MARINA 2011.

Methods	Modulation of Atherosclerosis Risk by Increasing dose of n‐3 fatty Acids (MARINA) RCT, parallel, 4 arms (n‐3 PUFA 3 different doses or olive oil placebo), 12 months Summary risk of bias: low
Participants	Non‐smoking men and women aged 45‐70 years N: intervention. 279 in 3 groups (G1 0.45 g/d n = 94, G2 0.9 g/d n = 93, G3 1.8 g/d n = 92); control: 88 (analysed G1 0.45 g/d n = 81, G2 0.9 g/d n = 80, G3 1.8 g/d n = 80, control 71) Level of risk for CVD: low Men: 38.7% intervention, 38.6% control Mean age in years (CI): G1: 55 (53, 56), G2: 55 (54, 56), G3: 55 (54, 57) intervention 55 (54,57) control Age range: 45‐70 Smokers: 0% intervention, 0% control Hypertension: 5.4% intervention, 5% control Medications taken by at least 50% of those in the control group: none Medications taken by 20%‐49% of those in the control group: none Medications taken by some, but less than 20% of the control group: statins, antihypertensives, HRT, thyroxine Location: UK Ethnicity: G1: white 80.9%, black 4.3%, Asian 6.4%, East Asian 4.3%, other 4.3% G2: white 78.5%, black 6.5%, Asian 10.8%, East Asian 0%, other 4.3% G3: white 85.9%, black 1.1%, Asian 2.2%, East Asian 4.3%, other 6.5% Control: white 77.3%, black 10.2%, Asian 6.8%, East Asian 2.3%, other 3.4%
Interventions	Type: supplement (fish oil capsules) Comparison 1: EPA + DHA vs MUFA Comparison 2: high EPA + DHA vs low EPA + DHA Intervention: 3 × 1 g oil gelatin capsule/day consisting of blend of EPA concentrate, DHA concentrate, refined olive oil and 0.1% peppermint oil. Providing a daily dose of: 0.45 g, 0.9 g, or 1.8 g per day (all with EPA/DHA ratio of 1.51). Dose: 1.8 g/d EPA + DHA (G3 used for outcomes) Control: 3 gelatin capsules/ day containing refined olive oil + 0.1% peppermint oil Compliance: measured by capsule counting and erythrocyte lipids for proportion of EPA/DHA @ baseline, 6 months, 12 months. 88.5% of participants consumed > 90% of capsules provided. EPA and DHA in erythrocyte lipids increased in dose‐dependent manner compared with placebo, indicating long‐term compliance with intervention. Length of intervention: 12 months
Outcomes	Main study outcome: endothelial function, arterial stiffness Dropouts: 38 intervention (13,13,12), 17 control Available outcomes: lipids, dietary intake, CRP, BP (supine and ambulatory – numeric data not provided, but study states that there were no significant differences between arms). Weight data not used as baseline is different between groups (FMD, arterials stiffness, carotid intima media thickness, heart rate variability, heart rate, endothelial progenitor cells reported but not used) Contact with authors: yes (many outcomes above provided in end of study report from authors)
Notes	Outcome data used G3 (highest dose) vs placebo for continuous outcomes and combined the 3 intervention groups vs placebo for dichotomous outcomes Study funding: Food Standards Agency
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "the random allocation sequence was generated with a computer program by using the process of minimisation to balance age, sex and ethnicity between treatment groups."
Allocation concealment (selection bias)	Low risk	Quote: "We enrolled eligible participants and the study database program allocated a serious of capsules to the participant. The treatments associated with the capsule codes were concealed from all investigators and associated clinical staff until the data analysis was complete. The code breaker was an employee of MedSciNet who constructed the trial database."
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Quote: "We enrolled eligible participants and the study database program allocated a serious of capsules to the participant. The treatments associated with the capsule codes were concealed from all investigators and associated clinical staff until the data analysis was complete. The code breaker was an employee of MedSciNet who constructed the trial database." "blends of the test fat with 0.1% peppermint oil to disguise the fish taste of the EPA and DHA" (peppermint oil in both intervention and control capsules)
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Quote: "We enrolled eligible participants and the study database program allocated a serious of capsules to the participant. The treatments associated with the capsule codes were concealed from all investigators and associated clinical staff until the data analysis was complete. The code breaker was an employee of MedSciNet who constructed the trial database."
Incomplete outcome data (attrition bias) All outcomes	Low risk	15% withdrawal, reasons for attrition reported
Selective reporting (reporting bias)	Low risk	Outcomes published match trials register. Registered September 2008, trial started June 2008, ended December 2010, main publication 2011
Attention	Low risk	No difference between groups
Compliance	Low risk	Statistically significant difference in erythrocyte omega 3 fats at 12 months between different arms
Other bias	Low risk	None noted