This Today.com story addresses an issue of desperate concern to parents of autistic children: the possibility of spotting signs of autism earlier than age two or three, which is the development stage at which autism is typically identified.
The story did a good job explaining the study’s design and how the benefits were measured. Using a number of sources to provide perspective, it also discussed the need for more research to better refine the test.
That said, the story would have been stronger if it had discussed the “big picture” of early markers for autism, and where this latest research fits in with that. It also needed more on the harms of a screening test like this.
Parents of children with autism may find hope in the possibility that progress is being made toward identifying early markers of autism. This makes it important that stories about this issue don’t over promise, and this one doesn’t. The story is careful to qualify the implications of the research results, and clearly delineates what they do and don’t mean at this stage of knowledge.
Although this article made it clear that parents of children with autism should not rush out and get MRIs of infant siblings, it did not mention the costs of MRIs, which may be prohibitive for parents without good insurance coverage.
Benefits in terms of detection are described, including limitations of the current state of knowledge. The article is careful to draw parameters around the immediate significance of the findings and to express the need for further research. Numbers are even provided for how accurate researchers would like the MRI to be in predicting autism in infants before the procedure could be recommended for siblings of autistic children.
No mention is made of the harms inherent in MRIs for infants. Some of the harms of MRIs have to do with the sedation that is required, which, ironically, may harm the developing brain.
Also, as with any screening test, there are the harms of “false-positive” and “false-negative” results. In this case, a false-positive finding would result in a healthy baby being diagnosed with autism. And for a false-negative result, a baby with autism wouldn’t be diagnosed.
The story describes the size of the sample, and the proportion of infants within the sample who had siblings with autism. The sample is fairly large and was followed over two years. All of this is explained in reader friendly language. The story also told readers not to rush out and request an MRI for autism screening, as it’s not ready for clinical use yet.
The story would have been stronger if it also had mentioned that there is not clear evidence that such early detection of autism will improve outcomes.
There is no disease mongering in this article, and it does a good job of establishing how prevalent autism is in the general population, and the rate of diagnosis between siblings.
The story includes quotes from independent researchers, and we did not detect any potential conflicts of interest.
One other study is mentioned having to do with changes in surface area of infant brains. No other information is provided about other advances that have been made regarding early markers for autism. This means readers do not have a context within which to judge how important this advance is compared to others.
No explicit discussion is provided about the availability of MRIs for this sort of detection procedure (and if any special, different tools or procedures are needed), but the story does make it clear that this idea is still under research, so it’s experimental for now.
The story doesn’t establish novelty. According to the study itself, this study “confirms and extends” previous findings, so it is not a new idea.
The story does not appear to rely on a news release.