select search filters
briefings
roundups & rapid reactions
before the headlines
Fiona fox's blog
Research published in Occupational & Environmental Medicine demonstrates that maternal occupational exposure to solvents may increase the risk for autism spectrum disorder.
A Before the Headlines accompanied this Round Up.
Dr Paolo Fusar-Poli, Reader in Psychiatry and Youth Mental Health, Institute of Psychiatry, Psychology & Neuroscience, King’s College London:
“This study suggests an association between exposure to solvents and child autism. However, since the effect sizes are very small these results should be interpreted extremely cautiously. Investigating risk and protective factors for severe mental disorders is a cornerstone not only of modern epidemiology but also of preventive interventions in psychiatry; it requires solid methods that can control for several biases and produce robust findings. We recently published a large-scale review on risk and protective factors for autism* which found only a few factors which were convincingly associated with an increased risk of autism spectrum disorder, such as maternal age or being overweight before or after pregnancy. Only through such large studies, termed ‘umbrella reviews’, can we provide robust, evidence-based knowledge that can inform clinical practice.”
Prof Alastair Hay, Professor (Emeritus) of Environmental Toxicology, University of Leeds, said:
“The results of this case-control study are interesting and add to the evidence base that autism spectrum disorder (ASD) is associated with exposure to agents in the environment; in this study, the link is between solvents and autism.
“Others (cited in this paper) have shown that a range of heavy metals, pesticides and air pollution, amongst other agents , are associated with the autistic spectrum. There are no environmental exposures where the evidence is good enough to say that a particular agent causes autism. As with this study it is only an association between exposure and the disorder.
“The study has been done very well. All the children (537 with ASD and 414 developing typically [the controls ]) underwent cognitive, social and medical investigations, so diagnosis is not an issue. Environmental exposures were assessed by questioning mothers about activities, with approximately 2/3 of mothers providing details about the fathers’ circumstances. The study found an association between mothers’ exposure to solvents, but not fathers, and the latter group may have been misclassified because they were not questioned directly.
“Exposures and their intensity were assessed by industrial hygienists and done well as one would expect given the authors of the study who are from NIOSH in the US, an agency, whose remit is assessing exposure to chemicals in the workplace.
“A potential weakening of the finding is that it was only significant for the children from mothers classed as moderately exposed to solvents, but not for those with ‘high exposures’. If there is a causal link between autism and solvents, you would expect this to be even more evident with the high exposure category.
“And, as the authors themselves note, the findings must be interpreted with caution. After statistical correction (for multiple comparisons) the results were no longer significant.
“Much more evidence is needed to confirm any link between solvent exposure and autism. Larger studies, looking at specific solvents, are needed, as solvents exert their toxic effects in a wide variety of ways.”
Dr Peter Jenkinson, Managing Director, CEHTRA (Consultancy for Environmental & Human Toxicology and Risk Assessment) UK, said:
“The study described in this press release reports that the authors have looked at 16 different agents, previously associated with autism spectrum disorder (ASD), and the estimated work exposures to mothers, fathers, and both parents combined. They report a statistical association between exposure of the mothers to ‘solvents’ and the incidence of ASD in their children. However, they also state that after correction for statistical bias (multiple comparisons) then the association was no longer statistically significant, which in essence means that it was never truly statistically significant in the first place. Plus, solvents is a generic term and in this case included paint chemicals, solvents, and degreasers in combination. Are we to understand that all solvents have the same mechanism of action? Clearly this is highly unlikely, particularly as many paints are these days water-based.
“Furthermore, actual exposures were not measured, they were estimated from the job histories of the mothers given during telephone interview. This information was interpreted separately and blinded by two industrial hygienists, whose estimates were not in particularly good agreement, ~60%. Nethertheless, the estimates were combined and used in the analysis. When the analysis was done on the extent of the exposure then the highest exposed group had a lower likelihood (odds ratio) of ASD than the lower exposed groups, such that “High cumulative exposure to solvents in mothers was not associated with ASD.”
“Finally, the odds ratios for the other 8 agents of concern where 5 or more mothers were exposed, whilst not statistically significant were all greater than 1, which suggests that there may have been a hidden bias tending to increase the odds ratio in all cases. It seems difficult to accept the interpretation of the data given by the authors of this paper as being scientifically valid.”
Dr Oliver Jones, Associate Professor of Analytical Chemistry, RMIT University, said:
“This paper is one of a growing number reporting an association between exposure of parents to environmental pollutants (in this case certain industrial solvents) and health effects (in this case autism) in their children. However, while the results might at first sound scary, it is well worth reading the whole paper. The association described is weak at best and was only found in one out of 16 groups tested. Even then, the authors themselves note that the link disappears after the data are corrected for statistical bias. Exposure to the compounds is also estimated rather than measured directly so it is impossible to draw any conclusions from this data in my view. Even the press release that goes with the article says that a “cautious interpretation is needed”.”
Dr Peter Jenkinson, Managing Director, CEHTRA (Consultancy for Environmental & Human Toxicology and Risk Assessment) UK, said:
“The study described in this press release reports that the authors have looked at 16 different agents, previously associated with autism spectrum disorder (ASD), and the estimated work exposures to mothers, fathers, and both parents combined. They report a statistical association between exposure of the mothers to ‘solvents’ and the incidence of ASD in their children. However, they also state that after correction for statistical bias (multiple comparisons) then the association was no longer statistically significant, which in essence means that it was never truly statistically significant in the first place. Plus, solvents is a generic term and in this case included paint chemicals, solvents, and degreasers in combination. Are we to understand that all solvents have the same mechanism of action? Clearly this is highly unlikely, particularly as many paints are these days water-based.
“Furthermore, actual exposures were not measured, they were estimated from the job histories of the mothers given during telephone interview. This information was interpreted separately and blinded by two industrial hygienists, whose estimates were not in particularly good agreement, ~60%. Nethertheless, the estimates were combined and used in the analysis. When the analysis was done on the extent of the exposure then the highest exposed group had a lower likelihood (odds ratio) of ASD than the lower exposed groups, such that “High cumulative exposure to solvents in mothers was not associated with ASD.”
“Finally, the odds ratios for the other 8 agents of concern where 5 or more mothers were exposed, whilst not statistically significant were all greater than 1, which suggests that there may have been a hidden bias tending to increase the odds ratio in all cases. It seems difficult to accept the interpretation of the data given by the authors of this paper as being scientifically valid.”
‘The CHARGE study: an assessment of parental occupational exposures and autism spectrum disorder’ by Erin C McCanlies et al. was published in Occupational & Environmental Medicine at 23:30 UK time on Thursday 27th June.
Declared interests
Dr Peter Jenkinson: I have no declarations of interest relating to this topic and research.
None others received.