Are two thirds of cancers really due to bad luck?

A paper published in Science has been widely reported in the media today. According to media reports, such as this one, the paper showed that two thirds of cancers are simply due to bad luck, and only one third are due to environmental, lifestyle, or genetic risk factors.

The paper shows no such thing, of course.

It’s actually quite an interesting paper, and I’d encourage you to read it in full (though sadly it’s paywalled, so you may or may not be able to). But it did not show that two thirds of cancers are due to bad luck.

What the authors did was they looked at the published literature on 31 different types of cancer (eg lung cancer, thyroid cancer, colorectal cancer, etc) and estimated 2 quantities for each type of cancer. They estimated the lifetime risk of getting the cancer, and how often stem cells divide in those tissues.

They found a very strong correlation between those two quantities: tissues in which stem cells divided frequently (eg the colon) were more likely to develop cancer than tissues in which stem cell division was less frequent (eg the brain).

The correlation was so strong, in fact, that it explained two thirds of the variation among different tissue types in their cancer incidence. The authors argue that because mutations that can lead to cancer can occur during stem cell division purely by chance, that means that two thirds of the variation in cancer risk is due to bad luck.

So, that explains where the “two thirds” figure comes from.

The problem is that it applies only to explaining the variation in cancer risk from one tissue to another. It tells us nothing about how much of the risk within a given tissue is due to modifiable factors. You could potentially see exactly the same results whether each specific type of cancer struck completely at random or whether each specific type were hugely influenced by environmental risk factors.

Let’s take lung cancer as an example. Smoking is a massively important risk factor. Here’s a study that estimated that over half of all lung cancer deaths in Japanese males were due to smoking. Or to take cervical cancer as another example, about 70% of cervical cancers are due to just 2 strains of the HPV virus.

Those are important statistics when considering what proportion of cancers are just bad luck and what proportion are due to modifiable risk factors, but they did not figure anywhere in the latest analysis.

So in fact, interesting though this paper is, it tells us absolutely nothing about what proportion of cancer cases are due to modifiable risk factors.

We often see medical research badly reported in the newspapers. Often it doesn’t matter very much. But here, I think real harm could be done. The message that comes across from the media is that cancer is just a matter of luck, so changing your lifestyle won’t make much difference anyway.

We know that lifestyle is hugely important not only for cancer, but for many other diseases as well. For the media to claim give the impression that lifestyle isn’t important, based on a misunderstanding of what the research shows, is highly irresponsible.

Edit 5 Jan 2015:

Small correction made to the last paragraph following discussion in the comments below. Old text in strikethrough, new text in bold.

76 thoughts on “Are two thirds of cancers really due to bad luck?”

  1. You say “We know that lifestyle is hugely important not only for cancer”.

    That’s exactly what we don’t know (apart from cigarettes). I fear you are assuming the answer that you’d like to be true.

    1. When you say “apart from cigarettes”, that’s one of the most important lifestyle factors. I assume you wouldn’t dispute that cigarettes are a hugely important risk factor for cancers and other diseases?

      I realise that the evidence on lifestyle factors comes mostly from epidemiological studies, and we both know that epidemiological studies can often lead us to erroneous conclusions. However, remember that some evidence comes from randomised trials, such as this one which showed a reduction in hard cardiovascular endpoints with a Mediterranean diet.

      Also, the evidence on overweight and cardiovascular risk is pretty strong and consistent. You could certainly have legitimate arguments about whether overweight per se is the cause, or whether it’s more related to lack of physical activity, but it doesn’t seem plausible to me that lifestyle isn’t implicated somehow.

      And as for the answer I’d like to be true, I can assure you that I would much rather the answer were that I could eat and drink as much as I like with impunity!

      1. Mummies’ clogged arteries take smoking, fatty foods, lethargy out of the mix

        By Tom Valeo, Times Correspondent

        Tuesday, April 23, 2013 4:30am

        You do everything right: You exercise every day, include lots of fruits and vegetables in your diet, never smoke, minimize the stress in your life and take medication to keep your cholesterol and blood pressure under control. You’re preventing modern life from ruining your heart, right? • Well, maybe modern life isn’t as much of a problem as merely living. CT scans of 137 ancient mummies from three continents show that our ancestors had plaque in their arteries, too, even though they never smoked, never tasted ice cream or pork rinds, and had no choice but to exercise vigorously every day of their lives.

        According to the study, which appeared recently in the Lancet, at least one-third of the mummies, who lived as long as 5,000 years ago, had arteries that had narrowed as a result of atherosclerosis — the buildup of fatty deposits in the arterial wall. Apparently the cardiovascular system has a tendency to clog up over time.

        “Our research shows that we are all at risk for atherosclerosis, the disease that causes heart attacks and strokes,” said Gregory Thomas, medical director of the MemorialCare Heart & Vascular Institute, Long Beach Memorial Medical Center, and one of the authors of the study. “The data we gathered about individuals from the prehistoric cultures of ancient Peru and the Native Americans living along the Colorado River and the Unangan of the Aleutian Islands is forcing us to look for other factors that may cause heart disease.”

        The diet of the mummies varied widely, but contained ample protein and vegetables (and presumably no cupcakes or pork rinds). Aside from the few Egyptian mummies who lived their lives as pampered royalty, these ancient people used their muscles constantly.

        Yet, the atherosclerosis was found in mummies who died in what we today would consider middle age (almost none made it to 60). And just as today, their arteries became more narrow as they got older. CT scans of modern people have demonstrated that after the age of 60 for men and 70 for women, some degree of atherosclerosis is all but universal. One large study found that teens ages 15 to 19 showed early signs of atherosclerosis, and 50 percent already had conspicuous accumulations of plaque.

        “All of us age in every tissue of our body,” says Dr. Donald LaVan, a professor of medicine at the University of Pennsylvania and a spokesman for the American Heart Association. “It’s just a question of how rapidly it happens. There’s nothing you can do to stop aging. All you’re trying to do is prevent it from advancing faster than it should.”

        The authors of the paper agree. “Although commonly assumed to be a modern disease, the presence of atherosclerosis in premodern humans raises the possibility of a more basic predisposition to the disease,” they concluded.

        So what can we do to thwart that predisposition?

        Above all, don’t smoke, says LaVan, and engage in regular physical activity.

        “After that, we’re in the realm of treating disease,” he says. “If your lipids are up or you have hypertension, take care of it. If you have problems with rhythm disturbances, that must be treated, too, because it impairs the ability of heart to pump efficiently. We’re looking at common sense here, but getting patients to do these things is tough.”

