The disparity in findings between our study and prior investigations may also reflect differences across ethnic groups in factors associated with risk for drug and alcohol use, such as perceived peer use (Callas et al., 2004; D’Amico and McCarthy, 2006; Korhonen et al., 2008; Kosterman et al., 2000), accessibility of substances (Beyers et al., 2004; Komro et al., 2007), religiosity (Brook et al., 1999; Jeynes, 2006; Turner-Musa and Lipscomb, 2007; Wills et al., 2003), and parental attitudes toward substance use (Beyers et al., 2004; Callas et al., 2004). There is some evidence, for example, that lower prevalence of substance use in African-Americans is attributable in part to higher religiosity (Heath et al., 1999; Nasim et al., 2007; Wallace et al., 2003) and lower perceived peer substance use (Heath et al., 1999; Vega et al., 1993). These differences in risk factors may translate into distinctions in the relative contributions of genetic and environmental factors to substance use onset, although this hypothesis has yet to be tested empirically.
