Covid-19: 'Vast majority' of positive rapid Covid tests are likely to be false, scientist claims - Money

The ‘vast majority’ of positive rapid Covid tests at school are likely to be false, Professor Jon Deeks,

Mass testing of secondary school pupils is costing £120,000 for every case it detects, it emerged today.

Data from March 11 to 18 shows 1,805 positive results from 3.8million tests taken. Given the current best-guess false-positive rate, only 645 of those are likely to be genuine infections.

‘At £20-a-test, that’s £120,000 per case found,’ Professor John Deeks from the University of Birmingham told The Daily Telegraph.

He told the newspaper the ‘vast majority’ of positive results from rapid coronavirus tests used at secondary schools are likely to be wrong.

Cases have fallen rapidly among children since the start of the year, with just 0.05 per cent of youngsters thought to infected at the moment — just one in 2,000.

Professor Deeks, head of the university’s Biostatistics, Evidence Synthesis and Test Evaluation Research Group, warned it becomes harder to distinguish the true size of outbreaks using lateral flow tests when cases plummet to such low levels.

Fears false positives will spike as cases drop has prompted the Government to U-turn on its lateral flow-testing policy.

The Department for Education updated its guidance yesterday to say all positive rapid test results taken at school have to be followed up with a lab-based PCR test.

If the PCR test is negative, students and staff will now be able to return to class and stop self-isolating, the advice says.

Previously only positive rapid tests taken by schoolchildren at home required a backup PCR test under the guidance.

All students in secondary schools in England were expected to complete three rapid Covid tests on their return to classrooms on March 8, spaced three to five days apart. Since then all tests have been taken at home twice a week.

Professor Deeks, who last week authored a report showing rapid tests may miss four in 10 asymptomatic people, told The Daily Telegraph he was ‘bewildered’ it took the Government this long to change its position.

But he added it was ‘better late than never’ that the number of children having to having to unnecessarily stay at home in isolation will fall.

Just 1,805 of the 3.8million lateral flow tests taken last week were positive, the figures showed. Pictured: A student takes a lateral flow test at Weaverham High School in Cheshire

Official data shows the test positivity rate among secondary pupils is around one in 2,000. Just 1,805 of the 3.8million lateral flow tests taken last week were positive.

Health bosses say around one in 1,000 positive results from rapid kits are false. This would be the equivalent of around 3,800 positives from 3.8million tests.

Experts say this is proof that most of the positive results would have been incorrect because when case numbers are low the likelihood of false positives increases.

But the ONS — which carries out gold standard studies — says prevalence of Covid among secondary school children in England is around 0.4 per cent.

This would suggest that the tests may in fact be missing lots of actual cases in schools.

It comes as Government data showed the proportion of pupils attending schools in England last week dropped as more children were forced to self-isolate at home.

Nine in 10 state school pupils were in class on March 25, down from 91 per cent on March 18 when all schools in England had fully returned, the Department for Education analysis shows.

Coronavirus cases are dropping in all age groups other than children and teenagers in the United Kingdom since schools reopened in England on March 8, dismissing fears cases would balloon in the wider population once children returned to classrooms

Rapid tests — which give results in as little as 30 minutes — are not genetically sequenced so they cannot be used to identify variants of concern, including the South Africa variant

Dr Susan Hopkins, strategic response director for Covid at Public Health England, said the lateral flow tests have an ‘extremely low’ number of false positives.

She told The Telegraph: ‘However, as the prevalence of Covid reduces to low levels in the population, the chance of a false positive result increases – though still remains unlikely.

‘During periods of low virus prevalence, we can mitigate against the risk of false positives by confirming a positive lateral flow test with another test — such as a PCR test performed in the laboratory.

‘This makes it highly unlikely that anyone would need to self-isolate unnecessarily, while ensuring that we can retain the benefits of rapid testing to detect true positive cases and for confirming when people are not infected.’

It comes as Department for Education data yesterday suggested that 220,000 pupils were out of class and self-isolating on Thursday last week due to potential contact with Covid, up from 169,000 pupils in the previous week.

Around 3.3 per cent of all pupils on roll — up to 264,000 children — did not attend school for Covid-related reasons on March 25, up from 2.5 per cent on March 18.

The DfE said the main reason for absence among pupils who did not attend school for Covid-related reasons was self-isolation due to contact with a possible case inside the school.

Geoff Barton, general secretary of the Association of School and College Leaders (ASCL), said: ‘Although there was a slight dip in attendance last week, this was always likely to happen as Covid positive cases emerged and triggered Covid protocols necessitating the self-isolation of close contacts.

