[How to care for homeless mentally ill men in the community - a randomised intervention study]

The aim of the present study was to clarify whether the mediation of homeless persons identified as mentally ill into the sociopsychiatric or the help-seeking system can be facilitated by a short-term intervention.

Probands with at least one psychiatric diagnosis according to ICD-10 (73 % of the sample) were randomized to the intervention group (manualized procedure with five brief contacts) or to the control group (no intervention). The course of the contacts and the further help-seeking behavior after 4 and 8 months were registered.

The very sporadic participation of the patients in the therapeutic discussion permitted a manualized procedure in individual cases only. The establishment of contact was very successful, but the problem-centered procedure only within limits. The intergroup comparison revealed no fundamental differences in the utilization of offers of support during the course of the study. Overall, however, the number of users tended to increase, with an increased contact frequency being recorded especially among those participating in therapeutic discussions.

Only a relatively small proportion of the clientele can be reached through short-term behavior therapy. Fundamentally a low-profile, primarily outreach-based offer founded on longer-term contact appears to be more promising.