RESUMO
Nearly 20% of the 208 million pregnancies that occur annually are aborted. More than half of these (21.6 million) are unsafe, resulting in 47,000 abortion-related deaths each year. Accurate reports on the prevalence of abortion, the conditions under which it occurs, and the experiences women have in obtaining abortions are essential to addressing unsafe abortion globally. It is difficult, however, to obtain accurate and reliable reports of attitudes and practices given that abortion is often controversial and stigmatized, even in settings where it is legal. To improve the understanding and measurement of abortion, specific considerations are needed throughout all stages of the planning, design, and implementation of research on abortion: Establishment of strong local partnerships, knowledge of local culture, integration of innovative methodologies, and approaches that may facilitate better reporting. This paper draws on the authors' collaborative research experiences conducting abortion-related studies using clinic- and community-based samples in five diverse settings (Poland, Zanzibar, Mexico City, the Philippines, and Bangladesh). The purpose of this paper is to share insights and lessons learned with new and established researchers to inform the development and implementation of abortion-related research. The paper discusses the unique challenges of conducting abortion-related research and key considerations for the design and implementation of abortion research, both to maximize data quality and to frame inferences from this research appropriately.