Prof. Peter Bearman is the Jonathan Cole Professor of the Social Sciences at Columbia University. His team has been delving in-depth into the California Department of Developmental Services (CDDS) data on autism.
There are some big caveats to using the CDDS data. These include:
The CDDS dataset is based on administrative prevalence. In other words, it is a listing of individuals who sought and were successful getting services. It is not a listing of all autistics in California and the standards of inclusion are not standardized over time and geography.
Popular VideoThe average American throws away 82lbs of clothes:
In terms of sheer size, it is probably the largest such dataset in the U.S.. So, taking the limitations to heart, it is worth taking a look at what one can do with these data.
The authors note this:
Popular VideoThe average American throws away 82lbs of clothes:
Use of administrative records of the California DDS for identification of autism represents a strength of the study, facilitating population-based analyses over 11 years of birth records from this populous and diverse state. However, inclusion as a case subject depends on seeking services and receiving a qualifying diagnosis, with previous reports estimating that 75% to 80% of people with autism in California register with the DDS.
The authors are: Keely Cheslack-Postava, PhD, MSPH, Kayuet Liu, DPhil, and Peter S. Bearman, PhD.
The team took the data and asked, is there an increased risk of autism for children based on how long the parents waited after a previous birth?
Consider second-born children. If a mother gets pregnant right after her firstborn, is the risk of autism the same, greater or less than if she waits? Based on what they found, the Columbia group would say that the risk is higher if the mother gets pregnant again shortly after giving birth.
Here is a blurb on the study:
INCREASED AUTISM RISK FOUND IN CLOSELY SPACED PREGNANCIES
An examination of California birth records found second-born children were more than three times more likely to be diagnosed with autism if they were conceived within 12 months of the birth of their older sibling. The farther apart pregnancies were spaced, the lower the risk of autism. The study, “Closely Spaced Pregnancies Are Associated With Increased Odds of Autism in California Sibling Births” published in the February 2011 issue of Pediatrics (published online Jan. 10) examined the odds of autism among more than 660,000 second-born children. Compared to children who were conceived more than three years after the birth of an older sibling, children conceived after an interpregnancy interval (IPI) of less than 12 months were over three times more likely to be diagnosed with autism. Children conceived after an IPI of 12 to 23 months were 1.86 times more likely to have been diagnosed with autism, and children conceived after an IPI of 24 to 35 months were 1.26 times more likely to have been diagnosed with autism.
One possible explanation for the increased risk of autism is that women are more likely to have depleted levels of nutrients such as folate and iron, as well as higher stress levels, after a recent pregnancy; however, these factors were not tested in the current study. Study authors suggest the finding is particularly important given trends in birth spacing in the U.S.; between 1995 and 2002, the proportion of births occurring within 24 months of a previous birth increased from 11 percent to 18 percent. Closely spaced births occur because of unintended pregnancies but also by choice, particularly among older women who delay childbearing. The study was funded by the NIH Director’s Pioneer Award Program.
Here is the abstract:
OBJECTIVE: To determine whether the interpregnancy interval (IPI) is associated with the risk of autism in subsequent births.
METHODS: Pairs of first- and second-born singleton full siblings were identified from all California births that occurred from 1992 to 2002 using birth records, and autism diagnoses were identified by using linked records of the California Department of Developmental Services. IPI was calculated as the time interval between birth dates minus the gestational age of the second sibling. In the primary analysis, logistic regression models were used to determine whether odds of autism in second-born children varied according to IPI. To address potential confounding by unmeasured family-level factors, a case-sibling control analysis determined whether affected sibling (first versus second) varied with IPI.
RESULTS: An inverse association between IPI and odds of autism among 662 730 second-born children was observed. In particular, IPIs of 36 months. The association was not mediated by preterm birth or low birth weight and persisted across categories of sociodemographic characteristics, with some attenuation in the oldest and youngest parents. Second-born children were at increased risk of autism relative to their firstborn siblings only in pairs with short IPIs.
CONCLUSIONS: These results suggest that children born after shorter intervals between pregnancies are at increased risk of developing
autism; the highest risk was associated with pregnancies spaced
Simply put, they claim that indeed there is an increased risk of autism if a follow-on pregnancy comes shortly after the first. In fact, the odds of having an autistic child are
Here is Figure 2 of the paper. This shows their computed odds ratio as a function of IPI—inter-pregnancy interval.
The authors conclusion is:
This study provides evidence of an inverse association between IPIs and autism risk, with a more than threefold elevated odds in pregnancies conceived within a year of a previous birth. This finding is particularly important given trends in birth spacing in the United States. Between 1995 and 2002, the proportion of births occurring within 24 months of a previous birth increased from 11% to 18%. Closely spaced births occur in some part because of unintended pregnancies but also by choice, particularly among women who delay childbearing. Therefore, additional research to confirm this association in other populations and to undercover underlying mechanisms is particularly critical.
As with any study like this, replication is critical. But, if there isn’t some unkown artifact at play here, this could point to more information on causation in autism.