A Statistical Investigation

On the WHO-says-bacon-as-bad-as-cigarettes business:

Experts attack claims that bacon is ‘as big a cancer threat as smoking’

Dr Ian Johnson, Emeritus Fellow, Institute of Food Research, said: “Although there is epidemiological evidence for a statistically significant association between processed meat consumption and bowel cancer, it is important to emphasise that the size of the effect is relatively small, and the mechanism is poorly defined.

“It is certainly very inappropriate to suggest that any adverse effect of bacon and sausages on the risk of bowel cancer is comparable to the dangers of tobacco smoke, which is loaded with known chemical carcinogens and increases the risk of lung cancer in cigarette smokers by around 20 fold.”

One of the things that always puzzles me about many of these epidemiological statistical studies is that many of them have been carried out by doctors. For example, Richard Doll was a doctor. So also was Ernst Wynder. And George Godber. And the WHO is behind the bacon scare.  And so what we often have are a bunch of doctors doing statistics, in which they probably had little or no training. They were operating “out of area”.

And why is it that somebody from Food Research responds to these statistical studies?

I’m always wondering what real, professional statisticians think about these studies. So today I set off to do some investigating.

The first thing I discovered is that Britain has a Royal Statistical Society. I wondered whether it would be trumpeting the bacon-causes-cancer scare, but there was no mention of it at all.

Eventually I found myself watching a few videos on the website, which were actually quite interesting. They were discussing what went wrong at the UK General Election earlier this year, which pollsters said would result in a hung parliament, but which the Conservatives actually won. I’d expected to see lots of mathematics, but there was none at all. And the participants were from polling companies, the BBC, and various newspapers. They all just talked, and made some interesting points.

I then watched a video of the current president of the Society giving what seemed to be a review of the state of the society in June 2015. He started off saying that the appearance of computer statistics packages had been widely predicted to be the death of the profession, but said that in fact statisticians were as much in demand as ever. He talked quite a lot about computers and computing languages, and mentioned Sir Ronald Fisher using a giant early calculating machine. There was no mention of Richard Doll or Bradford Hill. He also discussed the use of statistical methods in Africa, to keep track on various communicable diseases, and the difficulties in doing so.

He talked for about an hour, and I kept hoping/expecting him to talk about tobacco, alcohol, sugar, salt, and the flood of statistical studies we’re all too familiar with here. But there was not a word about any of them.

I also took a look at the society’s Code of Conduct, and was a little disturbed to find:

1. Fellows should always be aware of their overriding responsibility to the public good; including public health, safety and environment.

This might mean the Royal Statistical Society has been overrun by healthists and environmentalists, like almost everywhere else. But none of the videos I saw this afternoon seemed to be particularly health- or environment-oriented. So the jury has to be out on that. But I would think that honesty and integrity should have been the prime requirements in a code of conduct, not public health and safety.

But, while I’d heard several interesting talks, I still had no idea what the attitude of the Royal Statistical Society to the tidal wave of bacon-causes-cancer, cheese-as-addictive-as-hard-drugs statistical studies might be.

So I thought I’d just go straight to the top, and email the President of the Royal Statistical Society to ask him. I cited various of these scaremongering stories in the media, and said that I didn’t believe a word of any of them. My question to him was; “Do you not think that there is a considerable danger that the entire science of statistics is being brought into disrepute by the plethora of these scare stories?”

I have no idea whether I’ll get a reply from such an august figure, but he seemed to be a pretty down-to-earth Lancastrian. And since he had also listed his favourite restaurant as an Italian pizza outlet (lots of cheese, salami, bacon, ham!!) I deduced that he wasn’t exactly a vegetarian.

Anyway, I came away from the investigation with the impression that the society was a small (I had the impression that the president was addressing 50 or so people) but active one, and slightly fearful of its own extinction, and also wrapped up in what seemed to me to be silly questions about whether they were doing “mathematical statistics” or “statistical mathematics”.

I’ll let you know if I get a reply. I suspect that I’ll get a pretty fulsome reply if he’s as sick of all these scare stories as I am. And no reply at all if he’s a fully paid up healthist.


About the archivist

This entry was posted in Uncategorized and tagged , , . Bookmark the permalink.

