RE: st: Simplification of formula in logistic regression

["David Radwin" <dradwin@mprinc.com>]

Mikkel,

On why not to group continuous variables, please see
http://www.stata.com/statalist/archive/2011-03/msg00154.html and
references therein, particularly:

Wainer, H., Geseroli, M. & Verdi, M. 2006. Finding what is not there
through the unfortunate binning of results: The Mendel effect. Chance,
19(1):49-56.
http://www.amstat.org/publications/chance/2006/CHANCE%2019_1.pdf

David
--
David Radwin
Research Associate
MPR Associates, Inc.
2150 Shattuck Ave., Suite 800
Berkeley, CA 94704
Phone: 510-849-4942
Fax: 510-849-0794

www.mprinc.com


> -----Original Message-----
> From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-
> statalist@hsphsun2.harvard.edu] On Behalf Of Nick Cox
> Sent: Sunday, May 15, 2011 9:49 AM
> To: statalist@hsphsun2.harvard.edu
> Subject: Re: st: Simplification of formula in logistic regression
>
> Sorry, but I think you will continue find this "correct way" to be
> elusive.
>
> Nick
>
> On Sun, May 15, 2011 at 4:23 PM, Mikkel Brabrand <mikkel@brabrand.net>
> wrote:
> > If I want clinicians to use my model, it needs to be simple. I cannot
> expect them to use a piece of software to calculate the risk score and
it
> is virtually impossible to have it incorporated in the programs used at
my
> department. I therefore need to simplify it and make the variables
> categorized or dichotomous. I have previously used the trial and error
> way, and come up with a model that seems reasonable (and tested it in an
> independent cohort, and am now testing it in two external cohorts at
other
> hospitals). However, there must be a correct way to select the cuf-off
> levels, I just cannot find out how. I have asked most statisticians I
have
> met on my way, but no one seems to know how. I hoped that some of you
> might have a suggestion...
> >
> > Mikkel
> >
> > Den 15/05/2011 kl. 16.49 skrev Nick Cox:
> >
> >> I don't know what "statistically correct" would mean here. If you
> >> think your model is useful, there are no grounds for coarsening it.
If
> >> the implication is that clinicians can't understand or don't need to
> >> understand the internals of the formula you can think of
encapsulating
> >> the details in a Stata do-file or some equivalent in other software.
> >>
> >> A broad issue is that detailed models optimised to fit particular
> >> datasets often perform poorly on other data.
> >>
> >> Nick
> >>
> >> On Sun, May 15, 2011 at 3:43 PM, Mikkel Brabrand
<mikkel@brabrand.net>
> wrote:
> >>
> >>> I have performed a logistic regression analysis including five
> variables and one outcome. However, I would like to simplify the formula
> significantly for clinical use. So, instead of the formula been
something
> like -12.22+2.33*systolic blood pressure-1.21*temperature etc., I would
> like to make a scoring system where the score is calculated on basis of
> the measured values of the vital signs.
> >>>
> >>> An example could be something like this
> >>>
> >>> .................2 points..1 point...0 points...1 point.....2 points
> >>>
> >>> Pulse   ...........-30........31-50....51-100....101-200..201-
> >>>
> >>> Sys. BP.........-60........61-100..101-200...201-
> >>>
> >>> However, I have no idea how to find the optimal cut-off points. Do
any
> of you have a suggestion how to do this statistically correct?
> >>
> >> *
>
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/statalist/faq
> *   http://www.ats.ucla.edu/stat/stata/
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/