Brief psychodynamic-interpersonal therapy for adults with a history of self-harm: the SafePIT RCT

There are over 200,000 hospital attendances for self-harm per annum in the United Kingdom at an estimated annual cost of £133-162M. Systematic reviews prior to commencing the study suggested that brief psychological interventions are effective in reducing psychological distress after self-harm and reduce repetition of self-harm.

The SafePIT trial was designed to evaluate the effectiveness and cost-effectiveness of self-harm-focused psychological therapy plus standard care versus standard care alone.

Pragmatic, multicentre individually randomised controlled trial of brief psychodynamic-interpersonal therapy compared with standard care with internal pilot, cost-effectiveness and process evaluation.

People aged over 18 years who attend hospital after intentional self-harm with a history of ≤ 3 episodes in the last 12 months.

Individual psychodynamic-interpersonal therapy, delivered face to face or by video conferencing by liaison mental health nurses, over four (or fewer by mutual agreement) 50-minute weekly sessions with two optional boosters.

The primary outcome was time from randomisation to first repetition of self-harm leading to hospital attendance. Secondary outcomes (at 6 and 12 months) included rate of repetition of self-harm leading to hospital attendance; self-reported self-harm using questionnaires and Short Message Service; psychological distress and clinically significant improvement (Clinical Outcomes in Routine Evaluation - Outcome Measure); anxiety (Generalised Anxiety Disorder-7); hopelessness (Beck Hopelessness Scale); interpersonal function (Inventory of Interpersonal Problems-32) and quality of life (EuroQol-5 Dimensions, five-level version; Recovering Quality of Life; Clinical Outcomes in Routine Evaluation-6D).

The planned sample size was 770 participants. The trial closed to recruitment early in January 2023 at the end of the 12-month internal pilot, with 22 randomised participants, 12 allocated to psychodynamic-interpersonal therapy and 10 to standard care. Due to the early trial closure, trial follow-up was curtailed to 6 months, and analyses are restricted to descriptive statistics. Seven of 12 participants allocated to psychodynamic-interpersonal therapy started therapy, and four completed therapy. Participant-reported secondary outcomes were completed for nine (40.9%) participants at 6 months. Repetition of self-harm leading to hospital presentation could be assessed for 18 participants and occurred in two participants in the psychodynamic-interpersonal therapy arm (18.2%) and no participants in the standard care arm within 6 months of randomisation. Economic findings indicated no substantive changes in health-related quality of life, or primary and secondary care resource usage across arms or over time. Intervention costs were highly sensitive to assumptions regarding the number of patients that would be treated per therapist in real-world role out of intervention.

The study was unable to recruit the necessary sample size, preventing the trial from progressing. The trial met with several challenges.

Trial timelines coincided with the start of the second wave of the COVID-19 pandemic, causing substantial delays, difficulties with recruitment and, ultimately, its early closure. Although the trial closed early and with insufficient participants to proceed with full statistical analysis, our experiences and recommendations can inform future trial design and delivery.

Self-harm remains a major risk factor for suicide, and provision of cost-effective interventions for people who self-harm is a key part of the government's Suicide Prevention Strategy.

This synopsis presents independent research funded by the National Institute for Health and Care Research (NIHR) Health Technology Assessment programme as award number NIHR131334.