“Nearly One in Four Americans Will Die of Cancer” (NYT)–Unless You Are a Woman, a Non-Smoker, Middle-Class or over 35 . . .

Summary-Not long ago, a New York Times editorial repeated a statistic that you may have heard before: "one in four Americans are projected to die of cancer."  Fortunately, this is a complete fabrication. For most Americans–including the vast majority of Times' readers– the risk is far lower.


A recent New York Times editorial announced that the war on cancer is reaching "a state of crisis."   Citing the Institute of Medicine (IOM), the Times called for boosting funds that support cancer trials while also raising the academic rewards to encourage researchers to run clinical trials. Appealing to readers' pre-dawn terrors, the editorial concludes by declaring that "Nearly one in four Americans are projected to die from cancer. It is vitally important to find the best treatments for them."

It's a strong "kicker "(journalese for an ending). The only catch: it's not true.  One in four Americans are not expected to die of cancer. Current projections say that one in four American men born in 1985 will die of cancer. For a woman, the odds drop to one in five. And for most Americans, the risk is much lower.

Consider that at least 30% of cancer deaths are caused by smoking. If you don't smoke, it is far less likely that you will end your life battling cancer.

Why the Risk is Lower for Most Americans

Dartmouth's Steven Woloshin, Lisa M. Schwartz and H. Gilbert Welch broke out the possibility of dying of cancer by age, sex, and smoking status in an article published in the Journal of the National Cancer Institute in 2008.   It turns out that if you are a 50-year-old man who never smoked, the chances that, over the next 10 years, you will die of one of the cancers that are most likely to kill men ( prostate cancer, lung cancer, or colon cancer) are just 4 out of 1,000.  That is  a 0.4% chance. (The odds are higher, 5 out of 1,000,  that you will die in an accident). If  he is a smoker,  the odds for  a 50-year-old man more than quadruple, to 21 out of 1,000.

For a 50-year-old woman who doesn't smoke, the chances of dying of lung, breast,ovarian or cervical cancer over the next 10 years stand at  7 out of 1,000; if you smoke the odds rise to 20 out of 1,000. The authors of the study define a  "non-smoker" as someone who has smoked less than 100 cigarettes in her life.  A "smoker" is someone who has inhaled more than 100 cigarettes and smokes now  (any amount.) If you quit smoking, your odds of dying of cancer are cut in half ten years later.

Moreover, it's worth noting that the likelihood of dying of cancer are one in four for all 25-year old men (smokers and non-smokers lumped together) born in 1985. Back in 1975, projections suggested that a man born that year had only a 18% chance of dying of cancer, while a woman had a 16% chance. By 1985, the odds for a man had risen to 25%, and for a woman to 20%.  This is because we are living longer. In the past, a person was more  likely to be killed by heart disease or an infection before cancer caught up with him or her.  (Typically cancer tumors are diagnosed at age 67.)

Another way of putting this is to say that the longer you live, the more likely it is that cancer will cause your death. If a man lives to 70, the odds are 28 out of 1,000 that this much dreaded disease will kill him over the next ten years, even if he never smoked.  On the other hand, at that age the odds that  he will die of something over the next decade are nearly 300 out of 1,000.  By contrast, the likelihood that cancer will do him in is pretty slim.

A 70-year-old women who never smoked faces the same odds: 28 out of 1,000- that , sometime over the next ten years, "cancer" will be listed as the cause of her death.. But the chances that she something will kill her over the same span are 207 out of 1,000.  For smokers who make it to 70, the risk of death by cancer is far higher: 132 out of 1,000 for men, 80 out of 1,000 for women.

Finally, low-income Americans face a much greater risk. For example, poor women are about twice as likely to die of breast cancer as wealthier women. This is because they receive less medical care, and less sophisticated care. "Where patient care is equal, treatment and outcome are equal," explains Otis Brawley, MD, associate director of the Winship Cancer Institute at Emory University. Article published in the American Cancer Society's News Center.

Brawley begins with data on African-American women which shows that they are about twice as likely to die of breast cancer as white women. But this is a relatively new phenomenon.  Until about 1980, their survival rates were equal.

And genes "didn't change around 1980," Brawley observes. In other words, there is little reason to suspect that African-American women suddenly became more susceptible to breast cancer than white women.

What changed since 1980?  Treatment for breast cancer has advanced over the past 30 years. "Improved breast surgery techniques became more common, new chemotherapy drugs improved survival times and tamoxifen began reducing recurrence rates,"  says Brawley. ut "women too poor to have insurance or adequate income to pay for those advances didn't benefit from them.  "Because there is a larger portion of low-income earners among African Americans, it can seem that race is the reason for worse cancer rates, when it mainly is poverty," he adds.

