A hastily written summary of why the effectiveness of tests needs to be established before it can be rolled out.
There has been a lot of criticism over
the statement by the Government that when it comes to the virus which causes Covid-19
that “No test is better than a bad test.”
Comment threads on videos where the
statement is made including numerous statements arguing that such a statement
cannot be true, is impossible, etc. And
on the surface it sounds impossible. The
problem is that a badly designed test could be ineffective in any number of
ways. Assuming only two answers – yes or
no – to be a good test it would have to be better than a random match
probability (i.e. it would have to be right more than 50% of the time). A badly designed test could either over or
under count the number of cases. In the
case of over counting – yes there are potentially fewer bad outcomes – but when
there are limited resources false negatives can burn up resources quickly
leaving less available from confirmed positive cases. False negatives could be catastrophic, not
only for the patient, who won’t get the treatment and could die, but because of
the false sense of security created by being told that the results were
negative – the person could return into society and spread the virus to family,
friends, etc without knowing. It is
safer to assume people are exposed, and limit testing to confirmatory, rather
than investigatory testing (which may have to repeated again and again after
exposure) to keep hold of some resources for testing cases where clinical decisions
depend on the results. This is in effect similar to the idea of treating every wire as if is a live electrical wire unless absolutely and safely confirmed otherwise. This is why the
tests that have been purchased by the UK government need to be validated – we need
to see how sensitive they are, how reliable they are, and how quickly and
efficiently they can be performed in practice versus in theory.
Yes, front-line workers in the NHS should have priority for testing, but
we owe it to them to make sure we get it right when it happens. It is easy to criticise decision makers for
inaction – and no doubt mistakes have been made in every country dealing with
this crisis – but we have to remember that to a certain degree there is no
perfect outcome – especially when taking clinical decisions which for some
people will be the difference between life and death.
