The study by David Reardon and associates1 is seriously flawed and presents misleading conclusions. The authors compared 2 groups of women as if the only difference between them was whether or not they had had an abortion. However, many factors were unknown, for example whether the women in the control group were in stable relationships and whether they had planned their pregnancies. The women in the case group had not had a psychiatric admission in the previous year, but the reasons for their abortions were unknown. These women may have been troubled before they got pregnant. Women who are mentally ill, under stress and without social supports constitute the group most likely to need an abortion (because they cannot cope with a pregnancy or having a child), but these women will also have the most difficulty coping with an abortion.2 The fact that the most common diagnosis for those admitted (after their abortions) was psychotic depression strongly suggests that some of these women had a previous history of mental illness.
Reardon and associates bolster their alarmist conclusions by quoting a select group of studies that have supposedly also found major negative consequences, ignoring the thorough reviews by both the American Psychological Association3 and the Surgeon General of the United States (Koop CE. Surgeon General's report on the public health effects of abortion. Unpublished report to Congress, 1989), who found no evidence of harm.
It is surprising that CMAJ would publish such poor-quality research.
Gail Erlick Robinson Professor of Psychiatry University of Toronto Toronto, Ont.