Introduction Light-to-moderate alcohol consumption is associated with decreased mortality from cardiovascular diseases in most studies in the West. Data on the health effects of alcohol drinking in China, which involves mainly distilled spirit, are limited.
Methods A nationally representative prospective cohort study included 220 000 Chinese men aged 40–79 years who were recruited in 1990–1991 from 45 widely separated study sites. By 1 January 2006, over 40 000 deaths had been recorded at age 40–80 years. Cox regression was used to relate alcohol drinking to cause-specific mortality, adjusting simultaneously for age, area, smoking and education.
Results About a third of participants reported drinking alcohol regularly (ie, drank at least once a week), mainly involving spirit. The estimated mean alcohol consumption among regular drinkers was 372 g/week. There was a J-shaped association between amount of alcohol consumed and all-cause mortality, after excluding all men with prior disease at baseline and the first 3 years of follow-up, with those who drank <140 g/week having the lowest overall mortality. Regular alcohol drinking was hardly related to respiratory mortality, but J-shaped associated with mortality from IHD, stomach cancer and lung cancer, and positively associated with mortality from stroke, oesophageal cancer and liver cirrhosis and accidental deaths (eg, violence, suicide and trauma). There is no apparent evidence of reduced mortality from CVD among light-to-moderate drinkers.
Conclusion Among Chinese adults aged 40–79 there was a J-shaped relationship between alcohol consumption and overall mortality.