The misuse of health statistics by politicians

Rudolph Giuliani prostate cancer[This post contains links to the New York Times. If you position your mouse over a link, you can view the destination URL at the bottom of your browser.]

When former New York City mayor Rudy Giuliani was seeking the Republican presidential nomination in 2007, he used to give a campaign speech that referred to prostate cancer and health care. His sound bites were turned into a radio commercial and included the following:

I had prostate cancer five, six years ago. My chance of surviving prostate cancer — and, thank God, I was cured of it — in the United States? Eighty-two percent. My chance of surviving prostate cancer in England? Only 44 percent under socialized medicine.

What’s wrong with this picture? Several things.

The numbers themselves – the 82 and 44 percent — were incorrect. Chances of survival are typically stated as the prospect of living another five years. According to the National Cancer Institute, the five-year survival rate for prostate cancer in the US is 98.4%. For England (according to the United Kingdom’s Office of National Statistics), the number is 74.4%.

Where did the lowly 44% for England come from? Giuliani’s health care adviser started with the number of people who have prostate cancer and the number who die (called incidence and mortality rates): how many people have the disease in a given year and how many die from the disease in that year. From those numbers he came up with a five-year survival rate. This is not possible. “Five-year survival rates cannot be calculated from incidence and mortality rates, as any good epidemiologist knows,” according to the Commonwealth Fund.

Comparing apples to oranges

More importantly, comparing five-year survival rates for the US and England is fundamentally misleading. Prostate cancer is overdiagnosed in the US. Many men who receive the diagnosis do not have cancer or will never develop symptoms, let alone die from the disease. The estimate for the US is that 48% of men diagnosed with prostate cancer do not have a progressive form of the disease. In England, on the other hand, testing is performed after symptoms appear, so a diagnosis is much more significant and meaningful.

Another way to understand this is to consider a group of men who die of prostate cancer at age 70. If their symptoms do not develop until they are age 67+, and if they are not tested until they have symptoms, their five-year survival rate will be zero percent. That corresponds, hypothetically, to England and seems pretty dire. But suppose, through aggressive screening, those same men had been diagnosed at age 64. Their five-year survival rate would be 100%. That corresponds to the US and makes Americans seem so much healthier. But in both cases, all of these hypothetical men live to the same age of 70.

To say that survival rates are better in the US than in England means nothing if screening policies are significantly different in the two countries. A higher five-year survival rate does not mean Americans are better off. What counts is the annual mortality rate: how many men per 100,000 die from prostate cancer in a given year. It turns out, according to New Scientist, that England and the US are statistically in a dead heat for this number.

Unfortunately, both politicians and journalists prefer shocking sound bites to more complex facts. As I quoted from the New England Journal of Medicine in the previous post, “Misleading headlines, designed to attract larger audiences, can make life difficult for physicians who want to practice cost-effective medicine [or, I would add, act in their patients' best interest] but are beset by patients’ requests or demands for costly new therapies: the public reflexively mistrusts any apparent withholding of widely touted diagnostic or therapeutic interventions, even when they might do more harm than good.” The “more harm than good” that can result from a false diagnosis and unnecessary treatment of prostate cancer is impotence and incontinence.

Politics as usual

Also misleading about the Giuliani commercial was the “socialized” medicine slur. It’s true that Dennis Kucinich advocated a government-run health care plan that would include all adults and children. The other leading presidential candidates, however, made a point of saying Americans could keep the insurance they currently had. What goes unsaid by opponents of “socialized” medicine is the well-documented fact that survival rates for cancer patients are significantly lower for the uninsured.

The misleading nature of Giuliani’s numbers were immediately pointed out by FactCheck.org and the New York Times. The numbers ultimately came from a publication (City Journal) of the conservative think tank, the Manhattan Institute. When questioned about this, a spokeswoman for Giuliani told the Times: “The citation is an article in a highly respected intellectual journal written by an expert at a highly respected think tank which the mayor read because he is an intellectually engaged human being.” (So there!) When the Times asked if Giuliani planned to continue running the radio ad and to claim, on the campaign trail, that these incorrect numbers were true facts, the spokeswoman replied (defiantly): “Yes. We will.”

Such is the state of truth and politics these days. (A more recent example, on Planned Parenthood funding: “his remark was not intended to be a factual statement.”) No wonder we make so little progress in slowing the growth of health care costs.

Related posts:
Why is it so hard to reduce US health care costs?
Health care inequality: The US vs. Europe
Why the US doesn’t have universal health care
Democrats and Republicans: How they differ

Resources:

Image: Daily News

A Bogus Cancer Statistic, FactCheck.org, October 30, 2007

Julie Bosman, Giuliani’s Prostate Cancer Figure Is Disputed, The New York Times, October 31, 2007

Marianne Freiberger and Rachel Thomas, Spin doctors: Die another day, New Scientist, February 12, 2011 (subscription required)

H. Gilbert Welch, Overdiagnosed: Making People Sick in the Pursuit of Health

Share

2 Responses to The misuse of health statistics by politicians

    • Jan says:

      There’s been so much happening with health care reform (and politics) in the US in the last two+ years that I feel sorely under-informed about what’s happening in other countries. During the health care debate here, we were inundated from the left with stories of how much better health care was in other countries. And from the right, with how bad health care was in countries that had “socialized” medicine.

      I’m aware that huge changes to the NHS have been proposed in Britain since the election, but it feels like it would be a full time job to keep up with the latest developments. Now that I follow some voices from the UK on Twitter, I’m hoping to be a bit more informed.

      On overdiagnosis, @mellojonny wrote that the drivers for this practice are similar worldwide. Although that makes sense, I would have thought that in the UK – which in my mind is less privatized and thus less profit-driven – there would be less impetus for overdiagnosis. But if it’s patient-driven (“I want to be tested for every possible health risk”), then it’s difficult for doctors to resist, irrespective of the economics and politics of the health care system. We all read the same media scare stories about health worldwide.

      And of course the tendency for statistics to produce the evidence one wants to find is universal and nothing new.

      Thanks for the link, Anne Marie. The evidence you present certainly doesn’t increase trust in what we learn from government sources.