I thank Mintzes and associates for their comments. I should have stated that neither study by Nakhai-Pour and colleagues1 or Motherisk2 could make any definitive conclusions, although I stand by my comment that “if there is a true increased risk of spontaneous abortion caused by gestational use of antidepressants, it is very small.”3
The small risk referred to was the odds ratio (OR) of 1.68; it is well known among epidemiologists that ORs of less than 2 are problematic.3 Observational studies have inherent biases that cannot be removed with certainty. Although the authors addressed some of the issues, it is impossible to identify and remove the influence of all factors, since many remain unknown; there is still a possibility of a false-positive result.
In my commentary, I focused on antidepressants and spontaneous abortion and discussed Nakhai-Pour and colleagues’ study. I do not feel it is appropriate that Mintzes and associates mention other outcomes, quoting from a review that is no longer current.5 It is also inappropriate to talk about treatment efficacy, since I did not mention this in my commentary. At Motherisk, we conduct research to evaluate the safety and risk of a particular drug during pregnancy only, not to examine the efficacy of treatment. The latter should be left to the prescribing physician, who is frequently faced with an already pregnant woman taking an antidepressant and requiring treatment. Is it reasonable to advise such a patient to stop treatment and switch to psychotherapy?
I agree that women facing depression during pregnancy need accurate and unbiased information on the use of anti-depressants. After reviewing all the evidence, each woman, with her physician, needs to make an informed decision about whether to take the medication.