The Diet and Reinfarction Trial (DART) involved 2033 men (mean age 56 5 years) recovering from myocardial infarction. They were randomly allocated to receive advice or to receive no advice on each of three dietary factors: an increase in fatty fish intake; a reduction in fat intake with an increase in polyunsaturated fat:saturated fat; an increased intake of cereal fibre. Compliance was satisfactory with the fish and fibre advice, but less so with the fat advice. The men given fish advice had 29% lower 2-year all-cause mortality; the other forms of advice did not have any significant effects. The Diet and Angina Randomized Trial (DART-2) involved 3114 men (mean age 61·1 years) with stable angina, who were followed up for 3–9 years. Advice to eat oily fish or take fish oil did not affect all-cause mortality, but it was associated with a significant increase in sudden cardiac death (P = 0·018), and this effect was largely confined to the subgroup given fish oil capsules. Advice to eat more fruit and vegetables had no effect, probably because of poor compliance. The outcome of DART-2 appears to conflict with that of DART and some other studies; various possible explanations are considered. Nutritional interventions are not equally acceptable and should be tailored to the individuals for whom they are intended. Various distinct groups have a raised risk of CHD, and it cannot be assumed that the same nutritional interventions are appropriate to them all. Nutritional supplements do not necessarily have the same effects as the foods from which they are derived.