On Rolling Over and Autism (the Latest Parental Worry)

Over the last several years, I have had dozens of queries from concerned parents who had read some “red flag” websites that describe symptoms of autism. These parents were terribly worried that their babies did not make enough eye contact and that meant they would be autistic. There have been a few cases where the parent’s description was worrisome, but in most of the cases the problem was simply that the baby was very young. What those websites often neglect to tell you is that young babies normally show many of the behaviors that would be symptomatic of autism if they were older. The autistic toddler or preschooler continues to show some behaviors that are normal in earlier life; the typically-developing child stops doing things that belong to earlier development. Autism is primarily characterized by developmental delays, not by doing things that are never seen in the lives of typically-developing children. For example, the hand-flapping and toe-walking often noted as indications of autism are also seen in typically-developing children, but at an earlier period of development.

In the last few weeks, in addition to the usual questions about eye contact, I have noticed some new questions about motor development and especially about the way a baby rolls over. I see references to this issue on various chat groups and on Youtube too. They all refer back to a particular publication, and I want to talk about that.

In the mid-1990s, the researcher Philip Teitelbaum became interested in the idea that atypical movement patterns might precede and predict more obvious symptoms of autism. This was by no means a silly idea—many other neurological problems of childhood involve movement anomalies. It’s been known for a long time that infants who already have “handedness” (they use one hand consistently rather than alternating) or who crawl asymmetrically, with one side of the body doing more work, most often have some damage to one side of the brain. Cerebral palsy, which can interfere so badly with speech, may not be diagnosed at birth, but it may be seen in anomalies of movement in the first year. That autism might also involve problems of motor development was a reasonable hypothesis.

In order to see whether babies who were later diagnosed as autistic had unusual movement patterns later on, Teitelbaum advertised and announced his work in various newspapers (this was before much Internet). He asked for parents to send him home videos of their babies, so he could examine movement patterns in early life and see whether he could identify any unusual movements in those who were later considered autistic.

Teitelbaum and some co-authors published the results of this study in 1998 in the Proceedings of the National Academy of Sciences. They had found videos for 17 babies who were later diagnosed as autistic, and reported that these babies showed unusually movement patterns. Delays and asymmetries in sitting and walking were evident, and rolling over was described as not only delayed and awkward, but as involving a different use of head and legs to turn the body that is characteristic of typically developing infants. However, there has been little replication of this work, and there is ongoing discussion of the method that should be used to observe and describe motor behavior (see http://asb.brain.riken.jp/files/Threelessonsfrom.pdf ).

There are several points to be made here. One is that the rolling pattern described has not clearly been shown to be a precursor of autism, and in fact Teitelbaum was criticized for the lack of control or comparison information he provided. Failing to roll at all, past the usual time when this milestone is achieved, is probably a more significant issue.

Second, and very important, Teitelbaum’s work looked at the motor development of babies of whom the great majority slept in the prone, tummy-down position. Babies in the United States, Australia, and Great Britain today are very likely to be put to sleep supine, lying on their backs. This practice is known to delay motor development compared to what was seen years ago for prone-sleeping babies (see www.ncbi.nlm.nih.gov/pubmed/15934485 ). The sequence and timing of motor development Teitelbaum found in typically-developing babies were not what we would see today in typically-developing babies, so delays that Teitelbaum considered to be characteristic of later-found-to-be-autistic infants cannot be used to assess babies today.

Third, every baby has his or her own schedule of achieving motor milestones. There may be a long pause between two milestones, or one may come very rapidly after another--  as when babies hardly creep at all, but go quickly to walking. There is a normal range of development, not a deadline to get something accomplished. That means that only a rather sophisticated observer will be able to detect something out of sequence or dramatically delayed in motor development; Youtube won’t do the job!

Fourth, let me point out that if a baby does have unusual movement patterns, there can be many reasons for them other than autism, and it is important to seek medical care if such patterns are seen.

Conclusion?  The concern about how babies roll over has recently been whipped up on various websites devoted to encouraging parents to be anxious about the possibility of autism. Let me remind you that the present high reported rates of autism are probably caused primarily by changes in diagnostic approaches that include some minimal disturbances in the same category with debilitating autistic disorders. Also, it is clear that quite a few children who are diagnosed as autistic in the preschool period no longer have that diagnosis several years later. Finally, because parents who are worried about eye contact and rolling over also want to know about “early intervention”--  the best thing you can do for any baby is to have fun together and respond to the baby’s communications as best as you can. If you’re worried about movement problems, by sure you do plenty of tummy time. Minimize screen time, and that means the time you are on your phone in the baby’s presence as well as screens for the baby to look at. Talk to the baby a lot. All babies, typical or atypical, will benefit from these actions.