This article discusses the challenges of applying ethical principles in epidemiological research (including respect for people, non-maleficence, beneficience, and justice) to population research on domestic violence, an area where poorly designed research could put women in violent relationships at substantial risk. The article draws on experiences from the WHO Multi-country Study on Women’s Health and Domestic Violence Against Women, from the International Research Network on Violence Against Women, and one author's research in Nicaragua. Examples from a number of developing and developed countries are discussed.


Concerns include ensuring the safety of respondents in a context in which many live with their abuser, protecting confidentiality when breaches could provoke an attack, and ensuring the interview process is affirming and does not cause distress. The article concludes that the inherent risks involved in such research can only be justified if, during interviews, information on available services is provided and immediate referrals are made when necessary, if high-quality data are obtained (which requires minimising the under-reporting of violence), and if findings are used to raise awareness of, and improve services for, women who experience domestic violence.

(Free registration with The Lancet is required to view this article.)