Re: Covid-19 mass testing programmes - do we have a suitable test?
Raffle et al are correct to compare mass testing for Covid-19 with screening, rather than to call it a form of screening. Just as with screening, a number of criteria have to be met.
There are real problems with the tests available for Covid-19.
We don’t know enough about immunity to understand what the easily available antibody tests mean. There is insufficient evidence regarding what the antibody tests mean, and the duration of immunity. Tests for T-cell responses to the virus may tell us more; but at present they are difficult (and expensive) to undertake, so they are mainly confined to use for research purposes.
Viral culture – the gold standard for confirming the presence of viable, infectious virus – is difficult, expensive, and will never be widely available.
The test that is widely available uses reverse transcriptase polymerase chain reaction (Rt-PCR, or just PCR for short). This looks for sequence of nucleic acids – for viruses such as Coronaviruses and influenza, for RNA; for other organisms such as shigella toxin producing E coli O157, for DNA. The technology is exquisitely sensitive – able to detect very small quantities of genetic material; and highly specific, as long as there is no cross-contamination.
Sadly, however, it is only as good as the sample; and in the real world, roughly 30% of samples from people who are ill with Covid-19 give (false-) negative results (test sensitivity is about 70%). It will be much lower in people who are asymptomatic sensitivity; and the tests will not detect people in the pre-infectious stage of their incubation period. It is, therefore, unwise to rely on a test to “rule out” infection; and tests will miss a significant proportion of people who subsequently become infectious if used eg to screen returning travellers.
The other aspect of accuracy is the test’s specificity. Their specificity for detection of viral RNA is likely to be very high indeed. But PCR tests do not identify live, viable, infectious viruses. Just viral RNA. And this RNA is durable. It can stay in the body for weeks – possibly months – after all viruses have been killed by the immune system, giving “positive” results in people who are no longer infectious.
Now that we are routinely testing many more asymptomatic people – health and social care workers, and care home residents – as well as people with symptoms that could be caused by Covid-19 – we will detect more people with residual RNA rather than ongoing infection. True positive test results for viral RNA; but false positives for infectiousness.
We do not yet know how unlikely it is that somebody will be ill or potentially infectious to others with Covid-19 within any given time frame. Despite a recent report, reinfections seem to be uncommon as yet – although, if there is a limited duration of immunity, we can expect more cases to be reported. (It is worth noting that, in this report, the patient was asymptomatic, indicating they may have been protected against illness, albeit without sterilising immunity.)
When somebody has tested positive previously (and at some time been infected), we have no simple and reliable way of distinguishing a repeat infection from non-viable, leftover RNA, although it is likely that somebody who actually has current infection – and who may be infectious to others – will have a much higher quantity of viral RNA on any swabs.
PCR tests work by repeatedly doubling the amount of genetic material in the original sample, until there is a detectable quantity of it. Each doubling is referred to as a “cycle”; and the number of cycles or doublings before there is a detectable quantity of genetic material is called the “cycle threshold” (CT or Ct). The larger the amount of viral RNA there is in the sample, the smaller the number of cycles that are required before it can be detected. And since the number of cycles is the CT value, the lower the CT value, the more virus there was in the original sample, and the more likely it is thought to be that the case was actually infectious, rather than still carrying leftover RNA, which is not clinically significant.
Sadly, not all forms of PCR testing provide a measurable cycle threshold; and the privatised labs used for “pillar 2” testing in England cannot or will not routinely share CT values, making it much harder to assess the likelihood that an positive result represents infection is a repeat infection.
This matters, because the consequences of being deemed infectious are not trivial. If you test healthcare workers, and decide they are potentially infectious, they have to self-isolate for 10 days; and their household and other significant contacts for 14 days.
This can have a very serious impact on the workforce, with significant economic and personal implications. And, in care homes, it can again be very disruptive if residents are deemed to have re-infection.
We need greater clarity about the purposes of and strategy for testing.
And we need to be better able to make decisions about when we can dismiss “positive” results as being due to leftover RNA, rather than current infection.
1. Raffle AE, Pollock AM, Harding-Edgar L. Covid-19 mass testing programmes. BMJ 2020;370:m3262, DOI: 10.1136/bmj.m3262 (https://www.bmj.com/content/370/bmj.m3262 or https://www.bmj.com/content/bmj/370/bmj.m3262.full.pdf).
2. Wilson J, Jungner G. Principles and practice of screening for disease. WHO Chronicle 1968;22(11):473 (http://whqlibdoc.who.int/php/WHO_PHP_34.pdf).
3. Parry J. Covid-19: Hong Kong scientists report first confirmed case of reinfection. BMJ 2020;370:m3340, DOI: 10.1136/bmj.m3340 (https://www.bmj.com/content/370/bmj.m3340 or https://www.bmj.com/content/bmj/370/bmj.m3340.full.pdf).
4. Singanayagam A, Patel M, Charlett A, Lopez Bernal J, Saliba V, Ellis J, et al. Duration of infectiousness and correlation with RT-PCR cycle threshold values in cases of COVID-19, England, January to May 2020. Euro Surv 2020;25(32):2001483, DOI: 10.2807/1560-7917.ES.2020.25.32.2001483 (https://www.eurosurveillance.org/content/10.2807/1560-7917.ES.2020.25.32...).
Competing interests: No competing interests