Abstract
Many of the studies in this issue of Autism provide a humbling reminder of all that we don’t know about this set of disorders. Three of the studies examine issues that could be conceptualized as arousal disregulation, an area of increasing interest in attempting to understand the etiology and presentation of autism. Ozsivadjian and colleagues conducted a qualitative analysis of parents’ and children’s perspectives on the nature of anxiety in autism. They provide a rich contextual description from parents and children, which highlights the disabling nature of anxiety and the need to develop and test specific interventions. The study conducted by Lanni and colleagues (Verbal Ability, Social Stress, and Anxiety in Children with Autistic Disorder) suggests that we still need this kind of baseline conceptual work. They found that tests that are anxiety producing in typically developing children are not stressful or anxiety producing in children with autism. While a number of clinical and epidemiologic studies report that children with autism are more likely to be anxious, the present study provides evidence that we still have far to go to develop a good conceptual or empirical model of how to produce or measure that anxiety and associated stressors.
The third study examining arousal disregulation comprises the first longitudinal, population-based study of sleep problems in children with autism. Sivertsen and colleagues surveyed 3700 caregivers to measure both autism symptoms and sleep problems at two time points, finding more sleep problems among children with autism at baseline and more problems emerging over time relative to typically developing children in the cohort. While there are some measurement challenges associated with this study, the results still suggest the fluid nature of sleep problems and their tie to autism-specific symptoms.
Two studies in this issue highlight the challenge of finding tasks and measures that evoke a qualitatively different response in children with autism than in other groups. Schilbach and colleagues used a novel experimental paradigm to examine the effects of gaze cues on reaction times in children with autism. The authors report a difference in comparison group but not ASD group when the gaze cues are consonant or dissonant with the correct response. I might argue that the appropriate comparison would be the difference in the deltas, rather than showing that one difference was statistically significant and the other not, especially given that the magnitude of the differences was similar and in the same direction. This study perhaps provides evidence of another task that doesn’t discriminate between children with autism and children who are typically developing.
Similarly, Veness and colleagues studied early indicators of autism at 12 and 24 months of age. This carefully conducted longitudinal study prospectively followed a community sample to determine what differentiates children with autism from their typically developing peers and from children with other developmental disabilities. While many measures differentiated children with autism from their typically developing peers, few measures distinguished between autism and developmental disabilities or specific language impairments. One concern regarding the lack of difference is the lack of rigorous research diagnosis; a very real possibility, however, is that in a community (as opposed to clinically referred) sample, variables that show promise in lab settings lose their predictive power.
Two studies in this issue address broader issues related to the care of children with autism. Khanna and colleagues examined the psychometric properties of the Caregiver Strain Questionnaire among caregivers of children with autism. This measure has been well-validated among families of children with disruptive behavior disorders but has not been applied to families of children with autism. A growing body of research documents the considerable strain families of children with autism experience, but there is not consensus on the best ways to measure it. This study suggests that the Caregiver Strain Questionnaire is a promising and valid measure in this population, but may require further development to fully capture issues specific both to autism and to younger children.
Finally, Thomas and colleagues explore one of the oldest observed associations with autism, that of socio economic status. They find that wealth may be driving diagnostic practices, rather than be associated with autism per se. In wealthier school districts, autism was more commonly diagnosed, children received more evaluations, and children were diagnosed at a younger age. Hopefully this study will provide yet more incentive for focusing on improving community care and reducing disparities in diagnosis and treatment, rather than exploring potential etiological links between financial resources and autism.