55 Responses to A Statistical Investigation

  1. Roobeedoo2 says:

    Good luck, Frank, though the RSS got it’s Body Licence in 2009 from this lot:


  2. harleyrider1978 says:

    time for an oldie but a goody

    Epidemiologists Vote to Keep Doing Junk Science

    Epidemiology Monitor (October 1997)

    An estimated 300 attendees a recent meeting of the American College of
    Epidemiology voted approximately 2 to 1 to keep doing junk science!

    Specifically, the attending epidemiologists voted against a motion
    proposed in an Oxford-style debate that “risk factor” epidemiology is
    placing the field of epidemiology at risk of losing its credibility.

    Risk factor epidemiology focuses on specific cause-and-effect
    relationships–like heavy coffee drinking increases heart attack risk. A
    different approach to epidemiology might take a broader
    perspective–placing heart attack risk in the context of more than just
    one risk factor, including social factors.

    Risk factor epidemiology is nothing more than a perpetual junk science machine.

    But as NIEHS epidemiologist Marilyn Tseng said “It’s hard to be an
    epidemiologist and vote that what most of us are doing is actually harmful
    to epidemiology.”

    But who really cares about what they’re doing to epidemiology. I thought
    it was public health that mattered!

    we have seen the “SELECTIVE” blindness disease that
    Scientist have practiced over the past ten years. Seems the only color they
    see is GREEN BACKS, it’s a very infectious disease that has spread through
    the Scientific community with the same speed that any infectious disease
    would spread. And has affected the T(thinking) Cells as well as sight.

    Seems their eyes see only what their paid to see. To be honest, I feel
    after the Agent Orange Ranch Hand Study, and the Sl-utz and Nutz Implant
    Study, they have cast a dark shadow over their profession of being anything
    other than traveling professional witnesses for corporate hire with a lack
    of moral concern to their obligation of science and truth.

    The true “Risk Factor” is a question of ; will they ever be able to earn
    back the respect of their profession as an Oath to Science, instead of
    corporate paid witnesses with selective vision?
    Oh, if this seems way harsh, it’s nothing compared to the damage of peoples
    lives that selective blindness has caused!

    • harleyrider1978 says:

      Pretty well sums it up and all the WHO and IARCs bullshit amounts too.

      • harleyrider1978 says:

        Of course all post grads have to pass this course

        NIH Forms & Applications | grants.nih.gov


        National Institutes of Health
        Standard Form 424 (Research & Related) Grant Application Forms Includes … Continuation – Research Performance Progress Report for an NIH Grant. The RPPR is required …. HHS 568 (MS Word – 40 KB), 06/2015, Final Invention Statement.

      • churchmouse says:

        Thanks, I was just thinking of the same report.

        Next week we’ll get something on the order of ‘Cured meat helps you live longer’ from some other statistical organisation.

        We’ve seen it with caffeine, beer and wine — why not bacon and sausages?

        • harleyrider1978 says:

          remember eggs was hit in the 1990s we all should have learned what was to come back then especially with the EPA junk study on shs.

  3. harleyrider1978 says:

    2009 _ http://www.ncbi.nlm.nih.gov/pubmed/19279082 —– CONCLUSIONS:

    “The overall cancer incidence rates of both the vegetarians and the nonvegetarians in this study are low compared with national rates. Within the study, the incidence of all cancers combined was lower among vegetarians than among meat eaters, but the incidence of colorectal cancer was higher in vegetarians than in meat eaters”.

    Cancer incidence in vegetarians: results from the European Prospective…

    ncbi.nlm.nih.gov|By Key TJ , et al.

    stolen from mandy Vincent

  4. petesquiz says:

    Yesterday, when the story first came out I was appalled by the reporting on BBC Radio 5 Live when their news headline kept on saying that processed meat caused cancer even after interviewing a doctor who said that it only increased the risk by 18%. He even said that it was nowhere near as bad as smoking and that red meat was even less of a risk. As usual, no journalist asked the vital question, “What IS the risk of developing colorectal cancer?”

    So I did some searching and found this link to a nice table showing how the risk of developing colorectal cancer varies with age and what the % risk of getting it in 10, 20 or 30 years.


