Background: Self-harm (SH) is any act of intentional self-injury or self-poisoning, with or without the intention to die. People who SH are at high risk for future suicide and often suffer considerable emotional distress. Depression is common among people who SH and may be an underlying driver of self-harm behaviour. Treating depression in people who SH has the potential to reduce the risk of further SH and suicide. Self-harm is often repeated, and risk of repetition is highest immediately after an act of self-harm. Readily accessible brief talking therapies show promise in helping people who SH, but further evaluation of these approaches is needed. The Community Outpatient Psychological Engagement Service for Self-Harm (COPESS) is a brief talking therapy intervention for depression and self-harm. This intervention was shown to be feasible and acceptable in hospital emergency department settings, but accessibility was limited; thus, the investigators wish to develop and test a community-based version of this intervention. The COPESS project moves this therapy into a community setting, thereby increasing the number of people who can be helped, especially those in hard-to-reach groups and from disadvantaged areas.

Aim: To assess the feasibility of conducting a trial of the COPESS intervention in a community setting, for people with depression who self-harm, in relation to participant recruitment and retention. Therapy, acceptability and safety of the intervention will also be assessed.

Methods: Using a Single Blind Randomised Control Trial (RCT) design, the investigators will recruit a sample of n=60 participants with a history of SH within the last six months, who are also currently depressed. Recruitment will take place via GP practices for patients who: 1) seek consultation for self-harm; 2) have consulted for self-harm in the previous 6 months (determined by GPs' search of their own data-systems); or 3) self-refer via their GP. Following baseline assessment participants will be randomly allocated to either receive COPESS plus Treatment as Usual [TAU; n = 30)] or TAU alone (n=30). Follow-up assessments will take place at one, two and three months. Recruitment rates, attrition and data completeness will be monitored. Qualitative interviews with participants and stakeholders will further investigate feasibility.

Design A mixed-method evaluation of a single-blind randomised controlled trial (RCT) will determine the acceptability and safety of the COPES intervention, in preparation for a future large-scale trial of the efficacy of COPES for depression in people who SH. People who have depression and SH in the community will be randomised to receive either COPES plus treatment-as-usual (TAU) or TAU alone.

Recruitment rates and power calculations This study is intended as a preliminary evaluation of the feasibility of conducting a future large-scale RCT of the COPES intervention. As, the investigators are not conducting any hypothesis testing of effectiveness, or estimation of prevalence (other than attrition) a formal power calculation is not appropriate. A recruitment target of n=60 (n=30 per trial arm) will be used. This is a common and realistic recruitment target for feasibility trials, and will allow adequate precision in estimating rates relevant to trial outcomes (e.g. attrition, adverse events). This will allow an overall attrition rate of 30% to be estimated with 95% confidence intervals of +/- 12% or, 16% for a single arm. This sample size is adequate for estimating relevant analysis parameters, such as the standard deviation of effects, which are needed for determining the feasibility of a later efficacy trial. To avoid high attrition rates, the investigators will keep participants engaged via text and/or telephone reminders. The previous pilot of HOPE within the ED achieved an average recruitment of nine eligible referrals each month. Sixty-one percent of participants engaged with the therapy and attended at least one session. Rates of attendance for SH at GP practices and walk-in centres in the Liverpool area have been estimated at 70 per month. Based on these figures a recruitment rate of 3 to 4 participants each month with a goal of n=60 participants overall is achievable.

Research Sites The research will take place in Liverpool, in the North West of England. The North West has particularly high rates of SH. Between April 2011 and March 2014, there were 16,405 presentations for deliberate SH to emergency departments within Merseyside and Cheshire. Liverpool local authority had one of the highest number of attendances for SH in this area (15%, n = 2,489). Therefore, Liverpool provides a good location for undertaking work on SH.

Recruitment of GP Practices Lead GPs and/or practice managers at practices identified as suitable for inclusion in the COPES study will be invited to take part in writing. Invitation packs will be given by GPs and followed up by a telephone/skype/face time/face-to-face contact from a member of the research team to ascertain level of interest, and if appropriate, to arrange a meeting to discuss the study in more detail. The Clinical Research Network (CRN) will assist with recruitment of GP practices and have provided a letter of support (see Appendix B). Clinicians will be briefed on the COPES trial and provided with guidance on referring patients in to the study. To assist recruitment, clinicians will be provided with a brief screening checklist to help identify patients where depression-linked SH may be a concern.

Inclusion Criteria:

• Recent episode of SH (in the past six months, self-reported)
• A score of 14 or greater on the Beck Depression Inventory-II (BDI-II) corresponding to the BDI-II score where 14-19 = mild depression, 20-28 = moderate depression, >28 = severe depression
• Aged 16 years and over (SH is especially prevalent in adolescents and young adults and often this transitional age is neglected. This study will be offering at least TAU which young people may not otherwise access. To leave 16-18 year olds out of this study may be unethical and guidance from the British Psychological Society states that people aged 16 years and over can give informed consent)
• Help-seeking, defined as attendance at GP practices or self-referral into the trial

Exclusion Criteria:

• Non-English speaking
• Diagnosed with an intellectual disability as determined by review of clinical notes - the therapy has not yet been adapted for working with this population
• Experiencing severe problems with addiction to alcohol or illicit drugs
• Actively suicidal or psychotic and/or severely depressed and unresponsive to treatment as judged by clinical team
• Unable or unwilling to give written informed consent to participate in the study
• Currently receiving talking therapy for SH

No need to make data available to other researchers

Participants will be randomized to receive COPES plus TAU or TAU via a random sequence generation computer algorithm within STATA to generate (stratified) randomised (block) allocations. Randomisation will be carried out by the study statistician using an independent computerised randomisation system with a randomly sized block design with block sizes of 2, 4 and 6. The study population will be stratified for those receiving TAU and those receiving COPES plus TAU. Researchers completing study assessments will be masked to treatment allocation, facilitated by briefing members of the research team and participants on the need to avoid disclosure of treatment details. The trial will follow an Intent-To-Treat (ITT) protocol. Attrition and reasons for drop-out will be recorded where possible. Guidance will be sought from the patient advisory group on how to manage and minimise attrition over the study period.