Underreporting of induced abortion in survey research is a ubiquitous problem. The use of an indirect interview technique in which questions were asked about abortion in the context of unwanted pregnancy was described earlier as holding promise for increasing the response rate. This report reviews the mixed results from multicountry studies that used the indirect technique. Exploratory qualitative studies are needed to identify a setting-specific context for discussing abortion. Because the subject of induced abortion is inherently sensitive, survey measurement of this practice is less precise than that of other, less controversial maternal health-care practices. This lack of precision should not deter the pursuit of the study of this critically important public-health-care concern.