Dissent of the Day

A reader writes:

The US being on top in Cancer survival rates is a good thing.  My wife is a US based breast cancer survivor and believe me I sincerely appreciate the good work done here.  However, three things jumped right out at me when I looked at the list:

1) Cancer research and treatment are the "sexiest" things in American medicine.  They get big grants from both charity and government, and oncologists and cancer surgeons are among the highest paid specialists in the country.  However, a medical system shouldn't be judged by the quality of only one of the things it does, just as you shouldn't judge a bridge by the strength of only one of its trusses.

2) We have more than forty million people uninsured here.  I think it is reasonable to assume at least some number of cancers in that population remain undiagnosed at death.  This would have at least a small downward impact on our numbers.  Which brings us to:

3) There are five countries that are very, very close to our numbers. If you take into account number 2 above, they may actually beat us.  Yet, they have roughly half the per capita spending we do on health care, and manage to insure 100% of their population versus our 85%.

Another writes:

The statistics you linked are interesting, but do not tell the entire story.  Cancer survival rates are not always a very good measure of the the quality of health care that a person recieves, and here's why.

Cancer survival rates are based on the time from diagnosis to future point in time - say, 1 year, 5 years or 10 years, etc.  Because of this, they are subject to what researchers call "lead time bias."  Wikipedia has a much better explanation here than I can ever give, but in short it means that advances in cancer screening can artificially inflate the "cancer survival time."

Here's an example, involving prostate cancer.  U.S. male patients usually get screened for prostate cancer starting at around age 50.  Many European countries don't bother screening for prostate cancer at all, since many studies don't show any survival benefit (meaning people's lives aren't extended) to screening.  A hypothetical American male may find out at age 52 that he has prostate cancer - which is often a slow growing cancer.  Say he lives for another 20 years - which is not uncommon - before dying of something else, such as a heart attack.  His "cancer survival time" is now 20 years.  A hypothetical European man isn't screened for prostate cancer, but it is discovered when he is 65 during routine lab work.  He lives another 7 years before dying of a heart attack.  His "cancer survival time" is now only 7 years.  And so on, and so on.

As you can see, cancer survival rates can be inaccurate for measuring the quality of health care.