    So, for example a man of my age (call it 60 for ease of calculation) the risk of getting colorectal cancer in 10 years is 1.26%, with added bacon and sausage it now ‘leaps’ to 1.49%. I’m no statistician, but I’d say that’s not a significant increase in risk! So, it’ll be a full English Breakfast in the morning with Fried Bread done in Lard – let’s see if I can get that risk up to a full 1.5%!!

  5. Tony says:

    I hope you do get a sane reply from the Royal Statistical Society but I’m not optimistic. I seem to remember looking into this a few years ago and finding that they had been crushed underfoot long before the Royal Society suffered the same fate.

    Epidemiologists had been abusing statistical methods for decades but described themselves as ‘statisticians’. Much like the ‘environmental scientists’ did later, they demanded representation in the appropriate society. They probably persuaded the admin staff by promising money, influence and prestige. The actual members may have had little or no say.

    Once in place they dominated because there were more of them and they had connections and money. I saw a talk once where a statistician joked that if two statisticians were in conversation, you could tell which one was the extrovert because he was the one looking at the other guy’s shoes. So I suspect they were not all that good at fighting their corner.

    I should point out that I am merely an outside observer. I have no inside knowledge.

  6. Lepercolonist says:

    Maybe a retired statistician on a pension may give a more honest reply to these scaremongering studies ?

    • Frank Davis says:

      You’re probably right. While they’re still working, these people probably come under very strong pressure to conform, or else… Once they’ve retired, there’s no ‘or else’ to it, and they can say what they actually believe.

      But must we really wait for everyone to retire before they start telling the truth?

  7. waltc says:

    In an odd way, I like this scaremongering. If eating a bacon burger is as dangerous as smoking, then smoking is only as dangerous as eating a bacon burger. Maybe this helps to put things in perspective for the willing-to-be-scared.

    Then too this new scare was also derived by a meta-analysis which, before science became “science”, was widely disapproved of . Meta’s combine studies done by different methods, on different kinds of populations, in different countries, using different confounders (or, as pointed out,,in some cases, none) and with different numbers of subjects. A study of 12 people is given equal weight to a study of 12,000. Then too, researcher bias can eliminate studies from the mixture that they don’t like (ie, that would change the outcome and contradict the agenda). The latter was blatantly done by the EPA in its”landmark report” on secondhand smoke and was admittedly done by researchers on the link between smoke and heart trouble.

    If you’re looking for a solid statistician, is the estimable Brian Bond still around?

    • Frank Davis says:

      then smoking is only as dangerous as eating a bacon burger.

      Exactly. But how many people are able to draw that conclusion.

    • BrianB says:

      Still around, Walt! I come here every day, but I rarely post, as I find too often that the discussion has moved on to a new topic/thread by the time I have gathered my thoughts!

      With respect to the Royal Statistical Society, they no more represent the views of their members than do the many august societies who take positions on CAGW without consulting their members. I was a member for a year after graduating, but gave it up as I saw no particularly valuable return for my fees.

      I’m glad someone else shares my absolute hatred of meta-analyses! I have been going on about this for some considerable time, and my position is that they are just a form of cheating. They first select (cherry pick) a bunch of ‘studies’ with similar findings in terms of ‘risk’ statistics, but most of which failed to achieve statistical significance because of relatively small sample sizes. The ‘meta-analysts’ then combine the data from the selected studies, to calculate a ‘pooled’ (you’ll see this word in the WHO report) risk ratio that now has a super sized sample, and thus magically converts a bunch of statistically insignificant results into a single significant one.

      In other words, it is a technique for making a strong chain out of weak links!

      But it is deeply flawed. For a start, if you look at the history of meta analysis techniques, they have been driven, almost exclusively, by the epidemiology and medical statistics disciplines. This doesn’t surprise me, as I’ve long seen how these people are sufficiently arrogant to redefine statistics (or other disciplines) to suit their own agendas.

      Secondly, meta analysis is wide open to bias, especially in the selection of studies to ‘pool’, but also in the alleged ‘weightings’ that are applied to each study according to its own individual power. Any such weightings are only ever derived from other ‘estimates’, and so bring further ‘error’ into the calculations – error that isn’t included in the final significance testing. And, of course, they inevitably wind up with a relative risk statistic that is within the 1.1 – 1.5 range (as per various passive smoking meta-analyses – and as, too, with the WHO in their latest piece of fanaticism), which is so ridiculously close to the ‘no effect’ ratio of 1.0, that no other field of statistical analysis would consider it to be worthy of any response other than “go back and extend your study to try and get the significance level up from 95% to 99%”. We are living in an era of unscientific madness, when a 1.2 risk ratio, with 90% confidence intervals (I refer to the US EPA nonsense), can be concluded to prove causality. This is just dreadful!

