No Evidence for an Early Dementia Epidemic

By Neuroskeptic | August 7, 2015 9:13 am

Another day, another alarming brain-related story hits the news:

Dementia is striking victims earlier and death rates are soaring

Modern living has led to earlier dementia, says study

In fact the study in question doesn’t show that. The paper, published in Surgical Neurological International (see also) is from British researchers Colin Pritchard and Emily Rosenorn-Lanng.


The authors show that the number of deaths attributed to neurological diseases (including dementia) have risen in recent years. Using World Health Organization (WHO) data from twenty-one industrialized countries, Pritchard and Rosenorn-Lanng found that the rate of neurological deaths in the over 75s more than doubled (112% increase) in twenty years, 1990 to 2010.

That’s the finding. At first sight, this does seem a “Cause for Concern”, as the authors put it in the paper’s title. Neurological diseases are killing twice as many people! But that’s just one interpretation of these results.

Another interpretation is that people are just getting older. From 1990 to 2010, life expectancy increased and there are more old people now. Everyone knows that dementia is more common in old people. So an ageing population will, all else being equal, inevitably suffer more dementia.

Pritchard and Rosenorn-Lanng do address this point, but in my opinion, not succesfully. As ‘control conditions’, they consider deaths from cancer and heart disease in the 75+ group. These deaths have not increased from 1990 to 2010 – in fact they have fallen. The authors argue that since cancer and heart disease are diseases of old age too, the “people are just older” argument would equally well apply to those diseases, yet neurological diseases have increased more, so there must be more than just an age effect.

The problem is that this is comparing apples to oranges. For one thing, cancer and heart disease are often treatable, and we’re getting better at treating them. There have been lots of new drugs, treatments and screening programs for cancer and heart disease since 1990, so it’s no surprise that death rates fell. Dementia, on the other hand, is not treatable, nor are many other neurological disorders. Pritchard and Rosenorn-Lanng say

The disproportionate rises in [neurological deaths] compared to [cancer and heart disease deaths] might be attributed to more effective treatment of these conditions… However, would this account for the doubling of [neurological deaths] in most countries in such a relatively short‑time?

But this is hand-waving. We need data to say whether the ageing population can explain the rise in neurological deaths. We’d need to know the rates of dementia at different ages, for one thing. There’s no such data in this paper, yet the authors are willing to dismiss the possibility out of hand.

Another problem is diagnosis. Awareness of dementia is rising, and doctors today might be more willing to diagnose the condition than they were in 1990. This could plausibly explain some, or all, of the rise in recorded neurological deaths. Pritchard and Rosenorn-Lanng acknowledge the problem of diagnostic “fashions”, but they brush it aside, saying that “the WHO international data is the most reliable available”.

Perhaps it is the best available, but that’s not the question. The question is whether it is reliable enough, and there’s no way of knowing.

Having concluded that the rise in neurological deaths from 1990 to 2010 is probably real, Pritchard and Rosenorn-Lanng go on to speculate as to the possible causes. Except they claim not be speculating:

We avoid the temptation to speculate but point toward a number of studies indicating possible epigenetic factors influencing neurological morbidity,suggesting that possible nongenetic influences on gene expression, may be entertained.

This is right out of Susan Greenfield‘s playbook: she, too, once claimed that she was not speculating but merely pointing: “I point to the increase in autism and I point to internet use. That’s all“.

Actually, many aspects of this paper reminded me of the ‘autism epidemic‘ controversy, such as the dismissal of prosaic explanations in favor of scary hypotheses about environmental toxins.

Pritchard and Rosenorn-Lanng suggest that the factors responsible for the epidemic of neurological disorders might include

Increased population, economic activity, substantial rises in road and air travel; increased home technology involving background electromagnetic fields (mobile phones, microwave ovens, computers), which are unique to these later years and these possible environmental factors cannot be ignored, especially as they probably interact. This list of possible features might be described as “modern living”…

The headline-writers, of course, made much of this paragraph, but this is not science, it’s speculation.

P.S. Those headlines about how this study found dementia striking earlier? Outright wrong. Pritchard and Rosenorn-Lanng found that rates of neurological deaths in people aged 55-74 years were effectively unchanged, with an increase of just 2% in men and 1% in women.

ResearchBlogging.orgPritchard, C., & Rosenorn-Lanng, E. (2015). Neurological deaths of American adults (55-74) and the over 75′s by sex compared with 20 Western countries 1989-2010: Cause for concern Surgical Neurology International, 6 (1) DOI: 10.4103/2152-7806.161420

  • feloniousgrammar

    Not living in fear of epidemics is becoming a radical act.

