Table 1 Considerations for conducting research in oppressed populations

Health Promoting

For clinical effectiveness interventions, interventions should be health promoting and address a clearly population relevant illness. There should be a clear plan for how research findings will assist community, based on a priori agreement on how findings will be used. This requires active engagement with target groups and plans for how findings should be interpreted with target group members.

Opportunities to avoid exposures should be afforded to participants

Exposures may include exposures to the target diseases or to human rights violations. This should include education (e.g. condom use, needle provision and needle cleaning); assistance with known human rights promoting activities or possible escape; and, the avoidance of remaining in exposed status.

Planned efforts to determine the human rights status of a particular population

This may include formative research and background reviews of human rights NGO literature.

Community participation commensurate with level of political oppression

There are few reliable ways to measure oppression. Due to the large geographical and cultural heterogeneity, there may not be a unified voice and so we need minimum standards for Community Advisory Boards (CABs).

CABs must include mixed levels of education, oppressive situations, age and gender

They should be able to inform on the importance of the research and therefore, acceptability. CABs may need to be outside target areas (i.e. Burma)

The more oppression, the more reason for research required

Assuming researchers cannot improve oppressive setting, the conducting research may be placing participants or colleagues at risks. Oppressive conditions may make population 'captive'

HR violations should be captured

Good research captures all health outcomes determined of interest in clinical event forms. In certain conditions we can anticipate HR violations. Case report forms should inquire of HR violations. This may allow for subgroup analyses.

Report an Ethical methodology

Ethics and human rights issues are not black or white issue (ethical or not ethical), but require explanation to determine challenges that exist in the research settings. It may well be appropriate to advocate for a limitations section of the article that addresses what could not be accomplished. This could be advocated by the Consolidated Standards for Reporting of Randomized Trials Group (CONSORT), a group that advocates minimum standards of reporting clinical research. It is only through acknowledgement of these issues that the field can be advanced.

Measure long-term impact, for informing "customary research"

We need to determine if the research has improved conditions, or worsened it? Has 'Western' research improved or reduced the quality/expectations of indigenous research? Finally, we need to determine if post-trial access was accomplished and what care was provided for post-trial injuries related to participation.