I should add that studies do suggest that skin color can add to disparities in how much care patients receive. The vast majority of doctors are white, and research suggests that African American patients and white doctors may view each other with some suspicion. At the very least, cultural differences can  make communication difficult.

But Brawley points out that among U.S. military personnel-who all have access to the same quality of care-survival rates among African American breast cancer patients and white breast cancer patients have remained the same since 1980.

Why is the New York Times Trying to  Scare Its Readers?

Just how many of the NYTs' readers are low-income, African-American men who smoke? Not many.

How many  are 30 to 55 year old middle-class or upper-middle-class women who don't smoke?  Many more. Those women face a chance of dying of cancer in the next 10 years that ranges from less than 1 out of 1000 to 13 out of 1000, depending on their age. The longer they live, the greater the risk. We all will die of something. So if a person beats heart disease, it becomes more likely that he will face cancer. But  a 75-year-old woman should not lie awake worrying about this terrible disease. There is a 1 in 3 chance that she will die before she turns turns 85. The odds that cancer will be the cause of her death are just are 33 out of 1,000.

The bottom line: New York Times readers who don't smoke do not face a one in four chance of dying of cancer at any age. I don't know why the Times would throw that  number out there. But I would guess that the person who wrote the editorial believes the statistic. Perhaps he read it in another newspaper.

Finally, let me suggest that rather than pouring more  money into chasing an elusive cure for cancer, we should be putting those dollars into two things that we KNOW will greatly reduce death by cancer: a major campaign to help Americans stop smoking — complete with free drugs and nicotine patches –and an all-out effort  to ensure that all Americans, poor as well as rich, receive the cancer treatments that we know are most effective. (Giving more Americans health insurance is one thing —making sure that everyone receives the same level of care, regardless of income or skin color, is something else.

The Following charts from "The Risk of Death by Age, Sex, and Smoking Status in the United States: Putting Health Risks in Context" Steven Woloshin , Lisa M. Schwartz , H . Gilbert Welch can be found here.

click on chart to enlarge


16 thoughts on ““Nearly One in Four Americans Will Die of Cancer” (NYT)–Unless You Are a Woman, a Non-Smoker, Middle-Class or over 35 . . .

  1. You ask why the New York Times is trying to scare its readers? I think citing the IOM and calling for boosting funds that support cancer trials and raising the academic rewards to encourage researchers to run clinical trials could give you an answer.
    Tens of thousands of scientists pushing a goal of finding the tiniest improvements in treatment rather than genuine breakthroughs, that fosters redundant problems and rewards academic achievement and publication above all else.
    We have produced an entire generation of investigators in clinical oncology who believe that the only valid form of clinical research is to perform well-designed, prospective, randomized trials in which patients are randomized to receive one empiric drug combination versus another empiric drug combination.
    Proof of efficacy of a cancer treatment such as chemotherapy requires a randomized trial in which it has been shown that the group treated with chemotherapy experienced significantly increased survival when compared to that of an untreated group. This has never been done.
    Most claims for the efficacy of a chemotherapeutic agent comes from trials showing shrinkage of tumors or from comparison of survival rates of unmatched groups over time. Unless tumor shrinkage is accompanied by evidence of increased survival, the treatment cannot be claimed to be effective.
    Additionally, in clinical trials, many patients are excluded because they could not complete the rather arduous treatment. So randomized comparisons are of healthier treated patients against all the controls, rendering a lot a trials flawed.
    Until the controlled, randomized clinical trial approach has delivered curative results with a high success rate, the choice of physicians to intergrate promising insights and methods remains an essential component quality cancer care.

