3 replies to this topic

[srimadh]

Dear All,

I am doing a retrospective study on Rheumatoid arthritis ... say screening the databases for specific tests like RA factor, CRP and ESR, etc. But, to compare with the conrols ... what should be perfect control for the RA study?

Can it be a sample without the symptoms ( according to ACR ) or can it be any other disease other than RA in the whole gamut of rheumatic diseases. I know that the control should be age matched, gender matched, and also ethnicity matched. But, which group of the patients or people can be taken as controls?


Please help me with this ...

Thank you.

With Regards
Srimadh

[bob1]

Preferably age etc matched people without rheumatoid symptoms.

[srimadh]

Mr. Bob,

You mean to say I can take anything that which is not Rheumatoid in symptoms.

A person with just a normal fever (not rheumatic) or

A person with any other simple ailment ....

But for the comparison ( with controls ) the same parameter (say RA factor or CRP) should be done even in the non rheumatoid person ( healthy ) ?

[bob1]

Technically, so long as they don't have rheumatoid symptoms, then they should be OK as a control... this is the problem with doing retrospective or meta studies, is that the data is designed to get out the information you want, so sometimes you have to make compromises.

You should really really talk this over with your supervisor - they are the person who (should) know what they want and how to get it.  You may also need to talk to a statistician and maybe an epidemiologist.