Abstract: | Summary Four hundred people who came to a hospital casualty departmentduring one year after taking an overdose were randomly assignedto an experimental task-centred casework service or to the routineaftercare service. Half the sample was followed up four monthsafter overdose and the remainder 18 months after it. Clientsin the experimental group were more satisfied with the servicethey had received and showed more immediate improvement in socialproblems. A task-centred method (agreement on target problems,work on tasks and adherence to a time-limit) proved feasiblein about half the experimental cases. |