R help - Jun 2010 - Recoding dates to session id in a longitudinal dataset

Hi,

I'm fairly new to R but I have a large dataset (300000 obs) containing
patient material. Some patients came 2-9 times during the three year
observation period. The patients are identified by a unique idnr, the
sessions can be distinguished using the session date. How can I recode the
date of the session to a session id (1-9). This would be necessary to obtain
information and do some analysis on the first occurence of a specific
patient or to look for trends.

Thanks

JP Bogers
University of Antwerp

	[[alternative HTML version deleted]]

---------- Forwarded message ----------
From: John-Paul Bogers <john-paul.bogers@ua.ac.be>
Date: Sat, Jun 26, 2010 at 10:14 PM
Subject: Re: [R] Recoding dates to session id in a longitudinal dataset
To: jim holtman <jholtman@gmail.com>


Dear Jim,

he data concerns HPV screening data.
The data looks as follows
pat1 sampledate1 HPV16 0.3
pat2 sampledate2 HPV16 0
pat3 sampledata3 HPV16 0.5
pat1 sampledate4 HPV16 0.6
pat4 sampledate5 HPV16 0
pat2 sampledate6 HPV16 0
pat1 sampledate7 HPV16 0

What I would like is

pat1 1  HPV16 0.3
pat2 1  HPV16 0
pat3 1  HPV16 0.5
pat1 2  HPV16 0.6
pat4 1  HPV16 0
pat2 2  HPV16 0
pat1 3  HPV16 0

I would like to recode sampledate (real date, in date format) to session
sequence (first sample of this patient, second sample of this patient, ....)

I hope this makes it clear.

Thanks

JP

PS: I answered this as a reply to your private mail, how do I get this on
the mailinglist?
 On Sat, Jun 26, 2010 at 7:59 PM, jim holtman <jholtman@gmail.com> wrote:
> It would be useful if you could provide an example of what the data
> looks like now and what you would like it to look like; otherwise it
> is impossible to help.
>
> On Sat, Jun 26, 2010 at 8:37 AM, John-Paul Bogers
> <john-paul.bogers@ua.ac.be> wrote:
> > Hi,
> >
> > I'm fairly new to R but I have a large dataset (300000 obs)
containing
> > patient material. Some patients came 2-9 times during the three year
> > observation period. The patients are identified by a unique idnr, the
> > sessions can be distinguished using the session date. How can I recode
> the
> > date of the session to a session id (1-9). This would be necessary to
> obtain
> > information and do some analysis on the first occurence of a specific
> > patient or to look for trends.
> >
> > Thanks
> >
> > JP Bogers
> > University of Antwerp
> >
> >        [[alternative HTML version deleted]]
> >
> > ______________________________________________
> > R-help@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.
> >
>
>
>
> --
> Jim Holtman
> Cincinnati, OH
> +1 513 646 9390
>
> What is the problem that you are trying to solve?
>
	[[alternative HTML version deleted]]

Hi Jim,

Thanks for the answer.
What I actually want is a session sequence 1,2,... per patient. This would
be very useful to look at trends of HPV infections from the first to the
second sample etc. It would also allow me to extract the HPV data of the
first sample (session 1).

Thx
JP


On Sat, Jun 26, 2010 at 10:23 PM, jim holtman <jholtman@gmail.com> wrote:
> Here is one way of doing it:
>
> > # data by patient and then sequentially number the data
> > x
>    V1          V2    V3  V4
> 1 pat1 sampledate1 HPV16 0.3
> 2 pat2 sampledate2 HPV16 0.0
> 3 pat3 sampledata3 HPV16 0.5
> 4 pat1 sampledate4 HPV16 0.6
> 5 pat4 sampledate5 HPV16 0.0
> 6 pat2 sampledate6 HPV16 0.0
> 7 pat1 sampledate7 HPV16 0.0
> > x.s <- split(x, x$V1)
> > # now put in the ids
> > do.call(rbind, lapply(x.s, function(.pat){
> +     .pat$V2 <- seq(nrow(.pat))
> +     .pat
> + }))
>         V1 V2    V3  V4
> pat1.1 pat1  1 HPV16 0.3
> pat1.4 pat1  2 HPV16 0.6
> pat1.7 pat1  3 HPV16 0.0
> pat2.2 pat2  1 HPV16 0.0
> pat2.6 pat2  2 HPV16 0.0
> pat3   pat3  1 HPV16 0.5
> pat4   pat4  1 HPV16 0.0
>
>
> On Sat, Jun 26, 2010 at 4:14 PM, John-Paul Bogers
> <john-paul.bogers@ua.ac.be> wrote:
> > Dear Jim,
> > he data concerns HPV screening data.
> > The data looks as follows
> > pat1 sampledate1 HPV16 0.3
> > pat2 sampledate2 HPV16 0
> > pat3 sampledata3 HPV16 0.5
> > pat1 sampledate4 HPV16 0.6
> > pat4 sampledate5 HPV16 0
> > pat2 sampledate6 HPV16 0
> > pat1 sampledate7 HPV16 0
> > What I would like is
> > pat1 1  HPV16 0.3
> > pat2 1  HPV16 0
> > pat3 1  HPV16 0.5
> > pat1 2  HPV16 0.6
> > pat4 1  HPV16 0
> > pat2 2  HPV16 0
> > pat1 3  HPV16 0
> > I would like to recode sampledate (real date, in date format) to
session
> > sequence (first sample of this patient, second sample of this patient,
> ....)
> > I hope this makes it clear.
> > Thanks
> > JP
> > PS: I answered this as a reply to your private mail, how do I get this
on
> > the mailinglist?
> > On Sat, Jun 26, 2010 at 7:59 PM, jim holtman
<jholtman@gmail.com> wrote:
> >>
> >> It would be useful if you could provide an example of what the
data
> >> looks like now and what you would like it to look like; otherwise
it
> >> is impossible to help.
> >>
> >> On Sat, Jun 26, 2010 at 8:37 AM, John-Paul Bogers
> >> <john-paul.bogers@ua.ac.be> wrote:
> >> > Hi,
> >> >
> >> > I'm fairly new to R but I have a large dataset (300000
obs) containing
> >> > patient material. Some patients came 2-9 times during the
three year
> >> > observation period. The patients are identified by a unique
idnr, the
> >> > sessions can be distinguished using the session date. How can
I recode
> >> > the
> >> > date of the session to a session id (1-9). This would be
necessary to
> >> > obtain
> >> > information and do some analysis on the first occurence of a
specific
> >> > patient or to look for trends.
> >> >
> >> > Thanks
> >> >
> >> > JP Bogers
> >> > University of Antwerp
> >> >
> >> >        [[alternative HTML version deleted]]
> >> >
> >> > ______________________________________________
> >> > R-help@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.
> >> >
> >>
> >>
> >>
> >> --
> >> Jim Holtman
> >> Cincinnati, OH
> >> +1 513 646 9390
> >>
> >> What is the problem that you are trying to solve?
> >
> >
>
>
>
> --
> Jim Holtman
> Cincinnati, OH
> +1 513 646 9390
>
> What is the problem that you are trying to solve?
>
	[[alternative HTML version deleted]]