A Cochrane review of omega-3 and ADHD in children and teens, covering related fatty acids too, found with high certainty no medium-term difference in parent-rated symptoms versus a dummy pill (Gillies, 2023). Reviews from 2013 (ages 3 to 18) and 2017 (children) found a small benefit, of unclear importance (Sonuga-Barke, 2013) and too small to add much (Pelsser, 2017).
What did a Cochrane review of omega-3 and ADHD find?
A 2023 Cochrane review pulled together 37 trials, 30 of them in its pooled analysis, with more than 2,374 participants (Gillies, 2023). Everyone in it was a child or teen aged 6 to 18 with a diagnosis of ADHD. The supplements were omega-3 and related fatty acids, given for anywhere from two weeks to six months, in trials run in many countries.
The main result is clear. Over the medium term, the review found high-certainty evidence that the supplements made no difference compared with a dummy pill. That held for parents' ratings of overall ADHD symptoms, for inattention, and for hyperactivity and impulsivity, each looked at separately. "High certainty" is the review's own rating of how confident it was in that result.
There was one more hopeful thread, and it's much weaker. On a separate measure, which simply sorted each child as improved or not improved, the review found low-certainty evidence that the supplements may help over the medium term. That result rested on just 3 studies and 191 children. The same measure didn't show a clear difference over the short term or the long term. So a single, thinly supported result points one way, while the much larger and better-supported results on symptom ratings point to no difference.
On safety, the review found moderate-certainty evidence that overall side effects didn't differ between children taking the supplements and children taking a dummy pill. The number of families who dropped out of the trials was also likely similar in both groups (Gillies, 2023).
What did earlier reviews of fatty acid supplements find?
A 2013 review pooled randomized trials in children and teens aged 3 to 18 who had a diagnosis of ADHD or scored above set cutoffs on symptom questionnaires (Sonuga-Barke, 2013). It paid close attention to who was doing the rating. It compared ratings from the people closest to the treatment with ratings from people who probably didn't know which children got the real supplement. Under those probably blinded ratings, fatty acid supplements still showed a small effect, one unlikely to be down to chance. The authors were careful about what that means, though. They wrote that the clinical significance of the effect remains to be determined.
A 2017 review stepped back further (Pelsser, 2017). It looked only at earlier pooled analyses, one of them the 2013 review, built from trials where neither families nor researchers knew who got the real supplement or the dummy pill. All the children in those trials met criteria for ADHD or an equivalent; the review doesn't give an age range. In the analyses it reviewed, the effect of fatty acid supplements was small when parent and teacher ratings were combined. Teacher ratings alone showed no clear difference, and parent ratings alone showed a small difference that could have been down to chance.
The authors' conclusion was direct. They judged the average effect too small to contribute much to ADHD treatment, and wrote that fatty acid supplements should not as yet be advised as general ADHD treatment (Pelsser, 2017).
Over the medium term, the 2023 Cochrane review found high-certainty evidence of no difference on parent-rated symptoms in children and teens aged 6 to 18 (Gillies, 2023).
What the research on omega-3 can't tell you yet
The Cochrane review studied children and teens aged 6 to 18 with ADHD (Gillies, 2023). The 2013 review studied ages 3 to 18, with ADHD or high scores on symptom questionnaires (Sonuga-Barke, 2013). The 2017 review studied children who met ADHD or equivalent criteria (Pelsser, 2017). Their results are not findings about adults.
The trials themselves had real weaknesses. The Cochrane review flagged small sample sizes, different rules for which children could take part, wide differences in the type and dose of supplement, and short follow-up times (Gillies, 2023). Some trials used omega-3, some used omega-6, and some used both, for four to 16 weeks, at doses that varied widely from trial to trial (Pelsser, 2017). The 2013 review also noted that it had little power to detect publication bias, where studies with weaker results go unpublished (Sonuga-Barke, 2013).
If you're thinking about food more broadly, Food and ADHD: does what your child eats make a difference? covers what's been studied there, and the nutrition section brings the related articles together.
Try this today: Write down any supplement your child takes now, including the amount, so you have it ready for your next appointment.
Short follow-up matters too. Supplements were given for two weeks to six months (Gillies, 2023), so the long-term picture is thin. Pelsser (2017) also pointed out that only families interested in a diet-based approach may join such a trial, so the results may apply mainly to children whose parents are open to it.
Talk to your child's clinician about their situation before making any decision about supplements.
The numbers Stats and what the terms mean
The statistics behind the plain-language claims above.
- Gillies, 2023: parent-rated overall ADHD symptoms over the medium term, supplement compared with dummy pill, standardized mean difference −0.08, 95% confidence interval −0.24 to 0.07; 16 studies, 1,166 participants.
- Gillies, 2023: whether a child counted as improved, over the medium term, risk ratio 1.95, 95% confidence interval 1.47 to 2.60; 3 studies, 191 participants.
- Gillies, 2023: overall side effects, risk ratio 1.02, 95% confidence interval 0.69 to 1.52; 8 studies, 591 participants.
- Sonuga-Barke, 2013: symptoms under probably blinded ratings, standardized mean difference 0.16, 95% confidence interval 0.01 to 0.31.
What the terms mean:
- Standardized mean difference is how far apart the two groups' average scores were, in a common unit. Roughly, 0.2 is small, 0.5 is moderate and 0.8 is large. A minus sign shows the direction of the change, not its size.
- Risk ratio compares how likely an outcome was in the supplement group with the dummy-pill group, where 1 means no difference.
- 95% confidence interval is the range the true result probably falls within; when it crosses zero (or 1 for a risk ratio), the result could be down to chance.