        1. The pharaohs and other rich Egyptians who got mummified after death worked hard all their lives and never over-indulged in rich foods? Really?

      2. 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!

    2. That’s exactly what we don’t know (apart from cigarettes). I fear you are assuming the answer that you’d like to be true.

      As a cancer clinician (breast cancer surgeon and researcher), I have to concur. Adam is doing what lawyers would refer to as assuming facts not in evidence.

      In actuality, whenever it’s studied, with uncommon exceptions (tobacco, for instance, stands out as the biggest, baddest gorilla of an exception in the room that is related to cancers of the, head and neck cancer, esophagus, bladder, and pancreas) the effect of modifiable lifestyle factors on cancer risk has been found to be disappointingly modest to nonexistent.

      Diet, in particular has been disappointing as a modifiable risk factor, except for cancers of the bowel, and dietary manipulations to prevent cancer have had disappointingly and mixed results.

      1. So we all agree smoking is an important risk factor for many cancers, and that diet is a risk factor for bowel cancer.

        I assume we’d also all agree that excessive alcohol consumption can increase the risk of some cancers?

        And then there is also evidence that physical exercise can reduce cancer risk. Or do you dispute that?

        I think I stand by my statement that lifestyle is important. Bear in mind I’m not just talking about cancer: most of these things also have a bearing on cardiovascular risk, and excessive alcohol consumption has all sorts of other risks too.

        But I think we’re also straying slightly from the original point of my blogpost: while the relative contributions of bad luck and lifestyle may be hard to pin down, the paper that’s been in the news today doesn’t remotely help us to do that.

    3. Dr. David, are there compelling data in the majority of the lifestyle studies that reject the importance of lifestyle (lets say)? If so, I would be truly glad to know about them (no challenges here!).
      Very respectfully, I cannot see bias confirmation in stating that we know the huge importance of lifestyle in many hard health outcomes.
      Cheers, Daniel

  2. Oh dear! You reference to the Mediterranean diet paper is unfortunate, Have you actually read it? It’s one of the papers that I chose as an example of totally unjustified hype in glamour journals

    “The diets had no detectable effect on myocardial infarction, death from cardiovascular causes, or death from any cause. The only difference was in the number of people who had strokes, and that showed a very unimpressive P = 0.04.”

    1. OK, that paper isn’t the strongest paper in the whole world ever, but I would make 2 important observations.

      First, it’s consistent with a wide body of epidemiological evidence. When we see the RCT evidence and epidemiological evidence converge like that, it’s stronger than either of them on their own.

      Second, the study was stopped early because it had already shown significant benefit on the primary endpoint, which was a composite of the various CV events. That means it was inevitable that the power of the study to detect significant differences for the secondary outcomes you mention was limited. I note that all endpoints still showed hazard ratios of < 1. The DSMB for the study obviously felt that the evidence was strong enough that it would have been unethical to continue the study further. That's not a decision I would wish to criticise.

  3. It’s a decision I’d certainly wish to criticise. P = 0.04 (roughly meaning “worth another look”) on one outcome. Nothing on others. It has to be one of the worst papers I’ve had the misfortune to read.

    1. On both the stroke outcome and the primary outcome, the P value for the Mediterranean diets combined vs the control diet was P = 0.005 (see table 3 of the paper).

      I acknowledge that stopping the trial under those circumstances is a tricky decision. Personally, I think the DSMB made the right call. If you don’t, well, we’ll have to agree to differ.

  4. Interesting blog Adam – I’m not commenting on the stats or numbers but here are a few thoughts:

    Carcinogenesis is a multistep process. There is risk and luck/bad luck. Risk can be increased by exposure to carcinogens or hereditary factors.
    We don’t know everything – we find out what we can and minimise the risk – but for people with cancer I think an awareness of risk/luck is also not a bad thing – guilt/blame is negative.

    The role of DNA replication, cell division and carcinogenesis is interesting – DNA damage can be ‘fixed’ as mutations and replicated before DNA repair has had a chance in tissues or circumstances where replication is higher.

  5. Thanks for going into more depth on this study than the cheap, simplistic and incorrect headlines. The main stream media has, again, done a terrible job in reporting this finding (which is interesting but hardly rocket science given what is already well known about cell division, mutations and error correction in cancer). The general public (and many scientists – although some may understand it too well and abuse it) have a horrible time with relative risk and contributions of multiple variables where 1 + 1 doesn’t necessarily equal 2. It’s January 2nd so either 2015 is starting off as badly as 2014 ended in quality of reporting, or we can blame the poor folks who pulled the short straw over the New Year shift.

    We can’t do anything about random events, but we can do something in reducing risk (as long as the Dr. Oz/tabloid style sensational diets, quack science and miracle cures are ignored). Moreover, efficiency of DNA damage detection/repair and detection of abnormal cells by the immune system as well as intrinsic cell death self-defence mechanisms may well be modifiable and can certainly be influenced by genetics.

  6. Mr. Statistician, please address this issue: what percent of people die from cancer vis-a-vis the percentage of people who USED to die from cancer? People have always died from cancer, but were more likely to die from death by saber-toothed tiger, starvation, plague, combat, pneumonia, flu. So now if a person doesn’t get hit by a bus, there is a larger chance of dying from cancer. Or so it seems.

    1. Yes Alice, I think that’s spot on. One of the biggest risk factors for many cancers is old age, so if you’ve dodged all the sabre-toothed tigers and the plague and live to a ripe old age, you’re far more likely to die from cancer than if you die in your 30s.

    2. Smoking Down – Cancer Up
      Average % Cigarette Smoking
      1950s = 44%
      2000s = 24%…

      U. S. National Cancer Institute Manipulates Cancer Statistics…

      …therefore we better see how cancer increase on statistics of Norway Cancer Research

      Cancer rates per 100 000 person
      1956: male = 171,1 female = 163,7
      2006: male = 362,2 female = 294,3

      Lung, trachea
      (rates per 100 000 person)

      male (cancer increased in 3 times)
      1956-60 = 11,7
      2006-10 = 35,8

      female (cancer increased in 10 times!)
      1956-60 = 2,7
      2006-10 = 25,1
      ‘Nearly 80% of people diagnosed with lung cancer in 2012, are non-smokers.’
      1. The mechanisms of cancer are not known. It is NOT possible to conclusively attribute a cause to effects whose mechanisms are not fully understood.
      2. Lung cancer has 40 suspected concomitants, that is, 40 different suspected causes randomly interacting in different amounts for each individual. Though it is true that there is a prevalence of smokers among lung cancer patients, that may be due to the prevalence of other concomitants in the “smoking personality.” If that is the case, quitting smoking would not affect the onset of lung cancer. And those who have quit smoking and reached an old age cannot claim that this is due to abstinence from tobacco.
      3.The ever-present, endlessly repeated “facts” from the anti-tobacco groups and pharmaceutical industry are wholly mistaken when they say, “if it was not for smoking, lung cancer would be a very rare disease.” Lung cancer IS a very rare disease. According to the WHO’s own statistics, it affects less than 1% of the population. That means that lung cancer is a very rare disease even among smokers.
      Please check the related links:

      ‘For more than forty years, hundreds of thousands of medical doctors have been deliberately lying to you.’

