Limited prosocial emotions (LPE) specifier in conduct disorder and offending behavior: findings from a 10-year prospective longitudinal study of juveniles in residential care.

BACKGROUND: Since the introduction of the Diagnostic and Statistical Manual of Mental Disorders (DSM)-5, a limited prosocial emotion (LPE) specifier has been added to the conduct disorder (CD) diagnosis in addition to the age of onset specifier. It was suggested that this would identify a subgroup with severe antisocial and/or aggressive behavior with serious current and future (mental health) impairment. Research in recent years has shown that this is indeed a subgroup with severe antisocial behavior; however, mental health problems do not appear to differ from those of youth with CD without LPE. Most research to date has been cross-sectional. However, longitudinal research is urgently needed to better understand the predictive value of the LPE specifier. The aim of the current longitudinal study is to examine future offending behavior of youth with CD with compared to youth without the LPE specifier. In addition, the predictive value of the categorical LPE specifier and the dimensional LPE score will be examined beyond factors that are strongly associated with future offending (i.e., gender, age, and prior offending).

METHODS: Adolescents and young adults (12-25) with CD (assessed with the Schedule for Affective Disorders and Schizophrenia for School-Age Children-Present and Lifetime Version [K-SADS-PL]) with (N = 61) and without (N = 75) the LPE specifier (assessed with the Callous-Unemotional [CU] dimension of the Youth Psychopathic traits Inventory [YPI]) (in line with Jambroes et al., 2016) were compared on sociodemographic characteristics, mental health problems and offending behavior. Future (general and violent) offending was based on official conviction data.

RESULTS: Our results showed that youth with CD with and without the LPE specifier did not differ in self-reported and informant-reported mental health problems. However, youth with CD with the LPE specifier showed more offending behavior and personality pathology at baseline. In addition, the categorical LPE specifier was associated with future general offending, but not with future violent offending. The dimensional LPE score was associated with both future general and violent offending. However, after adjustment for gender, age, and prior delinquency, these associations disappeared, with the exception of the association between the dimensional LPE score and violent offending, which remained significant even after controlling for gender, age, and prior violent offending.

DISCUSSION: In conclusion, there seems to be evidence of a relationship between limited prosocial emotions and future offending behavior in youth with CD. This relationship, however, should not be overestimated, as there are other (static) factors (e.g. gender and prior offending behavior) that also have a strong influence on future (violent) offending behavior. Still, from a clinical point of view, a dynamic factor like prosocial emotional skills is a good focus for reducing the risk of future offending behavior.