Systematic error and confounding: Meta-analyses of alcohol and disease

This is a collaborative work that was led by Dr. Kaye Fillmore from the University of California, San Francisco campus until her death in early 2013. It was initially funded for a two-year period by a prestigious challenge grant from the US National Institutes of Health. Following this, the team was successful in applying for a three-year continuation to conduct meta-analysis and explore sources of variation in the relationships reported between levels of alcohol consumption and different disease outcomes. This new grant focuses on additional disease outcomes, including breast cancer, dementia, coronary heart disease, diabetes, stroke and all-cause mortality. There will be a particular focus on exploring reasons for heterogeneity in study findings and the continuing focus on methodological biases which may result in the false appearance of health benefits from light to moderate drinking.

Funding body: US National Institutes of Health

Progress to date: We have completed systematic reviews and meta-analyses on alcohol use and all-cause mortality, prostate cancer, coronary heart disease and diabetes mortality. Papers are published or in press on all-cause and prostate mortality. The findings indicate that (a) there may be no net benefits to health from drinking within low-risk drinking guidelines, and (b) there is a dose response risk of prostate cancer with alcohol consumption.