© Springer International Publishing Switzerland 2016
Johnny L. Matson (ed.)Handbook of Assessment and Diagnosis of Autism Spectrum DisorderAutism and Child Psychopathology Series10.1007/978-3-319-27171-2_99. Diagnosing ASD in Very Early Childhood
(1)
Louisiana State University, Baton Rouge, LA, USA
Keywords
AutismEarly childhoodEarly identificationYoung childrenInfantsToddlerDiagnosisEvaluationDifferential diagnosisDiagnosing ASD in Very Early Childhood
Autism spectrum disorder (ASD) is a common, neurodevelopmental disorder defined by deficits in social and communication skills as well as the presence of restricted, repetitive, and stereotyped behaviors, interests, or activities (American Psychiatric Association [APA], 2013). According to the most recent Autism and Developmental Disabilities Monitoring (ADDM) Network study, the current prevalence of ASD is estimated at 1 in 68 children in the USA (Centers for Disease Control and Prevention [CDC], 2014). The Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5) criteria, frequently used in the diagnosis of autism, conceptualizes ASD based on the presentation of symptoms in two domains: social communication and social interaction, and restricted, repetitive patterns of behaviors, interests, or activities. Items within the social communication and social interaction domain include (1) deficits in social-emotional reciprocity, (2) deficits in nonverbal communication, and (3) deficits in forming, maintaining, and understanding social relationships. Individuals being evaluated for ASD must evince all three items to be considered for diagnosis. The restricted, repetitive patterns of behavior, interests, or activities domain includes (1) stereotyped and repetitive language use, motor movements, or object manipulations; (2) strict adherence to routines, ritualized patterns of verbal or nonverbal behavior, and/or resistance to change; (3) restricted, fixated interests that are atypical in intensity or focus; and (4) aberrations in sensory reactivity or abnormal interest in sensory input from the environment. Diagnosis requires evidence of at least two of these four symptoms. Symptoms from both domains must be present early in life (APA, 2013).
Currently, formal diagnosis of ASD is rarely made before 3 or 4 years of age in the USA. However, researchers have shown that most parents recognize signs and symptoms of ASD in their children during the first or second year of life (De Giacomo & Fombonne, 1998; Goin-Kochel, Mackintosh, & Myers, 2006; Landa & Garrett-Mayer, 2006; Rogers, 2009; Wetherby et al., 2004; Zwaigenbaum et al., 2007). Research findings indicate that behavioral markers of ASD are apparent before 2 years of age and that diagnosis at 2 years old can be made reliably. ASD diagnoses made in children at 2 years old are stable and have been found to persist a year later in 90 % of cases studied (Lord, 1995; Wetherby, Watt, Morgan, & Shumway, 2007; Zwaigenbaum et al., 2007).
Although nearly a decade of research supports the reliability of identifying ASD at 2 years of age, there are several factors that contribute to the lag in formal diagnosis we are experiencing. The first involves a lack of knowledge and experience in both parents and professionals. Parents may be limited in their knowledge about typical development in young childhood, particularly if the child experiencing developmental delays or atypicalities is their firstborn. Parents may also dismiss any developmental concerns as problems their child will “grow out of” with time (De Giacomo & Fombonne, 1998). This outlook would be responsible for delaying professional consultation . De Giacomo and Fombonne (1998) found that the mean age of parental concern for children who are later diagnosed with ASD was 19 months but that it took approximately 5 months after that concern for parents to seek professional advice.
Further, pediatricians and general practitioners are typically the first professionals to be consulted with these developmental concerns (De Giacomo & Fombonne, 1998; Wetherby et al., 2004). These primary care providers may not have the expertise necessary to distinguish parental concerns as emerging ASD symptoms (De Giacomo & Fombonne, 1998; Goin-Kochel et al., 2006). For instance, Shah (2001) studied medical students’ understanding about various aspects of ASD using a ten-item questionnaire addressing diagnostic criteria, causes, symptomology, treatment, and outcome. Results indicated that fourth-year students averaged fewer than half correct responses on the questionnaire. Lack of information about autism may contribute to professional hesitation in addressing concerns about child development (Goin-Kochel et al., 2006; Shah, 2001).
