No effect of<i>n</i>-3 long-chain polyunsaturated fatty acid (EPA and DHA) supplementation on depressed mood and cognitive function: a randomised controlled trial

The researchers tested whether daily supplementation with a combination of two omega-3 fatty acids—eicosapentaenoic acid (EPA, 630 mg/day) and docosahexaenoic acid (DHA, 850 mg/day)—could improve mood and cognitive function compared to a placebo (olive oil capsules). The total dose was 1.48 g/day of EPA + DHA, which is roughly equivalent to eating one to two portions of oily fish per day or taking standard fish oil supplements.

The primary outcome was depressed mood measured by the depression subscale of the Depression Anxiety and Stress Scales (DASS) after 12 weeks. Secondary outcomes included:

Beck Depression Inventory (BDI) scores

Anxiety and stress subscales of the DASS

General Health Questionnaire (GHQ) scores

State anger (State-Trait Anger Expression Inventory)

Daily mood diary ratings (10 mood descriptors, rated 6 times/day for 3 days)

Attentional bias toward threat words (visual probe task)

Cognitive function (various computer-based tasks)

The comparator was an identical-looking placebo containing 2.36 g of olive oil, plus mixed tocopherols (vitamin E) and orange oil for flavouring. Both active and placebo capsules were made from fish gelatine and were identical in appearance.

**Sample size:** 218 participants entered the trial; 190 completed the full 12 weeks (97 in the EPA + DHA group, 93 in the placebo group)

**Population:** Adults aged 18–70 years with mild-to-moderate depressed mood (DASS depression score 10–24, where scores 21–27 are classified as "severe" and ≥28 as "extremely severe")

**Setting:** Primarily recruited through general practice surgeries in Bristol, UK, plus university events and local media advertising

**Key inclusion criteria:** No more than moderate consumers of EPA/DHA (equivalent to ≤1.5 portions of oily fish per week); not taking antidepressant medication; not pregnant or breastfeeding; no medical conditions contraindicating the supplements; fluent English speakers

**Demographics:** Mean age approximately 45 years; approximately 70% female; predominantly white British; mixed socioeconomic backgrounds

**Depression Anxiety and Stress Scales (DASS) – depression subscale:** 14 items, scored 0–42, higher = worse depression. Primary outcome at 12 weeks.

**Beck Depression Inventory (BDI):** 21 items, scored 0–63, higher = worse depression. Commonly used in clinical trials.

**DASS anxiety and stress subscales:** Each 14 items, scored 0–42.

**General Health Questionnaire (GHQ):** 12-item version, measures general psychological distress.

**State-Trait Anger Expression Inventory (STAXI) – state anger subscale:** Measures current anger intensity.

**Mood diary:** 10 mood descriptors (e.g., happy/cheerful, drowsy/sluggish, clear-headed) rated on 100 mm visual analogue scales, completed 6 times daily for 3 consecutive days.

**Visual probe task:** Computer-based test measuring attentional bias toward threat-related words (e.g., "cancer," "failure") vs. neutral words. Faster reaction times to probes appearing behind threat words indicate attentional bias.

**Cognitive function tasks:** Computer-based tests including simple and choice reaction time, digit vigilance, numeric working memory, and word recognition memory.

**Compliance verification:** Plasma fatty acid concentrations measured at baseline and 12 weeks to confirm actual absorption of EPA and DHA.

**Design:** Double-blind, randomised, placebo-controlled, parallel-group trial. This is the gold-standard experimental design for testing causal effects of an intervention.

**Randomisation:** Participants were randomly assigned to either EPA + DHA or placebo using a computer-generated randomisation sequence. The allocation was concealed from both participants and researchers (double-blind). The randomisation was stratified by sex and baseline DASS depression score (mild vs. moderate/severe).

**Blinding:** Both participants and all research staff (including those administering tests and analysing data) were blinded to group assignment. The capsules were identical in appearance, and orange oil was added to both active and placebo capsules to mask any fishy taste if capsules were bitten. Blinding was tested at the end of the study—participants were asked which treatment they thought they received. Results showed that blinding was successful: participants could not reliably guess their group assignment.

**Duration:** 12 weeks of supplementation. This is a reasonable duration for testing effects on mood, as most antidepressant trials show effects within 6–12 weeks. However, some omega-3 trials have used longer durations (6–12 months).

**Screening:** Two screening visits, 2–4 weeks apart, to exclude those with transient mood disturbance. This is important because mood fluctuates naturally, and a single screening might include people who are temporarily down but would recover without intervention.

