>Greetings,
>I have a quick query regarding handling assay results which are below the
>assay's limit of detection. For statistical purposes a value must be
>assigned to such samples. Does anyone have any suggestions on which is the
>most appropriate value to use. Those I most commonly see used are either
>the limit of detection itself, or 50 percent of the limit of detection
>(presumably assuming that of all such samples half will have a 'true' value
>between zero and 50 % LOD, and the other half between 50-100% of the LOD)
>
>Suggestions and comments all welcome.
>
>Thanks,
>
>RMD
>
>Richard M Dixon BVMS CertVR MRCVS
>University of Glasgow, Dept Vet Clinical Studies,
>Bearsden Road, Glasgow G61 1QH, Scotland.
>Tel ++ (0)141 330 5730 Fax ++ (0)141 330 6996
Off the top of my head:
1. Your final presumption is invalid.
2. Depending on the circumstances, one should either report the measured
value per se (preferably plus/ minus the SD) or state that the value lies
below the lower limit of detection, stating how this defined.
3.The measured value may be negative (50% of measurements of zero dose will
fall below zero assuming a normal distribution). In certain statistical
analyses it is important that negative results be reported as such even
though a negative mass or concentration is impossible. Reporting a negative
result (or a result less than the detection limit) as zero has various
biasing effects which I won't discuss in detail here. However a possible
problem is that calculation of a <0 result entails extrapolating the
response curve into a region where one has no calibrants; hence negative
dose estimates can be unreliable.
4. I have passed your message to my colleague Phil Edwards who will enlarge
on some of these points.
Roger Ekins
Molecular Endocrinology
UCL Medical School
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|