And in a second message to Chris:
>I do not think we received accurate info from the lab on the number of
>false positives.  We were only talking to a rep.  From the no.  of replies
>to my post and their content I believe he meant to say 99.8% accurate,
>does this seem about right
 
99.8% seems much more realistic.  Even that's a very high rate - in some of
the human diagnostic assays I've done in the past I reckon up to 5% were
false positives.
 
>to you.  I did have a false negative 6 years ago, he had tested neg 2
>years running before turning positive and I am shore he had no contact
>with positive ferrets.  Chris.
 
False negatives usually result from one of two reasons.
 
1) sometimes the virus version of the capturing antigen in the assay may
not have elicited an immune response, even if all the other different
antigens on the virus did so.  In this case, even though the animal/person
has a strong immune response, that particular version of the assay will not
pick it up.
 
Also, if it is early in the course of infection, the antibodies may not
have had time to develop to detectable levels.
 
2) Sometimes the overall immune response is weak for quite a while after
infection, or even absent.  If the infecting microorganism is not
particularly virulent in these stages the patient can live on quite
happily with no antibodies to indicate their infected status.
 
There are other reasons for false negatives.  One that comes to mind
occurs when the virus-infected cells shed soluble forms of the capturing
antigen/protein, or even whole viruses, which bind to the antibodies (after
all, this is the job of the antibodies).  This means that the binding sites
on the antibodies are already occupied, and thus the capturing antigen in
the assay is blocked from binding to the antibodies.  Result; an apparent
absence of antibodies in the test sample, even though they're really there.
Theoretically, it should be possible to break up these antigen/antibody
complexes before completeing the rest of the assay, but few if any
commercial tests I've seen actually do this.
 
I suppose the bottom line to all this is that the vagaries of the immune
system are such that a positive result does not necessarily mean that the
disease organism is (or was) present, and a negative result does not mean
that it isn't (or wasn't).  Caveat emptor!
 
Harko.
 
Harko Werkman,
Frog Research Group,
Department of Biological Sciences,
University of Newcastle,
Callaghan,
NSW. 2308.
Australia.
Ph: (+612) 4921 6253
Fax: (+612) 4921 6923
Email: [log in to unmask]
[Posted in FML issue 2954]