Dr Brown and Colleagues Reply

See letter by Shi and Li and article by Brown et al

This work may not be copied, distributed, displayed, published, reproduced, transmitted, modified, posted, sold, licensed, or used for commercial purposes. By downloading this file, you are agreeing to the publisher’s Terms & Conditions.

Dr Brown and Colleagues Reply

To the Editor: We thank Shi and Li1 for their comments on our meta-analysis2 regarding the association between antenatal selective serotonin reuptake inhibitor (SSRI) exposure and child autism. Methodological standards for meta-analyses note that, because the unit of observation is the study and not the article, care should be taken to include study subjects only once.3 Shi and Li comment on our decision to exclude, from the quantitative synthesis, 2 of 3 Danish studies with overlapping data sources and study periods.4-6 We address their comments below.

First, Shi and Li query why we included the retrospective cohort study by Hviid et al4 over that of Sørensen et al,5 which had a larger sample size. While we agree that statistical power is critical in the study of rare outcomes, our goal was to disentangle the effect of antenatal SSRI exposure from that of maternal mental illness. We therefore prioritized study quality—specifically, control for confounding by indication—as the criterion for inclusion in the quantitative synthesis. In addition to sociodemographic variables, Hviid et al controlled for maternal mental illness and other prescription drug use. Missing from Sørensen and colleagues’ multivariable models were maternal affective disorders and other prescription drug use.

Second, Shi and Li note that a previous meta-analysis7 rated Hviid et al5 and Sørensen et al6 as having "high quality" using the Newcastle-Ottawa scale.8 However, this scale does not explicitly address confounding by indication. The Systematic Assessment of Quality in Observational Research (SAQOR) tool that we utilized specifically assesses whether mental illness, other prescription drugs, and all other confounders are controlled for.9 This most likely explains the difference in quality ratings between the scales.

Third, Shi and Li question our exclusion of Gidaya et al,6 since this study would have been included in pooled estimates for case-control studies, rather than cohort studies along with Hviid et al.4 We maintain that inclusion of both studies would have resulted in "duplication" of Danish study subjects3; any bias in these data would have unduly influenced conclusions in a given analysis. We prioritized Hviid et al because of their superior control for confounding by indication; their mental illness variables included not just depression, anxiety, and schizophrenia but other potential indications as well, such as personality disorders and eating disorders.10,11

Fourth, Shi and Li propose that an alternative approach would be to conduct sensitivity analyses wherein the chosen study is replaced with an excluded study. However, given that both Sørensen et al5 and Gidaya et al6 had incomplete control for confounding by indication, the results would be less robust. In fact, Hviid et al4 was the only study, of 8 evaluated, that had adequate control for distorting influences.

In summary, we emphasize the importance of deliberate consideration of confounding by indication in studies of medication use in pregnancy and the use of appropriate quality assessment tools, such as SAQOR, which allow authors to do so in meta-analyses. Such robust approaches are critical in the creation of appropriate clinical recommendations for women with mental illness in pregnancy.

References

1. Shi Z-Y, Li Y-M. Association between antenatal exposure to selective serotonin reuptake inhibitors and autism: a need for further analysis. J Clin Psychiatry. 2017;78(7):e839.

2. Brown HK, Hussain-Shamsy N, Lunsky Y, et al. The association between antenatal exposure to selective serotonin reuptake inhibitors and autism: a systematic review and meta-analysis. J Clin Psychiatry. 2017;78(1):e48-e58. PubMed doi:10.4088/JCP.15r10194

3. European Network of Centres for Pharmacoepidemiology and Pharmacovigilence (ENCePP). ENCEPP Guide on Methodological Standards in Pharmacoepidemiology: Guidance on Conducting Systematic Reviews and Meta-Analysis of Completed Comparative Pharmacoepidemiological Studies of Safety Outcomes. London, UK: ENCePP; 2016.

4. Hviid A, Melbye M, Pasternak B. Use of selective serotonin reuptake inhibitors during pregnancy and risk of autism. N Engl J Med. 2013;369(25):2406-2415. PubMed doi:10.1056/NEJMoa1301449

5. Sørensen MJ, Grønborg TK, Christensen J, et al. Antidepressant exposure in pregnancy and risk of autism spectrum disorders. Clin Epidemiol. 2013;5(1):449-459. PubMed doi:10.2147/CLEP.S53009

6. Gidaya NB, Lee BK, Burstyn I, et al. In utero exposure to selective serotonin reuptake inhibitors and risk for autism spectrum disorder. J Autism Dev Disord. 2014;44(10):2558-2567. PubMed doi:10.1007/s10803-014-2128-4

7. Man KK, Tong HH, Wong LY, et al. Exposure to selective serotonin reuptake inhibitors during pregnancy and risk of autism spectrum disorder in children: a systematic review and meta-analysis of observational studies. Neurosci Biobehav Rev. 2015;49:82-89. PubMed doi:10.1016/j.neubiorev.2014.11.020

8. Wells GA, Shea B, O’ Connell D. The Newcastle-Ottawa Scale (NOS) for Assessing the Quality of Nonrandomized Studies in Meta-analyses. Ottawa, Ontario, Canada: Department of Epidemiology and Community Medicine; 2003.

9. Ross LE, Grigoriadis S, Mamisashvili L, et al. Quality assessment of observational studies in psychiatry: an example from perinatal psychiatric research. Int J Methods Psychiatr Res. 2011;20(4):224-234. PubMed doi:10.1002/mpr.356

10. Rinne T, van den Brink W, Wouters L, et al. SSRI treatment of borderline personality disorder: a randomized, placebo-controlled clinical trial for female patients with borderline personality disorder. Am J Psychiatry. 2002;159(12):2048-2054. PubMed doi:10.1176/appi.ajp.159.12.2048

11. Kaye WH, Nagata T, Weltzin TE, et al. Double-blind placebo-controlled administration of fluoxetine in restricting- and restricting-purging-type anorexia nervosa. Biol Psychiatry. 2001;49(7):644-652. PubMed doi:10.1016/S0006-3223(00)01013-1

Hilary K. Brown, PhDa,b

hilary.brown@wchospital.ca

Neesha Hussain-Shamsy, MHSa

Yona Lunsky, PhDb,c

Cindy-Lee E. Dennis, PhDa,b

Simone N. Vigod, MDa,b

aWomen’s College Hospital, Toronto, Ontario, Canada

bDepartment of Psychiatry, University of Toronto, Toronto, Ontario, Canada

cCentre for Addiction and Mental Health, Toronto, Ontario, Canada

Potential conflicts of interest: The authors have no financial relationships relevant to this article to disclose.

Funding/support: The first author is funded by a Canadian Institutes of Health Research Postdoctoral Fellowship (MFE-135409).

Role of the sponsor: The sponsor had no role in the planning or conduct of the original study or in the interpretation of the results.

J Clin Psychiatry 2017;78(7):e840

https://doi.org/10.4088/JCP.17lr11535a

© Copyright 2017 Physicians Postgraduate Press, Inc.