Exploring the Ethics of Observational Research: The Case of an HIV Study in Tanzania.

Background: Observational studies have generally been viewed as incurring minimal risk to participants, resulting in fewer ethical obligations for investigators than intervention studies. In 2004, the lead author (AN) carried out an observational study measuring sexual behavior and the prevalence of HIV, syphilis, and herpes simplex virus type 2 (HSV-2), among Tanzanian agricultural plantation residents (results reported elsewhere). This article uses an ethical lens to consider the consequences of the observational study and explore what, if any, effects it had on participants and their community.

Methods: Using a case study approach, we critically examine three core principles of research ethics-respect for persons/autonomy; beneficence/nonmaleficence; and distributive justice-as manifested in the 2004 observational study. We base our findings on three sources: discussions with plantation residents following presentations of observational research findings; in-depth interviews with key informants; and researcher observations.

Results: The observational research team was found to have ensured confidentiality and noncoercive recruitment. Ironically, maintenance of confidentiality and voluntary participation led some participants to doubt study results. Receiving HIV test results was important for participants and contributed to changing community norms about HIV testing.

Conclusions: Observational studies may act like de facto intervention studies and thus incur obligations similar to those of intervention studies. We found that ensuring respect for persons may have compromised the principles of beneficence and distributive justice. While in theory these three ethical principles have equal moral force, in practice, researchers may have to prioritize one over the others. Careful community engagement is necessary to promote well-considered ethical decisions.