We read, with interest, the recent series of articles in BJOG1–5 pertaining to a comprehensive longitudinal study on the effects of varying levels of prenatal ethanol exposure (PNE) on an array of cognitive measures in 5-year-old offspring. The study revealed no significant PNE-induced differences for a number of outcome measures (IQ/general intelligence, executive function, sustained attention) associated with low (described as 1–4 drinks/week), moderate (5–7 drinks/week) or binge (>5 drinks in one session) drinking during early–mid pregnancy. The study did, however, reveal modest reductions in sustained attention for children whose mothers had consumed relatively higher levels of alcohol (>9 drinks/week).

Although we acknowledge that the authors advise that women abstain from drinking during pregnancy despite their findings, we would like to take issue with the research team’s findings and their complacency with regard to the potential adverse consequences facing children whose mothers regularly drink during pregnancy. Three lines of research provide the basis for our concerns, and we outline them below.

  • 1
     Moderate PNE affects social behaviour in humans and animal models. The complex interplay between social relationships and executive functioning in adolescence6 may explain the lack of effect of PNE found in the target article, as the age at which the children were tested was only 5 years (preschool age in Denmark). The authors do briefly discuss the age of the cohort, but only with respect to the validity of the psychometric instruments they used in the study.
  • 2
     Recent research, using animal models, has shown that moderate PNE affects executive centres in the brain (prefrontal cortices), and this corresponds to alterations in social behaviour. The effects on cognitive function during the course of the child’s later development may become more evident, in particular following the formation of meaningful dyadic social relationships. Human studies have demonstrated that, although not as pronounced as is the case in fetal alcohol syndrome, there are detectable effects of moderate PNE on executive function in older children.7
  • 3
     It is becoming increasingly clear that moderate PNE is a risk factor for alcohol abuse potential. This has been demonstrated in humans,8 and is now being increasingly investigated in animal models. This, in conjunction with the reported effects on social cognition, represents a cause for concern for the PNE-exposed child as he/she enters adolescence and adulthood.

Finally, we would like to suggest that the null results found in the series of studies should be treated with a greater degree of caution, in particular considering that potential extraneous risk factors associated with continued cognitive development are ignored, and the data challenge such a large body of extant evidence. The authors do briefly refer to the danger of accepting null results, but only insofar as it pertains to having sufficient statistical power to rule out a type II error. To summarise, we believe that the implied message from the authors of the reports concerning low-level or moderate drinking during pregnancy may result in an overly complacent attitude, particularly in the absence of further follow-up data from the cohort.


  1. Top of page
  2. References