Distinguishing the Truth From Lies: Autism

Controversies in Neuroscience: Autism

Clinical Neuroscience

Controversies in Clinical Neuroscience: Autism Spectrum Disorders

Controversies in Clinical Neuroscience: Autism Spectrum Disorders

Although the U.S. Centers for Disease Control and Prevention (CDC, 2014a) and numerous medical organizations universally debunk the notion that vaccines contribute to the prevalence of autism, some sectors of the public refuse to let go of this belief and have even employed tactics designed to shut down opposing views ("Silencing debate," 2007). The emotionally-laced rhetoric infesting the debate over autism etiology, however, is a sign of the level of concern parents are increasingly expressing. This anxiety seems to be justified in part by recent data showing that 1 in 68 children, 8-years of age, suffer from autism spectrum disorder (ASD) (CDC, 2014b, p. 6). This means that close to 60,000 of the nearly 4 million children born each year within the United States (CDC, 2014c) will be diagnosed with ASD during childhood. To better understand this debate and other controversies surrounding ASD this essay will discuss symptoms, suspected and known etiology, recommended screening strategies, diagnostic criteria, and the interventions currently being utilized.

Autism Spectrum Disorders

Autism is not a single condition or syndrome, but an umbrella term for a variety of child neurological impairments with varying degrees of severity (Leonard et al., 2010). In 1980, the Diagnostic and Statistical Manual-III (DSM-III) employed the term 'spectrum' to describe the various conditions related to autism. Subsequent revisions included progressive developmental disorder not otherwise specified (PDD_NOS), childhood disintegrative disorder, Rett syndrome, and Asperger syndrome, which are all called autism spectrum disorder under DSM-V (ASD) (CDC, 2014b, p. 4).

The expansion of the definition of autism to include the various neurological conditions afflicting children and adults has contributed to rapidly increasing prevalence estimates for this disorder (Rice, 2011). Between 2002 and 2006, the number of children being diagnosed with ASD increased 57% according to the Autism and Developmental Disabilities Monitoring (ADDM) Network, but the reason(s) for the increase remains largely unknown. A portion of the increase can be explained by greater clinician awareness, better screening methods, and the inclusion of the milder forms of ASD.

ASD Etiology

The suspected causes of ASD are quite numerous and include both heritable and environmental toxins, but all causes lead to neurobiological deficits (Gadad, Hewitson, Young, & German, 2013; Roberts et al., 2013). The vaccine preservative, thimerisol, has been eliminated as a potential etiological factor because no change in autism prevalence rates was associated with its elimination from child vaccines at the end of the 20th century (CDC, 2014a). Mental retardation and epilepsy are frequent ASD comorbidities, possibly reaching 30 and 44%, respectively, which suggests overlapping or common etiologies (Gadad, Hewitson, Young, & German, 2013). In addition, some researchers have suggested that distinct etiologies exist for early and late onset ASD.

The suspected brain areas involved in ASD are the cerebellum, cerebrum, brain stem, and amygdala, but Gadad and colleagues (2013) cautioned that the quality of research in this area suffers from small sample sizes and a lack of scientific rigor. For example, little is known about contributions from brain areas that have yet to be investigated or whether symptoms are related to localized or diffuse abnormalities. The overall impression created by the research literature is that ASD is not a collection of diseases with a common etiology, but a collection of overlapping symptoms with a variety of causes. This would explain why researchers have found variable reductions of cerebellar Purkinje cell size and number among ASD brains (Gadad, Hewitson, Young, & German, 2013, p. 2-3). Accordingly, the inhibitory GABAergic neurotransmitter system would be affected in some ASD children. This possibility is supported in animal models for ASD where the activity of candidate genes, such as Mecp2, TSC1, TSC2, and Fmr1, has been genetically altered. Not only does altered gene function affect the GABAergic neurotransmitter system, but ASD-like symptoms are also induced. The other neurotransmitter systems believed to be involved in ASD are serotonin, dopamine, and glutamate.

Autism Diagnostic Criteria and Recommended Screening Guidelines

The primary ASD diagnostic criterion, according to the latest version of the DSM, is social communication and interaction deficits that are not situational-dependent and which persist over time (CDC, 2014d). More specifically, a child with ASD will have difficulty mirroring emotional and intellectual content within social situations, communicating verbally and non-verbally, fostering relationships with family and friends, and engaging in normative social interactions with other children. Additional criteria are provided for evaluating ASD severity, including the presence of repetitive behaviors and interests, inflexibility, hyper- or hypo-reactivity to sensory stimulation, age of onset, level of impairment in normative daily activities, and intellectual impairment.

The diagnostic criteria for ASD are less controversial than whether children should be screened for ASD. The American Academy of Pediatrics recommends routine screening of children for ASD symptoms at 9, 18, 24, and 30 months of age, with the hope of an early diagnosis, intervention, and improved outcome (Lipkin & Hyman, 2011). These recommendations were originally based on the sensitivity and predictive value of the Modified Checklist for Autism in Toddlers (M-CHAT). Since the M-CHAT was originally developed, several ASD screening tools have been developed with greater sensitivity and accuracy. However, Campos-Outcalt (2011) argues instead that the strength of the scientific evidence fueling these recommendations does not meet the standards typically required for mandating screening or any other intervention. Such standards are typically based on positive systematic reviews conducted by disinterested researchers, but in 2011 no such publication existed.

A recent systematic review of ASD screening methodologies revealed a lack of well-controlled studies showing ASD screening reduced the age of diagnosis (Daniels, Halladay, Shih, Elder, & Dawson, 2014), which would support Campos-Outcalt's (2011) position; however, this argument is flawed. For example, the risk magnitude for ischemic heart disease has been used in the past to justify the use of statin drugs, with moderate and severe risk patients being treated, but a recent randomized, controlled trial revealed this simple risk model is sorely inadequate and statin use could benefit low-risk patients as well (Ridker & Wilson, 2013). The best practice recommendations for statin drugs were formulated long before a systematic review could be published, but the lives of countless patients were improved and lengthened as a result. As long as the ASD screening methods do not pose a significant risk of harm to children and there is a chance of an early diagnosis and treatment, then it makes no sense to delay universal screening until a positive finding is revealed by a systematic review.

Treatment Approaches

Reichow and colleagues (2012) performed a systematic review of the research literature for early intensive behavioral interventions (EIBIs) and came to the conclusion that a number of studies have published encouraging data, but the strength of the findings were undermined by a lack of randomization. Only one randomized, controlled trial could be identified, while the other four studies included in the review were clinical controlled trials (CCTs). The four CCTs were subjected to a meta-analysis, thereby creating a sample containing 203 subjects. Significant improvements (p < .05) were found for adaptive behaviors, IQ, expressive language, receptive language, daily communication skills, socialization, and daily living skills. The mean effect sizes were all above g = .42, which suggests EIBIs provide a significant benefit.