‘On the whole, however, attendance levels have borne up remarkably well, and great credit is due to schools and their staff for all they have done.

‘There is now the natural firebreak of the Easter holidays and we will see where we are at the beginning of next term.

‘Everybody will obviously hope that the relaxation of Covid restrictions in wider society will not lead to an escalating rate of positive cases in schools, with the disruption this causes to education.’

A DfE spokesperson said: ‘Attendance in schools remains higher than at any point during the autumn term, as students and staff continue to follow the protective measures set out in our guidance to reduce transmission of the virus.

‘With the testing programme now in full swing and millions of tests being conducted each week, we are only seeing a small increase in the numbers of students testing positive and self-isolating.

‘We are grateful to everyone who continues testing twice-weekly from home and self-isolating where that is necessary, playing their part in keeping everyone as safe as possible.’

HOW LATERAL FLOW TESTS ARE ONLY TRUSTWORTHY WHEN ADMINISTERED BY TRAINED STAFF

Lateral flow tests are only accurate at diagnosing coronavirus when administered by trained professionals, studies have repeatedly shown.

The tests, which give results in as little as 15 minutes, use swabs of the nose or throat. Samples are then mixed in a testing liquid and put into a plastic cassette which can detect the presence or absence of coronavirus and then produce an image of a line, the same way as a pregnancy test, to indicate whether it is positive or negative.

The Department of Health and NHS are instructing people to use the tests on themselves, despite manufacturers of some kits saying they shouldn’t be used as DIY swabs.

Both the swabbing procedure and the use of the test cassette can easily be done wrong and affect the accuracy of the test.

If the swab isn’t done for long enough, or deep enough into the nose or throat, it may not pick up fragments of virus. Medical professionals are also able to use nasopharyngeal swabs, which go right to the back of the nostril, whereas this is not advised for people who test themselves.

And if the sample isn’t properly inserted into the cassette the result might be wrong, or people may misread the display when it produces a result.

SELF-TESTING CUT ACCURACY FROM 79% TO 58%

A University of Oxford and Public Health England evaluation of the Innova lateral flow test, which is being widely used in the UK, found its sensitivity – the proportion of positive cases it detected – fell from 79 per cent to 58 per cent when it was used by untrained members of the public instead of lab experts.

Based on this evaluation, officials pushed ahead and used it for a real-world self-testing trial.

PILOT IN LIVERPOOL FOUND FEWER THAN HALF OF POSITIVES

When the same Innova test was trialled on members of the public in Liverpool – with people taking their own swabs and trained military staff operating the tests – the swabs picked up just 41 per cent of positive cases.

In the study the rapid tests detected 891 positive results, compared to lab-based PCR swabs that found 2,829 positives in the same group. This means 1,938 people got a wrong negative result from the rapid test.

The study didn’t compare this to professionally done rapid tests, but the manufacturer Innova claims its test is 95 per cent sensitive in lab conditions.

…BUT TESTING DONE BY MEDICS IN SLOVAKIA ‘REDUCED INFECTIONS’

Despite rapid lateral flow tests getting bad press, officials in Slovakia used them on 5.2million people – almost the entire population of 5.5m – in a trial that a study later estimated to have cut the country’s infection rate by 60 per cent.

The tests used were between 70 and 90 per cent accurate and all the swabs and evaluations were carried out by trained medical workers. They used deep nasopharyngeal swabs, that go to the back of the nose, whereas self-testing generally relies on a swab of only the nostril.

London School of Hygiene & Tropical Medicine researchers said that the scheme successfully weeded out coronavirus cases that wouldn’t have been found otherwise, slashing the number of cases by over half in a week during a lockdown.

HOW RAPID TESTS ARE DIFFERENT TO LAB-BASED PCR SWABS

Lateral flow tests are an alternative to the gold standard PCR test – known scientifically as polymerase chain reaction testing – which is more expensive and more labour-intensive but more accurate.

PCR tests also use a swab but this is then processed using high-tech laboratory equipment to analyse the genetic sequence of the sample to see if any of it matches the genes of coronavirus.

This is a much more long-winded and expensive process, involving multiple types of trained staff, and the analysis process can take hours, with the whole process from swab to someone receiving their result taking days.

It is significantly more accurate, however. In ideal conditions the tests are almost 100 per cent accurate at spotting the virus, although this may be more like 70 per cent in the real world.