      Finally there is a simple rule behind any statistical analyses, and that is that the sample used in a study must be demographically representative of (ie identical to) the population from which it is drawn, and to which any findings would be later attributable. Whilst some (but by no means all) of the individual studies in the meta will have adhered to this rule, the bigger problem comes in when the samples are ‘pooled’, because we now have a (big) sample that has been drawn from many disparate populations – as Walt has pointed out – and thus represents no single population. Thus the golden rule of sampling is broken. Oh but they will, again, come up with various weighting factors, in order to convince us that they ‘control’ for population differences, but the mathematical reality is that they have an invalid sample, and no population-level inferences should be drawn from any of its statistical results.

      It’s all a bit of a vogue at the moment, advocate-researchers carrying out statistical analyses in Epi studies, and over-hyping any small effect statistics that drop out of a data dredge, and achieve the poorest (95%) level of statistical significance. Sadly our political and media classes are just. too mathematically illiterate ever to hold these fanatical cheats to account.

      I could go on (and on!), but it’s just so bloody frustrating!

      • Frank Davis says:

        I find too often that the discussion has moved on to a new topic/thread by the time I have gathered my thoughts!

        That’s my fault for posting almost every day. But discussions do sometimes continue in the comments for some days afterwards.

        I could go on (and on!),

        Well, if you want to write at length, I’d be more than happy for you to write a guest article. Although it would of course only be us lot who read it.

        Incidentally, wasn’t the Enstrom and Kabat study a meta analysis?

        • harleyrider1978 says:

          Frank if I remember right the enstrom study ran for 39 years as an ACS cps2 study. The preliminary results didn’t meet the agenda so the ACS pulled funding and issued a proclamation that from then on they could pull funding at anytime for any reason after kabat and enstrom raised hell about defunding due to bad results for the Nazis. So BT came in and said we will pick up the tab and as far as I can remember they kept up with the results via those study questionaires thru out those years,but some with a 5-10 year span between answers coming back.

        • BrianB says:

          No, the E&K passive smoking v. lung cancer study was just one very large longitudinal study – with a huge sample. I believe that they did, later, undertake a meta analysis into passive smoking and heart disease, although IIRC it was more of a literature review than a calculation of pooled risk.

          Of course, where the Enstrom & Kabat study gained ‘meta’ notoriety, is that it never made it into the selections for any of the big meta analyses (EPA, US Surgeon General, SCOTH) that supposedly proved the causal link between ETS and Lung Cancer. Had it done so, the ‘pooled’ results would have been markedly different – which, of course, they all knew only too well!

          I would like to write a guest article – when I feel sufficiently inspired. “Although it would of course only be us lot who read it” – fine by me, since I consider you all to be my friends.

  8. smokingscot says:

    Thing is they’re claiming 35,000 deaths through processed / red meat WORLDWIDE.

    Personally I’d have expected them to concentrate on far more immediate threats, but I do acknowledge they’re not in the business of dealing with issues that affect the 3rd world. Seems the way to get the headlines is to go after the developed world, or rather the middle class developed world.

    Taken a quick gander at the WHO original document (or at least the one that came up first at Google), and it strikes me as really very vague and wishy washy.


    I have a great historical interest in hot dog sausages and note that they claim cooking them on the BBQ or under a grill makes them more toxic. (I’d also assume the same may be true of lightly frying (in its own fat) salami). Yet they haven’t a clue if simply heating hot dog sausages in water actually neutralises or at least dissolves all the nasty chemical additives.

    And here I face a slight problem. We’ve had processed meat for yonks. We’ve been salting it, smoking it and drying it since time began – and frankly naff-all ever happened. What we did was simply stored when we had plenty to see us through the winter.

    So it’s not the processing that causes the problem, it’s the chemicals and E numbers and such that the food industry adds. Do they add this stuff because they want to, or do they add them because some pratt legislated that they should?