  2. Maggie- I am fairly new to your site, and have enjoyed a number of your articles, but I find this one a bit tendentious. A couple of points:
    First, regarding the headline that the 1 in 4 claim is wrong: Ok, so it’s 1 in 5 for women instead of 1 in 4. If we figure that about half of Americans are men and half are women, the overall number is probably somewhere around 1 in 4.5. Is that really so different? Is funding for cancer research not worth it if the average American born in ’85 only has a 22.5% chance of dying of it instead of a 25% chance?
    And what about the cancer patients who are lucky enough to survive their ordeal? Some cancers are highly treatable, even curable, but many of the therapies are toxic, disfiguring, or associated with significant side-effects. Research into new therapies can reduce these burdens. It seems to me, for example, that the research showing lumpectomy with radiation is (usually) just as good as radical mastectomy represents an important advance, even if it doesn’t increase the cure rate.
    Second, regarding your hypothetical 50 year olds: Obviously the likelihood of a 50 year old man dying within 10 years of 3 pre-selected cancers is not directly comparable to the proportion of men or women who will, when they ultimately die, die of cancer. So regardless of what the numbers are, there just isn’t a way for this kind of analysis to refute the claim you are supposedly refuting.
    But I use the word tendentious because the study you cite uses several factors that will artificially lower the mortality rate. They selected lung cancer as 1 of the 3 cancers that they track for both men and women, but then stipulate that these are non-smokers. Lung cancer is not a common cancer in non-smokers. Thanks to Pap smears, cervical cancer is also not a common cancer in America anymore (it’s the 8th most common cause of cancer death in American women according to wikipedia.) Prostate cancer is slow growing and not usually diagnosed before age 50. So the number of men who die of prostate cancer before age 60 is very different from the number who will ultimately die of it. In fact, that’s true of most cancers. As you note in the article, most people who die of cancer are older than 60. So why follow 50 year olds and only count deaths that occur within 10 years? Maybe it’s to intentionally underestimate the cancer death rate.

  3. Jenga & Word
    Jenga — Thanks much..
    Word–Welcome to HealthBeat.
    The big difference regarding risk of cancer is not between men and women. The big difference is between smokers and non-smokers.
    I would have thought that smoking mainly increased your chances of dying of lung cancer. But when I looked at the research, I discovered that wasn’t true.
    Your chances of dying of breast cancer, abdominal cancer, kidney cancer–etc., etc., are higher if you smoke.
    Woloshin, Schwartz and Welch focused on the cancers that cause the most deaths for men and women because other cancers account for a very small percentage of cancer deaths. And, they were trying to make the book & charts readable for a lay audience.
    They are also highly respected reasearchers–I’ve never seen anyone questioning their numbers in peer-reviewed medical literature.
    And they don’t focus on people who are 50 or under–their charts go up to people age 75– and look at what happens to them between 75 and 85. (A very small percentage of the population lives past 85).
    I quoted what happens to 50 -year -olds–as well as 70-year-olds–becuase the vast majority of the Times’ readers (and my readers) are well under 70. So I assumed they would be most interested in what happens to people in their peer group.
    Click on the link that leads you to the charts. (end of post) I suspect you will be surprised (as I was when I first saw them) to see how low my risk, or your risk, is of dying of cancer at any age (I’m assuming you don’t smoke.)

  4. Greg–
    I’m afraid that, to a large degree, you are right.
    Though, since I have worked at the Times, I would also say that Times’ editors are pretty concerned about their health — heart disease, cancer, etc.
    They are not as susceptible to these diseases as some other poorer, more obese Americans, but they do worry.

  5. Thanks for the warm welcome.
    You caught me- I hadn’t really looked at the charts closely and having now done so, I agree. My risk is surprisingly low. What I found even more surprising is that even a 75 year old smoker has a less than 50% chance of dying within 10 years from any cause, not just cancer, or heart disease.
    But, I still don’t think it’s fair to say that the 1 in 4 claim made by NYT is “a complete fabrication”. It seems that it’s an accurate number for men and nearly accurate for women. Just because the average NYT reader has a below average risk doesn’t mean they can’t cite the overall average.

  6. Word–
    point taken.
    But since the risk of dying of some type of cancer are so much greater if you smoke, lumping smokers and non-smokers
    together is very misleading.
    Also, most of the adults in this country who still smoke are poor.
    The poor are sicker than the rest of us and at greater risk–something we tend to ignore. Most people don’t realize that the poor die six years earlier than affluent Americans. A six-years difference in expected life-span within one society is extraordinay (you don’t see this in other developed countries.)
    We should be putting much more money into public health and public schools.
    But fear of cancer has spawned a very lucrative industry: equipment makers like GE, drugmakers, hospitals and some doctors make a fortune. The American Cancer Society takes money from the medical industrial complex that profits (and profiteers) from our fear of cancer.
    This is why they put out numbers like 1 in 4. And no one at the NYT scratches his head and says : “Wait a minute–can that be true? When I think of older friends and relatives who have died, one out of four didn’t die of cancer . . .”
    Meanwhile, the worried well tend to be middle-class and upper-middle class Americans who really don’t need to be so worried. Their risks are lower. They will live longer. If they took some of that energy and worried about the poor . . .
    That’s why I think that publicatons like the NYT (our paper of record) should getting the truth out there by analyzing the numbers rather than just throwing out a scary number that virtually everyone will think applies to them.

