Predicting Doom—Guest Post by Thomas Galli

Some treatments are more efficacious than others.
Some treatments are more efficacious than others.

I am not a statistics wizard; an engineer, I value the predictive power of statistics. Indeed, if one can precisely control variables in the design of an experiment, statistics-based prediction of future material properties is remarkably accurate. The joy of predicting end strength for a new carbon nanotube concrete mix design in minutes versus days melts the heart of this engineer.

This predictive power has a foreboding downside. It attaches to other projections, including those used by the medical profession to forecast life after diagnosis with late-stage cancer. Unfortunately, I have first-hand experience with this. I was granted but 6 months of remaining life nearly 11 years ago! My doom was predicted with certainty, and for a while, I believed it.

In the dwell time between treatments, I searched for methods used to generate projections of doom. Each patient’s type, stage, age, ethnicity and race were reported to the National Cancer Institute upon diagnosis. Deaths were also reported but not the cause of death. Nothing was captured on complicating health problems like cardio-pulmonary disease, diabetes or other life-threatening diseases. The predictive data set appeared slim.

My battle turned while mindlessly searching web pages of the American Cancer Society. Ammunition in the form of a powerful essay from the noted evolutionary biologist Stephen Jay Gould—“The Median Isn’t The Message”—contained the words: “…leads us to view statistical measures of central tendency wrongly, indeed opposite to the appropriate interpretation in our actual world of variation, shadings, and continua.”

The statistician seeks to aggregate and explain. I’d forgotten that I was in a “world of variation,” was but one data point in about 1.4 million Americans diagnosed in 2004. I might be “the one” on the right-shifted curve prohibiting intersection with the x-axis.

There was one benefit from my encounter with predictive doom. I found hope—something no statistician can aggregate or explain.

Gould survived 20 years beyond his late-stage, nearly always statistically fatal, abdominal cancer diagnosis. Ironically, he passed after contracting another form of unrelated cancer. A distinguished scientist, Gould eloquently described the limits of science and statistics by suggesting that “a sanguine personality” might be the best prescription for success against cancer. There is always hope, with high confidence.


Editor’s Note I have long been interested in working with physicians who routinely make end-of-life prognoses. The concepts of rating such judgments are no different than, say, judging how well climate models predict future temperatures. I mean predictions should be rated on their difficulty. I haven’t yet discovered docs willing to conduct these experiments, but if anybody happens to know somebody, let me know.


  1. DAV,

    You can see how bad it’s gotten when my enemies attack not only my own posts, but those of my guests as well.

  2. Thank you, Mr. Galli, for your personal story. I am so glad that you have survived your cancer so successfully. I am also grateful that you found just the right inspiration to continue therapy.

    However, although I am sure that Stephen Jay Gould is still considered a “noted evolutionary biologist”, I am equally certain that he should be “notorious” rather than “noted”. Even aside from Gould’s amateurish understanding of psychometric data and its analysis (a fact more or less notorious among psychometricians), just for example, in his most infamous work, ‘The Mismeasure of Man’, he falsified the key data he used to support his case:

    “… Our analysis of Gould’s claims reveals that most of Gould’s criticisms are poorly supported or falsified.”

    And Mr. Gould made a LOT of money from his ‘science’. According to the AP in (I think) early March 2006 (I couldn’t quickly find the original AP report), a lawyer for his widow (she was suing the same doctor who had saved him from a previous cancer in 1982) claimed that “Dr. Gould earned $300,000 a year from speaking engagements alone,” that “a seven-figure income was his norm” and that when he died “he was about to enter into a book contract for more than $2 million.” Matt, take note.

    I couldn’t be happier that you found life-saving value in Mr. Gould’s writing. That essay he wrote about his own cancer is really noteworthy, and has inspired others, perhaps many, many others, as well as yourself. That is wonderful!

    But Mr. Gould was (I hope) not really a “noted” evolutionary biologist.

  3. I think the whole point is to give you and your family an estimate of time so that you can make plans and prepare for such a very serious and difficult situation. If all goes well, that’s great. But it is unfortunately wise to prepare for the worst.


  4. JMJ,

    This is the key point. Without going into details, it is best if the doctor gives the best picture given his available information and does not try to slant his guess one way or another in an attempt at anticipating his patients’ actions. It’s the patients’ decisions here that count based on the prediction, not the doc’s.

    These often conflict, especially in screening, where it is usually best for a doc to insist that a patient be screened for a rare disease. But it is not always best for any patient. See the classic posts for breast cancer screening for an idea.


    This is relevant, in an odd sort of way.

    Seismologists in Italy cleared of manslaughter charges. Experts making predictions again.

  5. Surely what the physician needs to provide is a sense of the range of survival chances. That cannot be summed up just in the average, mean, mode or whatever – these figures are of little help in planning your future without some knowledge of the range, and how the figures are skewed within it.

  6. Just to add nearly all engineering is also this. You do not want to know how your average bridge will behave – you use your understanding of the range of outcomes to design one where it is safe (enough) anywhere within that range.

  7. Meh.

    My Mom was diagnosed with stage 4 breast cancer and basically died right on schedule with the majority of people with this diagnosis. The only failure in this post was the apparent inability to understand what the statistics are telling you(?). The most common cancer statistic is the 5 year survival rate and there are associated graphs that tell you the death rates per year for people with a similar diagnosis. Of course it doesn’t tell you exactly when a specific individual will die, or even if they will die from the diagnosis.

    You get stage 4 breast cancer, and you are likely to die within 5 years. Period. Wishing this wasn’t the case changes nothing. The fact that there are survivors changes nothing. The non-survivors aren’t writing any blog posts. Sometimes chemotherapy and other treatments can actually cure you, mostly not.

    I speak only for myself here, but I am not very enthralled with the “war on cancer” and the rah rah “I survived” hopey changey stuff that treats cancer like it was a sporting event or something. It is a nasty terrible disease and an absolutely horrible way to die. Sorry for the downer message, but this is reality folks. The Hospice organization should be commended for all the good work they do with terminal cancer patients.

    I recommend the Pulitzer prize winning book “The Emperor of All Maladies” for anyone interested in the history of the fight against cancer.

  8. Ray,
    I think we know exactly what causes cancer in a technological sense. Cellular DNA damage that causes cellular uncontrolled growth. Cells refuse to recognize the signal to stop multiplying. These groups of cell growths (aka tumors) spread until one group grows in an unfortunate location and causes organ failure which leads to death.
    The belief that there would be a silver bullet that prevents these mutations turned out to be naive. In one sense, the same forces that cause evolution (random mutations) and natural selection is also the same forces behind cancer (specific mutations that lead to death). Cellular damage can occur from many different sources, such as cigarette smoke or leaking radiation after the local nuke plant had a minor safety issue such as Chernobyl.

Leave a Comment

Your email address will not be published. Required fields are marked *