    Oh and I strongly suspect a whole load of Eskimo’s will be equally disparaging of what’s utter overblown rot. Their diet consists almost entirely of animal protein, with blubber a real delicacy.

    By the way I saw a talking head from the WHO trying to spin this. The name on the screen was DAN LOOMIS. Done a search on that name and bog-all came up, so what we have is a bunch of overpaid nonentities with too much time on their hands.

    Now what does that remind me of? Maybe the stuff we flick off our fags.

  9. harleyrider1978 says:

    WHO: Questionable Science And Conflicts Of Interest‏

    by David Atherton28 Oct 20155

    The World Health Organization (WHO) has recently added Britain’s breakfast favourites such as bacon, sausages and other red meat to the list of the ‘encyclopaedia of carcinogens’. All processed red meat is to join the premier league of cancer causing agents such as smoking, asbestos, alcohol and argon gas.

    However, there is evidence the WHO is not being entirely rigorous or scientific throughout a whole range of sciences.

  10. harleyrider1978 says:

    Only one thing missing from their claims,High taxation is what put them in poverty not smoking.

    Smoking a cause of poverty for many in Calendar region

    Smoking could be the reason that thousands of families in our region live in poverty, it has been claimed.

    New figures show a third of households which include an adult smoker in Yorkshire and northern Lincolnshire are below the poverty line.

    Action on Smoking and Health (ASH) says quitting smoking would lift 118,000 people in the Calendar region out of poverty.On average a smoker spends £2,158 a year on tobacco.

    Local authorities are working alongside the NHS to bridge the health inequalities gap by reducing smoking rates. People from disadvantaged background appear to be the most affected as they live in high smoking rates areas.

    Smoking puts poorer families under significant financial pressure. This in turn places a burden on local services and negatively impacts the local economy. Our services to help people quit not only save lives but they put money back into the pockets of the poorest households in our community.

    – Dr Andrew Furber, Director of Public Health for Wakefield Council


    • harleyrider1978 says:

      Smoking puts poorer families under significant financial pressure.

      Smoking Taxation puts poorer families under significant financial pressure.

  11. harleyrider1978 says:
  12. harleyrider1978 says:

    Long-time anti-meat vegetarian activist-researcher on WHO meat-cancer panel

    I’ve been writing about National Cancer Institute’s meat-hating vegetarian Rashmi Sinha since 2000.

    Here is the WHO media release. Note the WHO panel’s analysis is not yet published — so science by media release release

    Sinha was on the WHO panel along with, no doubt, many other meat anti-meat researcher-activists. Here’s a 2005 column I wrote about her for FOXnews.com. It references a 2000 column about her as well. She’s been attacking meat since at least 1994.


  13. harleyrider1978 says:

    too good to let pass

  14. Jonathan Bagley says:

    Hi Frank,
    David Spiegelhalter, Professor for the Public Understanding of Risk at U of Cambridge writes very informatively about the bacon/cancer issue. I recommend his website – many excellent article. He often appears on R4 More or Less.


    • petesquiz says:

      Excellent clip! If only our journalists would learn to understand what the absolute risk is before spouting their scaremongering headlines, the world would be a much better place.

      In all of the misinformation and bad science surrounding all of the topics here on Frank’s blog I blame the media for giving oxygen to these spurious claims without even attempting to deconstruct them as Prof Spiegelhalter did above. Without the scandalously poor journalism most of the charlatans and snake oil salesmen of so-called Public Health bodies would have been laughed out of court long ago.

      • slugbop007 says:

        The journalists that I have contacted in Montréal so far are too lazy to do any research on the people that they interview. They never investigate these people’s academic/professional credentials, then they write verbatim whatever baloney they spew. One guy claimed to be a Professor and an air quality research expert, another claimed to have a PhD in medicine, and another is the recipient of the Luther Terry award. The Professor is an Assistant Professor, the PhD recipient just finished her first cycle in Family Medicine and the Luther Terry shmoo just recently called some guy with a PhD in Economics ‘one of Canada’s leading researchers’. Even when I called into question the credentials of these people and their exaggered claims with the journalists that I contacted no alarm bells registered in their brains. Maybe they have stock shares in pharmaceuticals?

    • Frank Davis says:

      Thanks for that. I think I may use those links for tonight’s post!

      And it does look like the real statisticians are fighting back, after all.

