Statistics from Altmetric.com
QUESTION: In patients with deliberate self poisoning, does brief psychodynamic interpersonal therapy (PIT) reduce suicidal ideation, severity of depression, and further episodes of self harm and increase patient satisfaction?
A university hospital emergency department in Manchester, UK.
119 adults who were 18–65 years of age (mean age 31 y, 55% women), presented with an episode of deliberate self poisoning, lived in the catchment area, were registered with a general practitioner, and did not need inpatient psychiatric treatment. 80% of patients completed 6 month assessments.
After stratification by history of self harm, patients were allocated to 4 fifty minute sessions of PIT (n=58) or to usual care (n=61). PIT consisted of identifying and helping to resolve interpersonal difficulties that caused or exacerbated psychological distress. The therapy was described in a standardised manual.
Main outcome measures
Suicidal ideation (Beck Scale for Suicidal Ideation). Secondary outcomes were depression symptoms (Beck Depression Inventory), patient satisfaction (10 point scale, with higher scores indicating higher satisfaction), and further episodes of deliberate self harm.
Analysis was by intention to treat. After adjustment for baseline values, patients in the psychotherapy group had less suicidal ideation (p<0.005) and less severe depression (p=0.037) than patients in the usual care group (table⇓ ). The difference in depression scores was no longer statistically significant after adjustment for marital status. Patient satisfaction was higher in the psychotherapy group than in the usual care group (p=0.015) (table⇓). Unadjusted rates for repeated self harm were lower in the psychotherapy group than in the usual care group (p=0.009) (table⇓).
In adults who have deliberately poisoned themselves, 4 sessions of psychodynamic interpersonal therapy reduced suicidal ideation and further episodes of deliberate self harm and increased patient satisfaction.
Rates of hospital attendance after self harm are about 400 in 100 000 per year in the UK, and in people who have committed suicide, 1 in 4 attended hospital after a non-fatal act in the previous year. Under the circumstances, the evidence for the effectiveness of interventions is disappointing.1
Guthrie et al struggled with some familiar problems, and despite their best efforts, many exclusions and refusals occurred; in the end they included only 23% of presenting patients. We cannot be sure how generalisable their findings are, although patient baseline characteristics were typical for the UK.
Final numbers were respectable but were none the less relatively small, and possibilities exist for bias. For example, the treatment and control groups differed in marital status and past psychiatric history. The authors adjusted for some potential confounders in their analysis, but not all. The apparently large effect of psychological therapy on the repeated self harm rate needs to be viewed with caution.
The results of this trial are encouraging because they add to the evidence that brief psychological therapies improve outcomes after self harm.2 Those who are sympathetic will accept this as further evidence that patients with such a high burden of problems and risk of suicide should be offered treatment. Promising therapies (like the one evaluated here) are brief, have a strong focus on practical problem solving and interpersonal difficulties, and are delivered in a format that patients find acceptable.
For the sceptical, the evidence remains less than rock solid. For them, we still need large multicentre trials to test the real world effectiveness of psychological therapies before we can argue for their routine inclusion in clinical services.
The intervention group received psychotherapy that was based on a model developed by Hobson. The therapy is described in a standardised manual, and involves identifying and helping to resolve interpersonal difficulties that cause or exacerbate psychological distress. Therapy was delivered by nurse therapists in the patients� homes. Four weekly 50 minute sessions were offered. Treatment fidelity and adherence was ensured by weekly supervision, audiotaping of interviews, and use of a standardised rating scale.
The control group received usual care, which consists of an assessment by a casualty doctor or junior psychiatrist in the emergency department. Approximately one third of patients are routinely referred for follow up as a psychiatry outpatient, a few are referred to addiction services, and the rest are advised to consult their own general practitioner. No patients are routinely referred to psychotherapy or psychology services.
1 Hobson RF. Forms of feeling: the heart of psychotherapy. London: Tavistock Publications, 1985.
2 Shapiro DA, Startup MJ. Raters� manual for the Sheffield psychotherapy rating scale. Sheffield: MRC/ESRC Social and Applied Psychology Unit, University of Sheffield, 1990.
The manual cited in reference 2 is available from:
The Psychological Therapies Research Centre
School of Psychology
University of Leeds
17 Blenheim Terrace
Leeds LS2 9JT, UK
E-mail Fax +44 (0)113 233 1956
Sources of funding: North West Regional Health Authority and NHS Research and Development Levy.
For correspondence: Dr E Guthrie, School of Psychiatry and Behavioural Sciences, University of Manchester, Rawnsley Building, Manchester Royal Infirmary, Manchester M13 9WL, UK. Fax +44 (0)161 273 2135.
A modified version of the abstract and commentary appears in Evidence-Based Medicine.
↵† Information provided by author.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.