Causes of death: 1900 and 2010

An editorial in the 200th anniversary issue of the New England Journal of Medicine looks at mortality and health through the centuries, and includes this chart of causes of death from the turn of the last century, which makes for quite a comparison. We're doing great on kidneys, but hearts not so much.

During the 20th century, heart disease, cancer, and other chronic conditions assumed more dominant roles (see bar graphTop 10 Causes of Death: 1900 vs. 2010.), although outbreaks of infectious disease — from eastern equine encephalitis (1938) and kuru (1957) to legionnaires' disease (1977), AIDS (1981), and multidrug-resistant tuberculosis (1993) — necessitated ongoing vigilance against microbes. New concerns also came to medical attention, from the terrifying consequences of thermonuclear war (1962) to the indolent but devastating effects of environmental pollution (1966) and climate change (1989). Optimism about prospects for the health of future populations persisted but remained tempered by concern about the pathologies of civilization. An obesity epidemic, feared in 1912, has come to pass. Our previously steady increase in life expectancy has stalled and may even be reversed (2005).

The Burden of Disease and the Changing Task of Medicine (via Beth Pratt)


  1. Well, it’s great to see that the DPT shot seems to have eradicated death by Diptheria!

    And that uh… what was I saying… oh yes, that senility has not been killing us as much as back in the day.  (I’m not recognizing Alzheimer’s as a “new, improved” name for senility.)

      1. “senility” was also largely caused by syphilis back in the day, which has since become treatable.

  2. Nice.  This is almost 100% good news.  I was surprised by the large drop in death by accident.  Probably due to those meddling liberals and their seat belts.

    1.  Or the fact that we all sit at desks getting ourselves heart disease, diabetes, blood clots and 3rd femurs.

  3. Saying that we’re not doing so great on heart disease (or cancer, which is a more pronounced difference) is actually based on a WYSIATI confirmation bias from limited information exposure (see “Thinking, Fast and Slow” featured on BoingBoing recently – and totally awesome).

    At the very least this graph would need to look at the average age of death vs. age of onset as well – it’s quite probable that heart disease *would* have been worse 110 years ago if people lived long enough (e.g., failed to succumb to the other myriad attributable causes for departing mortality) to actually contract it and progress to the point of complication.  Note that the sampling and measurement criteria have also almost certainly changed in all cases – such as the fact that people no longer die of “senility”, and many cancers may not have previously been diagnosable.

    At the very least, close examination of the source data and methods would need to be shared in order to trust this information as more than a series of curious coincidences and artifacts.

    Also, totally read that book.

    1. Well said. Yes, if heart disease and cancer treatment extended people’s lives on average by 10 years, they would still die. And the older you are, the more likely you are to get heart disease. Thus, effective treatment of other diseases could increase the number of people who die from heart disease and cancer, even though the treatment of heart disease and cancer has drastically improved.

      1. Yep… and if and when we are able to mostly prevent and cure heart disease then cancer would be the one to off most of us. And when the glorious day comes when we totally beat that cancer’s ass (well… there is a lot of cancer asses to beat, we do have given some of them already a nice beating…) it will be interesting to see what will then be the “big killer”.

    2. According to my dad a MD, absolutely a lot of people use to die before they got old enough to get cancer or heart disease. He said is we all lived long enough we would all get cancer.

    3. Don’t forget the change in and standardization of coroner reports.  This didn’t happen until the 1920s, 1930s or so and changed what people were dying from by giving it a new name.  

      Yes I’m lazy, you want fact checking?  DIY.

  4. Who knew suicide was unknown before the 20th Century?  I guess there’s something to be said for a short and painful life — less likelihood you’ll become jaded by it. 

    1. Or they could have just labeled suicide deaths as accidents in their results back then.  Remember, committing suicide is a sin, so attributing deaths to accidents instead of them choosing to off themselves would probably be a family’s way of rationalizing it so they don’t believe their dear family member gets sent to Hell.

      Also, not religious, just thought I’d point that out considering.

    2. Depends on where it was counted, and how high it ranked. For example, if it were counted under “accidental deaths” in 1900, for whatever reason.

      Or maybe, it wasn’t a large enough cause to make it into the top 10?Look at the 2010 chart, and you’ll see that AIDS, for example, isn’t there – though it’s quite notorious in our time. Doesn’t mean it’s unknown – just that it’s not a leading cause.

