The well-respected NHS Choices website has done an excellent job here in pointing out inaccuracies and exaggerations in some media coverage of the recently-published DHA Oxford Learning and Behaviour (DOLAB) study, and in presenting scientific details of this research in a clear and accessible way.
There are, however, some points in this article that merit comment.
Firstly it is unfortunate that the article refers throughout to 'fish oil' - when in fact the omega-3 DHA used in this study was derived from algae (a sustainable source, and also suitable for vegans).
The article also states that 'extensive research has previously been conducted into the effects of omega-3 fatty acid supplements on cognitive ability and behaviour', and that 'previous reviews of conducted studies have not found strong evidence for an effect on “brain power or behaviour” in adults or children'
This is just not true. 'Extensive' research of this kind has NOT previously been done in either adults or children - and certainly not with children from the general school population.
This misperception probably arises from the fact that a huge number of supposed 'trials' with absolutely no scientific validity whatsoever have previously been given extensive media coverage, despite their repeated exposure as Bad Science in numerous heroic efforts by Ben Goldacre. Given that these were NOT randomised controlled trials, however, and were neither peer-reviewed nor published in any recognised journals, NHS Choices could surely be expected to know the difference.
To put the record straight:
With respect to reading:
With respect to behaviour:
Having given such otherwise accurate and well-balanced coverage to the DOLAB study (which clearly needs replication now, as the authors emphasised) it does seem very unfortunate that the NHS Choices article should nonetheless help to perpetuate the myths that:
(a) extensive research in these important areas has already been done
No, it has NOT, and there remains a clear and urgent need for such research, as highlighted by a recent Cochrane review (Gillies et al 2012)
(b) reviews of what limited evidence there is do not show benefits of long-chain omega-3 for child behaviour problems
This is simply untrue. While more research is undoubtedly needed, a recent meta-analysis found modest benefits for omega-3 for reducing ADHD-type symptoms in children from a variety of populations. (Bloch and Qawasmi 2011)
See also the actual research study (open-access):
NHS Choices - 'Fish oil can make children less naughty'
Parents should give their children a daily dose of fish oils if they want to boost their brain power and stop them being naughty the Daily Express claims.
This advice is highly premature. The news follows research into supplements containing DHA, which is the omega-3 fatty acid found in fish, seafood and algae. The trial looked at the effects of DHA on reading, memory and behaviour in children.
In the study, children aged seven to nine who were underperforming in reading were given either DHA supplements or a placebo (dummy pill) for 16 weeks.
The researchers actually found that, overall, DHA had no beneficial effect on reading ability.
In a restricted analysis on a group of the children who had the poorest initial reading ability, those who were given DHA showed a small improvement compared with those given placebo.
DHA had no effect on memory or behaviour as rated by teachers, though it did improve some aspects of behaviour rated by parents.
This was a well-conducted study that had many strengths in its study design, but that only achieved very modest results in terms of improving reading ability.
If you are worried about your child’s reading ability, spending time reading with them will achieve better results than giving them any kind of food supplement.
This study was conducted by researchers from the University of Oxford and was funded by Martek Biosciences Inc, who also provided the supplements and placebo for the trial.
The study was published in the peer-reviewed medical journal PLoS One.
The claim made by the Daily Express that giving kids fish oil can “stop them being naughty” is clearly not supported by this research. The research only found small improvements in certain aspects of parent-rated behaviour, but no effect on teacher-rated behaviour. The newspaper’s claim may lead readers to question any other scientific claims it makes. Both The Guardian and the Daily Mail provide far clearer and more accurate coverage of this science.
Neither did the researchers recommend that children should be given a daily dose to “boost brain power”, as the Express suggests. A positive effect in terms of reading ability was only found in a small subset of children with identified difficulties.
Fish oil supplements are not safe or suitable for all children – such as children with a weakened immune system or an allergy to seafood. This study looked at supplements, which were used by the researchers because they provide fish oil in regulated doses. A similar effect could be achieved by including more oily fish in your child’s diet (read more about how much oily fish is recommended).
Fish oil supplements are popular in the UK and have been extensively studied. They do have some well-documented benefits, such as reducing the risk of heart attacks in people with a history of them.
But previous reviews of conducted studies have not found strong evidence for an effect on “brain power or behaviour” in adults or children.
This was a randomised controlled trial that aimed to look at the effect of omega-3 fatty acids on reading, memory and behaviour in primary school children.
The study used supplements of docosahexaenoic acid (DHA), which is the omega-3 fatty acid found in fish, seafood and some algae. The DHA in this study was derived from algal oil (and was therefore suitable for vegetarians).