      ‘Nicotine – The Zombie Antidote’… See More

      — with Christer Carl Olof Nylander.

  7. Hi Adam,

    I have two comments on this study and I’m wondering what your thoughts are:

    1) Looking at Fig 1, it appears that the relationship between risk (R) and divisions (D) is not linear but close to R ~ square root (D). This seems to conflict with the conclusion that each division represents an additive risk.

    2) Divisions is equal to lifetime cell count. Therefore risk could be related to any random characteristic of each new cell, not necessarily the division process. For instance, each new cell could have a random mix of protein concentrations and some of those mixtures result in cancer. The observed results would be the same.



    1. Interesting questions, Matt. Since the authors kindly provide their raw data in the supplementary material, I had a look at whether a square root relationship might fit better. It does, but the difference is really really small (R = 0.8037 for the linear relationship, R = 0.8041 for the square root relationship). I don’t think from the size of the dataset we have (which, let’s remember, is based on only 31 data points) we can meaningfully choose between those possibilities. I think assuming a linear relationship for the sake of simplicity when there is no strong evidence to the contrary is entirely reasonable.

      From a statistical point of view, I think you’re correct about the data being equally consistent with mutations occurring in the cell division process and some other factor occurring independently of that process. However, you’d have to ask someone more expert in cell biology than I am about whether those possibilities are equally biologically plausible.

  8. Why do you say the research was badly reported?

    This is from the press release:

    “If two-thirds of cancer incidence across tissues is explained by random DNA mutations that occur when stem cells divide, then changing our lifestyle and habits will be a huge help in preventing certain cancers, but this may not be as effective for a variety of others,” says biomathematician Cristian Tomasetti, Ph.D., an assistant professor of oncology at the Johns Hopkins University School of Medicine and Bloomberg School of Public Health. “

    “many forms of cancer are due largely to the bad luck of acquiring a mutation in a cancer driver gene regardless of lifestyle and heredity factors. ”

    “They found that 22 cancer types could be largely explained by the “bad luck” factor of random DNA mutations during cell division. The other nine cancer types had incidences higher than predicted by “bad luck” .”

    1. Hi Bobs

      I must confess to being a little flummoxed by your question. This whole blogpost was all about explaining why I thought the paper had been badly reported.

      But perhaps I didn’t explain myself very well? Was there a specific part of the post you thought was unclear?

      1. Sorry I was unclear. I agree with your entire post, except for your criticism of the media. Headlines said things like: “2/3rds of cancers explained by bad luck.”

        But these are essentially the words of the authors (as my quotes show). Maybe science journalists should be more critical but their headlines didn’t distort the claims of the paper. One claim is about the variation and the other one is about the actual explanation of cancer.

        1. Well, I think there are 2 issues here. The first issue is that the media did distort the claims of the paper. The media reported that 2/3 of cancer cases were due to bad luck. That’s not what the paper claimed. The paper claimed that 2/3 of cancer types were due to bad luck.

          That’s a subtle, but nonetheless important, difference.

          The second problem is that some of what was claimed in the paper was itself overblown. Now, you may argue that a journalist who is not a qualified statistician would not be able to tell. That’s true. But a good journalist should know that you can’t believe everything you read in the papers and should seek some opinion from relevant experts before writing the story if they don’t have the skill themselves.

    2. The greatest fear is that junk science lifestyle risk studies will be tossed out for the junk science they always have been!

      The rise of a pseudo-scientific links lobby

      Every day there seems to be a new study making a link between food, chemicals or lifestyle and ill-health. None of them has any link with reality.

      Manufacturing the science to meet the agenda, in black on white. Does anyone still have doubts?

      ”Bal laughs when asked about the role of scientific evidence in guiding policy decisions. “There was no science on how to do a community intervention on something of this global dimension,” he says. “Where there is no science, you have to go and be venturesome—you can’t use the paucity of science as an excuse to do nothing. We created the science, we did the interventions and then all the scientists came in behind us and analyzed what we did.”

      Read under the title :
      Tobacco Control: The Long War—When the Evidence Has to Be Created

  9. As one of the reporters who wrote about the study, here’s my perspective.

    Your statement “For the media to claim that lifestyle isn’t important, based on a misunderstanding of what the research shows, is highly irresponsible” is false, based on your own evidence.

    The Independent article you linked to quotes Vogelstein: “This study shows that you can add to your risk of getting cancers by smoking or other poor lifestyle factors. However, many forms of cancer are due largely to the bad luck of acquiring a mutation in a cancer driver gene regardless of lifestyle and heredity factors. The best way to eradicate these cancers will be through early detection, when they are still curable by surgery.”

    You also stated about the random mutation risk hypothesis:

    “The problem is that it applies only to explaining the variation in cancer risk from one tissue to another. It tells us nothing about how much of the risk within a given tissue is due to modifiable factors. You could potentially see exactly the same results whether each specific type of cancer struck completely at random or whether each specific type were hugely influenced by environmental risk factors.”

    But the authors addressed this point in the study:

    “We next attempted to distinguish the effects of this stochastic, replicative component from other causative factors—that is, those due to the external environment and inherited mutations. For this purpose, we defined an “extra risk score” (ERS) as the product of the lifetime risk and the
    total number of stem cell divisions (log10 values). Machine learning methods were employed to classify tumors based only on this score (see the supplementary materials). With the number of clusters set equal to two, the tumors were classified
    in an unsupervised manner into one cluster with high ERS (9 tumor types) and another with low ERS (22 tumor types)

    “The ERS provides a test of the approach described in this work. If the ERS for a tissue type is high—that is, if there is a high cancer risk of that tissue type relative to its number of stem cell divisions—then one would expect that environmental
    or inherited factors would play a relatively more important role in that cancer’s risk (see the supplementary materials for a detailed explanation). It was therefore notable that the
    tumors with relatively high ERS were those with known links to specific environmental or hereditary risk factors (Fig. 2, blue cluster).”

