Re: Spontaneous Publishing and Academic Miscarriages (SPAM)
I appreciate Professor Healy's concern that the risks of SSRI's
regarding birth
defects might be unfairly minimized. Certainly, if you are the mother of
a
child with a birth defect, regardless of the cause, it is a serious
concern and
not to be dismissed lightly. However, I did not intend for my comments to
be
interpreted as minimizing the risk; rather, I intended to place the risks
in
context both in terms of the magnitude of the risk (which is estimated to
be
relatively low compared to other known teratogens such as isotretinoin
which
can affect more than 20% of exposed pregnancies) and in terms of the
concomitant risks of non- or under-treatment.
Professor Healy mentions our California data on pregnancy outcomes
with
prenatal exposure to paroxetine. This is a perfect example of the
difficulty of
drawing conclusions from studies with inadequate sample sizes. Our data on
paroxetine were drawn from an ongoing open cohort study with an increasing
but still extremely small sample size. Preliminary results were published
in
abstract form several years ago (Unfred et al, Teratology 63:321-4, 2001)
and updated results were provided for and included in the meta-analysis
recently published by Wurst et al (Birth Defects Res Part A, Clin Molec
Teratol,
epub ahead of print, 2009). These same data were also included in a
published paper on the cumulative experience with paroxetine and cardiac
defects across several teratology information services (Einarson et al, Am
J
Psychiatry.165:749-52, 2008). As evidenced by the very wide confidence
interval and the lack of statistical significance as shown for our data on
cardiac defects in the Wurst et al paper, the contribution of this data to
a
better understanding of the relationship between paroxetine and cardiac
defects we deemed most appropriately evaluated in the context of this
comprehensive meta-analysis.
Finally, I wish to reiterate that my comments in this editorial and
elsewhere,
consistent with the joint guidelines recently released by the APA and
ACOG,
are intended to support the most appropriate treatment of the mother and
fetus in each individual situation, recognizing that there may be risks of
some treatments, as well as risks of inappropriate, under- or non-
treatment.
Competing interests:
See list of sources of grant funding
from original editorial
Rapid Response:
Re: Spontaneous Publishing and Academic Miscarriages (SPAM)
I appreciate Professor Healy's concern that the risks of SSRI's
regarding birth
defects might be unfairly minimized. Certainly, if you are the mother of
a
child with a birth defect, regardless of the cause, it is a serious
concern and
not to be dismissed lightly. However, I did not intend for my comments to
be
interpreted as minimizing the risk; rather, I intended to place the risks
in
context both in terms of the magnitude of the risk (which is estimated to
be
relatively low compared to other known teratogens such as isotretinoin
which
can affect more than 20% of exposed pregnancies) and in terms of the
concomitant risks of non- or under-treatment.
Professor Healy mentions our California data on pregnancy outcomes
with
prenatal exposure to paroxetine. This is a perfect example of the
difficulty of
drawing conclusions from studies with inadequate sample sizes. Our data on
paroxetine were drawn from an ongoing open cohort study with an increasing
but still extremely small sample size. Preliminary results were published
in
abstract form several years ago (Unfred et al, Teratology 63:321-4, 2001)
and updated results were provided for and included in the meta-analysis
recently published by Wurst et al (Birth Defects Res Part A, Clin Molec
Teratol,
epub ahead of print, 2009). These same data were also included in a
published paper on the cumulative experience with paroxetine and cardiac
defects across several teratology information services (Einarson et al, Am
J
Psychiatry.165:749-52, 2008). As evidenced by the very wide confidence
interval and the lack of statistical significance as shown for our data on
cardiac defects in the Wurst et al paper, the contribution of this data to
a
better understanding of the relationship between paroxetine and cardiac
defects we deemed most appropriately evaluated in the context of this
comprehensive meta-analysis.
Finally, I wish to reiterate that my comments in this editorial and
elsewhere,
consistent with the joint guidelines recently released by the APA and
ACOG,
are intended to support the most appropriate treatment of the mother and
fetus in each individual situation, recognizing that there may be risks of
some treatments, as well as risks of inappropriate, under- or non-
treatment.
Competing interests:
See list of sources of grant funding
from original editorial
Competing interests: No competing interests