Ali et al. recruited two cohorts of heavy drinkers (>60 units per week for men or >40 units for women): 291 with decompensated alcoholic liver disease and 208 with no evidence of liver disease (controls). Data were collected regarding drinking behaviour and the presence of liver disease in parents and siblings of cases and controls. Analysis of these data did not reveal a strong familial predisposition to alcoholic liver disease; the authors believe environmental factors are probably more important.