    You may disagree with the ERS methodology, but it would help for your to explain why it is wrong than just dismissing it without acknowledging that the authors tackled that issue.

    In addition, Science’s own news article on the study plainly stated that two-thirds of cancers were caused by chance. I consulted this article in writing my story, and it gave my confidence my interpretation was correct. This is from the article:

    “Here’s how it works: Take the number of cells in an organ, identify what percentage of them are long-lived stem cells, and determine how many times the stem cells divide. With every division, there’s a risk of a cancer-causing mutation in a daughter cell. Thus, Tomasetti and Vogelstein reasoned, the tissues that host the greatest number of stem cell divisions are those most vulnerable to cancer. When Tomasetti crunched the numbers and compared them with actual cancer statistics, he concluded that this theory explained two-thirds of all cancers.”

    In conclusion, please give a refutation that actually addresses the points the media made and the study itself. I would like to learn when I am mistaken, but your criticism doesn’t accomplish this purpose.



    1. Hi Bradley

      Thanks for stopping by. You’re right, the Independent article does include the information that lifestyle still makes a difference, but it’s all a question of emphasis. Here’s the first paragraph of that article:

      Most cases of cancer are the result of sheer bad luck – not unhealthy lifestyles, diet or even inherited genes, new research suggests today.

      What sort of message do you think that sends? Sure, the more nuanced information is further down the article, but I think this is a great example of what Ben Goldacre calls “The caveat in paragraph number 19”. The message most ordinary readers take away from the article is what’s contained in the headline and first paragraph, not what’s in the detail further down the article.

      You ask some very good questions about the ERS methodology. I did toy with the idea of putting that in the original post, but decided against it on the grounds of brevity and because it wasn’t directly relevant to the main argument. But since you ask, here goes.

      One problem is that it starts with the assumption that there is a linear relationship between log cancer risk and log number of cell divisions. While their data seem to be consistent with that, they have certainly not proved that it’s an established fact. However, the conclusions they draw entirely depend on that assumption. If there are other reasons why some tissues may deviate from the linear relationship other than susceptibility to specific environmental causes, then their analysis fails to prove what it says it proves. And it really seems entirely plausible that there would be other reasons. Why should two tissues with completely different biochemistry have identical cancer risks for the same number of cell divisions?

      Another problem is that for some cancers, they’ve separated out those due to known risk factors and those that aren’t: for example, they’ve treated lung cancer in smokers as a separate data point to lung cancer in non smokers. The authors claim that they used “machine learning methods” (a rather odd term to describe a bog standard cluster analysis) to separate the tumour types into R and D types based only on their ERS scores, and therefore that because the D tumours just happened to be the ones that were associated with known risk factors, the analysis supported their hypothesis. However, by treating cancers of the same tissue as different data points depending on whether they were associated with a risk factor, they effectively cheated, and their analysis included the information about risk factors, so it’s not surprising that they got the results they did. It’s really noteworthy that of their 9 D tumours, 6 of them had been previously classified as being due to risk factors. So their classification of cancers into R and D types is at least as much an artifact of the way they analysed the data as the biological characteristics of the tumours themselves.

      You’re quite right that Science supported what was claimed in the study in their news article, but in the immortal words of the recently departed Mandy Rice-Davies, they would, wouldn’t they?

  10. Isn’t one of the underlying problems in the interpretation of the results of Vogelstein and Tomasetti’s work the very definition of cancer? For a medical doctor, cancer means metastatic disease (or at least disease that has a high probability of becoming metastatic). This is a clinical, reductionist definition. But to address the question of the origins of cancer, one needs to broaden this definition to biology. And this is a completely different story, much more fuzzy, much more complex and usually a very long one. Just think about in situ tumors, all the undiagnosed cancers, spontaneous remissions, etc.

    Vogelstein and Tomasetti only looked at clinically diagnosed neoplasms, the visible tip of the “plasms” iceberg. But the causes of cancer are buried in the huge invisible part. And there is one thing we know for sure now: the predominant protective role played by the immune system ( the spectacular results obtained by non-targeted immunotherapies illustrate this well). We also know that the state of the immune system depends strongly on environmental and behavioral factors.

    If so, isn’t the status of the immune system THE MAJOR link between environment and behavior and the risk of cancer?

    Vogelstein and Tomasetti’s work did not look at any of this. This is OK, I guess because the scope of the paper is quite limited. But the message relayed by the media is wrong and dangerous since it devalues prevention and health promotion.

    It’s like saying that the more times an airplane takes off and lands, the higher the risk of crashing. And to remain alive we need to fly, Of course it’s true. But it’s a very partial view, because what prevents airplane crashes are ALSO all the other smaller accidents that build up to air disasters.Vogelstein and Tomasetti’s work only looked at crash statistics.

    1. Lifestyle factors of people with exceptional longevity.

      Einstein College recently studied folks who lived past age 95. The reluctantly reported result: “People who live to 95 or older are no more virtuous than the rest of us in terms of their diet, exercise routine or smoking and drinking habits.”

      Einstein College press release:… … 78&pt=news

      Did you notice in link above that they just state that the very old smoked about as much as did people who died younger, with no detail given, although detail is given regarding eating, boozing, exercise, and so on? Well, when it came to publishing the abstract with the National Institutes of Health, they ignore smoking results entirely! They do say that smoking was studied, but make no mention whatsoever that smoking was not shown to impair longevity: again, as with the press release, precise detail is given regarding other studied factors, but when it came to smoking — the holy taboo of all holy taboos — they simply couldn’t bear even to mention their own finding!

      Here it is: the official NIH abstract:

  11. A bit more cancer biology:
    Cancer is a multistep process.
    Cancer incidence versus age suggests 6 or 7 ‘hits’ i.e. 6 or 7 different events.
    A series of hits are required which may occur over a number of years.
    Some hereditary factors may result in earlier onset because a first step along the pathway has already occurred.

    Replicating cells are particularly sensitive to exposure to DNA damaging agents (including carcinogens).

    Mutations in genes involved in processes such as cell survival, cell proliferation and cell motility can lead to the development of cancer.

    There are many relevant cancer biology factors include genetic and epigenetics. As mentioned by JW above ‘efficiency of DNA damage detection/repair and detection of abnormal cells by the immune system as well as intrinsic cell death self-defence mechanisms may well be modifiable and can certainly be influenced by genetics.’

