The McGorry omega-3 prevention trial: what happened next

For about a decade, one of the most cited findings in early psychosis research was a small Austrian trial suggesting that omega-3 fatty acids could substantially reduce the risk of converting to a full psychotic disorder in young people at clinical high risk. The result was striking. The headlines were striking. The follow-up story turned out to be more complicated — and in some ways more useful — than the original.

The original Vienna trial

In 2010, G. Paul Amminger and Patrick McGorry published results from a randomised controlled trial of 81 young people meeting CHR criteria. Participants received either 1.2 g per day of long-chain omega-3 fatty acids (EPA + DHA from fish oil) or placebo for 12 weeks, with follow-up over 12 months. The result, published in Archives of General Psychiatry, 2010, was that the omega-3 group had a transition rate of 4.9% compared with 27.5% in the placebo group — an enormous absolute difference.

The trial received deserved attention. If it replicated, fish oil — cheap, available without prescription, with virtually no side effects — would be one of the few genuinely preventive interventions in psychiatry.

The seven-year follow-up

In 2015, Amminger's group reported long-term follow-up of the original cohort (Amminger et al., Nature Communications, 2015). The protective effect appeared to persist: only 9.8% of the omega-3 group had developed a psychotic disorder, compared with 40.0% of the placebo group, over a median 6.7-year follow-up. This made a stronger claim — that a brief 12-week intervention could meaningfully shift trajectories years later.

NEURAPRO: the bigger replication

The Vienna trial was small, with only 81 participants and 22 transitions. To know whether the effect was real, the field needed a larger, multisite replication. McGorry's team led that effort with NEURAPRO, a six-site trial across Australia, Asia, and Europe with 304 participants.

NEURAPRO's primary results were published in JAMA Psychiatry, 2017. Both the omega-3 and placebo groups received high-quality cognitive behavioural case management. At 12 months, transition rates were 6.7% in the omega-3 group and 5.1% in placebo — essentially identical, with no statistically significant benefit from omega-3.

What may explain the difference

NEURAPRO's authors and several commentators have offered several plausible explanations:

• Lower overall transition rate. The placebo group in NEURAPRO had a much lower transition rate (5.1%) than the Vienna placebo group (27.5%). If everyone is improving, there is little room for additional benefit.
• Active background care. Both arms in NEURAPRO received intensive cognitive behavioural case management — itself an active treatment. The Vienna trial offered less structured background care.
• Small-trial effects. The original effect size was unusually large for any preventive intervention. Small trials are well known to produce inflated effect estimates that shrink in larger replications.
• Population shifts. CHR populations have changed over time as services catch lower-risk presentations earlier.

What this means for fish oil

The current honest position, supported by guideline groups including NICE, is that omega-3 fatty acids are not recommended as a stand-alone preventive treatment for psychosis based on current evidence. They appear safe, are generally well tolerated, and may have modest benefits for cardiovascular health and depression. They are not, on present evidence, a reliable way to prevent transition to psychosis.

For people with established schizophrenia, the picture is similarly modest. Several trials of omega-3 augmentation have shown small effects on symptoms in some subgroups, but no consistent transformative benefit. See our omega-3 in schizophrenia article for that body of evidence.

What the story teaches about prevention research

The McGorry omega-3 saga is a small clinic-trial textbook in a few chapters:

• Single small positive trials are rarely the final word. They are the start of an investigation, not the end.
• Pre-registered, multi-site replications matter. NEURAPRO was designed exactly to test the original hypothesis rigorously.
• Background care matters. When everyone receives reasonable treatment, room for additional benefit shrinks.
• Negative results are scientific successes. NEURAPRO did not "fail" — it did its job, which was to give the field a clearer picture.

What is being investigated next

Several lines of research continue to look at potentially preventive or modifying interventions in CHR populations: cannabidiol (CBD), N-acetylcysteine, certain anti-inflammatory medications, family-focused therapy, and intensive integrated early intervention. None has yet produced a transformative effect, and the field has largely moved from a "predict and prevent transition" framing to a more flexible staged-care model focused on present distress and functional recovery.

Practical takeaways for families

• If a young person has been told they meet CHR criteria, fish oil is fine but should not replace evidence-based clinical care.
• The best supported interventions in CHR remain CBT, family work, and treatment of comorbid depression, anxiety, and sleep problems.
• Avoiding heavy cannabis use is one of the few clearly modifiable risk factors with consistent evidence.
• Stay engaged with the early intervention service — sustained contact catches changes early.

The bigger picture

The omega-3 story is a hopeful one even though the replication did not pan out. It showed what is possible when a research community takes a positive finding seriously, designs the right replication, and reports honestly when the result doesn't hold. That is how science gets us closer to interventions that genuinely work.

This article is for educational purposes only and is not medical advice, diagnosis, or treatment. Always consult a qualified mental health professional. If you or someone you know is in crisis, call or text 988 in the US, or your local emergency number.