In spite of the valuable data set that our two family studies offer, some possible limitations must be mentioned. In order to have a sufficiently large number of cases to evaluate the odds ratios for any particular disorder, analyses were based on the presence of a disorder at any time during the developmental period examined. Although steps were taken to adjust for this variation, children evaluated a larger number of times would have more opportunity to exhibit the disorder than those with fewer evaluations. Alternatively, this may be viewed as an opportunity to increase the reliability of assessment. While future publications will address the stability of diagnosis across waves using latent growth models, the purpose of the present analyses were to address the issue of familial/genetic loading and parental diagnosis on outcome of offspring during the child and adolescent periods.
