The truth about lateral flow devices – how reliable and accurate are they? – Telegraph.co.uk
Some of the fears over reliability concern the processing of the tests, with an ongoing review finding that the rapid tests have an overall sensitivity of 76.8 per cent, but that drops to 73 per cent when used by trained healthcare staff, compared to 58 per cent when used by self-trained members of the public.
This means there is a higher chance of false negatives when the tests are used by self-trained users until they develop more experience.
A well-placed Whitehall source admitted: “They are highly effective when used daily to test infectiousness and have helped to detect thousands of asymptomatic cases that otherwise would have not have been picked up, but people shouldn’t consider a negative LFD result as a guarantee that they are Covid free”.
Sir John Bell, regius professor of medicine at the University of Oxford, believes this message must be better communicated to the general public.
“The guy in the street doesn’t know what infectiousness means and doesn’t know the difference between a LFD and PCR test, and frankly doesn’t care.
“LFDs are designed to tell you whether you are infectious for the next 12 hours: no longer than that. Someone who tests negative with a LFD cannot then say, ‘Oh right, I’ll go to a party tomorrow night’. They need to be tested again the next day. They only work if used daily.
“PCRs give a longer window of confidence. The best way to think about LFDs is that they give a red light but not a green light. The upside is, every positive test you get is free. The downside is, how many people were negative who then went off and did stuff they shouldn’t have done?”
Research carried out by Sir John and Professor Tim Peto, co-leader for the Infection Theme of the Oxford Biomedical Research Centre, suggests that the detection rate is better than Prof Deeks’s research suggests.
Agreeing that they “are not a silver bullet” and only provide “an extra layer of detection,” they insist the Liverpool pilot actually only missed 10.5 per cent of cases with high viral loads, not 30 per cent.
They suggest the false positive rate was in the region of two cases for every 1000 (0.2 per cent) while the Government puts it at 0.32 per cent. Its guidance suggests that an overall sensitivity of 76.8 per cent for the Innova test means you will have less than 15 false negatives per 10,000 people tested.