**Statistical approach:** Primary analysis used analysis of covariance (ANCOVA) adjusting for baseline DASS depression score, sex, age, and recruitment method. Intention-to-treat analysis was used (all randomised participants analysed in their assigned groups, regardless of whether they completed the study). Missing data were handled using multiple imputation.

**What this design can prove:** A double-blind RCT with good compliance and adequate sample size can establish causal relationships. If the supplement worked, we would expect to see a statistically significant difference between groups on the primary outcome. The design controls for placebo effects, natural recovery, and regression to the mean.

**What this design cannot prove:**

It cannot rule out effects in specific subgroups (e.g., people with major depression, those with very low baseline omega-3 levels, or those taking concurrent antidepressants)

It cannot test whether longer supplementation (>12 weeks) might have effects

It cannot test whether higher doses might work

It cannot test whether EPA alone (without DHA) or DHA alone might have different effects

It cannot test effects on preventing depression (only treating existing symptoms)

**Major methodological strengths:**

Double-blind, placebo-controlled design

Objective compliance verification (plasma fatty acids)

Two screening visits to exclude transient mood disturbance

Adequate sample size (190 completers)

Multiple validated outcome measures

Successful blinding confirmed

**Major methodological weaknesses:**

12-week duration may be insufficient for some people

Mild-to-moderate depression only—cannot generalise to major depression

Participants were not taking antidepressants, so cannot assess augmentation effects

Single-centre study (Bristol, UK) limits generalisability

Some participants recruited through university events (younger, more educated sample)

**Primary outcome – DASS depression score at 12 weeks:**

EPA + DHA group: mean score 8.4

Placebo group: mean score 9.6

Adjusted difference: -1.0 (95% CI: -2.8 to 0.8; p = 0.27)

**Not statistically significant.** The confidence interval includes zero, meaning the true effect could be anywhere from a 2.8-point benefit to a 0.8-point harm.

**Secondary mood outcomes (all non-significant):**

Beck Depression Inventory: EPA + DHA 10.2 vs. placebo 11.3; adjusted difference -0.8 (95% CI: -2.9 to 1.3; p = 0.45)

DASS anxiety: EPA + DHA 5.2 vs. placebo 5.6; adjusted difference -0.3 (95% CI: -1.8 to 1.2; p = 0.69)

DASS stress: EPA + DHA 12.8 vs. placebo 13.8; adjusted difference -0.7 (95% CI: -2.7 to 1.3; p = 0.49)

General Health Questionnaire: EPA + DHA 12.2 vs. placebo 12.9; adjusted difference -0.5 (95% CI: -2.2 to 1.2; p = 0.56)

State anger: EPA + DHA 11.0 vs. placebo 11.2; adjusted difference -0.1 (95% CI: -1.0 to 0.8; p = 0.82)

**Mood diary (non-significant):**

No significant differences on any of the 10 mood descriptors (happy, drowsy, clear-headed, etc.) across the 3-day diary period

**Cognitive function (all non-significant):**

Simple reaction time: no difference

Choice reaction time: no difference

Digit vigilance: no difference

Numeric working memory: no difference

Word recognition memory: no difference

**Attentional bias (non-significant):**

Visual probe task: no difference in attentional bias toward threat words between groups

**Compliance:**

Plasma EPA and DHA concentrations increased substantially in the active group (EPA increased ~3-fold, DHA increased ~1.5-fold), confirming good compliance and absorption

87% of participants completed the trial (190/218)

Dropout rates were similar between groups (EPA + DHA: 12%; placebo: 14%)

**Meta-analysis update:**

The authors added their results to a meta-analysis of 12 previous RCTs (total n = 1,328). The updated meta-analysis confirmed an overall negligible benefit of omega-3 supplementation for depressed mood, with a standardised mean difference of approximately -0.10 (95% CI: -0.25 to 0.05), which is considered a very small effect.

The observed difference between groups was 1.2 points on the DASS depression scale (0–42 range). To put this in context:

A clinically meaningful improvement on the DASS depression scale is typically considered 5–7 points

The 95% confidence interval (-2.8 to 0.8) suggests the true effect could be anywhere from a small benefit (equivalent to about 1 week of natural recovery) to a small harm

The effect size (Cohen's d) was approximately 0.10, which is considered "very small" by conventional standards—roughly equivalent to the difference in mood between a Tuesday and a Wednesday

For comparison, standard antidepressant medications (SSRIs) typically show effect sizes of 0.30–0.50 in major depression, and even those are considered modest.

