The challenge of evaluating the effectiveness of treatments for deliberate self-harm.

Deliberate self-harm (DSH) is the most important factor associated with risk of suicide. It also poses a considerable burden on health and social services. Repetition of DSH after hospital presentation is common, with most studies reporting between 12 and 25% of patients repeating within a year and again presenting to hospital (Owens et al. 2002) and many more being involved in repeat episodes which do not bring them to clinical attention (Guthrie et al. 2001). Repetition itself increases the risk of a fatal outcome (Sakinofsky, 2000). The results of both psychosocial and pharmacological interventions to prevent repetition have been fairly disappointing (Hawton et al. 1998, 2000), although the findings of treatment trials have not all been negative. Dialectical behaviour therapy had a substantial initial impact on repetition of self-harm, although this did not persist in a treatment study of female patients with borderline personality disorders (Linehan et al. 1991) and depot flupenthixol reduced repetition in frequent self-harmers (Montgomery, 1987). A recent trial of brief psychotherapy was more effective than treatment as usual (Guthrie et al. 2001) and a meta-analysis of trials of problem-solving therapy suggested beneficial effects for depression, hopelessness and specific problems (Townsend et al. 2001).