The ethics of cluster-randomized trials requires further evaluation: a refinement of the Ottawa Statement.

OBJECTIVES: The Ottawa Statement is the first guidance document for the ethical and scientific conduct of cluster-randomized trials (CRTs). However, not all recommendations are straightforward to implement. In this paper we will reflect in particular on the recommendation on identifying human research subjects and the issue to what extent the randomization process should be disclosed if there is a risk of contamination.

STUDY DESIGN AND SETTING: The Ottawa Statement was thoroughly evaluated within a multidisciplinary research team, consisting amongst others of epidemiologists and ethicists.

RESULTS: Patients in a CRT may also be considered as research subjects if they are indirectly affected by the studied interventions in a CRT. Second, health care workers are research subjects in CRTs but have a different moral status compared with ordinary research participants. This different status has implications for withdrawal and the choice of the primary objective. Third, modified informed consent for CRTs may be obtained when researchers can demonstrate that disclosure of the randomization process would affect the validity of a CRT.

CONCLUSION: Recommendations of the Ottawa Statement on identifying the research subject and providing informed consent can and should be refined.