Author/Authors :
Similن، نويسنده , , Minna and Ovaskainen، نويسنده , , Marja-Leena and Virtanen، نويسنده , , Mikko J. and Valsta، نويسنده , , Liisa M.، نويسنده ,
Abstract :
Dietary pattern analysis has recently been used in nutritional epidemiology. However, there has been little research on nutrient co-occurrence patterns in foods although they are useful for understanding associations of nutrients in foods. In this study, nutrient content data of Finnish foods were submitted to factor analysis. From the food composition database Fineli®, 530 basic food items and 106 common nutrients were selected. Nutrient patterns were analysed in two data sets, nutrient values (/100 g) and nutrient densities (/1 MJ), with principal component solutions of factor analysis. Four patterns, explaining 36–39% of total variance in nutrient content, were identified for both solutions. The patterns in the nutrient value solution were characterised by: (1) fish, meat, dairy products, legumes, seeds, nuts; (2) vegetable fats; (3) staple foods; and (4) offal foods (liver, kidney). In the nutrient density solution, the patterns were characterised by: (1) vegetables and berries; (2) low fat fish, meat and dairy products; (3) mushrooms and offal foods (liver, kidney); and (4) vegetable fats. These patterns were consistent with prior knowledge of nutrient composition. The basic structures of nutrient content among foods seem to have held constant over recent decades. The study also attests to the benefit of increased consumption of vegetables, fruit and cereals.
Keywords :
Pattern , Factor Analysis , Food-based dietary guidelines , FBDG , FCDB , Food composition database