Hello:
I have been working on DVT risk assesment in order to see if there
are variability on this item that can help explain variations in DVT
prophylaxis usage. I use clinical vignettes that cover 8 clinical scenarios
(2 for each risk level) that were administered to 9 physicians that are
unaware of the real purpose of the study.The physicians rated each clinical
vignette in 4 risk categories (Low, moderate, high, very high)
I plan to check the agreement between the risk assesed by physicians
to a gold standard (a consensus of three internists and one hematologist
using a clinical practice guideline).
I want to know if it is possible to compare all observation between
them and with the gold standard using kappa or another statitistic that can
take into account the effect of chance agreement, on each vignette. Or I
have to compare each physician rating for each vignette with the gold
standard?
someone could suggest another analysis strategy?
Thank you in advance
|