The ageing population is one of the biggest challenges of the 21st century and it doesn’t get talked about enough. In England and Wales, dementia has been the leading cause of death among women since 2012. As recently as 2010, dementia was not even in the top five causes of deaths for men, but it will not be long before it overtakes heart disease to become the leading killer of men too.

Suffering the diseases of old age is the price you pay for tackling the diseases of middle age. Better medical treatment and healthier lifestyles create more geriatrics further down the line and it is unclear how western health and welfare systems are going to cope with this. A study in the Lancet last week attempted to address the issue. By far the biggest risk factor for dementia is age, but it identified fourteen other risk factors, up from twelve in the journal’s previous analysis in 2020. 

Ploughing through the enormous literature of observational lifestyle epidemiology, the authors conclude that 45 per cent of dementia cases worldwide are preventable, in theory at least. They include hearing loss and bad eyesight. Are these not just symptoms of old age? Apparently not. The authors assure us that epidemiologists have concluded that they are independent risk factors. And indeed it seems that the use of hearing aids by people who are a bit deaf helps prevent the onset of dementia. 

Other risk factors include heart disease, hypertension, heart disease, physical inactivity, diabetes and obesity. But are these all truly independent risk factors? Physical inactivity can cause obesity which can cause diabetes, hypertension and heart disease. Which is the cart and which is the horse? Don’t worry, say the authors of the Lancet study, because “most studies in these meta-analyses adjusted for health conditions”.

Air pollution is also a risk factor, they say, although most of the studies they cite report small or statistically insignificant effects. No matter. For the purposes of the Lancet, these associations are assumed to be real and causal. 

The authors’ faith in observational epidemiology suddenly — and quite tellingly — breaks down when they discuss alcohol. There is a lot of evidence that moderate drinking significantly reduces dementia risk. A meta-analysis published in 2022 found that moderate-to-heavy drinkers are 38 per cent less likely to get dementia than people who have never drunk. The Lancet authors mention that meta-analysis, but attempt to dismiss it with the claim that “some people who are counted as non-drinkers were previously heavy drinkers.” This is merely an assertion and, even if true, it would only be relevant if ex-drinkers were at greater risk of getting dementia. In fact, the meta-analysis separated former drinkers from lifetime abstainers and found “no evidence of differences between life-time abstainers and former drinkers in terms of dementia risk”. Other studies have found the same thing. The “sick quitter” hypothesis is only of interest if the quitters are sick but, at least in terms of dementia, they are not. 

That doesn’t stop the authors from using this zombie argument three more times in their brief review of the evidence. They accept that epidemiological studies “usually find a J-shaped dose–response, such that not drinking is associated with increasing dementia risk compared with light drinking”, but they immediately speculate that this is “probably because many non-drinkers have previously had high alcohol consumption or other illnesses that prevent them from drinking”. They then claim that “studies that correct for previous high alcohol consumption have reported that there is no excess mortality in the non-drinking group” and yet the study they cite to back this up shows nothing of the sort. It actually finds clear physiological evidence “in support of long-term health benefits related to alcohol consumption”

The authors then mention a recent study from Korea which found, as usual, that “maintaining mild to moderate alcohol consumption is associated with a decreased risk of dementia”. Once again, the Lancet authors are quick to speculate that “some non-drinkers might have been former heavy drinkers”. They then conclude that a “lack of clear evidence exists that not drinking alcohol increases the risk of dementia” and then add for a final time, just in case the reader had not got the message, that “observational evidence of excess risk for non-drinkers might be due to people who have previously drunk large amounts”.

They have no such quibbles about evidence that suggests heavy drinking causes dementia, of course. That evidence is actually much flakier — the aforementioned meta-analysis found “no consistent evidence to suggest that the amount of alcohol consumed in later life is associated with dementia risk” — but the Lancet authors conclude that drinking more than 21 units a week is a risk factor for dementia based on a study which found that people who drink until they pass out are twice as likely to suffer from dementia than moderate drinkers!

Heavy drinking probably does cause dementia, although you’d have to drink a lot more than 21 units a week, but the refusal to acknowledge that teetotallers are greater risk than most drinkers is pig-headed. Doctors are never going to recommend that non-drinkers start drinking. A medical journal that portrayed alcohol as in any way beneficial would be considered ideologically unsound in the current year. And so we are left with a soft Lysenkoism in which objective facts must be denied for the greater good. 

The same is even more true of nicotine, which is not mentioned at all in the Lancet study. The authors are keen to stress that epidemiological studies show that smokers are at greater risk of dementia (and, therefore, that smoking must cause dementia), but there is substantial evidence that nicotine confers all sorts of cognitive benefits that could be harnessed to tackle the disease. Research is occasionally commissioned to investigate this further, but there has been a distinct lack of urgency and any positive findings would be resisted by anti-vaping activists who portray nicotine as a “brain poison”.

An economic modelling study in the same issue of the Lancet saw “public health” academics retreat into their comfort zone. Having identified their fourteen risk factors, they sought to calculate the pseudo-economic cost savings that could come about from preventive action. By a remarkable coincidence, all the anti-dementia policies happen to be things that they have been itching to do for other reasons: minimum pricing for alcohol, higher tobacco taxes, food reformulation, low emissions zones and mandatory cycle helmets. The only surprise is that they didn’t manage to find an excuse to crowbar in some anti-gambling legislation.

It is undoubtedly true that more people would have dementia if nobody smoked. It is probably true that having a nicotine pouch with your daily drink gives you some protection from dementia. But none of this can be said out loud. According to the Lancet, smoking, obesity and “excessive alcohol” are responsible for just 4 per cent of the world’s dementia cases, but that will not stop them being the bogeymen of an ageing society. The diagnosis may be different but the prescription remains the same.