Alcohol Variability and Blood Pressure: Results from the Usual Care Group of the Multiple Risk Factor Intervention Trial (MRFIT)
Background: Previous studies suggest associations between pattern of alcohol consumption and blood pressure levels. However, there is little information on years-long patterns and blood pressure. Aim: Assess relationship of year-to-year variability in alcohol intake to blood pressure in alcohol users of the Usual Care (UC) group, Multiple Risk Factor Intervention Trial. Methods: 6,438 men from the MRFIT Usual Care group. Data on alcohol variability, blood pressure (BP) and confounding variables from five 24-hour dietary recalls and BP measurements at years 0,1,2,3, and 6. The intra-individual coefficient of variability (CV) in alcohol intake was computed for three sub-groups, defined by mean of reported intake: sub-group 1, less than 2 drinks/d; sub-group 2, 2-3 drinks/d; sub-group 3, more than 3 drinks/d. Multiple linear regression models --dependent variable blood pressure --were adjusted for mean alcohol intake, age, ethnicity, education, BMI, cigarette smoking, serum cholesterol, special diet, mean calories (in per day models), and mean dietary sodium and potassium intake. Results: Alcohol variability was inversely and significantly related to systolic and diastolic blood pressure in the whole group with alcohol expressed both as grams per day (e.g., for SBP: β= -0.026488, z=-2.132) and per cent energy (e.g., for SBP: β= -0.028594, z=-3.371). In sub-group 1 (n=5,128), the association was inverse but not significant; in sub-group 2 (n=1,073), the association was inverse and significant; in sub-group 3 (heavy drinkers) (n=237), the association was direct but not significant. Conclusions: Whereas year-to-year variability in reported alcohol intake was inversely related to blood pressure in men reporting less than 3 drinks/d, in heavy drinkers this evidence of erratic intake was associated with higher blood pressure.