Predictive Validity of Early Cumulative Risk and Protection for Health-Risking Behavior in Young Adulthood

Introduction: Several studies recognize that the accumulation of risk factors across multiple domains such as community, school and family significantly increases the likelihood of adolescent problem behaviors. The extent to which early protection buffers the effects of early risk is less well understood. This study investigates the predictive validity of cumulative risk and cumulative protection established in middle school for health-risking behaviors in late adolescence and young adulthood (e.g., early substance use, antisocial behavior).

Method: Data are from the Community Youth Development Study (CYDS), a test of the Communities That Care (CTC) planning system conducted in 24 communities. Using the CTC Youth Survey (CTCYS), we measured 19 risk and 13 protective factor scales in grade 6. Previously established cutoff values for the CTCYS scales were used to dichotomize 1986 CYDS control participants at elevated risk or elevated protection. Dichotomized scale scores were summed to create early cumulative risk (range 0-19, mean=6.6, SD=4.6) and cumulative protection (range 0-13, mean=7.1, SD=3.5) scores for each student. We used logistic regression to evaluate the predictive role of early cumulative risk, cumulative protection, and their interaction on later outcomes. Sociodemographic covariates were included to increase estimate precision. Preliminary analyses focused on predicting grade 12 substance use (dichotomized lifetime and past 30-day alcohol, tobacco, marijuana use); subsequent analyses will examine antisocial behavior, mental health, and educational attainment into young adulthood.

Results: Preliminary logistic regression analyses showed that higher cumulative risk in grade 6 significantly increased the likelihood of lifetime cigarette (p<.01), alcohol (p<.01), and marijuana use (p<.01) in grade 12, but early cumulative protection did not, nor did it moderate the effect of risk. Findings for recent substance use were complex. There were main effects of early risk (p=.03) and early protection (p=.01) on past 30-day cigarette use. However, the effect of early risk on past 30-day cigarette and marijuana use was moderated by cumulative protection (p=.04, .05). There were no main or interaction effects for recent alcohol use.

Conclusion: Preliminary findings validate the use of CTCYS cut-points to establish elevated risk and protection in early adolescence. They also show that early cumulative risk—but not protection—consistently predicted lifetime substance misuse through the end of high school. However, early protection moderated the effect of early risk on recent substance use. Implications for policy and community prevention efforts will be discussed.