Have secular changes in perinatal risk factors contributed to the recent autism prevalence increase? Development and application of a mathematical assessment model.

Background: A 57% increase in the U.S. prevalence of autism spectrum disorders (ASD) for 8-year-old children born in 1994 versus 1998 was recently reported.
Methods: To quantify the possible contributions of given risk/predictive factors on the recent ASD prevalence increase, we formulated a mathematical model based on the baseline risk factor prevalence (RFP), the proportionate change in RFP (cRFP), and the magnitude of the association between the risk factor and ASD [estimated relative risk (RR)]. We applied this model to several pregnancy-related factors (preterm, very preterm, low and very low birth weight, multiple birth, cesarean delivery, breech presentation, and assisted reproductive technology use). RFP and cRFP estimates for each factor were obtained from U.S. population-based surveillance datasets. Estimated RRs were obtained from a series of systematic literature reviews.
Results: We estimate that each risk factor examined, alone or in various combinations, accounted for a very small proportion (<1%) of the ASD increase. Additionally, hypothetical scenarios indicate RFP, cRFP, and RR all need to be sizable for a risk factor to appreciably influence ASD prevalence.
Conclusions: Thus, although various pregnancy factors have been found to be associated with ASDs, the contribution of many of these factors to the recently observed ASD increase is likely minimal.
(Published by Elsevier Inc.)