Dietary and other determinants of lipoprotein levels within a population of 315 Dutch males aged 28 and 29

The objective of this study was to estimate the strength of the association between diet and serum lipids and lipoproteins within a free-living population. Three hundred and fifteen Dutch males aged 28 and 29 were located via the Army registry. Their diets were assessed through a dietary history questionnaire plus oral cross-check, and blood was sampled twice at a 1-week interval. Height, weight, skinfolds, and waist and hip circumference were measured to estimate amount and location of body fat. Physical activity, social status and smoking were estimated through questionnaires. The range of cholesterol values predicted from differences in dietary fatty acid and cholesterol intake as summarized by the Keys equation score was 5.11 to 6.29 mmol/l, while the actual levels ranged from 3.30 to 9.28 mmol/l. Multiple regression analysis showed a weak but significant association of the Keys score with total cholesterol (slope 0.83, P less than 0.05), LDL cholesterol (slope 0.71, P less than 0.05), and the HDL/total cholesterol ratio (slope -0.04, P less than 0.01). The percentage of energy from alcohol was strongly and independently associated with both total and HDL cholesterol (slopes 0.04 and 0.02, P less than 0.001). Body fat parameters were strongly associated with all serum lipids and lipoproteins. We conclude that the contribution of dietary differences to differences in serum lipids and lipoproteins within this population is real but small.