Objectives: To describe the hospital-treated prevalences for repeat deliberate self-poisoning (RDSP) and the demographic characteristics of the RDSP group, and to compare the RDSP and non-RDSP groups.
Design: Prospective longitudinal cohort study, with a one- to four-year follow-up.
Setting: The Hunter Area Toxicology Service (HATS), a regional toxicology treatment centre in New South Wales.
Subjects: 1238 consecutive DSP patients referred to hospital, 1992-1994, with follow-up through 1995.
Outcome measures: Deliberate self-poisoning (DSP) admissions within one year (RDSP-1), within six months (RDSP-6m), and within 28 days (RDSP-28d) of any other DSP admission by the same patient; length of stay; demographic characteristics; and drugs ingested.
Results: 175 patients (14.1%) repeated DSP during the study; 165 (13.3%) were classified as RDSP-1, giving a patient prevalence of hospitalisation in the range of 14.6 to 20.7 per 100,000 per year. Fifty-six RDSP-28d patients (33.9% of RDSP-1) accounted for 49.8% of the RDSP-1 admissions, and 123 RDSP-6m patients (74.5% of RDSP-1) accounted for 83.5% of RDSP-1 admissions. For RDSP-1, the male:female ratio was 1:1.9, with 35.7% unemployed, 29.1% pensioners and 15.8% married or in de facto relationships. RDSP-1 patients had a shorter length of stay (3 h), which was not clinically important. RDSP was more likely for the 25-34 years age group (odds ratio [OR], 2.24; 95% confidence interval [CI], 1.17-4.29) and the 35-44 years age group (OR, 2.12; 95% CI, 1.02-4.39) than the 10-18 years group, and more likely for women than men (OR, 1.69; 95% CI, 1.17-2.46). Being married/de facto reduced the risk for repetition (OR, 0.55; 95% CI, 0.31-0.96) compared with being single. Medications indicated for psychiatric illness were most commonly used for DSP.
Conclusions: Many patients who repeat DSP do so after a very brief interval and account for a disproportionate number of hospitalisations. Availability of psychiatric medications for DSP patients is a possible area of intervention.