[R] I have a question on nomograms.

[Frank E Harrell Jr]

Gabor Grothendieck wrote:
> On Thu, Mar 25, 2010 at 10:43 PM, Frank E Harrell Jr
> <f.harrell at vanderbilt.edu> wrote:
>> (Ted Harding) wrote:
>>>  http://www.nomogram.org
>>>
>>> They call this sort of interface "nomogram" too, and you can have
>>> a go at the one they offer for bladder cancer. This is not a
>>> graphical nomogram in the sense that you describe, rather it is
>>> a form-filling interface with boxes into which you insert
>>> information, and then press the button.
>> It is very curious that a site named nomogram.org would not contain a single
>> nomogram.  I have pointed this out to the author of the site, who maintains
>> that disagreements with dictionaries are inconsequential.
>>
> 
> It is strange although I think that in certain areas of medicine its
> common to use the term nomogram to mean the model underlying the
> nomogram as opposed to its graphical representation.
> 
> For example, in the following discussion the authors seem to be using
> nomogram  merely to denote any model that does not involve forming
> discrete levels of the inputs in the way that forming risk groups
> (high risk group, low risk group, etc.) would:
> 
> http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2777059/
> 

Gabor,

It is not very common to use it in that manner although I have seen it a 
few times.  But it is never correct, even according to a medical 
dictionary: http://medical-dictionary.thefreedictionary.com/nomogram

Frank


-- 
Frank E Harrell Jr   Professor and Chairman        School of Medicine
                      Department of Biostatistics   Vanderbilt University