BMJ. 1990 May; 300(6735): 1297-301.

Dietary intake and plasma lipid levels: lessons from a study of the diet of health conscious groups.

Thorogood M, Roe L, McPherson K, Mann J.

Department of Public Health and Primary Care, Radcliffe Infirmary, Oxford.

AIM--To re-examine the contentious relation between diet and plasma lipids within a population. DESIGN--Cross sectional sample from a large prospective cohort study of people eating different diets in Britain. Blood samples and diet records collected from subjects. SUBJECTS--Volunteers eating one of four distinct diets--namely, vegans, vegetarians, fish eaters who do not eat meat, and meat eaters. 52 Subjects selected from each group. METHODS--Examination of the relation between nutritional intake recorded in a four day dietary record and plasma lipid concentrations of subjects measured in blood samples collected previously. RESULTS--After controlling for age, sex, and body mass index, the correlation between plasma total cholesterol and the Keys score (which includes dietary cholesterol and saturated and polyunsaturated fat) was 0.37 (p less than 0.001). The mean saturated fat intake in all groups was low (6-14% of energy), but polyunsaturated fat intake was high, so mean total fat intake was generally above that recommended. A high dietary fibre intake was not associated with high carbohydrate intake. Plasma high density lipoprotein values were not associated with any measure of fat intake, but there was a significant correlation of 0.24 between high density lipoprotein values and alcohol intake. CONCLUSIONS--The nature rather than quantity of dietary fat is an important determinant of cholesterol concentrations. Health conscious individuals select a fat modified, rather than a low fat--high carbohydrate diet. National cholesterol lowering dietary advice should be reconsidered.