Assessment of risk for juvenile compared with adult criminal onset: Implications for policy, prevention and intervention

Most risk assessment instruments aim to predict juvenile offending to aid effective prevention policy. The aims of this study were to explore to what extent adult onset offending can be predicted and by which different risk measures, and to examine the extent to which some psychological dynamic factors may buffer the effect of risk factors so as to delay onset and therefore promote adult onset offending. Longitudinal data from the Cambridge Study in Delinquent Development (CSDD) were analyzed. 403 South London men, followed from ages 8-10 to ages 48-50, were divided, on the basis of conviction records, into late starters (LS = 38), early starters (ES = 129) and non-offenders (NO = 236). Four risk measures (antisocial behavior, family, socio-economic, and internalizing) were studied. The findings showed that late- onset offending, at age 21 or thereafter, was predicted by internalizing problems. Early onset offending before age 21 was predicted by antisocial behavior, family risk, and to a modest degree by socio-economic risk measures. Internalizing behavior was negatively related to early onset but positively related to late onset offending. This is the first longitudinal investigation to show that different factors predict early or late delinquent onsets. These findings suggest that including psychological dynamic factors in the assessment of risk not only improves the prediction of future offending, but also informs prevention policies and clinical intervention. Early and late onset offending deserve separate consideration and need different risk assessment processes and instruments not only to measure the level of risk but especially to reduce it. Internalizing factors as well as externalizing factors should be targeted in intervention programs.