I am a poor unhappy soul who has gotten bitten by the bug of looking
for decent methods for counting protozoan blood parasites (in my case
_Babesia canis_ . I'm planning at some point to try bisbenzamide and a
fluorescence-actived cell sorter, if I can find one. Currently I am
working on a very simple manual method.
My problem for both is the same. I can prove that the methods are
repeatable. But, having no standard method which I know is accurate,
how do I prove that they are accurate - since the term is defined as
deviating very little from the true value, and I won't know what the
true value is? I don't know that comparing with existing methods will
help, as (in my humble opinion) they really aren't very good for
quantitation, at least not for me, and all I will be able to show is
that the newer methods show less variation, and are better methods.
But still the accuracy thing eludes me.
Given that all tests were once new, and had nothing to be compared
with, I would have thought any clinical chemistry book or similar
would be bursting with guidelines for this kind of problem. No such
luck.
Presumably some of you have had similar experiences with flow
cytometry? Or did you just do what people in the parasite field seem
to do - count the sample ten times and find the method is repeatable,
and say, "Here is the method."
Any ideas?
Linda