    3.  I noticed that too.  I’m suspicious of 1900 data set vis a vis suicide. common sense says it would be higher in 1900 than present day.  Could be labeling (ie, “accident”=suicide), conscious omission from study, coroner consciously miscategorizing…  i mean, there’s no “crime/murder” category in either chart (the murder rate would be higher than the crime rate in 1900 or 2010, no?)

      while we’re being morbid, here’s a suicide fun fact: in Japan, the suicide rate much “higher” in part because police very often categorize as suicide murders that don’t have an evident perpetrator. (they don’t like to have “open” cases over there, it mucks with their high-90s success %.)  so they often won’t risk it: if the case looks dubious, why not just chalk it up to suicide.

  5. Cancer has really risen, both in absolute and relative (used to be half of Heart-disease now it’s about 1:1 …probably we’re better with Heart-disease now…?)  i knew a pathologist who was fond of saying: “live long enough, and you’ll get cancer”.  he’s dead now… (aneurism)

        1. That’s what I’m counting on! I decided dying is just sooo old fashioned, just need medicine to catch up.

    1. I often wonder if cancer isn’t necessarily more prevalent now, but just more often diagnosed. It used to always be there, now it’s just named.

      1. My guess more prevalent now – as noted by the guy’s funny dr above, people living longer.

      2. My guess is both… People live longer => greater chance of random malignant mutations over a lifetime. Medicine is improved => greater chance of finding a cancer in those cases where, in earlier days, you’d just say “Oh, he died of old age”…

  6. Before you jump on the “heart disease and cancer has risen”.. just remember a large portion of the previous deaths may have eventually LED to cancer or heart disease, but the little buggers didn’t last long enough to develop cancer or HD.   TB, GI infections and diptheria being nullified provide enough people to survive long enough to die of something else.

  7. The graphs show that with a lot of infectious diseases cured people get older and die of old-age causes (heart, cancer, senility).

    I think a much better comparison would have been between  cohorts – e.g compare death causes for 10-20 year olds for then and now (and for all other age decades).

    1. I was thinking the same thing – age should be a factor.

      A good question to ask would be, in each age group, what are the leading causes of death between 1900 and 2010. Add another axis – number of deaths in each age group, and I think you’ll find some interesting correlations…

  8. I’d love to see this broken down by age. Are more people dying of heart disease and cancer because they’re not dying due to accidents and infectious diseases, or are these “modern diseases” really getting worse with our modern way of life?

  9. Anyone here remember Ricky Jay’s little car ride monologue in the film “Last Days”?
    Then you might remember ye olden, fantastically peculiar term “death by misadventure”.

    1. That’s still a legal term in Britain. It’s one of the standard inquest verdicts- I believe it’s when the deceased did something that could reasonably be expected to kill them, but without intending to kill themselves. The case of it I can remember hearing of is when a guy tried to relieve himself on the third rail…

  10. Something seems wrong with the diabetes numbers, specifically the fact that it’s not on the list of deaths from 1900 even though it was a common human disease (it was called ‘diabetes’ by the ancient Greeks) and was a death sentence before insulin was discovered and put into widespread use in the mid 1920s.

    1. Just not in the top 10. What with senility and gastrointestinal disease taking such a lion’s share – most people probably didn’t last enough for their diabetes to kill them…

      1. I’m talking type one, what used to be called juvenile diabetes. That would kick in long before senility or gastro.

      1.  Oh and the scandals where Soylent Green deaths are covered up under the heading of being ‘recycled’ rather than dying.

  11. Where’s the (trans-fat) margarine sub-category of Heart Disease?
    Every celebrity (and everyone else) that had heart attacks before their time in the 20th Century, that’s the first thing I think of, they were told to eat “healthier” margarine, died because of it!
    (Not scientific, I know.)

    I only skimmed the margarine article,
    who first said that margarine was healthier than butter, and what was their “proof”?

  12. Suicide was quite common in the old days and there were plenty of deadly poisons and drugs available over the counter at the local pharmacy?  Want to OD on morphine? They’d sell the drugs and the needle. to a child.

    Keep in mind that there were no antidepressants or psychoanalysis, although there were plenty of “mental hygeine” regimens and sanitaria resorts for unhappy folks.

    Murder was also several times more common although often unreported.

    There was also a considerable stigma attached to cancer.

    Not to mention all the people that died from syhphillis, measles, chickenpox, etc.

    1. I don’t think that many people died of chickenpox. It’s not a common cause of death now in countries that don’t vaccinate against it, and doctors don’t prescribe antivirals for it- or anything other than medicines for the symptoms, which they probably had in 1900. The only difference I can see is that you were in more trouble if you scratched and it got infected.

