Medical statistics

A

AMH

Hello,
I am a surgeon, that means minimal knowledge in statistics. The problem I am
analyzing involves different doses of radiation to a tumor, I would like (if
possible) to be able to calculate the cut off dose for good clinical
response and the dose above which the probability of complications
significantly increase. The data comes as numbers for the doses and as
Yes/No (0/1) for complications. The clinical response can be a 2 items or 4
items field which ever feasible to use.
We have SPSS v9 and excel 2002.
Help much appreciated

Thanks in advance
 
K

Kevin H. Stecyk

Hi Doc,

To be honest, I am not positive I completely understood your message. I
understood your data is (x,y) where x is the number of doses, and y is
binary (0,1) for no complications and complications.

I didn't understand what you meant by "The clinical response can be a 2
items or 4 items field which ever feasible to use." If you are simply
referring to how the data is formatted, you can always draw it into Excel
and then reformat it to your preference.

My *guess* is that different sized tumors or tumors in different locations
or different types of tumors require different dosages? And then perhaps
the age, condition (health), size (height and weight of the patient), and
other factors will also affect the amount of dosage. Does your current
source data capture the relevant inputs? Could you modify your source data
to capture the appropriate inputs?

If it were me, I would quickly do a literature search to see what statistics
might already be available. And then to add to my own body of knowledge, I
would capture the relevant inputs and outputs and throw it into Excel. As
I began to accumulate some data, I would begin graphing the data to see what
it looks like. You can learn a lot just by looking at graphs. Then once
you begin to formulate your ideas, you can use XL or SPSS to do more
analytical statistics.

I don't believe XL is a rigourous and advanced statistical software package.
(I am NOT a stat guru.) I am sure you can buy add-ons. But since you also
already have an advanced statistical package, you might just want to use it.
However, I think XL is very easy to use in order to manipulate data and
graph data. As you are beginning to familiarize yourself with your data
set, I think Excel is an excellent software package to use.

I am sure that others will jump in and provide advice as well.

I hope my message has provided some food for thought.

Best regards,
Kevin




AMH...
 
J

JE McGimpsey

You're scaring me, man...

Not just that a surgeon doesn't understand clinical statistics, but that
you'd think about making determinations of clinical effectiveness using
a tool that you don't understand.

I'd suggest that rather than relying on well-intentioned amateurs, you
hire an expert in medical statistics (who should tell you that XL2000 is
not the appropriate vehicle) to set up your data and analysis.
 
G

Gord Dibben

John

I concur very strongly with your assessment.

One of the most unsettling posts I have seen.

Hope there are not too many cutters like this out there.

Gord
 
A

AMH

Dear everybody
Thanks for your concern; I think the question was misunderstood. It is not
that I will take an advice and go apply it to patients, that is IMPPOSIBLE
to do, even if I wanted, it was a personal curiosity to know what the
statistician do in these situations. Remember every body I am a surgeon, we
do not give radiation that is another specialty.

Thank for your concern.
 
Top