Identity Verified Thinker in Business / Industries / Pharmaceuticals
Dennis Lendrem
Dennis Lendrem
MSc in Medical Statistics, PhD from Oxford. Background in Pharmaceutical R&D. Contributor to popular magazines and trade magazines including Scrip Magazine, New Scientist and the Economist. Formerly Editor of the journal Pharmaceutical Statistics.
 
Posted in Business / Industries / Pharmaceuticals

The Will Rogers Phenomenon

Feb. 7, 2012 4:09 pm
I belong to no organized party. I am a Democrat.
Will Rogers (1879-1935) actor, humorist, political columnist

Will Rogers was born William Penn Adair Rogers on the Dog Iron Ranch in Indian Territory in what is now Oklahoma. One quarter Cherokee he later quipped that his ancestors didn't come over on the Mayflower but they "met the boat."

The youngest of eight children Rogers dropped out of school in 10th grade and began working on the ranch before travelling to Argentina to set up his own ranch. Here he and his partner lost everything and Will set off for South Africa. In South Africa he worked as a ranch hand before joining a circus where he did roping and pony tricks. Returning to the US in 1904 Will Rogers went on to become an actor, humorist and social commentator. One of the best known celebrities of the 1920’s and 1930’s, at the time of his death Will Rogers was the highest paid actor in Hollywood.

It's a great story. It's a phenomenal story.

But this is not the Will Rogers Phenomenon.

So what exactly is the Will Rogers Phenomenon?

Moving an element from one set to another set raises the average values of both sets.
The Will Rogers Phenomenon

What has this got to do with Will Rogers? Bear with me.

The Will Rogers Phenomenon occurs when both the following conditions are met:

- The element being moved is below average for its current set. Removing it will, by definition, raise the average of the remaining elements.

- The element being moved is above the current average of the set it is entering. Adding itto the new set will, by definition, raise the average.

In particular, the Will Rogers Phenomenon is often used in epidemiological statistics to illustrate how historical controls can generate spurious improvements in the medium-term prognosis of patients. These improvements can be wrongly interpreted as treatment effects. The Will Rogers Phenomenon refers to apparently 'improved' survival in cancer (or other) patients that comes about by either:

- reclassifying them into different prognosticgroups,

- recognising more subtle disease manifestations, or by

- using diagnostic modalities that allow the diseaseto be diagnosed at an earlier stage.

These will result in a ‘zero-time shift’ and apparent improvements in survival. Essentially the disease is diagnosed at an earlier stage leading to an apparent increase in the length of survival even in the absence of any real improvements in treatment or care.

The Will Rogers Phenomenon has become even more of an issue with the development of new imaging tools allowing detection of cancer metastases before they become evident clinically. As a result, more patients are classified into the more severe metastatic disease stage from the less severe single tumour stage. Such 'stage migration' results in an apparent improved survival of patients in both the less and the more severe disease stages.

In other words, moving a subset from one stage to another stage increases the average survival times of both sets!

Which brings us back to Will Rogers.

While Will Rogers spent most of his life in California he was proud to hail from Oklahoma. He famously joked:

"When the Okies left Oklahoma and moved to California, they raised the average intelligence level in both states."

Moving an element from one set to another set raises the average values of both sets.
The Will Rogers Phenomenon
 
There are currently no comments.
 
 
 
Latest Ebooks