An important task for epidemiologic research on
illicit drug use is to identify risk factors that, if altered,
might reduce the risk of drug dependence and related
drug problems, thereby leading to reduced prevalence of
these problems in the population. In this context, a
research report by Robins and Przybeck (1985) aroused
considerable interest. They analyzed retrospective survey
data from adults interviewed for the Epidemiologic
Catchment Area (ECA) program and found, in crosssection,
that lifetime prevalence of drug abuse dependence
problems was higher for drug users who had
reported an earlier onset of illicit drug use (under age 15)
as compared to drug users who had started in midadolescence
(between 15 and 17 years of age) or in later
years. This new evidence on occurrence of drug abuse
dependence problems has been joined with a concurrent
line of research linking very early drug involvement to