Defining fair and acceptable randomisation procedure in trials targeting vulnerable populations: Qualitative evidence from the POWER trial in Cameroon

Abstract Important ethical issues have been consistently highlighted in randomised controlled trials (RCTs). Although RCTs are widely used in non-clinical health research, there is little research on participants' perceptions of the fairness and transparency of RCTs. Data were collected as part of the POWER trial, which aims to test the effectiveness of health shock prevention for HIV prevention among women who engage in commercial and transactional sex in urban Cameroon. This qualitative study was carried out in 2 phases. In the pre-randomisation phase (phase 1), we conducted 25 focus groups and 8 in-depth semi-structured interviews to determine the most acceptable randomisation strategy for our study participants. In the post-randomisation phase (phase 2), 41 in-depth semi-structured interviews were conducted with participants to assess their perceptions and satisfaction with their group allocation status (e.g. treatment or control). We found that participants understood the rationale for randomisation of the intervention and were satisfied with the randomisation strategy. This latter finding was mainly due to i) the involvement of participants in the chosen randomisation method and ii) the fact that the control group would receive the intervention at the end of the study. It is important that researchers conduct similar studies before designing RCTs in vulnerable populations to minimise ethical issues related to the conduct of RCTs. Conducting pre- and post-randomisation qualitative research is an effective approach to improve RCT design and to assess and address potential problems with RCTs.