Risk factors associated with conduct disorder

Research Matters
Number 7 – July 2012


What are the late childhood risk factors associated with conduct disorder in early adolescence, and how are these risk factors specific to particular categories of conduct disorder?


Conduct disorder is a psychiatric syndrome that begins in childhood or adolescence, and is characterized by a persistent and repetitive pattern of serious rule-breaking and violation of the rights of other persons. Given the link between conduct disorder and more serious juvenile and adult criminality (Lacourse et al., 2010; Moffitt et al., 2008), gaining a better understanding of this disorder is essential for informing the design of early prevention and intervention strategies aimed at averting later offending.

Recent research (Lacourse et al., 2010) on Canadian children has shown that conduct disorder consists of a number of distinct categories or subtypes, which differ according to the type and severity of symptoms. Although most children (82.4%) did not have any type of conduct disorder, 13.9% were classified as having Non-Aggressive Conduct Disorder (e.g., theft, destruction of property, and serious rule violations), 2.3% were labeled as having Physically Aggressive Conduct Disorder (e.g., fighting, assault with a weapon), and 1.4% were categorized as Severe-Mixed Conduct Disorder because they displayed a combination of both serious aggressive and non-aggressive behaviours.

The next step in this line of research was to examine whether a variety of risk factors (e.g., personal, family, and environmental characteristics) measured in late childhood (ages 10-11) could predict which youths would display various symptoms of conduct disorder during early adolescence (ages 12-13).


The present study was based on a sample of 4,125 children (boys and girls) aged 12-13, who were selected from three cycles of the National Longitudinal Survey of Children and Youth (NLCSY; 1994-95, 1996-97, and 1998-99) and had previously been categorized as having no conduct disorder diagnosis or into any of the three conduct disorder subtypes noted above: Non-Aggressive, Physically Aggressive, and Severe-Mixed. Information was also obtained on the following personal, family, and environmental risk factors, which were assessed when the children were aged 10-11, with the exception of age (i.e., the present study examined whether the risk of having some type of conduct disorder differed between 13-year-olds and 12-year-olds):

Statistical analyses were conducted to examine the strength of the relationships between these risk factors and the various conduct disorder categories. In particular, the study examined whether the presence of a risk factor significantly increased the risk of being in a specific conduct disorder category, relative to the risk of having no conduct disorder.


Of all the risk factors that were considered in the analyses, only a subset was significantly associated with the three conduct disorder categories, as compared to children who were classified as not having any type of conduct disorder. Of this subset of significant risk factors, some were found to be common across different conduct disorder categories, while others were unique to particular categories.

For instance, being male significantly increased the risk of being in all three conduct disorder categories, as did having a non-intact or mobile family. In contrast, associating with deviant peers and experiencing coercive/ineffective parenting increased the risk of being in the Non-Aggressive category only, while physical aggression only raised the risk of being in the Physically Aggressive category, and hyperactivity/inattention uniquely increased the risk of being in the Severe-Mixed category.

The remaining risk factors (e.g., neighbourhood socioeconomic disadvantage and instability, family socioeconomic status) were not significantly linked to the risk of being in any of the conduct disorder categories, as compared to children with no conduct disorder.


Consistent with much of the available research literature, this study also suggests that broad socioeconomic variables (i.e., family and neighbourhood SES) do not appear to have a strong or direct impact on problem behaviour and later delinquency, and therefore indicates that resources could be devoted more to addressing individual- and familial-level risk factors. Accurately diagnosing conduct disorder categories may facilitate the development of intervention and prevention strategies targeted at the risk factors that are both common across categories and unique to each one.

While it is of course not possible to change static risk factors such as gender, the other significant risk factors identified in the present analysis are potentially modifiable. For instance, programs designed for youth in the Non-Aggressive and Severe-Mixed categories could have the common objective of helping non-intact and mobile families. However, the program for the Non-Aggressive type could also have unique components for changing coercive/ineffective parenting styles and reducing contact with deviant peers, whereas the specific part of the program for the Severe-Mixed category might focus on tackling hyperactivity/inattention. Moreover, in order to most successfully mitigate the impact of risk factors at ages 12-13, intervention should occur at a younger age (i.e., 10-11). Rigorous evaluations will also be required to assess the effectiveness of programs targeted at specific conduct disorder categories and their associated risk factors.

Finally, given that knowledge on identifying conduct disorder categories and their associated risk factors is still relatively new in Canada, future research should attempt to replicate and extend the present findings using samples of children and adolescents from a variety of neighbourhood contexts. Such a program of research should also try to shed some light on why physical aggression at age 10-11 only predicted the Physically Aggressive category and did not raise the risk of being in the Severe-Mixed category, given that the latter subtype displays both aggressive and non-aggressive behaviours. Finally, although previous research has investigated the links between conduct disorder subtypes in early adolescence and delinquency at age 14-15 (Lacourse et al., 2010), longer follow-up periods are warranted to examine the longer-term impact of various conduct disorder subtypes on later adult criminality.


Lacourse, E. (2012). Late childhood risk factors associated with conduct disorder subtypes in early adolescence: a latent class analysis of a Canadian sample. Research Report: 2012-2. Ottawa, ON: National Crime Prevention Centre, Public Safety Canada.

Additional references: Lacourse, E., Baillargeon, R., Dupéré, V., Vitaro, F., Romano, E., & Tremblay, R. (2010). Two-year predictive validity of conduct disorder subtypes in early adolescence: a latent class analysis of a Canadian longitudinal sample. Journal of Child Psychology and Psychiatry, 51(12), 1386–1394.

Moffitt, T.E., Arseneault, L., Jaffee, S.R., Kim-Cohen, J., Koenen, K.C., Odgers, C.L.,
Slutske, W.S., & Viding, E. (2008). Research Review: DSM-V conduct disorder — research needs for an evidence base. Journal of Child Psychology and Psychiatry, 49(1), 3–33.

For more information:

Lucie Léonard
National Crime Prevention Centre
Public Safety Canada
269 Laurier Avenue West
Ottawa, ON K1A 0P8
Telephone: 613-957-6362
Email: Lucie.Leonard@ps-sp.gc.ca

Date modified: