WHAT DOES EARLY ANTISOCIAL-BEHAVIOR PREDICT - A FOLLOW-UP OF 4-YEAR-OLDS AND 5-YEAR-OLDS FROM THE ONTARIO-CHILD-HEALTH-STUDY

Objective: To examine the predictive accuracy of antisocial behaviours among 4- and 5-year-old children for problem behaviours 4 years later (ages 8 and 9 years). Method: Data from the Ontario Child Health Stud y (1983) and Follow-Up (1987) are used. Predictive accuracy is concept ualized using positive predictive value (PPV) and sensitivity. The pre dictive accuracy of early antisocial behaviours for the 1987 outcomes is examined overall, by gender, by variable thresholds of predictor an d outcome by gender and by using contextual variables alone or in comb ination with antisocial behaviour recorded in 1983. Results: The predi ctive accuracy of 1983 antisocial behaviour for 1987 outcome is genera lly modest and differs by gender (better for boys for externalizing di sorder [PPV = 41%, sensitivity = 57%]; better for girls for internaliz ing disorder [PPV = 13%, sensitivity = 80%]; better for boys for condu ct problems [PPV = 54%, sensitivity = 21%]). Using either gender-speci fic thresholds or gender-neutral thresholds does not alter predictive accuracy in a consistent way, nor does the use of a single contextual variable. Use of a cumulative risk index increases PPV but decreases s ensitivity. Conclusions: The predictive accuracy of antisocial behavio ur in 4- and 5-year-old children over 4 years in a nonclinical communi ty population is limited. The clinical, research, and policy implicati ons of this work are discussed.