Background: There have been numerous studies investigating the association between omega-3 fatty acids (FAs) and depression, with mixed findings. We propose an approach which is largely free from issues such as confounding or reverse causality, to investigate this relationship using observational data from a pregnancy cohort.
Methods: The Avon Longitudinal Study of Parents and Children (ALSPAC) cohort collected information on FA levels from antenatal blood samples and depressive symptoms at several time points during pregnancy and the postnatal period. Conventional epidemiological analyses were used in addition to a Mendelian randomisation (MR) approach to investigate the association between levels of two omega-3 FAs (docosahexaenoic acid (DHA) and eicosapentaenoic acid (EPA)) and perinatal onset depression, antenatal depression (AND) and postnatal depression (PND).
Results: Weak evidence of a positive association with both EPA (OR=1.07; 95% CI: 0.99-1.15) and DHA (OR=1.08; 95% CI: 0.98-1.19) with perinatal onset depression was found using a multivariable logistic regression adjusting for social class and maternal age. However, the strength of association was found to attenuate when using an MR analysis to investigate DHA.
Limitations: Pleiotropy is a potential limitation in MR analyses; we assume that the genetic variants included in the instrumental variable are associated only with our trait of interest (FAs) and thus cannot influence the outcome via any other pathway.
Conclusions: We found weak evidence of a positive association between omega-3 FAs and perinatal onset depression. However, without confirmation from the MR analysis, we are unable to draw conclusions regarding causality.
Keywords: ALSPAC; Depression; Mendelian randomisation; Omega-3; Perinatal.
Copyright © 2014 The Authors. Published by Elsevier B.V. All rights reserved.