Although the sample is large, it was not recruited to be representative of a population, so our results may not be fully generalizable and may over-represent Caucasians males and non-heterosexual individuals. Furthermore, as the sample may not be representative of the general population, the interpretation of the gender results may not be generalizable. Although the participants are relatively young, there is always the inherent possibility of recall error when relying solely on self-report of past symptoms, both in terms of under and over-reporting, as well as misreporting relationships between drug use and the presence of psychiatric symptoms. The extent and direction of such errors is unknown.
