VOICES FROM THE FIELD. This is the second in an occasional series featuring stories and perspectives from the world of severe autism.
By Joseph D. Buxbaum, PhD
It is an open secret in the research community that much of what we know about the neuroscience of autism is very biased in critical ways. Most obvious are the many types of patient-based studies that rely heavily on individuals who are the least severely affected. For example, any time one encounters a report about functional neuroimaging and autism, one should ask oneself whether the task in the study could readily be done by anyone on the spectrum, irrespective of severity. The answer is often no, which means that the study was carried out with individuals that represent a portion of the spectrum (and there is little or no reason to believe that this subgroup is representative of all of the spectrum). Widely read review papers that summarize what we know about the ‘neuroscience of autism,’ rely heavily on such studies, so these biases are pervasive in both specialty and non-specialty literature.
More broadly, for any report, one should also ask whether more severely affected individuals, with frequent, significant comorbidities, including cognitive and language delays, hyperactivity, attentional issues, anxiety, difficulties with new environments, and struggles with following instructions are likely to be equally represented in the study. Here too, in many such studies, the answer is a clear no. In fact, recent results on the enrollment of more severely affected individuals in clinical trials shows that the enrollment of such individuals is actually decreasing over time (Stedman et al 2018).
One unanticipated consequence of DSM-5 is that we have lost the split between autistic disorder and Asperger’s Disorder/PDD-NOS, which provided a relatively easy means of providing some (albeit imperfect) sense of the levels of severity across participants. Unfortunately, ICD-11 is following suit, which means that large national and epidemiological studies throughout the world will also lose this critical information. We encourage all groups to include metrics of severity in any ASD study, as a necessary part of the report and we would like to see journals insisting on such information.
“We encourage all groups to include metrics of severity in any ASD study, as a necessary part
of the report and we would like to see journals insisting on such information.”
The Seaver Autism Center has multiple studies that include those who are most severely affected. To give one example, we carry out detailed clinical and neurobiological assessments on individuals with rare genetic disorders with very high risk for autism. Necessarily, we take all comers, irrespective of the level of severity. This means that we have to adapt all protocols to ensure that we can include those with the greatest need. To give two examples, using a neurobiological approach with evoked potentials provides a means of gathering EEG data in just a few minutes (Siper et al 2016). In addition, we have EEG caps with many electrodes, as is typical, but we also have versions with just three electrodes, which can be placed and removed in moments and have a minimal tactile impact.
A second example is around sensory processing issues. Much of the description of sensory processing in autism comes from self-report. Think of how widespread is the appreciation of Temple Grandin's experiences. For those who may not be effective at speaking for themselves, we rely on parent and caregiver report, which could be confounded in many ways. Paige Siper and Teresa Tavassoli developed an assessment battery where they present an individual with auditory, visual and tactile stimuli, and rate the responses in each modality as being either normosensitive, hypersensitive, hyposensitive, or sensory seeking (Siper et al 2017). We are using this observational battery (called Sensory Assessment for Neurodevelopmental Disorders, or SAND) across the severity spectrum. The good news is that the results SAND correlate well with parent/caregiver report, however, in the case of using sensory processing as an outcome, we would want something that is observational and more objective, hence the critical need for measures like SAND
We at the Seaver Center wish we could do better, and continue to be frustrated by our limitations — and the limitations in the field — to include those who are most severely affected in all studies, including critical intervention studies. Sadly, there is no question that this issue has been an obvious one for decades and yet the research community has so far failed to address it adequately. It falls, therefore, to the advocacy community to speak for the most severely affected and to demand that the community does much better with research and treatment in this most vulnerable population.
“Sadly, there is no question that this issue has been an obvious one for decades and yet the research community
has so far failed to address it adequately.”
I am personally very excited and enthusiastic about the National Council for Severe Autism.
Joseph D. Buxbaum is the Director of the Seaver Autism Center for Research and Treatment at the Icahn School of Medicine at Mount Sinai, New York.