Another obstacle for establishing earlier diagnosis is the heterogeneity in symptom presentation and symptom onset in children with ASD (Zwaigenbaum, 2010). Individuals with ASD vary in severity of symptom presentation; those children with a milder presentation often are not diagnosed until later in life when social demands increase and their deficits become more pronounced (Goin-Kochel, 2006; White, Keonig, & Scahill, 2007). Using the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, Text Revised (DSM-IV-TR) autism subtypes, Goin-Kochel and colleagues (2006) found that mean age of ASD diagnosis differed according to diagnosis; while the mean age of autistic disorder diagnosis was 3.4 years, children with PDD-NOS and Asperger syndrome were diagnosed significantly later (M = 4.2 and 7.5 years, respectively). Variability of onset and presentation creates complexities in the early detection of ASD. Finally, the existing ASD diagnostic criteria may not be appropriate when diagnosing very young children, and criteria specifically for infants and toddlers has not been established (Goin-Kochel et al., 2006; Zwaigenbaum et al., 2007). The behavioral repertoire of infants and toddlers who are later diagnosed with ASD may be too limited to evaluate for the symptoms illustrated in the diagnostic criteria of the DSM (De Giacomo & Fombonne, 1998).
Goin-Kochel and colleagues (2006) reported that over 40 % of parents of children with ASD were unsatisfied with the diagnostic process. Delays in diagnosis have been found to increase parental distress and coping difficulties. Delayed diagnosis would also postpone child enrollment into early intervention services (De Giacomo & Fombonne, 1998; Goin-Kochel et al., 2006). This is particularly problematic, as researchers have shown that early intervention has a greater benefit when provided before 3.5 years of age compared to after age 5. Early intervention beginning before 4 years of age has been associated with greater language, social, and cognitive improvements (De Giacomo & Fombonne, 1998; Wetherby et al., 2004). Working to improve the diagnostic process is essential for bettering the prognosis for both children with ASD and their families.
Studying ASD in Young Childhood
In order to improve early identification and diagnosis of autism in young childhood, we must first understand autism onset and presentation in this age population. Researchers and clinicians have employed several different retrospective and prospective strategies to study the emergence of ASD in infants and toddlers. It is from this line of research where early diagnostic measures, practices, and procedures arise. In this section, these strategies are discussed.
Retrospective Studies
Parent Report
Obtaining information regarding early autism emergence via parent report is a common method used in research and clinical practice. Retrospective parent report is efficient in collecting early history. Parents and caretakers observe their child’s behavior across time and across various settings allowing for a wide range of information to be gathered (Ozonoff, Heung, Byrd, Hansen, & Hertz-Picciotto, 2008; Zwaigenbaum et al., 2007).
Several limitations to studying ASD onset and early symptomology by parent report exist. First, parent report is prone to errors of memory and affected by the passage of time (Ozonoff et al., 2008; Rogers, 2009; Zwaigenbaum et al., 2007). The level of parental knowledge about typical child development also affects parent report. Parent ability to recall and recognize subtle social and communication atypicalities may be limited compared to clinicians or researchers specialized in ASD (Rogers, 2009; Zwaigenbaum et al., 2007). Recognition of symptom onset has been found to be particularly difficult for parents (Ozonoff et al., 2008). Parent report is subject to bias caused by knowledge of their child’s later diagnosis; parents may report more early behaviors and symptoms consistent with the diagnosis (Ozonoff et al., 2008; Zwaigenbaum et al., 2007). Last, assessors conducting parent report interviews must be trained to ask questions that are clear and full but are not leading. They must also be trained to provide probes when necessary, as parents often do not conceptualize symptoms in the same way as professionals in the field (Ozonoff et al., 2008).
Despite the limitations of parent report, studies have shown moderate consistency between the behaviors and symptoms parents report and those seen on videotape footage. Therefore, parent report is a relatively valid method of studying early onset and symptomology of ASD (Ozonoff et al., 2008).
Home Videos
Studying home videos offered the first opportunity for objective examination of the early behaviors exhibited by children who were later diagnosed with ASD (Rogers, 2009; Saint-Georges et al., 2010; Zwaigenbaum et al., 2007). Compared to parent report methods, studying home videos allows for more objective assessments by trained and unbiased viewers of early behaviors in children with ASD (Zwaigenbaum et al., 2007).