    The more we know about the mechanisms the better chance of new possibilities for prevention/reducing risk and for different treatments (and personalised medicine). This is going off topic a bit – so I’ll stop here.

    Back over to you – re the stats, this particular paper and its coverage in the media.

  12. Lifestyle is a known cause for SOME cancers. The other great embarrassment of the study and the commentary around it is that social responsibility and environmental degradation are left out of the discussion. Two fast growing cancers, breast and thyroid, are not behaviourally linked. Diagnostic advances explains some of the increased prevalence, but so too do environmental factors–the only KNOWN cause of thyroid cancer is radiation. We also know of many cases where cancer incidence increases in a population years, or decades, after a polluter moves into an area.

    It is super dangerous and irresponsible to claim cancer is caused mainly by luck. But it is ALSO irresponsible to single out “behaviour” for cancer in general when it works for some and not for others.

    1. You misunderstand me if you think I am trying to single out behaviour for cancer in general. Obviously cancer is due to a mixture of lifestyle choices, environmental exposures that we have little or no control over, genetics, and bad luck.

      All I am saying on the question of lifestyle is that it is wrong of the media to give the impression that lifestyle is unimportant. Maybe smoking doesn’t increase your risk of some cancers, but it sure as hell increases your risk of others. So IMHO it is quite wrong to give out the message that cancer is just a matter of luck so it doesn’t matter what lifestyle choices you make.

  13. – also it could be dangerous to extend these ideas to downplay the possible role of future lifestyle ‘non-choices’ and the continuing need for rigorous testing of new ‘environmental factors’, new foods, additives, cosmetics, medicines, pesticides, chemical pollutants etc – (and possible existing ‘hazards’ not yet identified)

    1. Yes, absolutely, Julie. Of course, there’s a lot of nonsense talked about the cancer risk of many environmental chemicals, but despite that it is an area where we need to remain vigilant and not start to think that testing chemicals before widespread use is unimportant.

  14. Hi Adam,

    Thank you for that very detailed reply. I am reading it over and over, and trying to think how the story could have been reported differently. Right now, I can’t think of anything that would substantially change it.

    I did everything by the book – talking with one of the researchers, having the story fact-checked by Johns Hopkins and two scientists not involved in the study before publication, and of course reading the study and the accompanying news article. The other media stories reported the facts similarly. So either a whole bunch of journalists read the article incorrectly in the exact same way, or something else is at issue.

    Let’s turn to the Independent story you discussed. In your reply, you pointed out that the first paragraph of the Independent story said:

    “Most cases of cancer are the result of sheer bad luck – not unhealthy lifestyles, diet or even inherited genes, new research suggests today.”

    And then you asked:

    “What sort of message do you think that sends?”

    An accurate message. It correctly summarizes the study. I don’t see how you get from there to your conclusion:

    “For the media to claim that lifestyle isn’t important, based on a misunderstanding of what the research shows, is highly irresponsible.”

    Yes, lifestyle factors are important, but no media account I’ve seen has denied it — including the Independent article you took to task. Lung cancer is the canonical example of a mainly lifestyle-caused cancer, presented in the study as well as in the Independent article.

    I hope you reword your description to correctly reflect what the Independent article actually said. It would be a fitting gesture in favor of accuracy for everyone, media and scientists alike.

    It seems you are concerned that people will take the study as a license to ignore environmental or dietary factors in cancer risk. But in my view, a science reporter’s job is to accurately reflect the state of the science, not to keep the public on message. There are plenty of people and groups for that purpose, and the media quote them often.

    And I find it hard to understand why information that pertains to a majority of cancers should be downplayed in favor of information about a minority of cancers. We all agree that a one-third variation in risk is substantial, and that is properly represented in the study and the media reports I’ve seen.

    You will rightly point out that with studies such as this one, informed perspectives from independent scientists are critical. And I managed to get two scientists to give their evaluations. But the process was unusually difficult, due to the impending New Year’s holiday. A lot of scientists were simply not available. I guess other reporters encountered the same obstacle.

    Perhaps scientists interested in media accuracy can help reporters find informed and independent experts before their stories appear. Most scientists I ask for help are not available for one reason or other. They may be on vacation, at a conference, or just busy with their work.

    BTW, the Independent article contains an error you didn’t mention: it refers to Science as a “medical journal.”

    Back to your reply:

    You wrote of the cancer risk/cell division hypothesis:
    “One problem is that it starts with the assumption that there is a linear relationship between log cancer risk and log number of cell divisions. While their data seem to be consistent with that, they have certainly not proved that it’s an established fact.”

    I agree that proof is not established. But the evidence appears to strongly point in that direction. I was particularly impressed with the results obtained in mice, by looking at cell divisions in the colon and small intestine. The pattern held up, even though in mice it is the small intestinal stem cells that divide more often than those of the colon. So it appears the higher rate of colon cancer in humans is not specific to that tissue.

    Of course, more studies are needed to see whether the conclusions really do hold up. And I’d like to see the numbers for breast and prostate cancer, as soon as those are available.

    You wrote of the D and R classification system:

    However, by treating cancers of the same tissue as different data points depending on whether they were associated with a risk factor, they effectively cheated, and their analysis included the information about risk factors, so it’s not surprising that they got the results they did.

    That is a point I’m going to further review, and will ask the paper’s authors about. My impression from the paper that they arrived at the results independently of the known risk factors. Perhaps these cancers in the same tissue could be distinguished by other means than risk factors?

    I hope you will continue to talk with reporters, and that scientists and reporters can resolve to reduce errors and distortions. It’s a good New Year’s resolution for us all.



    1. Hi Bradley

      On the first point about the message, I have acknowledged that media reports did include the information about lifestyle remaining important. My problem is the prominence that information is given. Did you follow the link I gave in my earlier reply to Ben Goldacre’s blog? I think he explains the problem of emphasis better than I do.

      Anyway, no matter how accurate the articles may have been if you read all the way through them, the way the headlines are written is misleading. Here are a few examples of how Joe Public interpreted the story:

      Hey smokers there has never been definitive proof smoking causes cancer. Another study supporting that it doesn’t

      Keep smoking! It’s all about luck, guys!

      Dr, why should I stop smoking? The internetss says cancer is just bad luck… It’s based on scientific research

      THANKYOU BBC!! Smoking doesn’t cause cancer. It’s just bad luck

      research proves Cancer is just plain bad luck, smoking not to be blamed

      Now, obviously all those people have misunderstood what the media reports were saying. But wasn’t that misunderstanding completely predictable from the way the research was framed?

