Aims/hypothesis The aim of the study was to examine the impact of statin or omega-3-acid ethyl esters 90 (omega-3 EE90; omega-3-acid ethyl esters 90 refers to a mixture of ethyl esters of n-3 fatty acids) on estimated cardiovascular disease (CVD) risk in community-based people with type 2 diabetes but without known CVD and not taking lipid-lowering therapy. Methods A central computer randomised 800 patients in 59 UK general practices to atorvastatin (n=401, 20 mg/day) or placebo (n=399) and omega-3 EE90 (n=397, 2 g/day) or placebo (n=403) in a concealed factorial manner. Participants with LDL-cholesterol <2.6 mmol/l, triacylglycerol <1.5 mmol/l and estimated 10-year CVD risk <20% were compared at 4 months.Results Mean (SD) age was 63.5 (11.7) years, HbA 1c 6.9 (1.1) % and known diabetes duration (median [interquartile range]) was 4 (2-8) years. Fifty-seven per cent were men, 90% white and 74% had an estimated 10-year CVD risk ≥20%. Of 732 patients with 4-month data, more allocated atorvastatin (n=371) compared with placebo (n=361) achieved LDLcholesterol <2.6 mmol/l (91% vs 24%, p<0.001) and had estimated 10-year CVD risks <20% (38% vs 26%, p<0.001). No differences were seen between those allocated omega-3 EE90 (n=371) compared with placebo (n=361) for participants achieving triacylglycerol <1.5 mmol/l (65% vs 60%, p=0.18) or estimated 10-year CVD risks <20% (34% vs 30%, p=0.18). There were no side effects of note. Conclusions/interpretation Many community-based diabetic patients without known CVD remain at high CVD risk despite statin treatment and require additional risk-reduction strategies. The impact of omega-3 EE90 on CVD risk will remain uncertain until clinical endpoint trial results are available.