R help - Oct 2008 - Dose Profile

Hi Everyone,

I have data in a long format e.g. there is one row per patient but each
follow-up appointment is included in the row.  So, a snippet of the data
looks like this:
      TrialNo Drug Sex Rand Adate1 Date1 Dose1 Time1 Adate2 Date2 Dose2
Time2  B1001029 LTG M 15719 30/04/2003 15825 150 106 29/08/2003 15946 200
227  B1117003 LTG M 15734 30/04/2003 15825 200 91 03/09/2003 15951 250 217
B138015 LTG M 14923 06/02/2001 15012 225 89 08/05/2001 15103 300 180
B112003 TPM F 14914 15/01/2001 14990 60 76 05/03/2001 15039 100 125
Of course, not everyone has the same number of follow-up appointments and so
there may be some column entries that are NAs.

What I'd like to do is a dose profile i.e. a plot of time on the x-axis
agaisnt dose on the y-axis for each patient ideally colouring lines
according to drug.  Does anyone know how I can do this?  Someone at work has
suggested that I use plot and then loess.smooth but I don't really know what
to do.

Thank you very much for your help,

Laura

	[[alternative HTML version deleted]]

Why don't you ask your "someone at work" to help, as you appear to
have
local resources handy?

-- Bert Gunter 

-----Original Message-----
From: r-help-bounces at r-project.org [mailto:r-help-bounces at r-project.org]
On
Behalf Of Laura Bonnett
Sent: Tuesday, October 28, 2008 9:33 AM
To: r-help at r-project.org
Subject: [R] Dose Profile

Hi Everyone,

I have data in a long format e.g. there is one row per patient but each
follow-up appointment is included in the row.  So, a snippet of the data
looks like this:
      TrialNo Drug Sex Rand Adate1 Date1 Dose1 Time1 Adate2 Date2 Dose2
Time2  B1001029 LTG M 15719 30/04/2003 15825 150 106 29/08/2003 15946 200
227  B1117003 LTG M 15734 30/04/2003 15825 200 91 03/09/2003 15951 250 217
B138015 LTG M 14923 06/02/2001 15012 225 89 08/05/2001 15103 300 180
B112003 TPM F 14914 15/01/2001 14990 60 76 05/03/2001 15039 100 125
Of course, not everyone has the same number of follow-up appointments and so
there may be some column entries that are NAs.

What I'd like to do is a dose profile i.e. a plot of time on the x-axis
agaisnt dose on the y-axis for each patient ideally colouring lines
according to drug.  Does anyone know how I can do this?  Someone at work has
suggested that I use plot and then loess.smooth but I don't really know what
to do.

Thank you very much for your help,

Laura

	[[alternative HTML version deleted]]

______________________________________________
R-help at r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.

Laura Bonnett wrote:
> Hi Everyone,
>
> I have data in a long format e.g. there is one row per patient but each
> follow-up appointment is included in the row.  So, a snippet of the data
> looks like this:
>       TrialNo Drug Sex Rand Adate1 Date1 Dose1 Time1 Adate2 Date2 Dose2
> Time2  B1001029 LTG M 15719 30/04/2003 15825 150 106 29/08/2003 15946 200
> 227  B1117003 LTG M 15734 30/04/2003 15825 200 91 03/09/2003 15951 250 217
> B138015 LTG M 14923 06/02/2001 15012 225 89 08/05/2001 15103 300 180
> B112003 TPM F 14914 15/01/2001 14990 60 76 05/03/2001 15039 100 125
> Of course, not everyone has the same number of follow-up appointments and
so
> there may be some column entries that are NAs.
>
> What I'd like to do is a dose profile i.e. a plot of time on the x-axis
> agaisnt dose on the y-axis for each patient ideally colouring lines
> according to drug.  Does anyone know how I can do this?  Someone at work
has
> suggested that I use plot and then loess.smooth but I don't really know
what
> to do.
>
>   
Hi Laura,
You can get a basic plot like this:

matplot(rbind(lbonnett$Time1,lbonnett$Time2),
 rbind(lbonnett$Dose1,lbonnett$Dose2),type="l",
 col=lbonnett$Drug,lty=1)

As you can see by running this, you will get a line for each patient, 
with the color of each line determined by the drug (you can select 
different colors, I was just being lazy). The problem you are likely to 
face is that there will be gaps in the lines if you have NAs. You can 
rejig matplot or write a similar function to get around this so that the 
lines are just drawn across the gaps if that is what you want.

Jim

you 'll may be interested in the drc package (a quick start would be 
http://www.bioassay.dk), where you get some nice plots and can fit 
appropriate models as well - very handy if you want to estimate 
effective doses.

hth.


Laura Bonnett schrieb:
> Hi Everyone,
>
> I have data in a long format e.g. there is one row per patient but each
> follow-up appointment is included in the row.  So, a snippet of the data
> looks like this:
>       TrialNo Drug Sex Rand Adate1 Date1 Dose1 Time1 Adate2 Date2 Dose2
> Time2  B1001029 LTG M 15719 30/04/2003 15825 150 106 29/08/2003 15946 200
> 227  B1117003 LTG M 15734 30/04/2003 15825 200 91 03/09/2003 15951 250 217
> B138015 LTG M 14923 06/02/2001 15012 225 89 08/05/2001 15103 300 180
> B112003 TPM F 14914 15/01/2001 14990 60 76 05/03/2001 15039 100 125
> Of course, not everyone has the same number of follow-up appointments and
so
> there may be some column entries that are NAs.
>
> What I'd like to do is a dose profile i.e. a plot of time on the x-axis
> agaisnt dose on the y-axis for each patient ideally colouring lines
> according to drug.  Does anyone know how I can do this?  Someone at work
has
> suggested that I use plot and then loess.smooth but I don't really know
what
> to do.
>
> Thank you very much for your help,
>
> Laura
>
> 	[[alternative HTML version deleted]]
>
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>   
-- 
Eik Vettorazzi
Institut f?r Medizinische Biometrie und Epidemiologie
Universit?tsklinikum Hamburg-Eppendorf

Martinistr. 52
20246 Hamburg

T ++49/40/42803-8243
F ++49/40/42803-7790