In their advisory report, the Committee of the Health Council of the Netherlands recommended to conduct the trials at several locations and in various cities simultaneously (Health Council, 1995). The study was, therefore, set up as a multicenter trial, both in order to recruit the necessary number of study subjects within a reasonable time-frame, to be able to practically manage the study and the treatment sites involved, and to provide a better basis for the generalization of the study findings. In addition, from the perspective of risk of public nuisance and manageability, large numbers of patients at one location had to be avoided.

For reasons of manageability and cost-effectiveness considerations, the optimal capacity for each treatment site was determined at 40-50 participants who would simultaneously receive co-prescribed heroin. Within each participating treatment site, the largest number of participants simultaneously receiving co-prescribed heroin would occur in the course of phase IIb of the study (see paragraph 2.2.1). Based on the power calculations and expected drop out rates, it was originally anticipated that a maximum of approximately 440 subjects would simultaneously receive co-prescribed heroin during this phase. Following the protocol amendment with regard to the removal of a treatment condition in the injectable heroin trial (see chapter 4), this number was anticipated to amount to 360 subjects. Hence, a total of approximately eight treatment units were required in the study.

In order to qualify for participation in the study, each city which showed interest, had to meet a number of selection criteria, which included the availability of a sufficiently large number of addicts who qualify for the study, treatment facilities that would qualify, and willingness to co-finance the study. The criteria to be met by the participating treatment sites included the existence of a methadone maintenance program for at least one year, and the willingness and sufficient equipment, in terms of staff and facilities, to comply to all aspects of the study protocol. Six cities met these selection criteria. The eight treatment sites eventually selected for the study were situated in Amsterdam (two units), Groningen, The Hague, Heerlen, Rotterdam (two units), and Utrecht.