Most major studies have found a U-shaped relationship between the level of alcohol consumption and all cause mortality, largely as a consequence of lower death rates from coronary heart disease (CHD) amongst moderate drinkers. Previous attempts to unravel the significance of this observation have focused on controlling for possible confounders, such as smoking, social class and the existence of previous ill-health in the group of abstainers. Our analysis of data from the Whitehall II study of British Civil Servants sought to determine whether psychological factors (GHQ, Hostility, Affect Balance, Social Supports) may be influencing the observed relationships between levels of alcohol consumption and some of the established risk factors for CHD. We found evidence of weak confounding only with respect to levels of apolipoprotein B (APoB) and as such have failed to provide compelling evidence that the U-shaped relationship between alcohol and CHD mortality could be easily explained by psychosocial confounding. At the same time we would not claim that the measures we have used are either flawless or exhaust the range of psychological variables that might plausibly influence physiological mediators of cardiovascular disease.