Flawed PCR tests: "Trust The Science"
It has been one of the most enduring Covid "conspiracy" theories: that the PCR tests used to diagnose the virus were picking up people who weren't actually infected.
Some even suggested the swabs, which have been carried out more than 200 million times in the UK alone, may mistake common colds and flu for corona. We were all told to "trust the science" (and the Government!) and take two weeks off in order to flatten the curve...
If either, or both, were true, it would mean many of these cases should never have been counted in the daily tally – that the ominous and all-too-familiar figure, which was used to inform decisions on lockdowns and other pandemic measures, was an over-count.
And many of those who were 'pinged' and forced to isolate as a contact of someone who tested positive – causing a huge strain on the economy – did so unnecessarily.
Scientists willing to give credence to such concerns have been shouted down on social media, accused of being 'Covid-deniers', and even sidelined by colleagues.
But now - with all eyes on Russia - are the mainstream media quietly ready to admit that these "conspiracy theorists" have been right all along?
Last month a report by the research charity Collateral Global and academics at Oxford University concluded that the PCR tests are not fit for purpose, stating that as many as one third of all positive cases may not have been infectious.
If they are right, that's a potentially staggering number – roughly six million cases.
The Oxford scientists branded the UK's testing programme – which cost an eye-watering £2bn-a-month – as 'chaotic and wasteful'.
It is, say these critics, not simply important that we learn from our mistakes.
For while testing will now only be routine offered to patients when they come into hospitals, or in other clinical settings, and to the vulnerable, PCRs will still be used to track the spread of the virus in the community. And should there be a resurgence, that number will once again inform policy.
Nearly two years on from the first lockdown - those two weeks to flatten the curve - how sure can we be that cases weren't overstated?
Key to understanding the issue lies in how PCR tests work and the Government decisions that dictated how they were used.
PCRs detect tiny fragments of Covid genes, known as RNA, in samples taken from the nose and throat. To do this, swabs are treated in a lab with chemicals to extract the genetic material.
There is such a tiny amount of RNA on the swabs that it has to be amplified in a machine before it can be detected. This is done by repeating a cycle of heating and cooling, which encourages the genetic material to make copies of itself.
The more times the cycle is performed, the more copies are made and the more likely it is the machine will detect the virus.
This technique has been used successfully for non-Covid viruses, such as HIV and hepatitis, and in crime scene forensics when looking for DNA. It's very good at working out whether minuscule amounts of genetic code are there or not.
But when it comes to Covid, there is a problem. The very small amounts could either be from a live virus – which means someone is potentially infectious – or dead fragments left over from a previously cleared infection.
And these dead fragments can linger for up to 90 days, according to studies.
Another concern lies in how the PCR tests were performed.
At the start of the pandemic, when only NHS staff and those admitted to hospital were being tested, the hospitals and a select few laboratories run by the now-disbanded Public Health England were processing them.
In early April 2020, the Government announced that it wanted to perform 100,000 tests a day, and farmed out the work to its newly established network of Lighthouse Labs.
Dubbed 'Covid mega-labs', as each had the capacity to process upward of 50,000 swabs a day, they were run by the Department of Health and Social Care in partnership with financial firm Deloitte, bringing on board academics.
Professor Alan McNally, a University of Birmingham microbiologist who helped set up the Lighthouse Lab in Milton Keynes, said the decision meant PCR testing methods were 'largely standardised'.
The labs performed 90 per cent of PCR testing for most of 2020, with the remaining ten per cent carried out by NHS trusts on staff and patients.
But the problems came in early 2021, when testing was scaled up further and more was farmed out to the private sector.
'There appears to have been little or no oversight of these new labs, and with different PCR methods and equipment being used,' says Prof McNally.
'In Milton Keynes, every test we performed was scrutinised and checked by experts, the quality was poured over every day and we were held to account.
'Clearly in some of the newer labs, that didn't happen. Cynically, one might say it almost turned into a money-making exercise for the private sector as we had lateral flows by then and everyone knew how do them.
'Why did we need expensive PCRs? The test results basically became meaningless.'
The Collateral Global report found there were huge variations in the methods being used to conduct PCR tests at labs across the country.
It analysed more than 300 Freedom of Information requests and found there were 80 to 85 different types of testing machines in use.
Each must be used according to the manufacturer's instructions, which recommend how many cycles of amplification should be conducted before a test is considered positive.
In some cases it was as little as 25 cycles, in others as many as 45. Some experts argue this is an important distinction.
If someone tests positive at a low cycle threshold they are likely to have a lot of viral fragments present in their sample – because it doesn't have to be amplified too many times to be detected – and very likely infectious.
The reverse is also true – a high cycle threshold can mean a positive result even if very little virus is present in the original sample.
The worry is that if some machines are running more cycles, they will picking up more 'positives' than others – and that many of those won't be infectious, or 'live' cases.
Dr Tom Jefferson, who led the analysis, believes 30 cycles is a good cut-off.
However the report found about one third of positive PCR results in some labs had undergone more than that this number of cycles.
Dr Jefferson claims this means these individuals who were subsequently told they had Covid were no danger to anyone.
'Covid press conferences were all about cases, hospital admissions and making comparisons with other countries,' Dr Jefferson said.
'In reality, comparisons even between hospital trusts may be difficult because the results depends on what test you use, what machine, what chemicals.
'The reason we wanted to spend billions identifying infectious cases was to stop or delay transmission. What the Government actually did was roll out tests on an industrial scale and found huge numbers of positives – which is hardly surprising if some are being run through 45 cycles.'
So what has the impact of all this testing been? Were cases numbers really overstated? The answer is, undoubtedly, yes. But it's hard to tell exactly by how much.