In a time when many people view psychiatric institutions mainly as evils which should be defeated, it might seem unnecessary to examine the present knowledge of milieu therapy and to search for lacunae which should be filled through new scientific activities. Several reasons make it still worthwhile. Firstly, we do not know how to cure all patients with a functional psychosis. New benefits from a more effective milieu therapy would be very welcome. Secondly, in spite of an extensive out-patient service, some patients still have to be treated in in-patient units. We therefore need knowledge about what types of milieu will promote the most favourable outcome for these patients, both in the short and the long run. Thirdly, a large group of psychotic patients will, in the future, be attending day hostels, half-way houses or collectives. Our need for knowledge about the positive and the negative milieu factors in these environments is indisputable. Lastly, all research concerning milieu treatment may be viewed as part of the more general study of the influence of environmental factors upon the course of functional psychoses. It may therefore give us new information about precipitating and possible aetiological factors.

From the title, one would expect this paper to limit it self only to long-term milieu treatment. It does not, and the reason is that to find the optimal duration of milieu therapy for different groups of psychotic patients, we have to be sure that we are studying milieus where the milieu therapy really works. At present it is this last question we are trying to solve.

In modem treatments of functional psychoses, we almost always use a combination of different therapeutic modalities, such as pharmacotherapy, individual psychotherapy, family therapy etc. This paper will deal with milieu therapy as part of such a combination of treatment methods. The possibilities of substituting pharmacotherapy with milieu therapy, will not be discussed.

The idea of a therapeutic milieu

The influence of the social environment upon the course of functional psychoses, has been known for a long time. Since World War II, this knowledge has been further developed through series of observations from hospitals, rehabilitation centres, families and in the community (Wing, 1978).