The purpose of the study is to evaluate the long-term efficacy of dialectical behavior therapy (DBT) in treatment of adolescents with deliberate self harm compared to enhanced usual care (EUC). This study follows-up 77 patients in the ages of 12-18 yrs who have been included in an RCT of DBT-A vs EUC. The main inclusion criterion for this study was repetitive self-harm behaviour. The patients were randomly allocated to receive 16 weeks of outpatient DBT or EUC in child and adolescent psychiatric clinics in Oslo. Participants have been assessed so far on six different time-points: baseline (before starting treatment), 9 weeks, 15 weeks, 19 weeks, 71 weeks and 3 years after start of the treatment. In the current project patients will be assessed a 7th time 10 years after treatment completion.

It is hypothesized that compared with participants who had received EUC in the original trial during their adolescence participants who had received DBT-A will:

A) report a significantly lower frequency of episodes of self-harm, both last year and over the extended 10-year follow-up interval.

B) be significantly less impaired with respect to social, family and occupational functioning and report a higher quality of life.

C) have retained significantly fewer diagnostic criteria of BPD and have less severe borderline features according to dimensional measures and have significantly fewer signs of emotion dysregulation.

Suicide attempts and non-suicidal self-harm, highly prevalent in adolescents, are behaviours strongly associated with severe emotion dysregulation, mental health problems and increased suicide risk. The original study, on which the present study is building, used a randomized controlled design, with a stratified block randomization procedure, and blinded pre-treatment, post-treatment and follow-up evaluations comparing 19weeks of DBT-A with EUC of equal length.

Of special interest to this follow-up study is to gain more knowledge on their capacity to regulate emotions, since failures in emotion regulation is an underlying mechanism of repetitive self-harm behaviours and several other difficulties, and since a main aim of DBT-A is to enhance adolescents capacity to use healthy emotion regulation strategies. The study is designed to carefully include user and family perspectives on the definition of specific outcomes and how to measure them, we will collect detailed data on important outcomes employing a combination of quantitative and qualitative methods, innovative approaches such as ecological momentary assessment and register data. Adding new data to the existing rich data will give us a basis to study the dynamic interplay between symptoms, emotions, and functional and dysfunctional behaviours and make new discoveries of mechanisms of therapeutic change, knowledge to further increase treatment response rates and individualize treatments and to inform further development of this treatment and its implementation in Norway and other countries.

The proposed study will investigate:

1. the 10-year post-randomization treatment outcomes of DBT-A compared with EUC with respect to: a) frequency of subsequent episodes of self-harm, b) severity of suicidal ideation, c) level of depressive symptoms and d) frequency of subsequent emergency room visits, hospitalizations and use of additional treatments due to risk of self-harm behaviour.
2. the 10-year post-randomization clinical course and stability of a diagnosis of BPD, its diagnostic criteria and borderline symptoms.
3. participants' adult mental health, social, family and occupational functioning and quality of life.
4. participants' adult strategies for and capacity to cope with stressful life-events and situations in their daily lives and effectively regulate emotions.
5. the association between participants' adult life use of skills and functional coping strategies and self-harm behaviour and other problem behaviours, and whether any substitution from onedysfunctional coping behaviour to another has occurred.
6. participants' evaluation of what components of the treatment was most helpful and led to the strongest and most persistent change in important functions and aspects of life.
7. the economic cost of treatments consumed and functional impairment and occupational disability over the 10-year post-randomization period and the long-term cost-effectiveness of DBT-A.

The inclusion criteria for the initial study were:

1. History of repeated self-harm (last episode within last 4 months) and
2. Age between 12 and 18 years and
3. Satisfied at least 2 criteria of DSM-IV Borderline Personality Disorder (BPD) (as measured by the SCIDII) (in addition to the self-destructive criterion) - alternatively at least 1 criterion of DSM-IV BPD plus at least 2 threshold level criteria and
4. Written informed consent from patient and parent(s) to participate in the study and
5. Patient fluently Norwegian speaking

Exclusion Criteria:

1. Psychotic disorders or
2. Anorexia nervosa or
3. Severe substance dependence disorders or
4. Mental retardation (IQ less than 70) or
5. Asperger syndrome/autism -