Background Notwithstanding the need to produce evidence-based knowledge on medications for pregnant women, they remain underrepresented in clinical research. Sometimes they are excluded because of their supposed vulnerability, but there are no universally accepted criteria for considering pregnant women as vulnerable. Our aim was to explore whether and if so to what extent pregnant women are vulnerable as research subjects.

Method We performed a conceptual and empirical analysis of vulnerability applied to pregnant women.

Analysis A conceptual analysis supports Hurst's definition of vulnerability. Consequently, we argue that pregnant women are vulnerable if they encounter an identifiably increased likelihood of incurring additional or greater wrong. According to the literature, this increased likelihood could exist of four alleged features for pregnant women's vulnerability: (i) informed consent, (ii) susceptibility to coercion, (iii) higher exposure to risk due to lack of knowledge, (iv) vulnerability of the fetus.

Discussion Testing the features against Hurst's definition demonstrates that they all concern the same issue: pregnant women are only vulnerable because a higher exposure to risk due to lack of scientific knowledge comprises an increased wrong. Research Ethics Committees have a responsibility to protect the vulnerable, but a higher exposure to risk due to lack of scientific knowledge is a much broader issue and also needs to be addressed by other stakeholders.

Conclusions The only reason why pregnant women are potentially vulnerable is to the extent that they are increasingly exposed to higher risks due to a lack of scientific knowledge. Accordingly, the discussion can advance to the development of practical strategies to promote fair inclusion of pregnant women in clinical research.