      • harleyrider1978 says:

        Frank I just saw a post on Facebook about the study groups were likely of the poorest health to begin with. Then it struck me that quite possibly they have always used the poorest health groups so they can have a distinct advantage in outcome based studies like these.

  15. Jonathan Bagley says:


    I’ve come across both Diggle and Spiegelhalter. They are not Public Health lackeys. Spiegelhalter gave the Million Women study alcohol/cancer claim a real kicking on R4’s More or Less a few years ago. The author refused to appear on the show – probably suspecting what was coming.

  16. slugbop007 says:

    You should tell that Lancastrian that there is nicotine in bell peppers, tomatoes, tomato sauce and tomato paste. Mmm, pizza.

  17. harleyrider1978 says:


    Red meat causes cancer? Bull!

    By Gene Hall

    Apparently, it’s time to beat up on beef again. And bacon. And processed meats.

    This time it’s the World Health Organization. Their International Agency for Research on Cancer (IARC) wants you to believe that eating ham, bacon, pastrami and salami are hazardous to your health. Right up there with exposure to arsenic, asbestos and cigarettes. They say processed meats cause cancer. They conclude that red meat “probably” does the same.

    Misguided, malicious, malarky!

    The report defies common sense. Dozens of studies show no correlation between meat and cancer. It ignores the fact that red meat is rich in protein and a good source of vitamins, iron and zinc.

    It doesn’t take into account the probability of getting cancer is based on a number of factors. Including age, genetics, socioeconomic characteristics and lack of physical activity.

    Eating red meat as dangerous as smoking cigarettes? Really?

    Looks to me IARC is smoking something. And it sure ain’t brisket!


  18. harleyrider1978 says:

    Soda Tax: Not the Answer

    Re “What a Big Tax on Soft Drinks Can Do” (editorial, Oct. 19):

    Discriminatory taxes on beverages do not improve public health, and they divert us from meaningful solutions to complex health challenges.

    In the few places where they have been tried, excise taxes that single out beverages have been shown to have no effect on obesity. West Virginia and Arkansas have longstanding soda taxes and continue to rank in the top 10 most obese states. States with no soda tax, such as Massachusetts and Vermont, are among the least obese states. Even a report by the European Commission that looked at the effect of taxes on common grocery items found “no robust conclusions on the impact of food taxes on public health.”

    Regarding a soda tax in Mexico, even the Times editorial board acknowledges there is no proof that public health improved as a result of the tax. What has happened is the closing of more than 30,000 neighborhood stores in part because of the tax, says Mexico’s National Association of Small Merchants.

    If we want to get serious about obesity and other complex health issues, it starts with education and collaboration to inform people how to maintain a balanced lifestyle. Taxes will not help.


    President and Chief Executive

    American Beverage Association



  19. harleyrider1978 says:

    Today’s Winnipeg Free Press poll. Red and processed meat is the cancer-causer of the day. Will you change your diet?
    No. Moderation is the key.
    81% (1587 votes)
    results bar
    Yes. It’s important to listen to scientists.
    13% (259 votes)
    results bar
    I’m a vegetarian.
    5% (103 votes)
    results bar
    Total Votes: 1949

  20. prog says:

    They’re claiming up to 34,000 ‘bacon’ related cancer deaths pa globally. That’s approx 0.004%. of total number of cancer deaths (8.2 million pa).

    I make it about 6 bacon related deaths pa in England and Wales (0.004% of c.160,000 cancer deaths). About 0.001% of all deaths pa (c.500,000).

    I’d wager PH scaremongering, an obvious cause of stress and reduced immunity, kills 1000s pa.

  21. harleyrider1978 says:

    Angry pedestrian screams at woman for smoking in Martin Place
    VIDEO watch it before it gets yanked down as youtube did already from chris snowdens link


  22. harleyrider1978 says:


    Vince Harden

    7 mins · Edited ·


    smile emoticon More on food-apparently some people’s mom’s did not tell them to not play with their food.Food fight has now broken out!

    [The BMJ investigation, released in September, asserted that the guidelines committee used “weak scientific standards” to make its recommendations. It also criticized several aspects of the new guidelines, such as “deleting meat from the list of foods recommended as part of its healthy diets.”]

