17
Dec

Should the Cox Proportional Hazards model get the Nobel Prize in Medicine?

Tweet about this on Twitter29Share on Facebook42Share on Google+3Share on LinkedIn0Email this to someone

I'm not the first one to suggest that Biostatistics has been undervalued in the scientific community, and some of the shortcomings of epidemiology and biostatistics have been noted elsewhere. But this previous work focuses primarily on the contributions of statistics/biostatistics at the purely scientific level.

The Cox Proportional Hazards model is one of the most widely used statistical models in the analysis of data from clinical trials and other medical studies. The corresponding paper has been cited over 32,000 times; this is a dramatically low estimate of the number of times the model has been used. It is one of "those methods" that doesn't even require a reference to the original methods paper anymore.

Many of the most influential medical studies, including major studies like the Women's Health Initiative have used these methods to answer some of our most pressing medical questions. Despite the incredible impact of this statistical technique on the world of medicine and public health, it has not received the Nobel Prize. This isn't an aberration, statistical methods are not traditionally considered for Nobel Prizes in Medicine. They primarily focus on biochemical, genetic, or public health discoveries.

In contrast, many economics Nobel Prizes have been awarded primarily for the discovery of a new statistical or mathematical concept. One example is the ARCH model. The Nobel Prize in Economics in 2003 was awarded to Robert Engle, the person who proposed the original ARCH model. The model has gone on to have a major impact on financial analysis, much like the Cox model has had a major impact on medicine?

So why aren't Nobel Prizes in medicine awarded to statisticians more often? Other methods such as ANOVA, P-values, etc. have also had an incredibly large impact on the way we measure and evaluate medical procedures. Maybe as medicine becomes increasingly driven by data, we will start to see more statisticians recognized for their incredible discoveries and the huge contributions they make to medical research and practice.

 

  • Anon
    • Keith O’Rourke

      Thanks for the Freedman paper.

      I have heard David Cox say this a number of times "Simpler analytic methods should be used first" (in Freedman's introduction) and also say a number of times he actually would prefer it not be called Cox Proportional Hazards model .
      David has worked at a Noble level for many years but pointing to one visible success that would not be widely misunderstood likely is problematic.

      (And folks are still waiting to hear the "Fat lady sing" on higher order asymptotic inference, I believe)

      • jtleek

        The purpose of the post was to call attention to one dramatically successful/influential statistical result that plays a major role in how we analyze/interpret/use medical data. There are of course a large number of other equally important examples.

        There are clearly well-founded criticisms of proportional hazards models, but it is hard to argue against the impact of the idea.

        • BillyJoeBob

          Consider the models that won their creators Nobel prizes in economics. Every last one is flawed (Black-Scholes FTW?) but every one represented a model that was less bad than all the others of the time.

          Cox's empirical process approach to survival analysis holds up very well.

  • Ken

    Have they been awarded at all? Bioinformaticians may be the next group to feel left out. The people who nominate would be similar to typical winners: they spend a lot of time in the lab, and they aren't likely to appreciate the talents of anyone who doesn't.

  • http://www.facebook.com/people/Paul-Lawrence-Hayes/100000758178679 Paul Lawrence Hayes

    “So why aren’t Nobel Prizes in medicine awarded to statisticians more often? Other methods such as ANOVA, P-values, etc. have also had an incredibly large impact on the way we measure and evaluate medical procedures.”

    You answer your own question there: those “fundamentally irrational” and “scandalously still taught” P-values etc. are an utter disgrace. They've damaged inference and science - especially medical science - and (orthodox) statisticians should consider themselves fortunate the scientific community hasn't awarded them the prizes they do deserve.

    • jtleek

      This seems a bit extreme.

      • http://www.facebook.com/people/Paul-Lawrence-Hayes/100000758178679 Paul Lawrence Hayes

        Sure. Such opinions - “It is a major scandal that orthodox methods continue to be taught at all to young statisticians, economists, biologists, and medical researchers; this has done irreparable damage in these fields for decades.” –E.T. Jaynes - may well seem a bit extreme if you're not familiar with the full context and justification.