PART II

STUDY DESIGN AND METHODS

Chapter 2

Methods

2.1 Study population

2.1.1 Target population

According to the advisory report of the Health Council of the Netherlands (1995), the study should be directed to the treatment of severely addicted heroin users who have not (sufficiently) responded to currently available medical interventions. To qualify for participation, patients are required to have repeatedly and unsuccessfully participated in treatment programs, aimed at - by means of medication in adequate dose levels and dose frequency - stabilization and the prevention of relapse (Health Council, 1995). The Dutch government added to this description that the study should be restricted to "older patients with a long addiction career, whose psychosocial condition is without perspective" (Ministry of Health, Welfare and Sports, 1995). According to the Committee of the Health Council, other categories of addicts would in principle qualify for a medical- scientific study with heroin as well. These addicts may be "difficult or impossible to reach", not (yet) as severely addicted to heroin, have not "repeatedly but unsuccessfully" participated in treatment programs aimed at relapse prevention and stabilization, and may still have a genuine possibility to stop their heroin use. Although neither the literature, nor past experiences contain convincing arguments against the prescription of heroin to this group in a controlled trial, the Committee of the Health Council argued that current knowledge provides insufficient arguments to justify, in ethical terms, the recruitment of patients who have not first tried the currently available treatment modalities in the first clinical trial with medically prescribed heroin.
Hence, the target population has been operationally defined in the study protocol (CCBH, 1997, 1999a) as the sub-population of chronic heroin addicts who have been treated repeatedly but unsuccessfully in methadone maintenance programs. In addition, the decision to limit the study population to this group of patients was based on the following considerations: (1) a considerable number of patients in methadone maintenance programs fit the description of the target group; these patients have been described earlier in this report as "not integrated" and "extremely problematic", (2) the problems of methadone maintenance patients in the Netherlands are quite similar to those of heroin users currently not in contact with the treatment system (Eland-Goossensen, 1997); this, in turn, may be related to the "low threshold" character of most Dutch methadone maintenance programs, (3) the registration systems of the methadone programs provide a clear sampling frame for the selection of study subjects, and (4) a combination of data from the methadone and the heroin registration systems can prevent the prescription of double (methadone and heroin) dosages. With regard to the validity of the study findings, the decision to limit the study population to treatment-resistant methadone maintenance patients may increase the internal validity.