If We Want to End This Crisis, We Need to Look Beyond PCR Testing


That is according to new research from academics at the University of Colorado (Boulder) and Harvard University's School of Public Health, published in the New England Journal of Medicine. You can access the full report here.


The PCR is Just One Test

The PCR testing process is regarded as the ‘gold standard’ of COVID-19 testing by media and policymakers alike because it is the most accurate of all the available tests. The PCR process is the best at telling us who definitely has the virus and who definitely doesn’t. Another way to phrase this is that the PCR minimises the probability of getting either a false positive or a false negative.

In the practical reality of outbreak management however, the PCR test is just one of a suite of tests, and each has its own strengths and weaknesses. The PCR may have the highest sensitivity, but that does not mean that it is the only test that we should consider. There are other factors that should influence our decision-making, as the authors of this paper have stated:


The key question is not how well molecules can be detected in a single sample but how effectively infections can be detected in a population by the repeated use of a given test as part of an overall testing strategy — the sensitivity of the testing regimen.

The authors are telling us that the media and the policymakers have confused the sensitivity of the individual test, with the sensitivity of the overall testing regimen. The PCR test may be the single best test, but that doesn’t mean that it it the best test in all situations or that it should be the only test that we use. What really matters to us is the performance and the sensitivity of the whole testing regimen, and at a national level, a regimen that is 100% PCR tests will underperform a regimen that has a more diverse and flexible testing toolkit.


Test Sensitivity vs Regimen Sensitivity

Consider a test that can be repeated daily with a sensitivity of 60% (where the probability of a false negative is 40%), versus a test with a sensitivity of 95% but that takes several days to return a result. When we repeat the measure daily, a 60% accurate test becomes 84% accurate on the second day, and 94% accurate on the third day.

NB: Accuracy after N tests = 1 – [prob(false negative)^N]


Figure 1: Sensitivity of Repeated Rapid Testing Regimens vs PCR


A single rapid test may have a lower sensitivity than a single PCR test, but if we can repeat the measure over several days, those rankings switch, so the repeated rapid testing regimen can reach a critical confidence level before the PCR test can return a result. Moreover, every rapid test result gives us additional information, so our confidence level rises each day, whereas with the PCR test we are left in the dark until the result is returned.

The CDC estimated that by June 2020, infections in the USA had been underestimated by about a factor of 10. The HSPC's SCOPI Study found a similar, though smaller effect in Ireland. The delay both in getting tested (due to supply constraints) and then in receiving the results, means that most infected individuals didn't get tested, and that many of those who did only found out they were positive after their period of infectiousness has passed.


In other words, despite very high analytic sensitivity of the diagnostic tests deployed for surveillance, today’s testing regimens have at best only 10% sensitivity to detect infections and are failing as Covid filters.

The authors illustrated the trade off between the sensitivity of the test and the frequency of testing in Figure 2 below.


Figure 2: High Frequency & Low Sensitivity vs Low Frequency & High Sensitivity Regimens

Explanation: An individual’s infection trajectory (blue line) is shown in the context of two testing regimens (circles are tests performed) with different sensitivities. The low sensitivity test is administered frequently and the high sensitivity test infrequently. Both tests detect the infection (orange circles), but only the low-sensitivity / high-frequency test detects it during the transmission window, which makes it a more effective filter. The window during which the PCR test detects infections before infectivity (green) is short, whereas the corresponding post-infectious but detectable window (purple) is long. This problem is averted by the rapid testing regimen.


Weaknesses of PCR Testing

The length of time between first symptoms to receiving a result is a major weakness of the PCR testing process. It slows the infection discovery process, it allows opportunities for the virus to transmit in the meantime, and it creates a lag in our PCR case data which impedes our policymaking. The PCR testing procedure is also exposed to more risks, as specimens must be collected, transported, and stored appropriately at a virology lab. The lab then needs enough staff, reagents, and equipment to run the tests accurately. Every part of every stage of this process represents a potential hazard that can slow the detection of the virus. Worse still, all this activity costs money and inevitably, PCR testing is far more expensive than repeated rapid testing.


Strengths of Rapid Testing

A rapid test can be administered on the spot and it can produce a result within an hour, giving the policymakers real time data about transmission, and giving the individual the information they need to get on with their life. Repeated rapid testing is more accurate than PCR tests at the population level because, as shown above, the number of rapid tests performed and the speed at which they are performed can compensate for the lower sensitivity of the individual test. Not only is repeated rapid testing a cheaper and more accurate testing regimen, it is also more flexible. Rapid tests can be administered at home in the morning before work or school, they can be used to screen individuals at entrances to public buildings (e.g. care facilities, shopping centres, or restaurants) or they can be used as part of an elimination protocol. Finally, the repeated rapid testing approach scales better, meaning it can detect the virus in the population in less time and at a lower cost than the PCR test.


Rapid Testing Makes Elimination Work

If a repeated rapid testing regimen was combined with a strict lockdown, safe and effective quarantine facilities, and a county-level ‘Green Zone’ protocol, the virus could be quickly detected and comprehensively isolated. If such a protocol was implemented nationally, most of the country could clear the virus within 8 weeks, leaving the nation’s surplus resources to be focused on the few remaining hotspots.

Elimination really is that simple. Who’s going to tell NPHET?