Physical Abuse, Sexual Victimization, and Illicit Drug Use: Replication of a Structural Analysis among a New Sample of High-Risk Youths

Article excerpt

Mounting evidence of serious adverse consequences of childhood physical and sexual abuse has important implications for public health officials and care providers. Given the potential impact on social policy of this area of inquiry, programmatic research is needed for the validation of theoretical models across populations and over time. This study, based on a sample of high-risk youths, replicates a structural model that specified the influence of child physical and sexual abuse on self-derogation and drug use. Results suggest that for male and female youths, physical abuse and sexual victimization had a direct effect on self-derogation and illicit drug use, and an indirect effect on drug use that was mediated by self-derogation. Moreover, structural coefficients for the model were found to be identical across gender. Subsequent analyses demonstrated the equivalence of the structural model of child physical and sexual abuse on drug use across two study cohorts. Implications for the identification and treatment of youths in high-risk groups are discussed.

A more complete understanding of the adverse developmental outcomes of child physical and sexual abuse has become a major priority among social science researchers and clinicians (Browne & Finkelhor, 1986; Finkelhor, 1984; Crit-tenden & Ainsworth, 1987). Particular interest exists in identifying the developmental effects of these traumatizing experiences among special high-risk populations of youths.

Most relevant research on the outcome of childhood maltreatment is descriptive in nature and lacks a guiding theoretical perspective that links child abuse to later emotional and psychological functioning and antisocial behavior (Browne 8c Finkelhor, 1986; Crittenden & Ainsworth, 1987; Widom, 1986). Information that is available supports the view that abused children have higher rates of later delinquency, crime, and drug use than children who do not have these victimization experiences (Hunner & Walker, 1981; Straus, 1984; Alfaro, 1981; Brown, 1982; Paperny & Deisher, 1983; Ben ward & Densen-Gerber, 1975; Sandgrund, Gaines, & Green, 1974; Russe & McBride, 1985).

Research on high-risk youths, such as juvenile detainees, is of particular importance, because these youths have an increased likelihood of future deviant behavior and because of the opportunity to introduce effective intervention strategies into the juvenile justice system. High-risk populations can also be expected to contain sufficient variability among key variables, including childhood maltreatment and drug use. This circumvents the statistical problem of restricted range, which often plagues research using cohorts of high school or college students (Hays, 1981; Judd & Kenny, 1981).

In an earlier investigation, Dembo et al. (1987) found empirical support for an integrated theoretical model of the effects of child physical and sexual maltreatment among a cohort of high-risk youths. The developmental damage view model encompassed the role of childhood maltreatment on adolescent drug use. Structural analysis found that sexual victimization had a primarily direct effect on drug use, whereas physical abuse had both a direct and an indirect effect on drug use. The indirect effect of physical abuse on drug use was mediated by the respondents' feelings of self-derogation. Further, the researchers found that the structural parameters for the model of sexual and physical abuse were invariant across gender, and that race did not affect the results. These findings have important implications for intervention with troubled youths, by highlighting factors placing youths at high risk of becoming involved in illicit drug use.

In this study, we attempt to confirm the validity of the model of abuse and drug use by Dembo et al. (1987) for a new cohort of high-risk adolescents. After exploring gender effects in the structural parameters involving abuse, self-derogation, and illicit drug use, we test the applicability of the theoretical model across multiple populations. …