{MEDSTATS} An issue involving independence and dependence - MedStats

The old Google Groups will be going away soon, but your browser is incompatible with the new version.
Message from discussion {MEDSTATS} An issue involving independence and dependence
Received-SPF: pass (google.com: domain of mkbeh...@gmail.com designates 209.85.213.182 as permitted sender) client-ip=209.85.213.182;
MIME-Version: 1.0
In-Reply-To: <99BF2055-86E3-47DC-BCED-72D25B91C...@me.com>
References: <9B75E7CF385CB94EAD6587DD96AC2D974660C...@SFPCEXMBX2.sanfordhealth.org>
	<99BF2055-86E3-47DC-BCED-72D25B91C...@me.com>
Date: Wed, 11 Jul 2012 14:34:12 -0400
Message-ID: <CAGioPdj9e9Y=c7Tf=5A0NAqvre4DGpt2y4SrUURuMKy+EMR...@mail.gmail.com>
Subject: Re: {MEDSTATS} An issue involving independence and dependence
From: M Behnke <mkbeh...@gmail.com>
To: medstats@googlegroups.com
Content-Type: multipart/alternative; boundary=14dae9340c4fef7bf304c49217f5

--14dae9340c4fef7bf304c49217f5
Content-Type: text/plain; charset=windows-1252
Content-Transfer-Encoding: quoted-printable

I would definitely not treat the legs as independent, because response to
treatment is very much dependent on patient factors such as immune status
etc that will affect both legs in the same way.

You could test this by calculating the correlation  in response between
right and left legs, but if (as I suspect) correlation is high, you're
stuck with deciding how to appropriately include their data (which is NOT
to choose the leg that had the response the PI wants ;)

Mik


On Tue, Jul 10, 2012 at 3:51 PM, Marc Schwartz <marc_schwa...@me.com> wrote=
:

> On Jul 10, 2012, at 1:53 PM, Thompson,Paul wrote:
>
> > I am involved as the statistician in a small trial of the use of a
> topical medication in a surgical intervention. The intervention occurs in
> the leg. We in the clinical research group assumed control of this trial
> after the PI (surgeon) started the trial. The endpoint is reoccurrence
> within the leg of the original condition.
> >
> > We discovered after the fact that the PI has begun to enroll both legs
> in some cases, treating each separately. The PI wishes to treat each leg
> independently. I do not. In addition, at the end of the trial, an unknown
> proportion of cases will  have two legs treated =96 to date, it is 1 of 1=
7
> cases or thereabouts. Of the 50 cases who are to be treated, we could
> possibly expect 3-5 to have 2 legs treated.
> >
> > I have contended that any publication of this trial will require
> evidence that the assumption of independence of the two limbs for a given
> person is a reasonable one. The PI wishes to assume independence simply o=
n
> his say-so.
> >
> > For those of you who have and are currently serving as referees, what
> say you? Must a strong case for independence be made? Or would you allow =
a
> trial with a simple assumption of independence to be considered?
> >
>
> Paul,
>
> Is this situation the same one that you posted on back in April?
>
>   https://groups.google.com/d/topic/medstats/ltydfqDfTE4/discussion
>
> If not, then perhaps the prior discussion might be apropos here and Prof.
> Altman's references in his reply in that thread might be of benefit.
>
> Regards,
>
> Marc Schwartz
>
> --
> To post a new thread to MedStats, send email to MedStats@googlegroups.com=
.
> MedStats' home page is http://groups.google.com/group/MedStats .
> Rules: http://groups.google.com/group/MedStats/web/medstats-rules
>

--14dae9340c4fef7bf304c49217f5
Content-Type: text/html; charset=windows-1252
Content-Transfer-Encoding: quoted-printable

I would definitely not treat the legs as independent, because response to t=
reatment is very much dependent on patient factors such as immune status et=
c that will affect both legs in the same way. =A0<div><br></div><div>You co=
uld test this by calculating the correlation =A0in response between right a=
nd left legs, but if (as I suspect) correlation is high, you&#39;re stuck w=
ith deciding how to appropriately include their data (which is NOT to choos=
e the leg that had the response the PI wants ;)</div>
<div><br></div><div>Mik</div><div><br><br><div class=3D"gmail_quote">On Tue=
, Jul 10, 2012 at 3:51 PM, Marc Schwartz <span dir=3D"ltr">&lt;<a href=3D"m=
ailto:marc_schwa...@me.com" target=3D"_blank">marc_schwa...@me.com</a>&gt;<=
/span> wrote:<br>
<blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1p=
x #ccc solid;padding-left:1ex"><div class=3D"im">On Jul 10, 2012, at 1:53 P=
M, Thompson,Paul wrote:<br>
<br>
&gt; I am involved as the statistician in a small trial of the use of a top=
ical medication in a surgical intervention. The intervention occurs in the =
leg. We in the clinical research group assumed control of this trial after =
the PI (surgeon) started the trial. The endpoint is reoccurrence within the=
 leg of the original condition.<br>

&gt;<br>
&gt; We discovered after the fact that the PI has begun to enroll both legs=
 in some cases, treating each separately. The PI wishes to treat each leg i=
ndependently. I do not. In addition, at the end of the trial, an unknown pr=
oportion of cases will =A0have two legs treated =96 to date, it is 1 of 17 =
cases or thereabouts. Of the 50 cases who are to be treated, we could possi=
bly expect 3-5 to have 2 legs treated.<br>

&gt;<br>
&gt; I have contended that any publication of this trial will require evide=
nce that the assumption of independence of the two limbs for a given person=
 is a reasonable one. The PI wishes to assume independence simply on his sa=
y-so.<br>

&gt;<br>
&gt; For those of you who have and are currently serving as referees, what =
say you? Must a strong case for independence be made? Or would you allow a =
trial with a simple assumption of independence to be considered?<br>
&gt;<br>
<br>
</div>Paul,<br>
<br>
Is this situation the same one that you posted on back in April?<br>
<br>
=A0 <a href=3D"https://groups.google.com/d/topic/medstats/ltydfqDfTE4/discu=
ssion" target=3D"_blank">https://groups.google.com/d/topic/medstats/ltydfqD=
fTE4/discussion</a><br>
<br>
If not, then perhaps the prior discussion might be apropos here and Prof. A=
ltman&#39;s references in his reply in that thread might be of benefit.<br>
<br>
Regards,<br>
<br>
Marc Schwartz<br>
<div class=3D"HOEnZb"><div class=3D"h5"><br>
--<br>
To post a new thread to MedStats, send email to <a href=3D"mailto:MedStats@=
googlegroups.com">MedStats@googlegroups.com</a> .<br>
MedStats&#39; home page is <a href=3D"http://groups.google.com/group/MedSta=
ts" target=3D"_blank">http://groups.google.com/group/MedStats</a> .<br>
Rules: <a href=3D"http://groups.google.com/group/MedStats/web/medstats-rule=
s" target=3D"_blank">http://groups.google.com/group/MedStats/web/medstats-r=
ules</a><br>
</div></div></blockquote></div><br></div>

--14dae9340c4fef7bf304c49217f5--
Create a group - Google Groups - Google Home - Terms of Service - Privacy Policy
Account Options
