Correlates of low density lipoprotein cholesterol: associations with physical, chemical, dietary, and behavioral characteristics.

The purpose of this cross-sectional study was to describe the relationship between low density lipoprotein cholesterol (LDL-C) and various physical, chemical, dietary, and behavioral attributes of 4768 white men and women aged 20-59 years from 10 North American populations sampled by the Lipid Research Clinics Program. For our analyses, women were separated into two groups: users and nonusers of hormones. A positive correlation coefficient between very low density lipoprotein cholesterol (VLDL-C) and LDL-C was observed for women but not for men. Fasting plasma concentrations of LDL-C were most strongly and consistently associated in each gender with age, the Quetelet Index of body mass, the number of cigarettes smoked per day, systolic blood pressure, and the levels of plasma glucose and uric acid (all positive associations); and with height and bilirubin levels (both negative associations). Gender-related differences were found in the associations between LDL-C and the other lipids and lipoprotein cholesterol fractions measured. The associations between LDL-C and individual nutrients computed from a 24-hour dietary recall were generally of a low-order magnitude. The strongest and most consistent among the statistically significant associations were those between LDL-C and total caloric intake and with carbohydrates and sucrose (all negative associations). Other statistically significant association between LDL-C and reported intake of nutrients were less consistent, e.g., LDL-C levels were negatively related to alcohol consumption in women, but showed a weak positive relation to alcohol intake in men. Multiple linear stepwise regression was used to assess the variation in LDL-C concentration attributable to each characteristic under study, adjusted for concomitant variation in any other variable. As in the univariable analyses, age was the strongest covariate of LDL-C; adjusting for all other terms in the stepwise regression analysis, cigarette smoking was identified as the next strongest determinant of LDL-C. Other measurements, such as body mass, clinical chemistries, and alcohol consumption, emerged as independent predictors of LDL-C levels in a pattern specific to gender and to hormone use.