Next we addressed whether the effect of parental EB exposure differed as a function of timing. These comparisons were limited to children exposed to parental EB during at least one developmental period. We compared the effect of exposure during a single period to the effect of exposure during the other periods. We also compared the effects of exposure during two periods; these could be contiguous or separated by a period of parental remission. These results (Table 3) indicate that the likelihood of AUD did not differ as a function of timing of exposure. Though we observed trends suggesting that earlier exposure was more pathogenic, confidence intervals always spanned unity.
