background  So far little is known about the cost effectiveness of long-term psychotherapy. Research into this matter is needed since cost effectiveness is being used increasingly as a criterion for determining to what extent psychotherapy costs should be reimbursed.
aim  In this article we describe possible ways in which the cost effectiveness of long-term psychotherapy can be assessed at an early stage of treatment.
method  The effectivity research is described in four steps: setting up a model, estimating costs, evaluating the effects, and estimating the effects. Then we examine how this research can be used to substantiate the cost effectiveness of longterm psychotherapy.
results Whereas in clinical studies evidence is obtained largely from randomised trials, emphasis in cost effectiveness studies is on estimating the real costs and effects on the basis of practice. For this reason results of (existing) naturalistic studies may in fact provide an adequate framework for the construction of ecomomic models. In the past partial models of this kind have provided useful new policy information. Of course, randomised trials are needed too if the effectiveness psychotherapy is to be firmly established.
conclusion  Even if the outcomes of randomised trials are not yet available, naturalistic studies and economic models based on these studies may provide adequate evidence to support current health policy. This means that we can now start dealing pro-actively with increasing societal demands for evidence of the cost effectiveness of pychotherapy.