      You write:

      “Perhaps scientists interested in media accuracy can help reporters find informed and independent experts before their stories appear.”

      Yes, absolutely! I certainly agree scientists should do that, and indeed I take part in an initiative designed to do exactly that. It’s really unfortunate that we didn’t pick up this paper and prepare a briefing on it while you guys were still writing the stories: I can only assume that it was something to do with it all happening over the holiday season.

      Very happy to join your new year’s resolution for scientists and reporters to talk more: it’s always good to talk!

        7 October, the COT meeting on 26 October and the COC meeting on 18
        November 2004.

        “5. The Committees commented that tobacco smoke was a highly complex chemical mixture and that the causative agents for smoke induced diseases (such as cardiovascular disease, cancer, effects on reproduction and on offspring) was unknown. The mechanisms by which tobacco induced adverse effects were not established. The best information related to tobacco smoke – induced lung cancer, but even in this instance a detailed mechanism was not available. The Committees therefore agreed that on the basis of current knowledge it would be very difficult to identify a toxicological testing strategy or a biomonitoring approach for use in volunteer studies with smokers where the end-points determined or biomarkers measured were predictive of the overall burden of tobacco-induced adverse disease.”

        In other words … our first hand smoke theory is so lame we can’t even design a bogus lab experiment to prove it. In fact … we don’t even know how tobacco does all of the magical things we claim it does.

        The greatest threat to the second hand theory is the weakness of the first hand theory.

      2. Judge doesnt accept statistical studies as proof of LC causation!

        It was McTear V Imperial Tobacco. Here is the URL for both my summary and the Judge’s ‘opinion’ (aka ‘decision’):

        (2.14) Prof Sir Richard Doll, Mr Gareth Davies (CEO of ITL). Prof James Friend and
        Prof Gerad Hastings gave oral evidence at a meeting of the Health Committee in
        2000. This event was brought up during the present action as putative evidence that
        ITL had admitted that smoking caused various diseases. Although this section is quite
        long and detailed, I think that we can miss it out. Essentially, for various reasons, Doll
        said that ITL admitted it, but Davies said that ITL had only agreed that smoking might
        cause diseases, but ITL did not know. ITL did not contest the public health messages.
        (2.62) ITL then had the chance to tell the Judge about what it did when the suspicion
        arose of a connection between lung cancer and smoking. Researchers had attempted
        to cause lung cancer in animals from tobacco smoke, without success. It was right,
        therefore, for ITL to ‘withhold judgement’ as to whether or not tobacco smoke caused
        lung cancer.

        [9.10] In any event, the pursuer has failed to prove individual causation.
        Epidemiology cannot be used to establish causation in any individual case, and the
        use of statistics applicable to the general population to determine the likelihood of
        causation in an individual is fallacious. Given that there are possible causes of lung
        cancer other than cigarette smoking, and given that lung cancer can occur in a nonsmoker,
        it is not possible to determine in any individual case whether but for an
        individual’s cigarette smoking he probably would not have contracted lung cancer
        (paras.[6.172] to [6.185]).
        [9.11] In any event there was no lack of reasonable care on the part of ITL at any
        point at which Mr McTear consumed their products, and the pursuer’s negligence
        case fails. There is no breach of a duty of care on the part of a manufacturer, if a
        consumer of the manufacturer’s product is harmed by the product, but the consumer
        knew of the product’s potential for causing harm prior to consumption of it. The
        individual is well enough served if he is given such information as a normally
        intelligent person would include in his assessment of how he wishes to conduct his
        life, thus putting him in the position of making an informed choice (paras.[7.167] to

  15. The paper claims that the lifetime risk associated with cancer types is linearly related with the cell division count log-wise. Why log-wise? Why a loglog plot? If there’s a baseline exponential process hiding behind it, which one is that? No mention in the paper.

    Even more worrisome is the ERS. This extra risk score multiplies the log of the risk by the log of the cell division count. Why multiply? If you’re regressing against a line, shouldn’t you take a residual against that?

    Overfitting is a serious issue. You’ve got *only* 31 points in 2d. If you’re free to cook up your ERS, you’ll always be able to prove almost anything. So the choice has to be principled. Again, why should we multiply the log of the cell division count by the log of the risk?

    I’d love it if the resident statistician(s) could answer these questions.

    1. An excellent question, Bobs. The ERS score is, as far as I can figure it out, actually kind of related to residuals, but it’s certainly a weird way of calculating it. I don’t understand why they didn’t just use an ordinary residual, and it’s not explained very well in the paper.

      You may well be right that it was “cooked up” because it gave the answer the authors wanted. I don’t know. If there is another explanation, I’d be fascinated to know what it is.

      1. I am baffled. The ERS is the area of the rectangle from a datapoint to the upper left corner (in the loglog plot). So it gives an equal score to points living on the same hyperbola. Perhaps my ignorance, but I am not aware of any physical process that traces a hyperbola in loglog space.

        Assuming the slope is 1/2 and the resulting square-root power law is right, then I wonder if this is related to some belief people have that clones of adult stem cells grow in neutral drift, hence as square roots of time. This could be quite interesting.

    2. Why not ERS=log(R/L) where R is the risk and L the lscd (lifetime number of stem cell divisions? That’s really just R/L and taking logs for convenience. Doesn’t that seem like the knee-jerk first thought at least? Their supplement didn’t motivate the product for me.

      1. They explain in the paper why they chose logR x logL instead of logR / logL, and that explanation at least makes sense. I’m not sure if the same considerations would apply to log(R/L), but given that the linear relationship is between logR and logL, it doesn’t seem very logical to take a ratio of unlogged R and L.

        However, none of that answers the question about why they do any of that instead of simply calculate a residual. It’s baffling.

        1. Perhaps cause residual depends on the specific fitted line and is not as simple for most readers. Log(R/L) seemed “intuitive” and simple, and could be motivated several ways: slope looks close to 1 – they don’t give the estimate I think – and we might expect it to be about 1. Whether we are regressing R on L or vice versa goes away with log(R/L) – it’s just distance to the line of slope 1. I’m not saying residual doesn’t seem natural for stats folks.

          That’s saying why log(R/L) might be tried, not why log(R)*log(L) makes any sense – where I’m also dumbfounded. My point for R/L was just to make obvious it sorts tumors the same as log(R/L) – unneeded here I expect, sorry (R/L might be “easier to think” for some).

          1. Since I have the data, I’ve calculated the slope. It’s very close to a half: 0.53 to be precise. That implies that there’s some kind of square root relationship. I wonder if there’s any biological reason why there might be a square root relationship?

