FAAT 2005.

Methods	Fatty Acid Antiarrhythmia Trial – FAAT Randomisation: RCT, parallel, 2 arms, (n‐3 EPA + DHA vs MUFA), 12 months Summary risk of bias: moderate or high
Participants	People with implanted cardioverter defibrillators (ICDs) N: intervention 200, control 202 Level of risk for CVD: high (patients with ICDs). Men: intervention 84.5%, control 81.7% Mean age in years (SD): intervention 65.7 (11.6), control 65.3 (11.7) Age range: unclear Smokers: intervention 15%, control 11.4% Hypertension: unclear Medications taken by at least 50% of those in the control group: ACE inhibitors, beta‐blockers Medications taken by 20% ‐ 49%: diuretics Medications taken by some, but < 20%: calcium channel blockers, amiodarone, sotalol, type 1 antiarrhythmics Location: USA Ethnicity: intervention 95.5% white, control 96.5% white
Interventions	Type: supplement/capsule Comparison: EPA + DHA vs MUFA Intervention: 4 ×1 g/d fish oil gelatin capsules, 2.6 g/d EPA + DHA (Pronova Biocare, quantities of EPA + DHA unclear). Dose: 2.6 g/d EPA + DHA Control: 4 ×1 g/d olive oil capsules, 4 g/d (in identical gelatin capsules, < 0.06 g/d EPA and < 0.06 g/d DHA) All were advised to use olive oil rather than the common plant seed oils for cooking, dressings, and sauces Compliance: pill counts and platelet phospholipid data suggested greater omega 3 intake in intervention participants. 35% were non‐compliers (36.5% intervention, 34.2% control) Duration of intervention: 12 months
Outcomes	Main study outcome: fatal ventricular arrhythmias Dropouts: intervention 13 deaths, unclear no. of dropouts, control 12 deaths, dropouts unclear Available outcomes: deaths, cardiovascular deaths, CVD events, deaths from heart failure, fatal arrhythmias, MI, angina Response to contact: yes (data provided)
Notes	Study funding: the study was supported in part by a grant from the NHLBI, NIH (HL62154)
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Computer‐generated randomisation tables for each collaborating site, stratified by site
Allocation concealment (selection bias)	Low risk	Author confirmed allocation was concealed from investigators
Blinding of participants and personnel (performance bias) All outcomes	Unclear risk	Study referred to as "double blind" and gelatin capsules (verum and placebo) were stated as being of identical appearance but no discussion of taste or smell. Author confirmed that investigators and patients were blinded.
Blinding of outcome assessment (detection bias) All outcomes	Low risk	VT and VF events were assessed blinded to allocation
Incomplete outcome data (attrition bias) All outcomes	High risk	Large numbers dropped out so some deaths, etc. may have been missed, 35% discontinued early due to non‐compliance but were assessed at study end, data censored for some participants
Selective reporting (reporting bias)	High risk	Trials registry data received September 2005, paper published November 2005
Attention	Low risk	Time and attention appeared similar between the 2 arms
Compliance	High risk	Pill counts and platelet phospholipid data suggested greater omega 3 intake in intervention participants. 35% were non‐compliers (36.5% intervention, 34.2% control)
Other bias	Low risk	None noted