Many studies like this don't exclude abstaining alcoholics that _used_ to be heavy drinkers, so you'll get strange results like "people drinking no alcohol are

Many studies like this don't exclude abstaining alcoholics that _used_ to be heavy drinkers, so you'll get strange results like "people drinking no alcohol are more likely to get liver cancer than moderate drinkers".