Background: The current literature concerning self-harm and suicide has shown firstly that self-harm is a major public health problem, secondly that people who self-harm are at increased risk of suicide and thirdly, that women prisoners have a higher rate of self-harm than women in the general population and than male prisoners. Women prisoners are therefore particularly high risk of self-harm and suicide and yet to date, no specific intervention has been developed for this group.Aims: This study intends to build on an intervention found to be effective in the community and to describe a methodology of conducting a therapeutic intervention in a prison environment. Method: The research utilized both qualitative and quantitative methods conducted over four phases. In Phase 1, semi-structured interviews were completed with women prisoners who self-harm and with prison staff. Data was analysed thematically. In Phase 2, the results from the analysis were then used to inform the modification of a Psychodynamic Interpersonal skills Therapy (PIT) model originally used in the community. During the third phase of the study, prison staff were recruited and trained to deliver the therapy. Finally, the fourth phase consisted of a feasibility and acceptability study with 64 women prisoners who had recently self-harmed. Thirty-two were randomized to the treatment group which consisted of four sessions of individual PIT therapy and 32 to the control condition which consisted of four sessions with a member of staff not trained in the therapy. Rates of self-harm were measured pre- and post-treatment. Standardized measures for depression, suicide intention, hopelessness and interpersonal skills were also administered. These data were further supported by interviews conducted with participants who completed the intervention, therapists and therapy supervisors. The methods used in the study were constantly evaluated and amended when required to describe a methodology suitable for this environment. Results: There was evidence of a lack of understanding of self-harm by prison officers. The trial phase of the study did not show evidence of efficacy for the intervention. There was an overall reduction in rates of self-harm pre- and post- treatment across the sample. Comparison of scores on outcome measures, pre- and post-intervention, showed a reduction on all tests in both treatment groups but this was not significant. It was not feasible to deliver the therapy as originally envisaged and substantial changes to the methods were needed. The results showed that prison staff had the skill and capacity to deliver PIT therapy but that the prison was not able to support such a role. As a result external therapists had to be brought in to the study. Conclusions: It is acknowledged that the study was not feasible as originally designed, but due to the small sample size we cannot say it was not effective and therefore, not worthy of further study.