Prev: Vintage analysis
Next: Drop down to R implementation of Proc Means in the Data Step
From: RolandRB on 10 Apr 2010 03:15
As you probably know, I have webbed an entire clinical reporting
system written in SAS and bash shell scripts. I even have bash shell
scripts that run SAS. I have webbed a graphical patient profiler as
well. I won it all. There is no problem with it belonging to a
previous employer. I have a contract to prove I took over the clinical
reporting system and the patient profiler I designed myself. All told,
there is about 2000 hours of work I have put into this. You can find
it at the following URL and follow the links to download it. There I
have supplied the entire source code for the clinical reporting system
and I have a working version of the patient profiler that is ready to
use (I am retaining the core code for this).

http://www.datasavantconsulting.com/roland/

It is all FREE to those who wish to use it for free. I do not want any
money from you unless you want to enter a support contract with me.
It has had hundreds of downloads but I rarely hear from people who use
it, perhaps because people think I will charge money in the future for
it. This is not the case. You do me a favor by making its use as
widespread as possible because by doing this it becomes "standard
software" in the academic community which means CROs and pharmas will
eventually use it and require a maintenance contract with me. But if
you use it it would be very helpful if I knew it and you sent me bug
reports and requests for clarification in the documentation so that I
can help it to be a better product. If there are enough users then
perhaps somebody could set up a mailing list to help other users.
Sometimes I might add a post to clarify some issues.





From: xlr82sas on 10 Apr 2010 18:54
On Apr 10, 12:15 am, RolandRB <rolandbe...(a)hotmail.com> wrote:
> As you probably know, I have webbed an entire clinical reporting
> system written in SAS and bash shell scripts. I even have bash shell
> scripts that run SAS. I have webbed a graphical patient profiler as
> well. I won it all. There is no problem with it belonging to a
> previous employer. I have a contract to prove I took over the clinical
> reporting system and the patient profiler I designed myself. All told,
> there is about 2000 hours of work I have put into this. You can find
> it at the following URL and follow the links to download it. There I
> have supplied the entire source code for the clinical reporting system
> and I have a working version of the patient profiler that is ready to
> use (I am retaining the core code for this).
>
> http://www.datasavantconsulting.com/roland/
>
> It is all FREE to those who wish to use it for free. I do not want any
> money from you unless you want to enter a support contract with me.
> It has had hundreds of downloads but I rarely hear from people who use
> it, perhaps because people think I will charge money in the future for
> it. This is not the case. You do me a favor by making its use as
> widespread as possible because by doing this it becomes "standard
> software" in the academic community which means CROs and pharmas will
> eventually use it and require a maintenance contract with me. But if
> you use it it would be very helpful if I knew it and you sent me bug
> reports and requests for clarification in the documentation so that I
> can help it to be a better product. If there are enough users then
> perhaps somebody could set up a mailing list to help other users.
> Sometimes I might add a post to clarify some issues.

Hi Roland,

Thanks for contributing this work to the SAS community.

I think you system is great for canned reports, which most pharma
companies feel is the holy grail.

I have used many of the smaller components of your system, ie like
macro optlength(Tthanks). I have also learned new techniques based on
your unix shells and you nice short tutorials. Some of the more
complex macros like unistats are a little to intimidating for me. I
also worry a little about macros whose argument list exceed the lines
of code needed for simple reports. I also worry a little about
improvements to ODS, ie like tagattr which could require substantial
changes to macros like unistats. We also need to at least keep our
eyes on LaTeX, perl and R.

Thanks Again

From: RolandRB on 11 Apr 2010 02:57
On Apr 11, 12:54 am, xlr82sas <xlr82...(a)aol.com> wrote:
> On Apr 10, 12:15 am, RolandRB <rolandbe...(a)hotmail.com> wrote:
>
>
>
>
>
> > As you probably know, I have webbed an entire clinical reporting
> > system written in SAS and bash shell scripts. I even have bash shell
> > scripts that run SAS. I have webbed a graphical patient profiler as
> > well. I won it all. There is no problem with it belonging to a
> > previous employer. I have a contract to prove I took over the clinical
> > reporting system and the patient profiler I designed myself. All told,
> > there is about 2000 hours of work I have put into this. You can find
> > it at the following URL and follow the links to download it. There I
> > have supplied the entire source code for the clinical reporting system
> > and I have a working version of the patient profiler that is ready to
> > use (I am retaining the core code for this).
>
> >http://www.datasavantconsulting.com/roland/
>
> > It is all FREE to those who wish to use it for free. I do not want any
> > money from you unless you want to enter a support contract with me.
> > It has had hundreds of downloads but I rarely hear from people who use
> > it, perhaps because people think I will charge money in the future for
> > it. This is not the case. You do me a favor by making its use as
> > widespread as possible because by doing this it becomes "standard
> > software" in the academic community which means CROs and pharmas will
> > eventually use it and require a maintenance contract with me. But if
> > you use it it would be very helpful if I knew it and you sent me bug
> > reports and requests for clarification in the documentation so that I
> > can help it to be a better product. If there are enough users then
> > perhaps somebody could set up a mailing list to help other users.
> > Sometimes I might add a post to clarify some issues.
>
> Hi Roland,
>
>   Thanks for contributing this work to the SAS community.
>
>   I think you system is great for canned reports, which most pharma
> companies feel is the holy grail.
>
>   I have used many of the smaller components of your system, ie like
> macro optlength(Tthanks). I have also learned new techniques based on
> your unix shells and you nice short tutorials.  Some of the more
> complex macros like unistats are a little to intimidating for me. I
> also worry a little about macros whose argument list exceed the lines
> of code needed for simple reports. I also worry a little about
> improvements to ODS, ie like tagattr which could require substantial
> changes to macros like unistats. We also need to at least keep our
> eyes on LaTeX, perl and R.
>
> Thanks Again

ODS is my major problem with the reporting macros. The requirement
from the CRO is that they can simultaneously output ascii reports with
RTF reports and I have achieved this plus the ability for it to
produce PDF reports. It has to do this by making multiple calls to
proc report. RTF output does not honor the split character in the data
cells, whereas ascii output does. Some of the column descriptors are
good for RTF but not for ascii. RTF column widths are easy to control
using outputwidthpct= but it does not work the same for PDF output so
a different call to proc needs to be made which avoids setting the
width for the narrower columns. It now works well enough for producing
canned reports, however.
 | 
Pages: 1
Prev: Vintage analysis
Next: Drop down to R implementation of Proc Means in the Data Step