Several concerns regarding the representativeness and standardization process in the home video method have been raised (Ozonoff et al., 2008; Rogers, 2009; Saint-Georges et al., 2010; Zwaigenbaum et al., 2007). Rather than collecting random samples of behavior, home videos are filmed for particular reasons in specific settings and may not represent behavior across all settings (Rogers, 2009). Further, it is possible that parents stop filming when children behave inappropriately or undesirably (Zwaigenbaum et al., 2007). Because of this, collecting data regarding ASD-specific behaviors (e.g., interaction patterns, object exploration, sensory response) may require additional parent report methods. Another issue involves the variability across families in the amount, content, and quality of home videos. Some families do not videotape their children at all, raising concerns of the generalizability of findings across the ASD population. Because home videos are made to preserve family memories, each family will have varying activities filmed in varying settings. This variability makes standardization difficult and time intensive (Ozonoff et al., 2008; Zwaigenbaum et al., 2007).
Despite limitations, the home video method has led to significant findings regarding the early distinguishability of ASD symptoms in infants (Rogers, 2009). Further, results from recent prospective studies have supported the findings of home video research. Therefore, home video analysis is viewed as a valid and reliable method of studying ASD emergence in infancy and toddlerhood (Saint-Georges et al., 2010).
Prospective Studies
Prospective studies test specific hypotheses utilizing experimental methods. These studies are designed to explore particular behavioral or biological constructs in a standardized manner (Zwaigenbaum, 2010). Multiple assessments are conducted across a long period of time on the same sample of infants and toddlers. This often involves frequent naturalistic assessments of participant development and less frequent “landmark” evaluations using standardized measures of language, cognition, and adaptive functioning. The developmental progression of high-risk (e.g., siblings of children with ASD) and low-risk children is compared. An endpoint of at least 3 years old is most often employed (Landa & Garrett-Mayer, 2006).
There are several advantages of employing a prospective design . Prospective studies are not subject to the same biases seen in parent report and home video research (Landa & Garrett-Mayer, 2006; Ozonoff et al., 2008). Prospective methods provide uniform data points and methods of data collection across participants. The longitudinal nature of prospective designs also allows for improvement in the understanding of developmental trajectories of individuals with ASD (Landa & Garrett-Mayer, 2006).
Several populations have been used to study ASD onset and symptom emergence prospectively. Among these populations are children with a sibling with ASD, children who fail population screeners, and children who have specific medical or genetic diagnoses that often co-occur with ASD (Landa & Garrett-Mayer, 2006). Using siblings of children with ASD in prospective studies offers a great deal of feasibility (Landa & Garrett-Mayer, 2006; Zwaigenbaum, 2010). ASD has among the highest recurrence risk of all neuropsychiatric disorders; the risk of having a child who will develop ASD when one child in the family has ASD is as high as 8 % and 35 % when the family has two children with ASD (Landa & Garrett-Mayer, 2006). Because of this, prospective studies on high-risk siblings can begin as early as prenatal periods offering unique opportunities to study neurobiological underpinnings of ASD (Landa & Garrett-Mayer, 2006; Zwaigenbaum, 2010). Studying siblings is not without limitations, however. The characteristics of the older sibling with ASD (e.g., severity level) may influence the likelihood of participation. Further, generalizability of results may be limited due to possible genetic differences in single-incidence versus multiple-incidence families (Landa & Garrett-Mayer, 2006; Zwaigenbaum, 2010).
Utilizing children who screen positive for delays in development offers the opportunity to study children with no family history of ASD; however, these studies rarely begin before the first year of life when delays cannot be reliably identified (Zwaigenbaum, 2010). Sampling biases may occur when using this high-risk sample because parents with concerns may be more likely to participate. Further, data collection beginning after the first screening restricts the age range that can be studied (Landa & Garrett-Mayer, 2006). Young children at a heightened risk for ASD due to a preestablished medical or genetic disorder may also be used in prospective studies. Among these disorders are fragile X syndrome, chromosomal aberrations, and tuberous sclerosis. However, these medical conditions are rare and are associated with unique presentation differences in ASD, making this population difficult to study and results difficult to generalize (Landa & Garrett-Mayer, 2006).
Weaknesses of prospective designs can be found in both the age of enrollment and the age of endpoint. Results may vary based on age of enrollment. If participants are enrolled at or after 1 year of age, there is a heightened risk of sampling bias because parents may be observing and may be concerned about behaviors consistent with ASD (Rogers, 2009). Variation may also stem from the determined endpoint of the study. Studies using a later endpoint (e.g., 60 months) may have higher rates of ASD compared to those using earlier endpoints. This would result from misclassification into the “typical development” group of children who have milder presentations of ASD such as those individuals who would have been classified as having PDD-NOS or Asperger syndrome in the DSM-IV-TR (Landa & Garrett-Mayer, 2006; Rogers, 2009). These milder cases of ASD are often diagnosed later when social demands increase and atypicalities become more pronounced.