**What the authors acknowledge:**

The study was limited to mild-to-moderate depression; results may not apply to major depressive disorder

The 12-week duration may be insufficient; longer trials might show different results

The dose (1.5 g/day) may be too low; some positive trials used 2–10 g/day

The ratio of EPA to DHA (approximately 1:1.3) may not be optimal; some evidence suggests EPA-predominant formulations may be more effective

Participants had relatively low baseline omega-3 intake, but some may have had adequate levels already

The study was not powered to detect very small effects

**Additional limitations a critical reader would note:**

**Sample size:** While 190 completers is reasonable, the study was powered to detect a moderate effect (Cohen's d = 0.40). Smaller but potentially meaningful effects would be missed.

**Population limits:** 70% female, predominantly white British, recruited from one UK city. Results may not generalise to other populations.

**Mood measurement timing:** The primary outcome was measured at a single time point (12 weeks). Mood fluctuates, and multiple assessments might have been more reliable.

**No dietary control:** Participants were not asked to change their diet, so background omega-3 intake from food could have varied.

**No assessment of baseline omega-3 status:** While plasma levels were measured at baseline, participants were not selected based on low omega-3 status. People with adequate levels may not benefit from supplementation.

**Industry funding:** The capsules were provided by Minami Nutrition SA, a supplement company. While the authors state the company had no role in study design or analysis, industry-supplied interventions can introduce subtle biases.

**No long-term follow-up:** Effects might emerge or disappear after the 12-week period.

**Cognitive tests may be insensitive:** The computer-based cognitive tasks may not capture real-world cognitive function.

For someone running their own n=1 experiment:

**What to test:**

EPA + DHA supplementation at 1.5 g/day (or a higher dose, e.g., 2–4 g/day, if you want to test the upper range used in some positive trials)

Consider testing EPA-predominant formulations (e.g., 2:1 EPA:DHA ratio) rather than the 1:1.3 ratio used in this study

Alternatively, test eating 2–3 portions of oily fish per week (salmon, mackerel, sardines, herring) vs. no fish

**Minimum meaningful duration:**

At least 12 weeks (as used in this study), but consider 16–24 weeks to allow for potential delayed effects

Measure mood weekly to track trajectories, not just endpoints

**What to measure (specific metrics):**

**Primary:** A validated depression scale weekly, e.g., PHQ-9 (0–27, higher = worse) or DASS-21 depression subscale. Aim for at least 8 weekly measurements before and during supplementation.

**Secondary:** Daily mood ratings on a 1–10 scale (e.g., "How depressed did you feel today?"), logged at the same time each evening

**Cognitive function:** If you want to test cognition, use a standardised test like the Stroop test, digit span, or reaction time tasks (many free apps available). Test at the same time of day, under the same conditions.

**Objective biomarker:** If possible, get a blood test for omega-3 index (red blood cell EPA + DHA as % of total fatty acids) before and after supplementation. This confirms you actually absorbed the supplement.

**Confounders to track daily:** Sleep quality (hours, quality 1–10), exercise (minutes), alcohol (units), caffeine (cups), stress level (1–10), medication use

**Key confounds to control for:**

**Seasonal affective disorder:** If you start in winter, mood may improve naturally in spring. Run the experiment at the same time of year, or use a crossover design.

**Life events:** Major stressors (job loss, relationship changes) will overwhelm any supplement effect. Note these in your log.

**Placebo effect:** Use blinding if possible (have someone else prepare identical-looking placebo capsules). If not, use a run-in period where you take placebo for 4 weeks to establish baseline.

**Dietary changes:** If you increase fish intake, you may also change other aspects of your diet. Keep a food diary.

**Exercise and sunlight:** Both affect mood independently. Keep these constant or track them.

**Menstrual cycle:** For women, mood varies across the cycle. Track cycle phase or run the experiment over at least two full cycles.

**Other supplements:** Multivitamins, vitamin D, St. John's Wort, and many other supplements can affect mood. Keep these constant.

**What a positive result would look like:**

A consistent decrease of ≥3 points on the PHQ-9 or ≥4 points on the DASS depression subscale compared to baseline/placebo

The improvement should be sustained for at least 4 consecutive weeks (not just a single good week)

Ideally, the improvement should be larger than your typical week-to-week variation (calculate your baseline standard deviation first)

For cognitive tests: a ≥10% improvement in reaction time or accuracy, sustained over multiple testing sessions

If using a crossover design: clear improvement during