  7. Just have to comment on health disparities–according to studies reviewed by the National Association of State Mental Health Program Directors (NASMHPD), people with major mental illnesses die 20-25 years before the average. While director of mental health and addictions for the State of Oregon, I commissioned a study that showed that people who had both a major mental illness and an addictions problems died at the average age of 44. Smoking, poverty, poor access to health care, some interactions with medications, and other factors contribute…is there a larger health disparity in the US?

  8. Approximately 2.4 million people die in the U.S. each year of which slightly less than 550,000 or 23% die from cancer. It’s our second largest killer after heart disease. I don’t know what the average age is among the people who die from cancer, how many of them smoked or what their socio-economic status was. As heart disease becomes better managed, fewer are dying from that cause. The bottom line is that close to 25% of deaths in the U.S. in any given year are from cancer.

  9. Barry–
    The bottom-line is that 30% of those deaths are related to smoking.
    And virtually all of the adults in the U.S. who stil smoke are poor.
    IF we cared, we would launch an anti-smoking campaign, giving out the nicotine patches and new drugs that now make it fairly easily for most people to quite smoking.
    This would cost far, far less than all of the money we spend on tests that don’t save lives or extend lives, terribly expensive drugs that give people only a few extra months of life . . .
    And we would have eliminated nearly 1/3 of the deaths by cancer.
    We also need an anti-smoking campaign in the shcools (complete with nicotine patches)–kids have begun smoking again.
    Finally, today’s cancer reasearch is not well co-ordinated (as the Times makes clear).
    Rather than pouring more money in it, we need to re-think and re-structure the research, with the NIH taking charge and making it collaborative, not competitive. (All researchers should be sharing their reserach).
    An article just came out in the Archives of Internal Medicine points out that if you are diagnosed with cancer, chances are 1 out of 2 that the cancer, or complications related to the cancer would kill youo.
    Yet, the article points out, most newspaper and magazine articles focus on cures, “sugar-coating” the truth– many cancers are incurable and many very aggressive treatments don’t work.
    We haven’t gotten as far in cancer research as many people think.
    Cliff Leaf wrote an outstanding piece for Fortune magazine 3 or 4 years ago about the state of cancer reserach. . . You probably can find it by Googling. He took a very honest look at what was happening.

  10. Maggie,
    I don’t know about that 1/3 reduction in deaths if people didn’t smoke. Maybe people who are more likely to die from cancer are also more likely to smoke and not the other way around (I’m not talking about lung cancer). As you point out poverty has a lot to do with both smoking and dying, but i’m not clear on which is the cause and which the effect.
    As to research, it must remain competitive at least to a certain degree. Competitiveness is one of the major drivers in science discovery. People actually dream about beating the other teams and getting the Nobel Prize. Not too much came out of the USSR collaborative system in spite of having top notch scientists.
    If something is responsible for one quarter of deaths, I say we step up the effort to find a solution, and I don’t mean palliative acceptance of defeat.
    Anti smoking campaigns are a very good thing too. Maybe we should start by withdrawing tobacco farmers subsidies and obliterate this death industry. Maybe we can pay them to switch to other crops like they do for poppies in other parts of the world.

  11. Margalit–
    Yes, people who are more likley to die of cancer are also more likely to smoke because poor people are more likely to die of cancer (even when insured they get subpar care) and poor people are also more likely to smoke.
    The heart of the problem is poverty, the stress that poverty causes, and the way we discriminate against the poor when providing health care.
    What we need is a new war on poverty.
    (The U.S. tobacco farming industry is shrinking on its own –only 2% of tobacco dollars now spent on U.S. tobacco, and most cigarettes smoked here use imported tobacco.)

  12. From a Prev Med specialist,
    Thumbs up! A nicely done post. Each generation seems to need a bogeyman. Back in the 50’s it was the Bomb. Now it’s MRSA or flesh-eating bacteria…both a problem, but uncommon. If we take our limited resources and apply them to the most common and highest risk–for which we have strategies that work–more people stay healthy and live longer.
    And we don’t Scare Ourselves To Death.

  13. Thanks Maggie-
    Applying rational principles to the cancer issue is very welcome. THANK YOU!
    It is only second to applying rational thinking to death and dying
    Dr.Rick Lippin

  14. Doc D, Greg, Dr. Rick–
    Thank you all.
    And Greg, thanks especially for finding to link for Leaf’s piece. It really is eye-opening.