A similar debate is going on in the BMJ at the moment in response
to the paper by Smeeth, Haines and Ebrahim. They point out the
serious limitations of NNTs.
My view is quite clear: NNTs are quite unsuitable for reporting the
results of trials and (especially) meta-analyses. Relative risks are
acceptable if the background risk is low. However the measure of
choice is the odds ratio and analysis should proceed on the log-
odds ratio scale unless evidence is produced that this scale is not
additive. (This is the default form for in generalised linear models for
binary data but others can be considered.)
There is a general confusion between the two phases of modelling
an effect and applying the results. Modelling should follow the
science. If we are going to pool results between different trials and
apply results studied in one population to another than we need
results that are as nearly constant from one to the other as is
possible. (Additive to use the statistician's jargon.) When we finally
come to make a clinical decision based on these results, then and
only then has the time arrived to translate the finding into clinically
relevant measures using as further inputs whatever is known about
the clinical state of the patient.
John Nelder some years ago pointed out to a similar confusion in
the field of quality control where workers were using the measure of
final interest to dictate the form of analysis rather than suing
additive models and translating the results at the point of
application.
In short our motto should be "additive at the point of analysis,
relevant at the point of application".
Regards
Stephen Senn
Date sent: Sun, 13 Jun 1999 03:37:47 PDT
Subject: NNT or no NNT: NEJM Editorial correspondence
From: padmanabhan badrinath <[log in to unmask]>
To: [log in to unmask]
Send reply to: padmanabhan badrinath <[log in to unmask]>
> Dear All,
> This week’s issue of the NEJM carries an exchange of views on NNT. In
> January the journal published the following article which showed that
> prophylactic mastectomy reduces the incidence of breast cancer by 90% in
> high risk probands with a family history of breast cancer.
>
> Hartmann LC, Schaid DJ, Woods JE, et al. Efficacy of bilateral prophylactic
> mastectomy in women with a family history of breast cancer. N Engl J Med
> 1999;340:77-84.
>
> As the authors had presented only the relative risk reduction, one of the
> readers has raised the issue of NNT in the editorial correspondence column
> (Ham RM et al. N Engl J Med, June 10 1999). They question the non use of NNT
> and provides a table of NNT for this article. In their response the authors
> of the original article defends their position. To quote Hartman et al,
> "Measurements of relative risk are currently the standard for reporting the
> results of trials of screening, treatment, and prevention. The number needed
> to treat has some advantages but important limitations as well. This number
> is not static but changes with the duration of follow-up, if the
> intervention has a durable effect". They provide another table showing how
> the NNTs vary depending on the duration of follow-up.
> I found the exchange useful and informative and might use it in future to
> present the opposing views on NNT while teaching our students. I am sending
> this message to the list for information but will be glad hear your views
> especially on the statement that "Measurements of relative risk are
> currently the standard for reporting the results of trials of screening,
> treatment, and prevention. The number needed to treat has some advantages
> but important limitations as well."
>
> Cheers and continue to have a good week end; we are already at the end of
> the second day of this week in the middle east!
>
> Badri
>
>
> Dr.P.Badrinath M.D.,M.Phil.,(Epid) PhD(Cantab)
> Assistant Professor and Epidemiologist,
> Department of Community Medicine,
> UAE University, PO Box 17666, Al Ain,
> United Arab Emirates.
> Tel: 00 971 3 5039 652
> Fax: 00 971 3 672022.
> [log in to unmask]
>
>
>
> ______________________________________________________
> Get Your Private, Free Email at http://www.hotmail.com
--------------------------------------------------
Professor Stephen Senn
Department of Statistical Science &
Department of Epidemiology and Public Health
University College London
Room 316, 1-19 Torrington Place
LONDON WC1E 6BT
Tel: +44 (0) 171 391 1698
Fax: +44 (0) 171 813 0280
Email: [log in to unmask]
webpage: http://www.ucl.ac.uk/~ucaksjs/
-------------------------------------------------
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|