January 9, 2012

Cancer clusters

Hugo Chavez, the President of Venezuela, has publicly speculated that some secret US weapon might be responsible for several Latin American heads of government getting cancer recently, which he said was “difficult to explain using the law of probabilities.” This is a perfect example of a phenomenon that is all too familiar to public-health workers: the cancer cluster. According to the American Cancer Society, more than 1000 cancer clusters are reported to state public-health departments in the US each year. How many of them turn out to be real? Approximately none.

There are two statistical phenomena here. Firstly, although individual cancer subtypes may be rare, cancer as whole is more common than most people realize. Secondly, we are very good at seeing patterns, even patterns that aren’t there.

The picture, from David Spiegelhalter, shows four 9×9 squares. Three are coloured entirely at random; one has a pattern. Before going on, decide which one you think is non-random.

The lower-right panel has an ‘obvious’ yellow cluster, but in fact the non-random one is the random-looking panel in the upper right.
A few cancer clusters, especially occupational ones, do turn out to have a common cause. Some important carcinogens have been discovered this way: historically, cancer of the scrotum in chimney sweeps, caused by soot; more recently, hepatic angiosarcoma from vinyl chloride and mesothelioma from asbestos. These are clusters of a single form of cancer, one that is rare in the absence of an environmental exposure.

More commonly, reported cancer clusters are a mixed bag of diseases, often relatively common ones, in a vaguely-defined group of people. On further review, not all the cases are confirmable. The cluster is attributed in some non-specific way to a popular villain such as a local factory.

The Latin American leaders cluster is unusual only in that it is so small. Luiz Inácio Lula da Silva had throat cancer diagnosed while President. Fernando Lugo had non-Hodgkin’s lymphoma (type not specified) diagnosed while President. Dilma Rousseff had a B-cell lymphoma diagnosed before she became President. Hugo Chavez hasn’t said what type of cancer he has. And this week it turns out that Cristina Fernandez, fortunately, doesn’t actually have thyroid cancer.

I must admit that there is in fact a cancer-causing technology developed in North America and further refined in both the US and Britain, which kills many people in Latin America. This technology is quite likely responsible for Lula’s throat cancer. But it’s not at all secret.

avatar

Thomas Lumley (@tslumley) is Professor of Biostatistics at the University of Auckland. His research interests include semiparametric models, survey sampling, statistical computing, foundations of statistics, and whatever methodological problems his medical collaborators come up with. He also blogs at Biased and Inefficient See all posts by Thomas Lumley »

Comments

  • avatar

    The top right one was obviously non-random: it’s the only one which doesn’t have two of the same colours touching on any edge (random chance ~0.0000004%), and is also the only one that has the sudoku-like property of having each of the 9 colours once in each row and column (and also in both major diagonals).

    12 years ago

    • avatar

      And looking at the post, I see it has the other Sudoku property as well (although I concede that wasn’t as obvious at first glance).

      12 years ago

    • avatar

      Also, apologies, but I think I put one too many zeros in that percentage :-)

      12 years ago