Testing 101: How Mass Rapid Testing Works, Part 2


Previously I discussed rapid antigen testing and the role it could play in our national pandemic response.

I presented a scenario in which serial testing schemes were being run in schools and workplaces, entrance testing was taking place at some public venues, and free rapid tests were available in pharmacies and other drop-in facilities. I argued that easy access to inexpensive testing could reduce transmission across the country, allowing us to put 18 months of restrictions behind us.

The key idea was that the simplicity of the test technology would make it easier for us to integrate testing into our normal day to day activities, thereby reducing the burden of the measure. With no need for appointments and no time lost to queues, we could get COVID-19 updates on demand, giving us clarity over our health status and the confidence needed to get on with our lives.

 

The analysis looked at the testing system through the eyes of the people using it, and in the context of the locations and venues in which the testing was taking place. In this sense, it was a ‘bottom-up’ perspective on testing.

But viruses are contagious, and we can only learn so much by studying behaviours and effects in isolation. To fully appreciate the strengths of a national testing program (as well as the weaknesses), we must zoom all the way out to see how the individual components perform in aggregate, at the macro level – from the ‘top down’.

And this is where the mainstream analysis of rapid antigen testing has so far come up short.

The academics and public health officials focused their analyses on the characteristics of the individual tests, inadvertently steering the national conversation into a tedious debate over test sensitivity. This left little time for the more practical matter of how and where the tests should be used.

As a result, the national discussion was never able to progress beyond brief mentions of a few individual use cases and towards a more comprehensive analysis of rapid testing in society and its potential to reduce transmission at a national level.

This piece will attempt to address that shortcoming.


The Top-Down Perspective on National Testing

Imagine a map of Ireland with every testing location – be it a workplace, a walk-in test centre, or a community venue – as a point on the map. These points can be thought of as the nodes or the COVID-19 sensors in a national testing network. With enough of these points dotted around the country, the testing network could span every town and community in the land, making testing accessible to all.

As the people go about their normal, unrestricted social activities, they will naturally come into contact with one of these testing points. If the test result is positive, they will isolate, and further transmission is prevented. If it is negative, they continue about their business. In this way, the network functions like an automated COVID-19 filtration system, identifying the virus and removing it from the population, and at minimal inconvenience to the people.

A higher volume of testing will naturally find infections in greater numbers and this will contribute to a general dampening of transmission across the country. It will also reduce the risk of sudden, high-speed outbreaks like the one we experienced in December 2020. We can think of this as a ‘sprinkler system’ effect, wherein both the volume of testing and its geographic distribution keep local clusters suppressed and prevent national waves taking hold.

An interesting outcome of this approach – one which is not identifiable from the bottom up – is that it allows us to implement serial testing at a national scale. In a high-volume testing system, most people will be getting tested at least a couple of times per week. From there it would be easy to tweak the system to ensure that the strengths of serial testing – an early warning system when transmission occurs, and peace of mind until then – could be experienced across the whole population.

It would have been next to impossible to design a process for serially testing the entire population from the ground up, yet by integrating testing into our day to day lives and changing our perspective a little, we realise that we have already achieved it!

 

When we zoom out, we can see that national testing isn’t just about testing more people in higher numbers. At this scale it’s about designing a system that can take us from the individual to the collective, from personal health to public health, from finding infections in people, to finding infections in the population.

This distinction is not arbitrary or semantic.

The traditional medical perspective focuses on the individual, so it sees an outbreak as the sum of the individual infections, and it builds a national response that is the sum of the individual treatments. But this perspective is hopelessly inappropriate when dealing with a large outbreak, as it misses the macro characteristics of both the virus’s transmission and our policy response.

A contagious outbreak behaves more like a systemic disease which has infected the whole nation, rather than a group of infected individual, and that is how we should approach the task. That means we need a national response with a truly national testing program – a program designed to test a nation.


Volume is Data and Data is Power

Perhaps the greatest strength of population-wide rapid testing is the amount of data it creates for us.

High-volume testing will naturally generate high-volume data on transmission, including the location of new infections, their severity, and the likely trajectory of the outbreak. The timeliness and improved detail in this surveillance data gives policymakers a more granular perspective on the outbreak, allowing them to allocate their resources with greater precision to get on top of, or even ahead of transmission.

Unlike other networks which need permanent physical infrastructure to function (e.g. telephone or internet), a rapid testing network is not fixed in place or singular in purpose. We can add or remove testing points wherever they are needed, and we can dial the volume of testing up or down at short notice. This means we can adapt our response to the shape of the outbreak in real time, giving policymakers more control over transmission, streamlining their decisions, and allowing them to set more ambitious goals for the nation.

It is worth emphasising that this intelligent, adaptive system and the top-down precision it gives us would not be possible with NPHET’s ‘PCR-only’ approach to testing. It is the simplicity of the individual rapid tests which allows faster testing and higher volumes, and this is what generates the high-quality data. In contrast, the PCR process is too expensive to run at this scale, and too slow to aid real time decision-making.

Ironically, the higher accuracy of the individual PCR tests leads to the lower accuracy of the overall policy response.


