Richard Dobbs

If we want herd immunity, we need mass testing

If we want herd immunity, we need mass testing
Chip Somodevilla/Getty Images
Text settings

At the start of the pandemic, we talked a lot about herd (or community) immunity. But talking about the journey to herd immunity became toxic as it was variously linked to high infection rates, sacrificing the elderly, and the NHS becoming overwhelmed.

The debate on herd immunity was restarted last week by Professor Karl Friston, of University College London, who told the Daily Telegraph that the 73.4 per cent vaccinated reached on Monday meant that 'based upon contact rates at the beginning of the pandemic and estimated transmission risk, this is nearly at the herd immunity threshold.' This is an outlying view: other academics questioned this analysis. Matt Hancock said that the government will continue to watch the real-world data.

But some rough scenarios on herd immunity can be helpful as we try to understand the way forward as we roll out vaccines. They also explain why the Prime Minister is right to ignore his sceptics and to continue to push for mass testing, alongside vaccines. Full disclosure: I have a vested interest in that I have a stake in some of the testing innovators.

Herd immunity relies on the suppression of the virus from a sufficient proportion of the population not catching the disease because they have been vaccinated or have antibodies from a previous infection. It is dependent on the rate at which each infected person passes on the virus. Let’s illustrate this with a simple example. Imagine that everyone infected will typically infect three other people if there was no immunity (the technical name for this measure is R0). However, imagine if two in three people already had immunity. The infected person would pass the virus on to three people but two would not catch the illness as they would have immunity. So, the infected person would only infect one other. The R-number would be 1 and the number of people infected would stay the same. We would have stopped the exponential growth in the number infected by having two thirds of the population with herd immunity.

So, the required level of herd immunity is dependent on the level of R0. For those interested, the maths is: Herd Immunity proportion = 1- (1/R0).

It should stress that any such calculations using averages are only directional. They don’t, for instance, adjust to reflect that some age groups transmit the virus more or less than others. Vaccinating those that transmit more (for instance because their job or social life has higher interactions) has more impact at suppressing the virus. So while vaccinating the elderly ‘sofa-sitters’ is essentially valuable at preventing death, it has less impact at suppressing the spread of the pandemic as this group has generally fewer contacts and so transmits the virus less. That being said, the rough calculations using averages generates some insights that can inform our debate on the way forward.

Prior to the first lockdown, we saw an R number of between 3 and 4 according to estimates by Imperial College. As very few people at that point had any immunity, that would suggest that the R0 was itself also over 3. If we assume a level of 3, we would have needed a herd immunity of around two thirds of the population. Indeed, Patrick Vallance suggested at the time that '60 per cent is the sort of figure you need to get herd immunity'. Clearly that was then a very distant prospect as without vaccines, the only way we could reach this level of herd immunity was by having over 60 per cent of the population catch Covid-19.

Today vaccines can also deliver immunity. We recently saw ONS report that around 55 per cent of the population tested positive for antibodies on a blood test in the week ending 14 March 2021.

Given the delay in these statistics and the lag between vaccination and the emergence of antibodies, the above graph will not reflect the 10 million vaccinated in England since the start of March. We might now have the proportion with immunity over 70 per cent. So it might be tempting to start celebrating the imminent arrival of herd immunity or call for an earlier end to lockdown measures.

However, there are two caveats that we need to take into account: the increased transmissibility of the now prevalent strain first identified in Kent and the less than perfect suppression from the vaccines.

At the end of last year, we saw a resurgence of the pandemic, partially driven by the new variant of the virus, first identified in Kent: this variant was, according to an estimates in the Lancet, up to 70 per cent more transmissible and perhaps slightly more deadly. This increase in transmission might suggest that the new R0 without lockdown measures could be over 5, suggesting we might need over 80 per cent infected or vaccinated to reach herd immunity.

Dr Anthony Fauci, chief medical advisor to Joe Biden (and Trump before him), has been raising the bar in relation to herd immunity, because of this increase in transmissibility. He said last December that he believes that it might take close to 90 percent immunity to bring the virus to a halt.

The second caveat is that the vaccines are not perfect at suppressing transmission.

If a vaccine is 90 per cent effective at suppressing transmission, that proportion needed for herd immunity when we unlock fully goes up from 80 percent to a little under 90 percent. That’s because people who get infected would still transmit.

But new variants, including the ones first identified in South Africa and Brazil, might reduce the effectiveness of the various vaccines at suppressing transmission further (especially for the AstraZeneca jab). If the vaccines end up only 70 per cent effective, we could find that herd immunity could never be achieved with a full removal of lockdown restrictions and that’s before we consider the proportion of the population that cannot or will not be vaccinated.

This above analysis suggests potentially bad news for our fantastic vaccine programme: while vaccines significantly reduce deaths, on their own they might never be enough to suppress the virus and allow us to return fully to normal. If we unlock fully, the virus could become endemic, with a high level of infection. While the vaccines will mean that we will not witness the high level of deaths we suffered before, Covid-19 could still be a nasty illness and we don’t yet know whether the vaccine prevents the long-Covid that one in seven currently experience 12 weeks after infection, according to an ONS survey.

But this is where the Prime Minister’s Moonshot of mass testing comes in. This Moonshot provides us all with access to the new lateral flow tests. While these tests do not identify those in their very early days of infection, unlike the traditional laboratory based PCR tests, they seem to be good enough at identifying those that are infectious. They also take less than 15 minutes and cost around a tenth of the cost of the PCR tests. They are therefore particularly good at identifying quickly the 30 per cent or so of individuals who are asymptomatic, that is have and can spread Covid-19 but without symptoms. The idea is that we could all test twice a week with these tests and if the test is positive, we and our household would stay at home and isolate and so do not spread the virus. They are already being used in schools, successfully.

If we could get mass testing and isolating to work, we would prevent many of those infected spreading the virus by taking them out of circulation. We don’t yet know the impact that could be achieved. But perhaps we might guess that if mass testing results in two thirds of those infected being removed from circulation so as to infect less than one other, and the remaining one third still infecting 5, we could bring the average R0 down to maybe around 2. That would mean that our required effective herd immunity would only be around 50 per cent which we have already achieved. Or if the vaccine is around 70 per cent effective at reducing transmission, we would need to have 70 per cent with herd immunity, again a level of herd immunity that is achievable. So mass testing and vaccines, could together allow us to remove all other lockdown measures.

The UK is already one of the global leaders in vaccines. If we got mass testing and vaccinations to work together, we could be the world leader in escaping lockdowns. It could also prepare us for any future variant or other pandemic. It’s time to have a proper conversation on herd immunity.

Written byRichard Dobbs

Richard Dobbs was a director of the McKinsey Global Institute. He is currently serving as a non-executive director on several boards, including the UK Statistics Authority, but writes in a personal capacity.

Topics in this articlePoliticstestingherd immunity