    2. Has the suicide rate decreased? I checked two sources briefly. It appears that the trend for suicide rate has increased slightly for all ages, while the suicide rate of the young has gone up significantly, especially with young men.

      This is consistent with what I understand as the major cause of suicide, which is social isolation. In the past, large families and closer social ties with friends and coworkers made it harder for an individual to become isolated. Today it is trivially easy. This is also consistent to why the developed world has a higher suicide rate than the developing world. More money spreads people apart, less money brings them together.

      1. In the old days if you were young man dissatisfied with life you were expected to hit the road.  by the time someone was 20 they might have been a lumberjack, crewed on freighter, and worked a half dozen other jobs.

  13. I love being able to see stats like this! So interesting..keeping the historical context in mind.

    People have already commented on some of this…

     Of course all the suicides are being lumped into accidents in the 1900 stats. Suicide was COMPLETLY taboo, religiously and socially, in many parts of the world in 1900. It still is of course, but it was then very much the case in Europe and North America. Suicide meant you were denied normal burial…which, to overwhelmingly religious families, meant banishment from the afterlife/the comfort of your dead family FOR ALL ETERNITY. Loving families would lie through their teeth so that members could avoid this fate . All families feared…quite rightly…the taint of suicide, and would have lied to preserve surviving family member’s  prospects for marriage, employment and professional advancement.

    I agree that cardiovascular disease and cancer are more common now. Some of that is surely our long, relatively safe, life. We avoid infection and trauma long enough to become prey to the slow killers. The 5300 year old ice man (Otzi) carried the signs of cardiovascular problems, so that health concern is nothing like new.

    I think it’s important to remember that cancer was  greatly stigmatized in even the very recent past, so likely greatly underreported in 1900. It was badly understood, a “wasting disease”, and a source of social embarrassment. Some forms mimicked STDs, whether they were related or not, and the common notion of everything from mental health to promiscuity to criminality to chronic disease “running in families” meant people routinely hid all variety of problems, including long-term health issues, from public view and even from doctors.

    I know of several cases in my Canadian extended family (both blood and in-law) where people denied or hid cancer until it killed them. I doubt these folks were unique to their time (the 40’s, 50’s and 60’s) and expect things were worse in the decades before. Autopsy was nothing like normal in 1900, so anyone whose cancer did not involve large, obvious tumours could easily have had their death attributed to something else.

    Was childbirth lumped into accidents too? I know childbirth was getting safer toward the end of the 19th century, but expect it still claimed a significant number of lives in 1900. Was it too embarassing to discuss/document?

  14. There’s a great table showing causes of death in a particular English village from the 16th or 17th century in the book of Jonathon Miller’s “Body in Question”. I don’t have it in front of me, but I do remember that “Planet” and “Rising of the Lights” were both more deadly than cancer.

  15. We’re probably doing much better on heart disease and cancer than 110 years ago. However, both cardiovascular problems and cancer get more likely due to aging, and people do get older today and die of fewer other things. In other words, cancer/cardiovascular issues where not the problem they are today 110 years ago, because other things did people in before that.

  16. Our previously steady increase in life expectancy has stalled and may even be reversed 

    In my looking at the issue (pouring over census data and trying to correct for things like wars  in order to engage an acquaintance MD in discussion) — ie., slightly informed but not rigorous — it looks like we actually hit the knee of the curve around 1950, at which time we started spending more and more (approching exponential) for less and less. One can always posit some kind of black swan or deus ex machina that will sweep away even the the concept of death, but the data actually suggest that, lacking such an intervention, we’ve been laboring under diminishing returns for over 60 years.

    In other words, if the Divine Singularity doesn’t show up, we’re all going to die (statistically speaking) about when we could be expected from current conditions, give or take a few years. Sorry Ray.

  17. Looking at the numbers themselves, the relatively minimal gain in heart disease from 137.4 to 192.9, considering the much more substantial gain in cancer during the same time period, is actually quite surprising.

    We’re living long enough for cancers to take hold, that seems clear.  But heart disease hasn’t mushroomed in the same way.

    We’ve just cut away most of the other killers via vaccinations and medication so it stands out.

    1. Sure seems like it should be included.  Even if it’s listed as homicide (also missing).

  18. The real question here is has cancer increased across the board, age-wise, or is it just the fact that we’ve extended our lifespan and the longer you live, the more chance you have of getting cancer.  It’d be really interesting to see cancer rates by age groups.

  19. It is an outrage that “all causes of death still add up to 100%. That is no progress at all!”

  20. Pie Graph: Seriously if you are representing and comparing percentages of a total, you use a pie graph. 

Comments are closed.