Background: Impulsivity (I) and anxiety (A) were hypothesized to be crucial clinical features in adolescents with at-risk behavior. We therefore classified them into sub-groups according to these two major dimensions. The study examined the relevance of these groups by describing their major diagnoses and behavioral characteristics.
Methods: During a 1-year period, all in-patients consecutively admitted for at-risk behavior, except those with a previous psychotropic treatment and/or schizophrenic disorders, were rated for anxiety and impulsivity, and categorized into four groups: impulsive and anxious (IA), impulsive and non-anxious (Ia), non-impulsive and anxious (iA), non-impulsive and non-anxious (ia). We assessed the main behavioral disturbances (suicide attempt, carving, violence, delinquency, substance abuse, and eating disorder) and the main current axis I disorder in each sub-group.
Results: A total of 69 patients were included. In the IA group 62% exhibited hypomanic episodes and 87% recurrent suicide attempts. In the Ia group all exhibited conduct disorders, 93% were males, 80% delinquents, and 100% violent with others. Both groups reported a high percentage of cannabis use (67%). The iA group exhibited anorexia nervosa (73%) with a major depressive episode. The ia patients were mainly non-violent, first suicide attempts with low risk.
Limitations: Long-term data are needed to assess the stability of these groups.
Conclusions: We found that sub-typing adolescents with at-risk behavior into four groups according to their level of anxiety and impulsivity was highly predictive of being suicidal with mood disorders (AI), delinquent with conduct disorder (Ia), anorectic or depressed (Ai), and with substance abuse associated only to impulsivity. It is likely that this sub-typing of patients may be useful for prevention and therapeutics. The impulsive-anxious group (IA) appears closely related to the soft bipolar spectrum. A replication and follow-up data are now needed.