Abstract

Background: Rates of violence in the USA have fluctuated widely over the past few decades. Theorists have examined period and cohort effects, but there appear to be no studies examining these effects on progression in developmental pathways towards violence.

Objective: To assess whether differences in progression among individuals in the Pittsburgh Youth Study are consistent with period or cohort effects.

Design: Multivariate logistic regression was conducted to examine differences between cohorts in the odds of progressing through the developmental pathway towards violence. Adjusted and unadjusted odds ratios (ORs) and corresponding 95% CI are reported.

Setting: Pittsburgh Pennsylvania, from 1987 to 2000.

Subjects: Two cohorts of male adolescents from the Pittsburgh Youth Study. The youngest cohort (n = 503) was followed from median ages 7 to 20, and the oldest cohort (n = 506) was followed up from median ages 13 to 25.

Main outcome measure: The odds of progression along a developmental pathway towards violence.

Results: There was no statistically significant difference between the cohorts in progression from minor aggression to physical fighting (OR = 1.13, 95% CI 0.77 to 1.65). However, after adjustment for major risk factors, the oldest cohort was significantly more likely to progress from physical fighting to violence (OR = 2.34, 95% CI 1.39 to 3.92).

Conclusions: These results provide initial evidence that cohort effects, which would be present early in development, do not contribute significantly to later differences in reported violence and raises the possibility of whether period effects can explain these differences.