  16. Suppose that you found that drivers that drove at 50mph through 30mph zones were five times as likely to have an accident as those who observed the speed limit. a) Nevertheless, most speed-limit violators would not have accidents b) for want of a better explanation chance would distinguish those who had accidents from those who did not c) you might well be able to say that many accidents (whether or not speed limits were observed) were due to chance in the sense that an element of bad luck was involved in every one d) of course association is not causation and you could think of possible (but not necessarily) plausible alternative explanations to that which has speed as a cause of accidents e) none of this is an argument against enforcing speed limits f) BIG YAWN

  17. Hi Adam,

    We will have to agree to disagree. My objection is that you indicted the media’s reporting by refuting claims that were not made in the article. And you haven’t corrected your error.

    Instead of providing examples of media behaving badly, you gave a few tweets that misconstrued the study. This is apparently supposed to prove the media is at fault. But by that logic, I can use Twitter to prove the media has irresponsibly reported on the Illuminati.
    iPhone = Eye Phone = Illuminati Phone. Siri spelled backwards is Iris, thats a part of the Eye. Apple is Illuminati. They’re watching you.
    HOLY SMOKES LOOK WHAT I FOUND! #IlluminatiConfirmed Telling us to “Believe in them” @ILLUMlNATI explain
    George Washington’s Warning To America About The Illuminati – Now The End Begins
    I liked a @YouTube video Disney Illuminati Satanism & Sex symbols Exposed

    Finally, I respect Goldacre and have read “Bad Pharma”. But his point doesn’t apply to the Independent article, because you inaccurately characterized it.



    1. Hi Bradley

      It seems you don’t see the connection between media reporting that stress the importance of bad luck and members of the public who take the message away that lifestyle isn’t important.

      Nonetheless, I take it seriously when you talk about correcting errors in my article. I don’t want to have errors in my article, and will happily correct if you can identify any. I’ve given this some thought, and perhaps you’re right that I’m wrong to state the media “claimed” lifestyle isn’t important. So I’ve corrected that.

      Your examples of crazy tweets are amusing, but your analogy fails because you can’t link all those tweets to a mainstream media article. But if you could, then I would probably be just as critical of the article that prompted them.

      1. Hi Adam,

        Thank you for making that correction. But it still needs backing up. You now say:

        “For the media to claim give the impression that lifestyle isn’t important, based on a misunderstanding of what the research shows, is highly irresponsible.”

        Yet you still have failed to produce any actual quote or passage in a media article to back up even that corrected statement. Such impressions aren’t in the Independent article, or you would have quoted examples by now. All you have given in evidence are cherry-picked tweets.

        You say, “It seems you don’t see the connection between media reporting that stress the importance of bad luck and members of the public who take the message away that lifestyle isn’t important. ”

        You haven’t *shown* a connection. The media articles the tweets linked to didn’t say or imply lifestyle isn’t important. That misconstruction exists in the minds of those who failed to understand what the articles said. And there will always be people like that.

        This is what my Twitter quotes were all about. They are as (un)scientific as yours, a collection of quotes selected to back an argument. They aren’t serious evidence. You didn’t quantify what percentage of all tweets on the subject your sample represented. What percentage got it right? You also didn’t confirm that the tweets were serious and not ironic. As a stats guy, I’m sure you’d fault others for omitting those points.

        So long as the media have accurately reflected the study’s findings, your true gripe is not with the press, it’s with the scientists who did the study and those in the public who misconstrue it.



        1. I provided a specific example of what was misleading about the Independent article in my first response to your first comment.

          You obviously don’t regard it as misleading. I do. As you said earlier, we will have to agree to differ.

          1. This is a fascinating (and detailed) discussion. My non-expert impression is that there has been as much criticism here of the original paper as of the media reporting of it. Apparently the authors have used some questionable techniques to go beyond their core claim about accounting for variability in cancer rates, AND the journal has glossed the paper in an eye-catching but misleading fashion.

            So long as the media have accurately reflected the study’s findings, your true gripe is not with the press, it’s with the scientists who did the study and those in the public who misconstrue it. [Bradley]

            To me this is the key issue: if a respectable scientific journal presents good research in a misleading way, to what extent are journalists obliged to unpick that? Is it enough to report accurately the authors/editors’ own summary of their findings, or do you have to be a Cochrane-level systematic reviewer to make a considered judgement on the methodological qualities of the paper before filing copy? I think the answer is somewhere in between, but I’m not certain where, and I sympathise with Bradley who seems to have done solid diligence before writing.
            regards, tom.

      2. The Media is slowly turning away from the Prohibitional mindset and statistical risk factor as a science in itself. By the way its way past time they did. Science by headline has been Tobacco Controls way of mass propaganda for over 20 years now. Other studies are coming out all the time destroying the fabled risk factor preachers.

        When epidemiology decides to include Toxicologists in their study set to prove its conclusions or findings then ad only then will epidemiology come out of its Junk Science roll.

        Activism is what finances epidemiology and if the outcome doesn’t meet the desired results grant money is quickly taken away and the study shut down. That’s exactly what the ACS does and even wrote a letter to such effect back in the 1990s when the Enstrom and Kabat study initial results didn’t back the ETS exposure propaganda being pumped out ad hominen everywhere to illegally impose smoking bans all over the world. So political is the anti-smoking fury none of it can be believed and should be tossed of as mad scientists out to create a Bogey man that never existed.

        Anyone for 3rd hand smoke junk science too!

        You’ve lost all credibility and it will be decades befor you ever get it back.

  18. My big question is how we know that the datapoints are independent enough to do a correlation on. There are a whole bunch of bone cancers in the correlation, did these figures all come from the same original paper, or use some same basic figure to create various estimates of stem cell replication. This sort of thing would inflate the R.

    1. Oh, that’s an excellent question, Natasha. I hadn’t thought of that. But you’re absolutely right: it’s a problem, particularly for the osteosarcomas. It’s noteworthy that the osteosarcomas are all at the bottom left of the graph, so if those data points are not independent (which they probably aren’t), then treating them as separate data points will indeed inflate R.

  19. Tom, thank you for understanding the journo’s dilemma. In my opinion, the media did a respectable job of reporting the cancer study’s conclusions. And most stories included comments from outside experts. This is as good as it gets in the popular press.

    Sorry, Adam.

    There are, of course, true examples of misreporting, such as the 2012 ENCODE study. Media accounts said it found that 80 percent of human DNA is active and needed. That came from a misleading press release, but the study itself didn’t say that. It said at least 20 percent and up to 80 percent of DNA performed some biochemical function. Whether it was needed or not (conserved by evolutionary pressures, in other words), wasn’t determined by the study.

    Reporters shouldn’t just write their story from press releases; they should read the study and ask questions about it. But many who should have known better were fooled, including the New York Times:

    If you want a behind-the-scenes messed up the messaging was on the ENCODE study, just read the account of an ENCODE author, Ewan Birney on his blog:

    Q. Ok, fair enough. But are you most comfortable with the 10% to 20% figure for the hard-core functional bases? Why emphasize the 80% figure in the abstract and press release?
    A. (Sigh.) Indeed.
    Originally I pushed for using an “80% overall” figure and a “20% conservative floor” figure, since the 20% was extrapolated from the sampling. But putting two percentage-based numbers in the same breath/paragraph is asking a lot of your listener/reader – they need to understand why there is such a big difference between the two numbers, and that takes perhaps more explaining than most people have the patience for. We had to decide on a percentage, because that is easier to visualize, and we choose 80% because (a) it is inclusive of all the ENCODE experiments (and we did not want to leave any of the sub-projects out) and (b) 80% best coveys the difference between a genome made mostly of dead wood and one that is alive with activity. We refer also to “4 million switches”, and that represents the bound motifs and footprints.

    So instead of confusing the media with two numbers, ENCODE sent a press release leading with one number that was widely misinterpreted. Creationists, of course, loved it.

    Here is my take on the issue at the time:

  20. Theres one factor nobody has mentioned as yet!

    The Sterilized world we live in and its effect on immune response.

    There have been a few childrens studies on the effects of childhood asthma and animal dander and other things that cause antigen reacations in children. The point is the sterile world has stunted immune responses and allowed who knows what to create an immune defense system that doesn’t respond well when other things come its way………..Im not talking about environmental factors but a immune system that didn’t have a fair chance to respond and grow in younger kids. You might call it living in a sterile bubble syndrome.

    But nobody seems to want to got there. When smoking rates were higher we had less asthma in children………..

    The inconvenient truth is that the only studies of children of smokers suggest it is PROTECTIVE in contracting atopy in the first place. The New Zealand study says by a staggering factor of 82%.

    “Participants with atopic parents were also less likely to have positive SPTs between ages 13 and 32 years if they smoked themselves (OR=0.18), and this reduction in risk remained significant after adjusting for confounders.

    The authors write: “We found that children who were exposed to parental smoking and those who took up cigarette smoking themselves had a lower incidence of atopy to a range of common inhaled allergens.
    “These associations were found only in those with a parental history of asthma or hay fever.”

    They conclude: Our findings suggest that preventing allergic sensitization is not one of them.”
    The Journal of Allergy and Clinical Immunology
    Volume 121, Issue 1 , Pages 38-42.e3, January 2008

    This is a Swedish study.

    “Children of mothers who smoked at least 15 cigarettes a day tended to have lower odds for suffering from allergic rhino-conjunctivitis, allergic asthma, atopic eczema and food allergy, compared to children of mothers who had never smoked (ORs 0.6-0.7)

    CONCLUSIONS: This study demonstrates an association between current exposure to tobacco smoke and a low risk for atopic disorders in smokers themselves and a similar tendency in their children.”
    Clin Exp Allergy 2001 Jun;31(6):908-14


    In 2008 this paper was produced in America and concludes that nictotine and hence active smoking and passive smoking leads to less asthma. It also gives the aetiology (causation) why nicotine and the biologial process that reduces asthma in recipients.

    The results unequivocally show that, even after multiple allergen sensitizations, nicotine dramatically suppresses inflammatory/allergic parameters in the lung including the following: eosinophilic/lymphocytic emigration; mRNA and/or protein expression of the Th2 cytokines/chemokines IL-4, IL-5, IL-13, IL-25, and eotaxin; leukotriene C4; and total as well as allergen-specific IgE. unequivocally show that, even after multiple allergen sensitizations, nicotine dramatically suppresses inflammatory/allergic parameters in the lung including the following: eosinophilic/lymphocytic emigration; mRNA and/or protein expression of the Th2 cytokines/chemokines IL-4, IL-5, IL-13, IL-25, and eotaxin; leukotriene C4; and total as well as allergen-specific IgE. ”

  21. Dear Adam – I thoroughly enjoyed reading your interpretation and impression of the New York Times article reporting on “Chance in Cancer” and all of the comments in the discussion area. We have a lot to learn from each other. The more we learn, the more we need to learn and add to the equation in order to come to solid conclusions. Everyone has a starting point, a belief system that leans in one direction or another. It’s hard to move our individual viewpoints and rationalizations for our conclusions.

    Control and GuiltSome people will continue to neglect their personal health, excuse lifestyle choices and simply give it up to “chance.”

    A Living ExperimentOthers, will participate dutifully in the world around them, selecting nutritional foods, herbs, “clean living” organics, instilling mental and emotional stress-reducing activities (yoga, breathing, reiki, etc.), become part of group, team and congregational organizations to connect with the family of mankind, behave with compassion by giving of themselves through charity and charitable activities, read up on studies dealing with vitamins, sunlight, and sleep patterns, circadian and biorhythms (surgery outcomes by day of week and time of day) etc. In all these ways, people feel they participate in extending life without disease through their actions, constructing their surroundings and interacting with their environment/people around them.

    Acceptance That Not Everything Is Under Our Control Then, after they have done all they can, forgive self and others for those things not within their control, realize each day is another chance at life; then, the element of “chance” or “accident” comes into play. To give a percentage on “chance” in the development of “cancer” when all complex factors is not input into the equation gives pause for me to wonder how the researchers arrived at their result.

    Our Complex Multi-factorial World Cannot Easily Be Studied As we add more data the conclusions will keep changing. Anyway, I choose to do everything in my power to live a clean, joyful and ethical life to improve my environment. Forgive my mistakes. Keep studying and understanding to the best of my ability what science has to offer. While, with reason believing we know just a fraction about the complexities of the universe.

  22. Sorry for being late,
    I’ve read the article by Tomasetti & Vogelstein only a month ago.
    Many criticize the results the authors are claiming but things are much easier. They pretend to use mathematics and their mathematics is wrong. A large number of mathematical mistakes make this paper mean nothing.
    Moreover the very question of comparative importance of different factors doesn’t make sense until the criterion
    of importance is defined.
    Maybe this could be useful:'s_point_of_view

Leave a Reply

Your email address will not be published. Required fields are marked *