As the researchers say, most previous trials of omega-3 in children have involved children with conduct or behavioural disorders, such as attention-deficit hyperactivity disorder (ADHD). They were researching whether omega-3 may be beneficial for children with no behavioural disorder but who were underperforming with their reading compared with their peers.
The research was conducted across 74 primary schools in Oxfordshire, and was open to any healthy children in these schools who were in the seven- to nine-year-old age bracket and who were in the bottom third on an age-standardised word reading test. The authors say that this would typically equate to a reading performance of around 18 months below what would be expected for their age.
Teachers also excluded children who had social or family circumstances that meant it would be inappropriate for them to take part in the trial, such as a recent death in the family.
A total of 362 children met study criteria.
Those randomised to take fish oil supplements (the intervention group) were given 600mg DHA per day, given in three capsules, and the comparison group received three capsules identical in taste and colour to the supplements, containing corn or soybean oil. Schools and parents were given full instructions for dispensing capsules and given a diary to record how many capsules were taken. Teachers, parents and researchers were all unaware of which treatment was being given (the trial was double blind).
The main outcomes of interest, assessed before the start of the study and at the end of the 16-week trial period were:
The researchers looked at the final 16-week scores minus the baseline scores, and compared the score changes between the DHA and placebo groups.
The research had high follow-up rates, with 359 of 362 children randomised completing the 16-week assessments.
Overall, for all children randomised, the changes in reading scores after the 16 weeks did not differ between the DHA (fish oil) and placebo groups (mean 1.5 score improvement in the DHA group and 1.2 score improvement in the placebo group). Therefore, DHA had no effect on reading ability in the whole group.
However, when the researchers restricted their analyses to the 224 children who had an initial reading ability below the 20th centile (two years below the standard expected for their age), DHA significantly improved reading performance compared with placebo (mean score improvement 2.0 versus 0.9 with placebo, p=0.04).
Of other outcomes, they found that DHA also improved some parent-rated behaviour problems compared with placebo, but not others.
Scores across the 14 scales that parents reported against began at around 50 to 60 units and dropped by about 2.5 more in the DHA group. About half of these drops were statistically significant. However, DHA had no effect on working memory or teacher-reported behaviour.
There was no difference in reporting of side effects between groups, and no difference between groups in compliance with taking the tablets.
The researchers conclude that “DHA supplementation appears to offer a safe and effective way to improve reading and behavior in healthy but underperforming children from mainstream schools”. They say that “replication studies are clearly warranted”.
This was a well-conducted randomised controlled trial that found that an omega-3 fatty acid supplement gave a slight improvement in reading ability over 16 weeks in primary school children who were underperforming in their reading ability. However, this should not be seen as a licence to print misleading and inaccurate stories, such as the claim made by the Express that giving children fish oil can “boost their brain power” and “stops them being naughty”.
The trial had a number of strengths, such as:
Even though the whole group analysis found no difference in reading ability between groups, the subgroup analysis of those with the poorest reading ability still included a large sample size and was representative of 62% of people in the study. It was also a pre-planned analysis. The trial had initially aimed to include only those with a reading ability below the 20th centile (about two years below expected level). However, the researchers found that setting the inclusion criteria at this level would result in only a small number of children taking part.
To increase the number of trial participants and give the study better power to detect differences in outcomes, they had increased the eligibility criteria to those with reading ability below the 33rd centile (about 18 months below expected level). However, it is difficult to tell from this how the 1.1 score improvement with omega-3 in this group compared with placebo would relate to the child’s overall performance and ability, and whether sustained improvement in reading ability would be seen if treatment were continued in the longer term.
The use of multiple outcome scales can lead to a selective reporting of those that were significant and the news reporting has led to further selective reporting of the positive outcomes. The child’s behaviour in the eyes of parents also improved on some scales with fish oils, but it is difficult to know how relevant these small changes in the standardised behaviour scales would be to the parents or children.
Overall, the researchers’ conclusion that “replication studies are clearly warranted” is appropriate. Extensive research has previously been conducted into the effects of omega-3 fatty acid supplements on cognitive ability and behaviour. Though previous large reviews of completed studies have found no firm evidence that supplements improve cognition or prevent dementia in older adults or improve behaviour in children with behavioural problems.
Research into the benefits – or not – of fish oil supplements is likely to continue. In the meantime, there are well-established methods of improving your child’s reading and behaviour, such as reading with them at home and making sure they get regular exercise.