We need Covid-19 mass-testing. But who will trust the government to deliver it?
As summer turns to autumn, Covid-19 cases have been increasing in the UK, and England has registered an increase in hospitalisations. The virus has already shown it can cause havoc for the NHS while infecting a relatively small proportion of the population.
Britain now has two options to transform its fight against Covid: we wait for a vaccine, and/or we radically ramp up virus testing. Chris Whitty, the government’s chief medical adviser, has warned that we may not have a vaccine before winter next year. Which leaves us with testing. For this to be fully effective, we need mass testing across large segments of the entire population.
In early April, more than 30 scientists, led by Julian Peto, a professor at the London School of Hygiene and Tropical Medicine, proposed that frequent testing of the entire population would be an effective way to rapidly identify virus carriers, who could then self-isolate. These tests may be the key to keeping schools, universities, businesses and restaurants open, where a person’s admittance could be conditional on proving a recent negative test. They might be the only way, in the medium term, that we’re able to regain some normality.
Leaked documents suggest the government is prepared to commit £100bn – approaching the entire yearly budget for NHS England – to a “moonshot” mass testing programme. Sage has provided a qualified endorsement of the strategy, if not the price tag.
For this to be effective, available tests must be evaluated transparently. The government has previously bought large supplies of Covid tests with no publicly available data on whether they worked. It’s also important that adequate income support is available for those who test positive and are asked to self-isolate, and that the public understands the differences between testing entire populations for infection and diagnosing people who may be sick with Covid.
Many rapid tests have been developed that are less accurate than the gold-standard tests used in the NHS. But if everyone could self-test, or be quickly tested on portable machines twice a week, these less accurate tests could still be an important part of controlling coronavirus. The Scottish government has purchased 300 machines capable of rapid testing. One large US company claims that its test costs only £4 and can return a result in 15 minutes; the US government has or
dered 150m of these.
Regulators usually prioritise the accuracy of a test, which is exactly what they should do in normal times. But during a pandemic, less accurate tests could be used much more frequently to quickly identify many more people who could spread the virus. A test result that is available within minutes is far more valuable than one that takes at least a day.
An accurate Covid test will have as few false positives (showing a positive result in someone who isn’t infected with Covid) and as few false negatives (showing a negative result in a person who does have Covid) as possible. Having few false negatives is more important when you’re trying to diagnose sick patients who need Covid treatments. But this doesn’t matter quite so much when you’re screening an entire population, when the tradeoff for using a less accurate test is that many more people can be screened and thus more positive people will be identified.
False positive tests may also result in unnecessary self-isolation until a second test result is available. Imagine that a Covid test generates one false positive reading for every thousand people. If it’s used on a population of a million, at least one thousand people will falsely test positive, in addition to the proportion of true cases the test identifies. The fewer true cases of Covid that are present in that population, the higher the proportion of positive test results will be people who don’t actually have Covid. So as we drive the proportion of people who are truly infected with Covid down, the likelihood of a test being a false positive goes up.
Without making testing compulsory, the government will need to persuade people that despite the risks and inconveniences of having a false positive test, testing is still a vital component of driving infection rates down as low as possible. Frequent testing would need to become part of our daily lives, much like washing our hands or covering a cough. In addition to full transparency about test performance we’d need rigorous, continuing quality control to prevent bad batches of tests giving false reassurance.
Yet we are not off to a good start. The government announced it had handed contracts to two UK companies to provide rapid laboratory-based virus testing in hospitals, but the details of the tests, their accuracy, and their costs have not been disclosed. University scientists evaluating the tests are told by the government they cannot discuss their work. After many months, the over-centralised, privately-run testand-trace system is referring people with symptoms to test centres more than 100 miles away. No-bid contracts have resulted in the withdrawal of defective tests. The leaked documents suggest that, far from learning from these mistakes, the government will ask the same private companies that failed to deliver a successful test-andtrace programme to set up this masstesting scheme.
Having everyone in the country test themselves twice a week would be undoubtedly expensive. But the costs will still be cheap relative to a second national lockdown. Still, a huge amount of work is needed to convince the public this is not just another prime ministerial rhetorical flourish, along with our “world-beating test-andtrace system”, and “return to normality … possibly in time for Christmas”. With the stakes so high, this time the government needs to not only get it right in practice, but to be seen to be getting it right by the public.
• David Hunter is the Richard Doll professor of epidemiology and medicine, University of Oxford