Brief Admission by self-referral (BA) is a standardized treatment model, providing patient-controlled and person-centered care. It was developed to reduce self-harm and compulsory care by promoting autonomy. Randomized clinical trials have not yielded significant between group differences with respect to inpatient care, including compulsory care. The major difficulty in evaluating BA is preventing the control group from cross-contamination, as in the implementation process of BA, all physicians, all inpatient and outpatient staff as well as managers need to be informed and undergo basic education regarding the intervention. As BA addresses a prevalent and frustrating issue in psychiatric health care, there is considerable risk that the approach leaks to the control group, reducing the possibility to detect between-group differences. In the current study this will be addressed through a register-based approach, comparing similar clinics, implementing BA at different timepoints over time. Individuals with traits of borderline personality disorder will be included and comparisons will be made with respect to compulsory care, voluntary inpatient care and mortality.

Inclusion Criteria:

• Living in the uptake area of Lund, Linköping, Helsingborg or Norrköping
• Age 18-65 years at the time of inclusion
• Admitted to hospital either diagnosed with ICD-10 diagnosis BPD F60.3, or admitted at least twice with Intentional self-harm X60-83 in combination with one of more of the following diagnoses ADHD (F90.0-F90.X), Bipolar disorder type 2, (F31.8W, F318A-F, F31.0, F31.8-9, F31.8W) or Autism, Atypical autism, Aspergers syndrome (F84.0, F84.1, F84.5) or Mild intellectual disability (F70.0-F70.9).

Exclusion Criteria:

• not fulfilling inclusion criteria

Individuals with repeated self-harm and traits of Borderline personality disorder