What the research actually says
So, what are we to make of this research? In studies such as this, one of the many pitfalls of exploring the association between an exposure variable (alcohol) and the outcome variable (dementia) is the presence of confounders. Confounders are a statistician’s worse nightmare – they are variables that may justifiably be associated with both the exposure and outcome variable.
In this study, abstainers with higher levels of illnesses such as heart disease, diabetes and obesity were those at highest risk of dementia. It may be that having a physical illness is the reason that these people abstained from alcohol. These same physical illnesses may also be the reason for developing dementia. In this way, physical illness unrelated to alcohol consumption could be a confounder for both abstinence and dementia. Drinking may just be a red herring…For those that do drink alcohol, people drinking above lower risk limits has, once again, been given a salutary warning. It is one that we should continue to heed – but we also need to get the message out there that alcohol-related dementia needs to be rediscovered and investigated again.