Comment on “Is It Safe to Provide Abortion Pills Over The Counter? A Study on Outcome Following Self-Medication with Abortion Pills”
We read with interest the article “Is it safe to Provide Abortion Pills over the Counter? A Study on Outcome Following Self-Medication with Abortion Pills” by Dr. K. Nivedita and Dr. Fatima Shanthini  and we agree that studying the implications of self-administration of abortion pills is a worthy research aim. However, the evidence presented does not support the authors’ conclusion that “unsupervised medical abortion can lead to increased maternal morbidity and mortality”.
The authors analysed the case sheets of 40 hospitalized women who reported self-administration of abortion pills, but this analysis lacks a meaningful denominator. Safety and efficacy cannot be known without quantifying the number of women in the catchment area who self-administered abortion pills and had complete and uncomplicated procedures and never sought follow up care. In fact, the data presented could indicate a very low failure and complication rate, depending on the denominator. By collecting data from only those women who present at the facility, the study introduces selection bias and will not be representative of the experiences of the wider population of women who self-administer abortion pills.
Tellingly, there were no deaths or hysterectomies in the series reported and 17.5% of the women analysed required either no treatment of simply additional misoprostol. This indicates that even those women who ended up in the hospital did not suffer major morbidity.
The author’s assertion of an increase in maternal morbidity and mortality is unjustified as there is no baseline from which to calculate such an increase. In fact, it is plausible and perhaps likely that the expanded availability of abortion pills over the counter reduces maternal morbidity and mortality by offering a safer alternative than invasive or violent methods.
The study also suffers from a number of ascertainment issues that make the results difficult to interpret. The authors rely upon women’s self-report of administering “abortion pills” without any validation of what specific pills, doses, and regimens were used. There is also no systematic collection of other remedies attempted. Therefore, the complications seen cannot be directly attributed to the use of mifepristone and misoprostol obtained at pharmacies.
The available data are insufficient to make any conclusions regarding the societal and policy implications of the availability of abortion pills in pharmacies. We cannot rely on this study to inform our understanding of the safety of abortion medications for self-use in India or elsewhere.
Laura J. Frye, MPH
Beverly Winikoff, MD
Access the Letter.