Despite these weaknesses , prospective studies of high-risk infants offer increased objectivity in studying infants and toddlers with ASD. Prospective studies also present a new opportunity to study behavioral and biological markers of ASD and have helped to improve the understanding of the neurobiology of ASD, the available early detection measures, and hold implications for early intervention.
ASD Symptom Emergence
Results from retrospective and prospective studies indicate that the most common onset pattern noted in children with ASD is gradual, where parents become increasingly aware of certain atypical symptoms or lack of progression through developmental milestones during the child’s first 2 years of life (Werner, Dawson, Munson, & Osterling, 2005). Though a third of parents recognize atypicalities before their child’s first birthday, most parental concerns begin between the first and second year of their child’s life (De Giacomo & Fombonne, 1998). A later diagnosis of ASD is associated with increasing concerns from 12 to 18 months of age, but not concerns at 6 months of age (Zwaigenbaum, 2010). Significant differences between infants and toddlers who are typically developing, developmentally delayed, and who later receive a diagnosis of ASD have been found at 13–15 months for socialization deficits , 16–18 months for repetitive behaviors, and 19–21 months for communication symptoms (Wetherby et al., 2007). This section describes specific symptoms across several developmental domains that characterize infants and toddlers who are later diagnosed with ASD.
Communication
Speech and language problems are the most common first concern for parents of children who are later diagnosed with ASD (De Giacomo & Fombonne, 1998). Researchers have shown that deficits in both verbal and nonverbal communication begin at 12 months of age in infants who develop ASD (Rogers, 2009). Verbal communication deficits can present as delayed development of language, abnormal use of language, or a complete lack of verbal language (Inglese & Elder, 2009). Although considered a part of typical language development, echolalia, or repeating words or phrases previously heard, is often observed beyond normal time limits in young children with ASD. Abnormal voice intonation when speaking is also seen (Chawarska, Klin, Paul, & Volkmar, 2007; Rogers, 2009). Infants and toddlers with ASD have difficulty expressing their wants, display more stereotyped vocalizations, produce less vocalizations directed towards other people, and have a reduced number of words and sentences compared to their peers. Receptively, young children with ASD present with deficits in following directions and have an inconsistent response to their name being called (Saint-Georges et al., 2010; Wetherby et al., 2004; Zwaigenbaum, 2010). In regard to nonverbal communication, infants and toddlers who are diagnosed with ASD engage in less or no communicative gestures (e.g., pointing; Saint-Georges et al., 2010; Wetherby et al., 2004).
Socialization
In cases of early ASD onset, social symptoms can be observed from several months after birth. Infants who are later diagnosed with ASD often do not assume an anticipatory posture before being held by an adult. Infants with ASD may show little social interest and a lack of responsiveness to social stimuli. This population shows deficits in interaction, and higher nonsocial attention, and demonstrates less appropriate emotions and facial expressions. Behaviors predictive of a later diagnosis of ASD in 12-month-old children include deficits in eye gaze, little social smiling, and delays in play skills and imitation (Wetherby et al., 2004; Zwaigenbaum et al., 2007; Zwaigenbaum, 2010). Joint attention, a behavior typically developed by age 12 months, is rarely observed in infants who develop ASD (Saint-Georges et al., 2010; Zwaigenbaum et al., 2007). Young children with ASD rarely share their interests and enjoyments with caretakers and show a lack of interest in their peers (Saint-Georges et al., 2010; Wetherby et al., 2004). Researchers have found that children who later receive an ASD diagnosis are less likely to attend to an adult feigning distress (Zwaigenbaum et al., 2007). By 2 years old, social symptoms become more pronounced. At 2 years of age, children with ASD interact poorly with other people, spend less time looking at others, show poor eye contact, demonstrate deficits in imitation, and express less positive affect (Saint-Georges et al., 2010; Werner et al., 2005).
