Developmental assessments of suicidal and substance use behaviors require the use of integrated multilevel bio-psycho-social models (Cicchetti, 1993; Hinshaw, 2002; Windle, 2004; Windle and Davies, 1999; Zucker et al. 1995). This method emphasizes that developmental outcomes are interdependent and influenced by multiple factors, such as genetic, biochemical, physiological, cognitive, and environmental variables. Using this method, assessments of the influences of these different types of variables are conducted concurrently to better understand suicidal and substance use behaviors. This developmental approach includes the principles of reciprocal causation such that an adolescent’s behavioral characteristics elicit particular behavioral responses from parents and peers, and these parents and peers influence the adolescent’s behavior, and interactions between the individual and their environment (e.g. gene-environment interactions and correlations). The goal of this approach is to examine patterns of interpersonal interactions and the behavioral outcomes that occur from these interactions across development (e.g. Caspi and Moffit, 1995; Caspi et al. 2003; Hinshaw, 2002; Sameroff, 2000). Testing these models requires a prospective design.
