Intended for healthcare professionals

Letters Lateral flow tests

Lateral flow tests need low false positives for antibodies and low false negatives for virus

BMJ 2021; 372 doi: https://doi.org/10.1136/bmj.n90 (Published 13 January 2021) Cite this as: BMJ 2021;372:n90
  1. Keith A Moyse, independent scientist
  1. Oxford, UK
  1. cowley2013-bmj{at}yahoo.com

The relative importance of the sensitivity and the specificity of lateral flow devices (LFDs) depends on the intended use.

LFDs that are being piloted for the detection of antigens need to have high sensitivity in order to detect the virus in those who should isolate. A false positive, on the other hand, merely risks someone self-isolating unnecessarily.

The reverse is true for LFDs used to detect the presence in the blood of antibodies against the virus. Such antibodies might give some level of protective immunity. In this case, having few false positives is crucial, as a false positive test might give the illusion of protective immunity where none exists. Low sensitivity of devices, on the other hand, would merely lead to erring on the side of caution.

Despite the above, evaluators of LFDs seem to have come to the opposite conclusions. Some LFDs designed to detect antibodies seem to have been rejected for home use by the National Covid Testing Scientific Advisory Panel, largely based on sensitivity rather than false positives.1

LFDs for use in the detection of active virus (discussed here23) have been evaluated to have high specificity but only modest sensitivity, 58% in the hands of self-trained users. This would seem to be the opposite of what is required. Nevertheless, these devices are described by evaluators as having the “desired performance characteristics.”4 Low sensitivity would mean that many cases would be missed.

If we are looking for protective antiviral antibodies, we don’t want false positives, but some false negatives could perhaps be tolerated. If we are looking to detect an active infection, we don’t want false negatives, but some false positives might be tolerated.

Some evaluators of LFDs seem to have the importance of the sensitivity and specificity requirements precisely the opposite way round.

Footnotes

  • Competing interests: None declared.

References

View Abstract