Cancer Care Study: The Shadow of Pharma

Check out the story in these two places:
http://gooznews.com/?p=3843
http://www.reuters.com/article/2012/04/09/us-cancercare-idUSBRE8380SA20120409

Thanks to Merrill Goozner at GoozNews, we read about a study recently published in Health Affairs by health economist Tomas Philipson at U-Chicago:
http://content.healthaffairs.org/content/31/4/667.full

This study purports to show that the US spends $70,000 per cancer case compared to $44,000 in Europe, and the US rate is climbing much faster than the European--but it's all worth it because US patients live an average of 11.1 years after diagnosis compared to the Europeans' 9.3.

Gooz reports that critics were all over this study as soon as it came out, for what I believe are quite valid reasons, indeed raising the question of how such a study came to be published in an excellent journal. There are two huge sources of error in a study done the way this one was.

One is called lead time bias. Suppose you have two patients with the same cancer at the same stage and there's no effective treatment for this type of cancer. One patient is diagnosed in 2012 and lives till 2016. The other patient is not diagnosed till 2014 and then dies in 2016. The cancer killed both patients at the same rate and at the same time. But because we diagnosed the first patient 2 years sooner, it falsely looks as if whatever we did added 2 years of life.

The second source of bias is becoming rampant in the US as we aggressively screen for more and more cancers, and is very well described in W. Gilbert Welch's wonderful book, Overdiagnosed. This is definitely true today for a lot of prostate cancer. Now that we are typically doing PSA screening on men starting practically from birth, we are diagnosing a lot of cases of prostate cancer. In olden days, if a man was diagnosed with prostate cancer, he probably had something seriously wrong with him because he managed to come to the physician's attention to begin with. Based on that population of patients we decided quite reasonably that "prostate cancer" was a big deal. But now we are finding very healthy men who have a few funny looking cells on a biopsy which they never would have had if their PSA test had not been a bit high, and if we followed those men without any treatment for decades, we'd see that these microscopic "cancers" never cause really serious disease. In other words, "prostate cancer" today means something quite different than what it did some years ago. But of course if you diagnose a lot of men with this sort of not-really-cancer, and see how long they live, you'll get very impressive survival statistics--which only proves that we do way too many screening tests, not that the money we spend on cancer is well spent.

So why is this important to us? As Gooz informs us, "Philipson is a fellow at the conservative American Enterprise Institute and at the Manhattan Institute, served in the administration of President George W. Bush and was a healthcare adviser to Sen. John McCain’s 2008 presidential campaign....The Philipson paper was supported in part by Bristol-Myers Squibb Co, whose cancer drugs include Yervoy. A drug for advanced melanoma, it costs $120,000 for a full course of treatment. Clinical trials showed that Yervoy produces a near-miraculous cure for some patients, with a median increase in survival of 3.6 months."

Golly gosh. I wonder of the sources of funding had anything to do with the conclusions of this study.