Authors' reply

We note with interest the comments raised by Voracek. He suggests that sex-hormone exposures in utero may play a role in the aetiology of autism, and that the second-to-fourth digit (2D:4D) ratio may be a marker for fetal androgen exposure. This seems to be a plausible hypothesis, and we believe that the potential association between the 2D:4D ratio and autism risk deserves further exploration. More importantly, studies on the direct effect of fetal sex-hormone profiles on autism risk are warranted.

However, the 2D:4D ratio was not included in our meta-analysis of potential prenatal risk factors for autism because it was not considered to be a prenatal exposure variable itself, although it likely represents the effects of prenatal exposures, in particular sex steroid hormones. There are many characteristics that become evident after birth that are likely due to prenatal exposures, but in our meta-analysis of risk factors for autism we focused only on those variables that could be assessed during the prenatal period (e.g. maternal medication use, parental age). Voracek speculates ‘that Gardener et al did not include this literature [2D:4D ratio] in their meta-analysis on the grounds that they categorised it under “medical hypotheses”’. This is not the case. Rather, we did not include the 2D:4D ratio because our article was limited to conditions assessed during the prenatal period, not their sequelae.