Tangent: a pet peeve of mine is plots of alcohol vs disease plots that suggest that one or two drinks a day is healthier than none. This is not the case.
The data sets that produce these plots almost never take into account why people stop drinking. When you do control for this, and in the process filter out people who stop drinking because of other severe health issues, it becomes obvious that not drinking at all is the healthiest option.
The data sets that produce these plots almost never take into account why people stop drinking. When you do control for this, and in the process filter out people who stop drinking because of other severe health issues, it becomes obvious that not drinking at all is the healthiest option.