    BMJ corrects controversial critique of US dietary guidelines report – Retraction Watch

    The BMJ has published a correction to a critique of the U.S. dietary guidelines report that has received…


  23. harleyrider1978 says:

    World Health Organisation, meat & cancer

    Posted on October 26, 2015 by Zoë Harcombe — 99 Comments ↓

    Today, 26th October 2015, the World Health Organisation declared the consumption of red meat as “probably carcinogenic to humans, based on limited evidence that the consumption of red meat causes cancer in humans” and declared processed meat as “carcinogenic to humans, based on sufficient evidence in humans that the consumption of processed meat causes colorectal cancer.” The red meat association was observed mainly for colorectal cancer.

    “The experts concluded that each 50 gram portion of processed meat eaten daily increases the risk of colorectal cancer by 18%.”

    From the headline “carcinogenicity of consumption of red and processed meat”, we’re already down to colorectal (bowel) cancer and “probably”.

    The press release is here. The Lancet article is here or here (it may not be on open view for long).

    So do we need to stop eating red meat and/or processed meat? Let’s dissect the headline more accurately:

    1) Where this data comes from

    The gold standard of evidence is a meta-analysis of randomised controlled trials – pooling together studies where an intervention was matched against a control group to see what impact A had on B. As far as I am aware, no intervention studies have ever been done testing the impact of 50 grams of processed meat per day as an isolated intervention, or any amount of processed or red meat as a sole intervention for that matter.

    We are thus looking at observational studies. This is where a large group of people (e.g. the Nurses’ Health Study or the Health Professionals Follow-up Study) are asked loads of questions and given health tests (blood pressure, weight, height, cholesterol ho ho etc) at the start of the study. This is called the baseline. These people are then followed for years to see what conditions they go on to develop.

    Researchers then look at the data to try to see patterns. No pattern = no journal article, so look hard! They may observe a pattern between people who consume processed meat and people who go on to develop bowel cancer. This is then reported in a journal article and it is all such articles that have been reviewed by the World Health Organisation.

    The first point to make, therefore, is that all of this is based on notoriously unreliable dietary questionnaires. Many ask what you ate yesterday or over the past 7 days. Here’s the European Prospective Investigation into Cancer questionnaire, one of the best possible questionnaires, as it asks for food intake over the past year. How accurate do you think yours would be?

    2) One’s diet vs. one food

    By singling out red meat/processed meat in this way, the whole diet and lifestyle of a person is not taken into account. There is a world of difference between the health of a burger/hot-dog/ketchup/white bun/fizzy drink guzzling couch potato and a grass-fed-steak eating/CrossFit/six-pack Paleo specimen.

    As I showed in this blog, the baseline for the processed meat eaters showed that they were far less active, had a higher BMI, were THREE TIMES more likely to smoke and almost TWICE as likely to have diabetes. This makes processed meat a MARKER of an unhealthy person, not a MAKER of an unhealthy person.

    Even if all the smoking/exercise/other conditions baseline factors are adjusted for, there is no possibility of adjusting for all the dietary factors that make up the couch potato vs. the Paleo buff. The whole diet is not adjusted for when the one line (meat) is targeted.

    3) Real food vs. processed food

    I’m a real foodie. I pretty much spend my life writing and talking about real food and the nutrition it contains. I am the first to say “Do eat real food; don’t eat processed food” and I include processed meat as processed food – something to avoid. However, this WHO report describes processed meat as “meat that has been transformed through salting, curing, fermentation, smoking or other processes to enhance flavour or improve preservation.”

    As Peter Cleave, Surgeon Captain, (1906-1983) said: “For a modern disease to be related to an old fashioned food is one of the most ludicrous things I have ever heard in my life.” To think that real meat, or meat preserved in natural ways, is bad for us is ludicrous. 1) You’d have to explain how we survived the past 3.5 million years, since Australopithecus Lucy first walked upright; especially how we survived the ice age(s). 2) You’d have to explain why all the nutrients we need to live (essential fats, complete protein, vitamins and minerals) are found in meat if it were trying to kill us at the same time.

