In an article in an earlier issue of this journal (Beenackers, 1982a), we showed that not all people who approach an institution for mental health care to ask for help stand an equal chance of indeed recieving the help they are seeking. Certain categories of people stand a relatively good chance whilst other categories stand a relatively poor chance. In other words, the one is subject to positive discrimination, and the other to negative discrimination. We then gave an example of this type of discrimination using the findings of our own research: this was the positive discrimination of people seeking help with relationship problems. We restricted ourselves in the article to pointing out that there was indeed discrimination, but did not discuss the question of how such discrimination in the health care sector can be avoided. This is the issue in this article. The problem is viewed first in the light of the literature, and subsequently on the basis of our own research. This study is at the same time an example of the use of program evaluation, i.e. a method of evaluation whereby a program is treated as a whole and evaluated and if necessary modified at set intervals. This is done using certain predetermined criteria which the program is required to meet. In the study in question, the aim of the criteria was to eliminate the discrimination that had been determined in the past and that was described in the preceding article. This article explains how the task was approached and what results were achieved.