Omega-3 fatty acid supplementation in patients with recurrent self-harm

Placebo-controlled trial in adults aged 16 to 64 years presenting with self-harm and a history of prior self-harm, recruited from the accident and emergency department of Beaumont Hospital in Dublin, Ireland. The active n-3 EFA arm included 22 participants, with a mean age of 30.5 years; 7 were male and 15 were female.

The active regimen was EPAX 5500 omega-3 capsules providing 305 mg EPA and 227 mg DHA per capsule, four capsules daily for a total of 2128 mg/day of EPA plus DHA, taken in the morning for 12 weeks. Participants continued standard psychiatric care and were compared with a corn-oil placebo containing a 1% EPA/DHA mixture.

Omega-3 supplementation was associated with better psychological outcomes than placebo over 12 weeks, especially for depression and perceived stress. BDI scores improved significantly in the n-3 EFA group (P=0.004), and more patients achieved greater than 50% and greater than 70% symptom reduction (both P=0.001); perceived daily stress also fell more than with placebo (P=0.021), with a significant 50% response (P=0.006) and greater DHUS improvement (P=0.027). Suicidality outcomes also favored treatment, with no suicidal ideation reported on the OAS-M suicidality sub-scale in the n-3 EFA group (n=14) and no completed suicides during the study, although self-harm episodes were not reduced (14 episodes total, 7 per group; P=0.65) and HRSD and impulsivity/aggression outcomes were similar to placebo. Tolerability was good, with only mild gastric discomfort and fish-like taste reported and no discontinuations due to adverse events.

This was a small, single-center study with only 22 participants in the active arm and a 12-week follow-up, which limits precision and generalizability. Several outcomes were null, including self-harm episodes, HRSD, and impulsivity/aggression, so benefits were mainly seen on surrogate psychological measures rather than hard clinical endpoints. Adherence was incomplete in some participants, and the trial was not large enough to determine effects on suicide mortality.