    Meat needed to be naturally preserved with salting, curing, drying, smoking etc or we would have needed to binge on the kill and risk dying of starvation before the next kill. The WHO report should have separated traditional ways of preserving meat from modern manufactured processing (where sugars and chemicals are added – just read the label). Similarly – if there is any harm in red meat, it will be because manufacturers have got involved and fed the poor animals grains, which they cannot digest and then pumped them with drugs to medicate the resulting illness. (Chris Kresser presents the view on nitrates here, if you’re interested).

    This should be a call to action to get back to your butcher, know him/her by name, know where your meat comes from, know how s/he prepares bacon & hand-made sausages and enjoy the health benefits of real food while supporting the grafters who provide it.

    4) Association vs. causation

    Even allowing for the weakness of observational studies, and the unreliability of dietary questionnaires, and the notion that food consumption can be a marker not a maker of health, and the whole dietary intake that has not been taken into account and the ignorance of the chasm between real and processed food, this is still association, not causation.

    I always wish that these huge and expensive studies would ask what colour socks the participant is wearing. I bet I could find an association between red sock wearing and one type of cancer if I looked hard enough. Would the headline be red socks cause cancer?!

    5) Relative vs. absolute risk

    The press release headlines with “each 50 gram portion of processed meat eaten daily increases the risk of colorectal cancer by 18%.” Crikey. 18%! Put that bacon sarnie down now (see – don’t blame the bacon for what the white bread & ketchup did!) This, however, is the game that all of these observational study research press releases play and it’s disgraceful scare-mongering.

    Shall we look at the absolute risk?

    Cancer Research UK has terrific statistics on all types of cancer. I’ve just looked at the UK. They do have data for other countries if you want to do your own rummage. The incident rate for all people in the UK, age-standardised (you pretty much won’t see bowel cancer before the age of 50 – look at the age data), in 2011 was 47 per 100,000 people.

    47 per 100,000 people.

    You would need to know 2,128 people, including enough older people, to know 1 person who developed bowel cancer in the UK in 2011.

    Now – let’s do that relative vs. absolute risk thing.

    Assuming that everything the WHO did had been perfect and that there really was an 18% relative difference between those having 50g of processed meat a day and those not (and assuming that nothing else was impacting this), the absolute risk would be 51 people per 100,000 vs. 43 people per 100,000.

    Now where’s the bacon and egg before my CrossFit session?!

    The likely harm of this report:

    The Lancet article does at least have the decency to mention the nutritional value of red meat: “Red meat contains high biological value proteins and important micronutrients such as B vitamins, iron (both free iron and haem iron), and zinc.” That’s still a bit of an understatement. Try both essential fats; complete protein; and the vitamins and minerals needed for life and health.

    What will be the consequences of this report scaring people away from real meat? It takes approximately 250g of sirloin steak to get the daily 10mg of zinc; over a kilo of the same steak to get the recommended daily iron requirement – and in the right form for the body. How about over 20 eggs to get the same iron intake? Still in a useful form to the body. Or 4.5 kilos of brown rice to get iron in the wrong form for the body?

    What do I take from this report? There is a heck of a lot of bad science coming out the World Health Organisation, an organisation that should know better, but then there have previous cases of not knowing better.

    Nothing has changed from my fundamental belief that human beings should eat real food (especially grass-fed, naturally reared meat and naturally preserved meat). Avoid processed food, including meat processed by fake food companies. And take every observational study that doesn’t know these five points above with a hefty pinch of salt.

    • Some French Bloke says:

      the processed meat eaters showed that they […] had a higher BMI, were THREE TIMES more likely to smoke […] This makes processed meat a MARKER of an unhealthy person, not a MAKER of an unhealthy person.

      Quote from one earlier Zoe Harcombe article called Animal protein as bad as smoking?: “Smoking presents an absolute risk – provide the same numbers for my grass-grazing roast dinner if that’s what you’re claiming.”

      Pity that Zoe doesn’t seem to have yet gotten wise to the fact that the big “marker to maker” switcheroo was applied to smoking back in the day (and is still going strong, unfortunately).

      • slugbop007 says:

        Zoe is just another Public Health zealot pretending to be a nutritional expert. Still banging on about obesity even after the latest studies show that it might not be as dangerous as previously claimed. She will probably sell lots of books.

  24. Pingback: Wednesday Morning Music Works | Library of Libraries

  25. Clicky says:

  26. Pingback: Scary Shit Settled Science | MalagaBay

No need to log in

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.