Dear Bob
An interesting problem. You state you need to stratify by GP clinic -
is there evidence of a big clinic effect? If not, the it may not matter.
The problem would be if, by chance, all the patients in one clinic
received the same treatment. Then, if you analysed using a fixed
effects model for GP clinic, you could not separate out the clinic
effect from the treatment effect.
One possibility would be to use a balanced incomplete block design
(i.e. only have two treatments per clinic). This would enable you to
get within clinic traetment contrasts.
I would also be interested to know whether one should analyse the
trial using a random effect for clinic, and if so whether the chance of
all patients in one clinic getting the same treatment was such a worry.
Best wishes
Mike
Date sent: Thu, 17 Jun 2004 09:17:47 +0100
Send reply to: Medical Statisticians interested in primary care <[log in to unmask]>
From: "Blizard, Robert" <[log in to unmask]>
Subject: randomisation/minimisation
To: [log in to unmask]
> I have been asked to help with randomisation in a General Practice
> trial. It is straight forward but has one complication. If anybody
> has experience of this, I would be grateful for your suggestions.
>
> The trial has three arms (A,B,C). The subjects need to be stratified
> by disease severity(hi,lo) and GP clinic. Each arm requires equal
> numbers and the ratio of patients in each severity level is unknown.
> Under normal circumstances, I would naturally prepare two block
> randomised lists, one for each severity level, in each clinic. The
> potential difficulty arises because the expected number of subjects in
> each clinic is 5.
>
> In perhaps the majority of clinic/severity combinations, there will be
> unfilled blocks, often insuffucient subjects to fill a single
> randomistion block. Am I right to be concerned about this ?
Professor Mike Campbell
Medical Statistics Unit
Institute of Primary Care and General Practice
University of Sheffield
Community Sciences Centre
Northern General Hospital
Sheffield S5 7AU
Tel 0114 271 5919
Fax 0114 242 2136
e-mail [log in to unmask]
http://www.shef.ac.uk/michaelcampbell/
http://www.shef.ac.uk/medical-statistics/
|