Restricted and Repetitive Behaviors
Restricted and repetitive behaviors are rarely seen before 12 months of age. In fact, several researchers have found that stereotypies and self-stimulatory behaviors in children who later develop ASD did not significantly differ from typically developing peers in the first year of life (Saint-Georges et al., 2010; Wetherby et al., 2004). By 12–18 months of age however, young children with ASD display aberrations in toy play such as less functional use and atypical exploration (e.g., spinning, rotating, repetitive actions, unusual visual regard; Rogers, 2009; Zwaigenbaum, 2010). Atypical reactions to sensory input may also be present in this age group (Zwaigenbaum, 2010). Though repetitive motor movements may be less prevalent in infant and toddlerhood, these behaviors may also be more difficult to differentiate from age-appropriate movements observed in typically developing infants. When assessing an infant for ASD, Zwaigenbaum and colleagues (2007) suggest looking at not only the type of repetitive behaviors present, but also the persistence, quality, frequency, and contexts under which the behavior occurs.
Other
Beyond the deficits observed within the core domains of autism, ASD has been found to affect several areas of development early in life (Zwaigenbaum, 2010). These associated symptoms include impairments in attention regulation, cognitive deficits, and hypoactivity (Saint-Georges et al., 2010; Zwaigenbaum, 2010). Temperamental difficulties are also seen in young children who later receive an ASD diagnosis (Rogers, 2009; Saint-Georges et al., 2010). Significant differences in motor development have been observed in young children with ASD. Researchers have found that infants and toddlers who develop ASD show signs of hypotonia as well as deficits in fine and gross motor skills in the first 2 years of life (Landa & Garrett-Mayer, 2006; Rogers, 2009; Saint-Georges et al., 2010).
Variability in Onset
From its first description (Kanner, 1943), autism has been conceptualized as a disorder present from birth; however, results from several longitudinal studies have shown that the course and presentation of ASD can vary tremendously between individuals, or even within the same individual over time (Saint-Georges et al., 2010; Volkmar, State, & Klin, 2009). Though most young children who are diagnosed with ASD present with the early symptom emergence discussed above, researchers have indicated that around 30 % of children experience late onset of ASD (Ozonoff et al., 2008). Of particular interest in this onset variability research is autistic regression . Regression can be loosely defined as a loss of previously acquired skills before age 3 years (Kalb, Law, Landa, & Law, 2010; Ozonoff et al., 2008). While some researchers require a loss of language skills to characterize regression, a child can experience regression over several different developmental domains (i.e., social, motor, and potentially cognitive skills). Researchers have found that approximately half of children who experience regression have both social and language skill losses, and up to one-third of children experience social losses only (e.g., loss of eye-to-eye gaze, decreased response to name; Kalb et al., 2010; Ozonoff et al., 2008).
The traditional conceptualization of regression involves a loss of skills following typical development. At 1 year old, children who experience this onset pattern have been found not to differ from typically developing peers in social-communication behaviors (e.g., eye gaze, social smiling, responding to name, vocalizing to others, points). However, by 2 years of age, these children show no differences from early-onset ASD cases in pointing, social gaze, response to name, and language development (Saint-Georges et al., 2010).
While it was previously thought that autism onset occurred either early or as part of a regression from typical development , recent research indicates that there may be several different onset patterns that are more prevalent than traditional regression (Ozonoff et al., 2008). For example, several young children with ASD experience typical developmental milestone progression before a “developmental plateau.” These children may display intact early social development and mild nonspecific delays until 2 years of age when progression stops. These children may exhibit mutual attention, emotional reactions, and social interest in games like peek-a-boo the same as children who develop typically. Differentiation in skills, however, becomes clear around 24 months of age (Saint-Georges et al., 2010; Tager-Flusberg, 2010). Some researchers have theorized that this trend represents an inability for children with ASD to transform basic social behaviors into more complex and multifaceted skills necessary for interpersonal competency (Kalb et al., 2010; Ozonoff et al., 2008; Saint-Georges et al., 2010). Another onset pattern identified in the research may actually be the most common of the late-onset patterns. A majority of children with regression demonstrate subtle developmental atypicalities prior to the loss of skills. Several researchers have characterized this pattern as a mixed onset showing both delays and losses (Ozonoff et al., 2008).
Differences in presentation and prognosis between children who experience varying onset patterns are unclear. Some researchers have suggested that children with regression have worsened communication, social, and behavioral outcomes compared to those with early-onset ASD. However, several researchers have found no significant differences in presentation or prognosis between individuals with early and late onsets (Kalb et al., 2010; Werner et al., 2005). What is clear is that ASD symptom emergence can occur as part of a later onset pattern; thus, conducting multiple assessments over the span of infant and toddlerhood is imperative for the early identification and diagnosis of ASD in young children (Kalb et al., 2010; Ozonoff et al., 2008; Rogers, 2009).
Related posts:
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
