STUDY QUESTION

Is maternal stress following the death of a close relative before or during pregnancy associated with the risk of infertility in daughters?

SUMMARY ANSWER

Compared with unexposed women, women whose mothers had experienced bereavement stress during, or in the year before, pregnancy had a similar risk of infertility overall, but those exposed to maternal bereavement during the first trimester had a higher risk of infertility.

WHAT IS KNOWN ALREADY

Animal studies have shown that prenatal maternal stress results in reduced offspring fertility. In humans, there is evidence that girls who have been prenatally exposed to stress have a more masculine behaviour and a slight delay in having their first child.

STUDY DESIGN, SIZE AND DURATION: This population-based cohort study, included 660 099 females born in Denmark between 1 January 1973 and 31 December 1993 to mothers of Danish origin and with at least one living relative in the exposure window, and followed the women through 31 December 2011.

PARTICIPANTS/MATERIALS, SETTING, METHODS: Overall, 13 334 women (2.0%) were considered prenatally exposed to stress because their mother had lost a spouse/partner, a child, a parent, or a sibling during pregnancy or in the year before conception. Infertility was defined as any record of infertility treatment or diagnosis of female infertility. We considered the date of onset as the date of the first appearance of any such record. The association between exposure and outcome was examined using hazard ratios (HR) with 95% confidence intervals (CI).

MAIN RESULTS AND THE ROLE OF CHANCE

Based on our definition, 40 052 (6.5%) women were infertile in the follow-up period (median age at the end of follow-up: 26.7 years, maximum age: 39 years). Overall, prenatal exposure to maternal stress was not associated with risk of infertility (adjusted HR = 1.04 [CI: 0.95-1.14]). However, women prenatally exposed during the first trimester had a higher estimated risk (adjusted HR = 1.40 [CI: 1.05-1.86]). These findings were consistent in subgroups defined by the relationship of the mother to the deceased and in several sensitivity analyses, including a sibling-matched analysis, and in analyses restricted to women who were married or cohabitating with a man, or to women born at term.

LIMITATIONS, REASONS FOR CAUTION: We did not have a direct measure of stress, but bereavement due to death of a close relative is likely to be very stressful. We based the timing of exposure on the date of the death of the family member, although the stress may well have started earlier. Infertility was also defined indirectly, and many women in the cohort were too young at the end of the follow-up to have been diagnosed. However, misclassification of the outcome was most likely non-differential, and the similar results from all sensitivity analyses suggest that it is unlikely that the effect observed in first trimester exposure would be due to chance.

WIDER IMPLICATIONS OF THE FINDINGS

Prenatal exposure to maternal stress in the first trimester may affect the later fecundity of daughters.

STUDY FUNDING/COMPETING INTERESTS: This study was supported by a grant from the European Research Council (ERC-2010-StG-260242-PROGEURO) to the ProgEuro project (http://progeuro.au.dk). O.P.-R. is partly supported by a fellowship from Aarhus University and a travel grant from Oticon Fonden. The authors report no conflict of interests.