Emotion dysregulation in adolescents: comparing community and clinical samples and their association with mental health and risk behaviors

Abstract Background Emotion dysregulation (ED) is a multidimensional construct associated with mental health disorders, however, its distribution and clinical significance in the adolescent general population remain underexplored. This study aimed to investigate the prevalence and impact of ED in community sample, and to compare these findings with those observed in a clinical sample. Methods A total of 611 students (mean age = 16.1 years) and 63 clinical outpatients with Borderline Personality Disorder, or Depressive, or Anxiety Disorder (mean age = 17.5 years) completed the Difficulties in Emotion Regulation Scale (DERS) and self-report questionnaires assessing depression, anxiety, impulsivity, risk behaviors, emotion regulation (ER) skills, social problem-solving, and childhood trauma. Students were classified into subgroups based on DERS percentiles and latent profile analysis (LPA). Results Based on percentile distributions of DERS Total score, the sample of students was stratified into low, moderate, and high-ED subgroups. A trend was observed linking higher levels of ED with increased psychopathological symptoms, impulsivity, childhood trauma, reduced adaptive strategies, and risk behaviors (i.e., binge eating, non-suicidal self-injury, and problematic internet use). The “high-ED” students exhibited clinical characteristics comparable to clinical sample, neverthless only 51% of them had ever sought mental health support. The LPA identified four distinct subgroups of students – “low-ED” (31%); “mild-ED” (31%); “moderate-ED” (24%); “high-ED” (13%) - showing different profiles of DERS subscales and association with clinical variables. The “high-ED” subgroup showed greater difficulties than those observed in clinical sample, particularly in relation to DERS strategies and DERS non-acceptance . Conclusions ED appeared to vary along a continuum of overall severity among students and it was strongly associated with clinical variables. Latent profiles were primarily distinguished by overall severity, while also showing subtle differences in the relative prominence of specific ER domains. The analysis of these profiles may inform the development of preventive interventions. Trial registration This study is conducted in the framework of a randomized clinical trial that was registered in ClinicalTrial.gov (ID: NCT04349709).