  • MichaelHelm

    55-74’s: “In
    1989–1991, the TC average rate was 495 pm, increasing
    to 503 pm over the period, equivalent to a 2% rise.
    Initially, the USA rate was 344 pm, and was 17th highest
    out of 21 countries but is now second highest at 627 pm,
    and had the biggest increase of all countries at 82%.”
    Not a lot of people, but …..

    • Neuroskeptic

      There was an increase in the under 75s in the USA, but I don’t think we can read too much into that; the increase was counterbalanced by decreases in other countries such as the UK. Out of 21 countries, one of them has to show the largest increase.

      • MichaelHelm

        I think the authors ask a good question (in a somewhat different context): “However, would this account for the doubling
        of TND in most countries in such a relatively
        short‑time, and for the remarkable rises in the USA?”
        It seems worth looking into, not just shuffling the cards, frankly. Maybe it’s just our reporting methodology, maybe it’s random, maybe we’re screwed. Wouldn’t it be nice to know?

        • Neuroskeptic

          It would be nice to know, but at this stage I’m not worried – we have two plausible explanations, 1) age and 2) changes in diagnosis. If there is a “real” increase in dementia rates then we should find out why, but I would be amazed if the real increase was a doubling.

  • David Roberts

    If the rise was due to more people living longer, then it should be normalized to victims per 1000 in that age group.

    • Neuroskeptic

      Agreed – age-normalized data would be extremely useful. Which is what I meant by: “We’d need to know the rates of dementia at different ages.”

  • PsyoSkeptic

    I have two comments. First, you only need a small shift in the success rate of treating the most common causes of death to have a large change in the proportion of incidence of less common causes of death when the causes are primarily age related. Second, and probably more importantly, it fails to note that a common outcome of surgery in the elderly (50%!!) is cognitive problems including dementia. It could very well be that the treatment of one disease is a cause of another.

    • vijay

      What’s the suggested causal link between surgery and cognitive problems, I’d be interested to know (also the citation for that statistic)? I’m asking because it could just be the same thing, right? increased diagnosis/attribution relating to healthcare contact? No?

      • PsyoSkeptic

        Anaesthesia is the link (should have put that in)… don’t have a link… ask any anaesthesiologist. They’ll tell you, if you’re over 70 and need general only do it if it’s life threatening.

        • vijay

          erm… my point was that if the elderly aren’t routinely assessed (by anesthetists or others) immediately before, and at a sufficiently distal time point /after/ anesthesia, there’s still the possibility that this is just increased ascertainment rather than a true difference.

    • rsconrad3

      Where are you getting that data about surgery?

  • Tim Shakespeare

    Great post. I was thinking ‘at least it’s not in mainstream media’ but just noticed it in the guardian Doh!

  • Pingback: NeuroBreak: Alzheimer's Data War, Coltrane for Epilepsy? - Tus dientes sanos - Salud y Vida()

  • Hanna Isotalus

    Well written, thank you.

    I agree that if there was an actual rise in dementia incidence in the ageing population that would be worth investigating. However, I don’t think it’s unlikely that there is. Of course we’d need better evidence of this than the paper you cite provides. Those who live longer now may very well be at a higher risk of dementia than those who lived longer before many modern advances in medical care took place – and that’s not only because of changes diagnostic practice.

    This hypothetical rise in dementia could be explained by the *at risk* populations living older. Those people who are at high risk of developing cardiovascular problems and cancer which, as you state, can now be treated and diagnosed easier, tend to also be at a higher risk of developing dementia. Smoking, high cholesterol, diabetes, obesity, inactive lifestyle, history of cardiovascular problem, and a number of other risk factors for disease that has become preventable, curable or manageable also increase risk of many of those neurological diseases we can’t treat – including the most common causes of dementia, e.g. Alzheimer’s disease and vascular dementia.

    Therefore, those people who live longer because of changes in medical practice may be more likely to develop dementia than those who would’ve lived longer despite medical developments.

    • Neuroskeptic

      That’s an excellent point. Obesity might well be a link since we know that it’s becoming more common.

    • van win

      If Doctor Malcolm Kendrick is correct, the cholesterol theory has gone out of the window. He has a website.

  • Danny Hoardern

    I heard a great “All In The Mind” episode recently (“The distracted mind”) which revealed multitasking depletes glucose in the insula (2m:55s). Seems best to learn to shut out distractions.

    Cannabis mitigates dementia and Alzheimer’s Disease…

  • Lady Bird

    Amazing article! Doing some research I usually use this tool it’s the best converter, which makes my work easier.



No brain. No gain.

About Neuroskeptic

Neuroskeptic is a British neuroscientist who takes a skeptical look at his own field, and beyond. His blog offers a look at the latest developments in neuroscience, psychiatry and psychology through a critical lens.


See More

@Neuro_Skeptic on Twitter


Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!

Collapse bottom bar