Is This Too Much Testing?

While a national testing program would represent a significant improvement over a national lockdown in every meaningful regard – mental and physical health, commerce, social activity, civil liberties – it would not be without its own inconveniences and irritations.

Waiting outside a pub for 30 minutes every time you wanted a few pints might be preferable to no pints at all, but all that hanging around would eventually get tiresome. And there are bound to be some unfortunate souls who would end up getting three, four or five tests in a single day, and the media would have a field day every time.

As with any other measure, we would have to expect adherence to fall over time, and with it, the effectiveness of the whole strategy. And then what?

The good news is that we would probably never get to that point. This approach might require a lot of testing in the short term, but there is little chance that it would ossify into a permanent restriction on our lives, as NPHET’s lockdown strategy has done.

And you don’t have to take my word for it.


Mass Testing Works Through Herd Effects

By cutting chains of transmission today, mass testing prevents clusters forming tomorrow and community transmission taking hold thereafter. These benefits multiply through the system, and modelling from Larremore, Mina and others has attempted to quantify the effect:


“surveillance testing of 75% of individuals every 3 days was sufficient to drive the epidemic toward extinction within 6 weeks and reduce cumulative incidence by 88%, and that other combinations also had successful but less rapid mitigating impacts”

So, if three quarters of the population were getting tested every three days, then the outbreak could be over in 6 weeks? That sounds promising to me.

Of course we must acknowledge that all modelling studies are based on assumptions, and that the real-world effectiveness of any policy would depend on its specifics and structure, and on the real-world conditions at the time it was implemented.

However, the message from this research is quite clear: high volume testing can be a very effective outbreak control measure. It reduces transmission, it can do so quickly and, given certain conditions, it has the strength to end an outbreak without the need for additional public health measures.

One of the key features of a successful mass testing program is that it displays herd effects at scale, as Mina and Anderson have pointed out in a separate paper:


“Notably, not every transmission chain needs to be severed to achieve herd effects. Mathematical models that incorporate relevant variation in viral loads and test accuracy suggest that with frequent testing of a large fraction of a population, a sufficient number of cases could be detected to create herd effects”

In other words, our mass testing program doesn’t have to be optimal in any way for it to succeed. We don’t need tests with 100% accuracy. We don’t need 100% adherence from the public. We don’t even need a testing program with 100% geographic coverage.

All we need from our testing program is that it can identify enough infections to keep the R number below 1. If it can do that, then it will tip the herd effects in our favour and from there, the outbreak will gradually burn itself out. The key to making the strategy work, as the papers above have shown, is that we run enough tests every day to reach that tipping point.

So yes, in the short run there would be a lot of testing, but if the program was implemented effectively and at the right scale, then we know the inconvenience would be temporary as it would lead to the permanent suppression of the virus, leaving us with easier decisions to make in every aspect of our lives.


If It Is Such A Good Idea, Why Has No One Tried It Before?

They have – in part, at times, and with varying degrees of success.

In the early months of the pandemic, the South Koreans saw the value of high-volume testing and responded quickly, investing in walk-through and drive-through facilities to make testing easily accessible to everyone. This helped them to keep the virus suppressed without lockdowns.

The UK government spent £billions on rapid antigen tests as part of their ‘Operation Moonshot’ program to mass test the country. They had the right idea, but their execution was poor. The Slovakians took a similar approach. Their effort was more structured than the UK’s experiment in Liverpool, but it took place over a shorter time period, so the benefits were only temporary.

Finally, the Chinese have run massive surge testing events (PCR, not rapid antigen tests) in which regional outbreaks are doused with testing capacity as quickly as possible. The scale of the program allows them to find the virus and suppress transmission quicker and more comprehensively, which shortens their periods of lockdown – something the Australian authorities should consider.

 

There is nothing controversial about mass testing. It is a new tool for a new problem, and the only challenge is to learn how to use it well – which is exactly what the nations listed above have been doing. Ireland could have stepped up to the challenge too, but the medical establishment wouldn’t allow it.

The medics decided that mass vaccination would end the pandemic and that we should all stay in lockdown until then. Since that was the plan, what did they stand to gain from an experiment with a new policy tool – especially one which they had publicly disparaged?

Sadly, the medics’ short-term professional convenience has come at the cost of Ireland’s long-term health and stability.

If mass testing could have eliminated community transmission, then it could have kept the virus suppressed for the last year without having to close whole sectors of the economy, isolate swathes of the population from each other, or postpone an inexplicable number of medical procedures.

Not only did the establishment condemn us to a long and unnecessary lockdown, their chronic apathy has left us exposed to future outbreaks and pandemics.

The tools and frameworks needed to fight this battle are the same tools and frameworks we’ll need to fight the next one, but by slowing the process of experimentation and adoption, the establishment has stood in the way of progress. The depressing truth is that we are in no better a position to defend ourselves from a novel virus than we were two years ago. We haven’t grown, we haven’t evolved.

So the next time another pandemic-potential virus is on the loose, Ireland will once again revert to the ‘lockdown until vaccinated’ strategy. We’ll have to. This generation of policy-makers has left us with no other options.