitochondrial dysfunction as a cause of mental retardation is unknown, but some studies suggest that mitochondrial abnormalities can be detected in a high percentage of patients with developmental delay, seizures, and hypotonia [Fillano et al., 2002; Marin-Garcia et al., 1999]. Mitochondrial genetics are too complex to be approached directly, because most patients are thought to have autosomal-recessive mitochondrial disease, and hundreds of genes are necessary for proper mitochondrial function. In cases of maternal inheritance, this problem can be addressed by full sequencing of the mitochondrial genome in muscle samples of affected individuals. Given that the mitochondrial genome contains only 16,500 base pairs (bp), this is a feasible goal in the short term. Investigators have begun to undertake whole mitochondrial proteomics approaches (i.e., combining immunocapture with mass spectrometry peptide fingerprinting) to determine the quantitative values for the specific polypeptides in these multiprotein complexes [Lib et al., 2003]. Missing subunits can be readily identified, and the post-translational modifications (e.g., phosphorylation) can be monitored [Schulenberg et al., 2004].


General Considerations

The known specific causes of mental retardation are too numerous to be listed here. The term mental retardation returns more than 1200 entries in the Online Mendelian Inheritance in Man site alone, and this catalogs only identifiable genetic causes. Mental retardation usually is classified by prenatal, perinatal, postnatal, and undetermined causes (Table 43-3). In most studies, the largest category of known primary causes is genetic or chromosomal [Leonard and Wen, 2002]. Many of the environmental causes, such as low birth weight and prematurity, are measured as risk factors and do not rise to the level of actual biomedical causes. Just as the prevalence varies, the percentage of mental retardation resulting from each group of causes varies by location and by definitions. In a California epidemiologic study of mental retardation, 75 percent of the known cases of mental retardation were due to chromosomal aberrations, whereas fewer than 3 percent were caused by any endocrine or metabolic abnormality [Croen et al., 2001]. In a Taiwanese study, 82 percent of all the chromosomal causes of mental retardation were a result of Down syndrome, but these data were collected before the clinical use of subtelomeric fluorescent in situ hybridization (FISH) probes [Hou et al., 1998]. In some regions of the world, cretinism from severe iodine deficiency occurs in up to 2–10 percent of the population of isolated communities [Delange et al., 2001]. Mild mental impairment from iodine deficiency occurs five times more frequently than cretinism, making iodine deficiency the most common preventable cause of mental retardation. Treatment during the first trimester has a significant effect on the frequency of cretinism [Cao et al., 1994]. In regions of mainland China with iodine deficiency, children score on average 10 IQ points less than cohorts in iodine-rich regions. This link between iodine deficiency and mental retardation also has a strong genetic component, because specific alleles for the deiodinase type II gene and the ApoE4 allele confer a significantly greater risk of mental retardation when the pregnant mother is iodine-deficient [Guo et al., 2004; Wang et al., 2000]. Despite the tremendous wealth of information about the causes of mental retardation, the cause remains unknown in most individuals. Genetic and epidemiologic approaches likely will continue to make progress toward unraveling and treating these currently unelucidated causes.

Table 43-3 Categories and Causes of Mental Retardation

Categories Causes
Prenatal Genetic


Unknown causes (most likely genetic but can be acquired)

Perinatal Birth asphyxia Infection (herpes simplex virus encephalitis or group B streptococcus meningitis) Stroke (embolic or hemorrhagic) Very low birth weight, extreme prematurity Metabolic (e.g. hypoglycemia, hyperbilirubinemia) Postnatal-environmental Toxins (e.g., lead) Infection (e.g., Haemophilus influenzae b meningitis, arbovirus encephalitis) Stroke Trauma (consider nonaccidental source) Poor nutrition Poverty Undetermined Familial   Nonfamilial

Genetic Causes

Some of the most significant advances in our understanding of the genetic causes of mental retardation have come from work addressing X-linked mental retardation, including mental retardation due to the fragile X syndrome or other genes.

Fragile X Syndrome

The fragile X syndrome, caused by inactivation of the FMR1 gene, has an estimated prevalence of 1 in 3000 males and is one of the most common causes of mental retardation [Crawford et al., 2001]. Expansion of the trinucleotide sequence CGG to more than 200 copies results in CpG methylation and inactivation of the FMR1 gene. Patients have narrow and elongated faces, large protruding ears, macro-orchidism, and joint hyperlaxity. Up to 20 percent have epilepsy, and most have complex partial seizures [Willemsen et al., 2004]. Carrier females and males with somatic mutations have various levels of intellectual impairment; studies demonstrate that the amount of the residual FMR protein detected in hair roots correlates well with IQ [Willemsen et al., 2003]. Research on the FMR protein has shown that it participates in the transport of mRNA and the regulation of protein translation in neuronal dendrites [Zalfa et al., 2003]. This function perhaps explains the paucity of dendrites seen in autopsy series of these patients [Greenough et al., 2001]. Although previous studies suggested that premutation male carriers (55–200 repeats) were asymptomatic, later work from Hagerman and colleagues [1988]

Buy Membership for Neurology Category to continue reading. Learn more here