Self-harm intervention, family therapy (SHIFT)
The Self-harm intervention, family therapy (SHIFT) trial is a large pragmatic intervention trial which has been undertaken over three sites (Leeds, Manchester, London). It has tested the effectiveness of a manualised family therapy intervention for young people with repeated self harm. This study is the largest complex intervention trial yet undertaken in child mental health.
Young people who self harm are at high risk of recurrence and of poor outcomes. They are a major public health priority.
The main aim was to look at the rate of repetition of self-harm leading to hospital attendance 18 months after randomisation. Secondary outcomes include rate of repetition at 12 months, cost-effectiveness and quality of life.
A pragmatic, individually-randomised control trial design has been used to compare family therapy with treatment as usual for adolescents aged 11 - 17 years who have self-harmed before.
Family therapy was delivered by qualified family therapists using a version of the Leeds Family Therapy and Research Centre Systemic Family Therapy Manual adapted for use with young people who self-harm.
Predominantly quantitative measures were used to capture effectiveness of the intervention. Measures were administered before, during and some time after the intervention.
The North London SHIFT team comprises family therapists from the Tavistock and Portman NHS Foundation Trust, Enfield and Islington on rotation between the sites. Every young person referred to Child and Adolescent Mental Health Services (CAMHS) in any of the sites by their GP, school or following admission to hospital for a self-harm incident were screened for eligibility.
The trial reached its target of 832 participants recruited on 30.12.2013, and is now in follow-up.
It is anticipated that trial results will be available soon.
Family therapists involved in the trial have published papers in a special edition of the Journal of Family Therapy published in April 2016.
An accompanying video is also available: