The horizon is not so far as we can see, but as far as we can imagine

Historical cancer indidence rates controlled for age

Anyone know a source for this data?  It would be much appreciated. Every source I find goes on about how increased cancer rates are largely because “we live longer”.  That’s clearly part of it, but how much?  Same thing with heart disease, anyone have that data?


Shorter Federal Reserve: The Economy Breathes Sort of OK if We Keep it On Life Support


Let’s Talk American Public Responsibility for Torture and Iraq


  1. charlie

    Seems like the sort of thing that would be *really* hard to figure out for the past.

    Also (and you probably know this but I’ll say it anyway) if you do find this data, make sure you distinguish between people living longer because their life span has actually increased, vs living longer “on average” simply because infant mortality decreased. A lot of people don’t realize that even in ancient times there were a lot of surprisingly old people, it’s just the average low span was low because of high infant mortality.

  2. Ian Welsh

    Yes, I’m aware. It’s why I think there might be some meaningful data. One set of interesting studies asked old people if they’d had a list of depression symptoms before they hit 30 and determined from that that depression rates were MUCH lower early in the 20th century (as much as 20 times lower.)

    Here’s the thing, I’m almost 100% positive that cancer rates were much lower 100 years ago, and same with heart attack rates — at every age range. But proving it is another matter.

  3. I would ask if cancer has increased not because we are living longer, but because we are not dying of other things. How many people who in the past died of tubercolsis, typhiod, measles, malaria, etc, etc, would have died of cancer had they lived long enough to do so?

  4. John Puma

    And … if you had reams of irrefutable data that our apparent higher cancer/heart attack rates were only an artifact “largely because ‘we live longer’ ” now …what would that allow you to say next?

    Try the CDC.

    John Puma

  5. nihil obstet

    I’d guess there’s also a diagnostic issue. We screen so heavily that we may even be identifying cancers that do not contribute to mortality. But say, before antibiotics, a person might well have cancer, but die of pneumonia before the cancer was ever diagnosed.

    It’s an interesting question. Be sure to update us if/when you find out.

  6. Dan Henry

    I tried looking for similar info a few years ago, could not find solid numbers. I did read a few arguments that the numbers dont exist/will never be solid, for most of the reasons mentioned above. There also the problem of biased diagnosis. Diseases can become quite “fashionable”.

  7. Hi Ian, I’ve not found anything for cancer BUT here is a link to a study on ‘dementia’ which shows that the popular assumption of: ‘it’s just people living longer is actually mythical’.

  8. Brian

    I’m not sure I can add much more than “and too” at this point. One difficulty is that most of the data that you want has only been tabulated since the 70’s. Confounding the data is that our detection methods are hugely more sensitive, the detection limit (for some cancers) is now perhaps on the few cell level.

    From the material i have read looking into this, I tend to think longevity and detection cover most of the increased incidence. Depending on the study, 1/2 to 2/3 of the cancer is from 60+ and I imagine it is not linear increase with age. So, even a small increase in lifespan could make a big difference.

  9. Formerly T-Bear

    It may be your best resource would be old actuarial tables used to determine risk. If anything is to be found there, some account of inexactitude of language be made e.g. consumption could be anything from tuberculosis to lung cancer to black lung which introduce imprecision in what information desired. Census records may also have some summary of death details relating to age at death and cause. Please keep posted on what you find – verrrry interesting *tricycle falls over*

  10. kathy Group CDC offers all kinds of statistical sorts.

  11. Julien

    The CDC’s Morbidity and Mortality Weekly report goes back to 1952. It might be a start:

    It’s extremely difficult to get accurate historical numbers, mostly because ‘cancer’ is actually is pretty diverse group of diseases, most of which would not have been diagnosed as such until very recently.

    I was looking for and could not find an article about prostate cancer showing up in several remains excavated from a medieval battle site in Europe, but instead stumbled upon this: Cancer Found in 2,000-Year-Old Mummy. Now, prostate cancer is a peculiar form of cancer, so it might not be representative of other forms.

    My understanding is that cancer is the essentially the default way to die. It’s cumulative genetic damage and since we’re oxygen breathers, that’s inevitable. The only way to not die of cancer is to die of something else first. We’ve become very good at that lately, hence what appears like more cancers.

  12. Cancer incidence data isn’t even comparable across countries, nevermind across centuries. And it’s unclear that cancer incidence has any relationship to cancer mortality in the presence of advanced screening techniques. For example, the U.S. does regular prostate screening via PSA tests for all men age 50 and up. The U.K. does not, prostate screening there consists of the doctor sticking his fingers up your groin to check the size of your prostate, and prostate cancer there is diagnosed only once the prostate is clearly enlarged. The U.S. has a much higher rate of prostate cancer than the U.K. — but it’s an artifact of detecting more prostate cancers to begin with. The U.S. also has a much higher rate of curing those prostate cancers than the U.K. — but the same percentage of white men of a given age die of prostate cancer in the U.K. as do in the USA. It turns out that the supposed advantage in prostate cancer treatment in the United States is a myth created by the fact that we detect far more cancers today than we did years ago — and that virtually none of those cancers that we’re treating (with treatments that generally cause impotence and/or sterility) would kill the person who has them before that person died of other causes, since the vast majority of prostate cancers are slow-growing and non-malignant and will never cause a problem before being large enough to be detected by digital examination.

    So compare incidence of prostate cancer today in the USA with the incidence of prostate cancer in the UK, and you find it’s not comparable due to the change in screening procedures. You can compare mortality rates across countries, assuming that cause of death is relatively accurately recorded in both, but even that’s not necessarily accurate across eras. A lot of the time a century ago deaths were attributed to “consumption” or “bilous humours” or other silliness of that sort when by modern diagnostic standards it could have very well been lung cancer or liver cancer.

    So is cancer more prevalent today than 100 years ago? 100 years ago was the tail end of the Coal Age, where people inhaled massive amounts of soot from burning coal for home heating and also got to inhale the smoke from coal-fired factory smokestacks. Somehow I doubt that coal smoke was any less carcinogenic than the stuff we’re subjected to today. In particular, the air in cities today is much cleaner than anything 100 years ago was, and smoking or chewing tobacco is far less common. In addition the “patent medicines” of 100 years ago often had nasty stuff in it that we now know is carcinogenic. My guess though is that 100 years ago you would have been more likely to die of pneumonia than of cancer — there were no antibiotics, remember?

    Now, if you were talking about 1,000 years ago… but the notion of accurate data from so long ago is just silly buggers.

  13. David Kowalski

    Disease names in the past were different. People did not have tuberculosis; they had consumption. There were also a lot of deaths in the 1800s from undiagnosed causes listed as “general debility” or something else. And then there were strange diseases. Lincoln’s mother died of “the milk sick.” It turns out that the family cow ate a poisonous weed and passed it on through its milk.

    I would also add the prevalence of childhood diseases as well as infant death. Look at many family histories from the 1800s and you find that at least one child (not an infant) died before reaching adulthood. The truism from the past was that people who reached 18 were quite likely to reach 50 before another dangerous period set in.

    The signers of the Declaration of Independence generally lived long lives. One lived to 95 and others to their 90s. One died at 42. I’m going through a table to compile an average and some other data on that.

  14. Ian Welsh

    I know of at least one study where they dug up corpses and looked for signs of tumors.

    As for the coal age, true enough, except that the vast majority of the population was rural, and the factories were concentrated in the cities.

  15. David Kowalski

    The Statistical Abstract lists the causes of death by age group for 2007. Heart disease is the leading cause of death from about age 50 through 84. it isn’t from 85 on. The New England Journal of medicine lists the top ten causes of death (but not by age) from 1930 to 2010. It says that heart disease peaked as a cause of death in the 1960s. It also adds that previous to 1929, cancers were often listed under the catchall diseases of the stomach,, lungs and … According to this source, causes of death before 1930 would be inaccurate.

    On a personal level, treatment and survival of the first heart attack or incidence of heart disease has improved a lot in the last 40 or 50 years. An uncle had a heart attack at age 54 and died in the early 1970s. My father had a heart attack at age 63 an lived before dying of heart disease at age 86. Lots of people experience and live with serious heart disease for 20 years or more. I hope to be one of them.

    The truism that most people who survived childhood, at least males, lived historically past age 50 is probably true. The 56 signers of the Declaration of Independence lived, on average to age 65.7.

  16. Celsius 233

    Harvard Professor Daniel Lieberman has an interesting read in the Guardian concerning cancer’s growth due to western lifestyles/affluence.

    He states, “… cancer was first named and described by the Greek physician Hippocrates (c460– c370BC).”
    In any event, it’s an interesting read…

  17. Celsius 233

    Not entirely off topic, yet another article by Daniel Lieberman in the same issue of the Guardian; this is about obesity;

  18. guest

    I would think it would be very important, if you do find such data, that it be broken down by the type of cancer.

    Prostate cancer supposedly will show up in most men, supposedly, if they live long enough, even though it might not be very lethal, although I imagine modern chemistry fucking with hormones along with longevity has probably made that a lot more common.

    Skin cancer is going to be very common in old folks of northern european backgrounds (my grandfather had it for his last couple years and no one was terribly concerned, and he had some heart disease by the time he died at 85, but what got him was leukemia, which they had a hard time diagnosing – the diagnosis came the same afternoon he died. Long and short, he was pretty old and after his second wife died he went from healthy to dead within 2 years, and all sorts of things were going downhill, I don’t know how you can pin that on one cause). But shorter lives would have disguised a lot of potential skin cancer back then. And changing racial make-up of the population will skew those numbers, as well as the shift from rust belt to sunbelt.

    At least in my family, it would be almost impossible to untangle the changing lifestyle factors from advances in medicine. Maybe half of my 19th century ancestors lived to 85 or 90, some with very tough lives (potato famine, etc). Others died at 40, and quite a few of the men died before 60. My 20th century grandparents and one parent, though, had crappy lifestyles (smoking, poor -or ratehr too rich – diet, some of them drunks, heart attacks starting in their 50s) and yet they all made it to about 75 without being invalids, with visits to the hospital after heart attacks every few years.

  19. Tim Cook

    Despite the all the difficulties trying to make sense of cancer rates, one thing I think should draw our attention is if there are any spikes, or rather cliffs (I don’t suppose a sudden rise is followed by a sudden fall) in cancer rates. Due to the long gap between exposure to carcinogens and cancer inducement, depending on the type of cancer, a noticeable jump should indicate some introduction of a new carcinogen years earlier. I’ve heard people tie the rise in cancer to the atomic bomb testing of the 1950s and 60s, and my hunch as a non-scientist is that there’s a relationship, but I’d like to see the studies, if they exist.

  20. Formerly T-Bear

    Completely OT.

    The Guardian carried this article concerning development of aid for stroke victims and cognitive problems arising from the trauma:

    Any further word on Sterling’s progress?

  21. LC

    The closest you’re likely to get is using the data that exists at NCI for the last few decades, but good luck finding anything older than that.

  22. glegars (@GaelleBXL)

    This site refers to a number of studies around the world linking cancer incidence to air pollution .. the part on “Cohort studies of long-term exposure” in particular may be close to what you are looking for ….

Powered by WordPress & Theme by Anders Norén