Background: There is controversy about the relationship between coffee intake and blood lipids in women and many studies fail to control for potential confounding factors.
Methods: In 1983 and 1984, 541 randomly selected premenopausal women were recruited from a list of licenced drivers, aged 42-50, within selected areas in Allegheny County, Pennsylvania. They were invited to participate in a 5-year study of biological and behavioural changes associated with menopause. At baseline and the first follow-up visit, lipoprotein, lipid and apoprotein levels, physical activity, dietary intake and coffee consumption were measured. Pearson correlations between coffee consumption and the various lipid values were calculated. Multivariate analyses, which controlled for the potential confounding effects of body mass index, alcohol intake, percent of calories from carbohydrate and fat, number of cigarettes smoked per day, physical activity level, and age in months, were also employed to investigate the relationships.
Results: Mean coffee consumption for the group was 3.35 cups of coffee per day at baseline and 3.02 cups per day at follow-up. Results of analyses of correlation between coffee consumption and blood lipids as well as multiple regressions to control for menopausal status at follow-up were non-significant (P > 0.05) for all blood lipid values with the exception of triglycerides, which were inversely related to coffee consumption at follow-up.
Conclusions: These results do not support a relationship between coffee consumption and lipoprotein, lipid or apoprotein levels with the exception of an inverse relationship between coffee consumption and triglyceride level at follow-up.