A recent dust-up over an ad that tries to make a simple statement out of complex statistics shows how difficult it is to have meaningful discussion and understanding in a sound-bite world when it comes to health care reform.
The ad, launched by the Independent Women’s Forum on August 18th, comes to the conclusion that because the five year survival from breast cancer in the United States is 83.9%, and the five year survival in the United Kingdom is 69.8%, 350,000 women in the United States “may not have survived if the US survival rate were that of Britain.” The implication is clearly stated: “(T)hat should give American women serious pause before reflexively touting the glories of a government-run health plan.”
If only the numbers truly represented what we know.
The basis of the argument was an article that appeared in the British medical journal Lancet Oncology in July 2008. The study was funded by the Centers for Disease Control in the United States and Cancer Research UK, both highly regarded public health and research institutions. Their research was focused on trying to make rational sense out of various statistics about cancer survival throughout the world.
You can imagine that this was nothing short of a herculean task. Different types of cancer registries, different reporting systems, different definitions, different rules for diagnosing cancer and many other factors figured into the mix.
The net result was a mind-numbing—but very laudable—paper that actually tried to show through some standardized process what the five year survivals were for patients initially diagnosed with several common cancers including cancers of the breast, colon, rectum, colorectal cancer, and prostate cancer.
The researchers also looked carefully at cancer rates in the United States, especially at the differences in survival rates among different states and between black and white populations.
So what did they find?
Actually, the numbers quoted above about five year survival are correct: the five year survival for breast cancer in the United States is indeed greater than it is in the United Kingdom.
But wait a minute. Is that also true for other countries that have government-run health plans? Uh, not really.
Canada’s numbers aren’t particularly worse than those seen in the United States, and clearly better than the breast cancer survival in New York City (77.4%). Japan doesn’t fair too badly either. And—horror of horrors—Cuba is slightly BETTER than the United States at 84%!!!! (In fact, surprisingly, Cuba also leads the pack on colorectal cancer survival in women and is second on the list for colorectal cancer in men.)
In fact, when you look at the overall numbers, there are other health care systems in the world that are universal that don’t do so badly when it comes to five year survival in comparison to the United States. And there are some that do much worse.
Now, I don’t want this blog to be misused and misinterpreted, such as suggesting that I think health care in Cuba is better than here in the United States (although there is data that shows they spend a small fraction per person on health care compared to the United States and have a life expectancy that is almost identical to ours). And I am not taking sides on the issue of whether a universal government run program does a better job than fee for service medical care. That is a political debate, and I will let the politicians and the public work on that one.
But I do think that at time statistics can be used whichever way someone wants to use them. You have to be careful how the numbers are presented, and the simple suggestion that taking one number (five year survival) times another number (the total number of breast cancer survivors in the United States) and coming up with another number (350,000 women) that suggests that many women would have lost their lives in another country because of their health system is in essence misleading and erroneous.
There are a lot of reasons that survival numbers for cancers differ among various countries and populations. It is true that we invest considerably more than other countries on health care technology, and this does lead to more access to more services, some of which are related to the early detection and better treatment of cancer. For example, if you detect cancer earlier, you may in fact detect more cancers than will actually cause harm to the patient (prostate cancer is the best example of this) and those cancers have to be treated.
You can also “bias” statistics by diagnosing cancers earlier in their course, which has the net effect of adding in some part to an appearance of longer survival (called lead time bias) for some patients with cancer, when in fact their disease will nonetheless progress leading to death at the same time in the future that would have happened had the diagnosis been delayed to a later date.
The authors had this to say about the issue:
“Cancer survival is a valuable indicator for international comparison of progress in cancer control, despite the fact that part of the variation in cancer survival identified in this study could be attributable to differences in the intensity of diagnostic activity (case finding) in participating populations….If overdiagnosis—which depends on diagnostic intensity—is more marked in one country than another, then it will certainly be harder for researchers to compare incidence, mortality and survival in those countries. But over-diagnosis has different connotations for health-care systems and patients. In each country, the health-care system will have to be funded, staffed, and equipped to cope with the diagnostic and therapeutic burden of all patients with cancer, however they are diagnosed. The heath-care system must make provision accordingly, and monitor the outcome of that provision; cancer survival is one such overall indicator.”
One of my colleagues, Dr. Elizabeth Ward, who is an outstanding epidemiologist, recently made some interesting observations.
You can’t look just at survival, but must also consider the larger picture including death rates from various cancers. According to Dr. Ward, death rates from cancer in the United States in 2004 for men and women were 147/100,000. In Canada, the number was 135, and in Britain it was 133. For breast cancer, the death rates in the United States were indeed much lower in the US (15.6) compared to Canada (17.6) and the United Kingdom at 25.4.
However, if you look at overall life expectancy, the numbers in the United States for men are 76 years, and 81 years for women. Canada does better at 78 and 83 respectively, and so does Britain at 77 years for men and 82 years for women.
The bottom line? Be careful which figure you choose before you start comparing one country to another.
Coincidentally, an article appeared in last week’s issue of the New England Journal of Medicine, titled “British Lessons on Health Care Reform.”
Needless to say, I suspect the Brits are a bit sensitive these days about all the bashing their national health service has taken from those in opposition to a single payer plan or public option as part of our health care reform proposals currently under consideration.
But it was the specific comments about cancer care that I found most interesting, and certainly relevant to this current discussion.
The authors point out that Britain has had to substantially increase its investment in its health care system, which is now 8.4% of their gross domestic product, compared to 16% in the United States.
They go on to talk about the methods Britain employs to control costs, which relies in part on determining the amount of money a treatment costs relative to its benefit for the patient:
“Most of the debate centers on the concept of cost per quality-adjusted life-year-gained and on where the funding cutoff is set. For instance, is an extra month of life for a patient with cancer worth $1000, $10,000 or $100,000? Admittedly, cancer survival is worse in Britain than in the United States, where there are far higher numbers of cancer specialists, better access to novel therapies, and more widespread informal cancer screening, but it is unclear which, if any, of these factors contributes the most to this cancer survival gap.”
So there it is: an admission in writing in a respected medical journal that cancer survival in Britain is less than the United States. But that’s not the whole story, since there are other countries where government controlled programs do a better job with respect to cancer survival, and surpass the United States in life expectancy.
Again, the purpose of this blog is not to tout one system or another. Rather, the intent is to set the numbers on the table, have an honest discussion, and inform rather than scare and inflame.
I just wish my expectations of civil discourse over an important issue weren’t set at such a high level.