What is elimination and what could defeat it?
In the first of a two-part series, Nikki Macdonald looks at the assumptions underpinning our Covid-19 strategy – and what if they’re wrong. Today: can we really eliminate the virus if people get infected without showing symptoms?
We started with ‘‘stamp it out’’. Then it became ‘‘elimination’’. Either way, the goal is to get this vermin of a virus out of New Zealand, and keep it there.
Last week, Prime Minister Jacinda Ardern announced that objective had already been achieved. Experts disagreed.
But one thing everyone agrees is that getting rid of the disease will be a nightmare if it can be easily spread by people who have no symptoms.
Without a vaccine or treatments, controlling Covid-19 relies on our ability to identify positive patients early, so they can be quarantined to prevent them passing on the infection and so any close contacts they might already have infected can be speedily found and isolated.
But that strategy falls over if people get the disease without knowing, and unwittingly spread it as they go about their normal business.
So the reassuring run of low numbers at the daily press conference could be obscuring a silent reservoir of disease waiting to be unleashed when distancing restrictions are relaxed.
So is asymptomatic transmission a thing, and if it is, is it fatal to our elimination strategy?
Covid-19 is already trickier to stamp out than its cousin Sars, because the best guess is that patients are most infectious in the first five days after they start showing symptoms.
Sars patients, by contrast, were most infectious between day 6 and day 11.
For Covid-19, that makes a tight window for identifying cases who will mostly have only mild symptoms, isolating that patient to prevent them spreading the disease, and tracing and isolating close contacts before they themselves start to show symptoms and infect others.
But if infected patients show no symptoms at all, they won’t be tested or isolated and the people they’ve hung out with – and potentially infected – won’t be traced. So that chain of transmission continues unchecked.
Otago University infectious diseases specialist Professor David Murdoch says one of the reasons flu pandemics are considered unstoppable is because they can spread without symptoms, so aggressive testing, isolation, and contact tracing can’t stop the disease running rampant.
‘‘If we did this with influenza, we would lose pretty quickly.’’
Children are also key spreaders of the flu, but don’t seem to be big sources of infection with Covid-19. And if silent spreading of Covid-19 was common, you would expect a higher number of cases with no known source, Murdoch says.
‘‘Just because they’re asymptomatic doesn’t mean people they infect would be, so you should see a significant number of cases that have no explanation. At the moment case numbers are exceptionally small. That for me is reassuring.’’
It is logical that people with no symptoms would be less infectious than people coughing or sneezing disease particles 6-8 metres.
Research into the contact tracing results of 100 positive cases in Taiwan included nine patients who had no symptoms. While the contact tracing found that the 100 cases infected a further 22 people, none of those secondary cases were infected by those who never developed symptoms.
Epidemiologist and infectious diseases doctor Ayesha Verrall agrees that, even under lockdown, asymptomatic transmission should show up through people with no symptoms infecting their housemates. But that has not been widespread.
‘‘At some point, we have to stop believing in fairies.’’
However, policymakers need to know for sure, so they can plan post-lockdown surveillance, she says. If there’s no asymptomatic transmission, we can focus on finding people with symptoms, but if silent spread is significant, wider population testing is needed.
‘‘I hope we can get clear on whether this asymptomatic thing is real, because asymptomatic transmission is a real concern for elimination.’’
But there’s another sting in Covid-19’s tail, that has received less attention – transmission of the virus before people show symptoms.
The Taiwan transmission research found the virus’ ‘‘attack rate’’ was 0.7 per cent in contacts who had mixed with the infected patient only before they showed symptoms, compared with 1 per cent in people who had contact with a case 1-5 days after symptom onset. No-one was infected through contact more than five days after someone showed symptoms.
Verrall says presymptomatic transmission is ‘‘definitely a thing’’ and was responsible for at least seven clusters in Singapore. New Zealand’s contact tracing regime investigates a case’s movements two days before symptoms emerge, which is in line with World Health Organisation guidelines. But the Taiwan study suggests contact tracers should start their investigations four days before symptoms showed.
The study concluded that ‘‘the high transmissibility of Covid-19 before and immediately after symptom onset suggests that finding and isolating symptomatic patients alone may not suffice to interrupt transmission, and that more generalized measures might be required, such as social distancing’’.
Some research has also suggested the virus can hang in the air long after an infected person has gone, potentially infecting people who would never be identified in contact tracing.
Otago University public health professor Michael Baker says anything that’s dropletborne, as Covid-19 is, can also be airborne. ‘‘But it’s about relative importance.’’
Even with intensely studied diseases such as flu, there’s still fierce debate about the relative role of droplets, aerosols and infected surfaces in spreading the disease.
But even if Covid-19 can be spread by people who haven’t yet got symptoms, or who never get symptoms, or it hangs around in the air, the fact that other countries, such as China, have stopped the disease mid-flight suggests none of those things is fatal to our goal of elimination, Baker says.
‘‘It’s a tough virus to stop, because it’s quite infectious, and quite a high proportion of cases are entirely asymptomatic and certainly people are infectious a couple of days before they
become symptomatic. So all those things make it tougher, but the net effect, even with those barriers, is not enough to make it uncontrollable, unlike influenza, which is considered a pandemic that you can’t contain.’’
What does elimination mean?
On April 27, Prime Minister Jacinda Ardern announced we had ‘‘currently eliminated’’ Covid-19. Experts disagreed.
Which raises the question – what does elimination even mean, and how will we know when we’re there?
Despite two days of zero cases, Ardern this week warned of the virus’ long tail. Low case numbers are no guarantee the disease won’t return.
The Japanese region of Hokkaido reported single-digit and zero case numbers before it emerged from its three-week lockdown on March 19. Less than a month later, it pulled the blinds back down, after it was hit by a second – worse – wave of disease.
Much-admired Singapore has also seen a resurgence, although that mostly relates to outbreaks in packed hostels for migrant workers, which does not have a New Zealand equivalent.
Baker says elimination is a well-established term in epidemiology, but it doesn’t mean eradication and it doesn’t mean no cases. Eradication means global extinction, to the extent you no longer need to vaccinate. The only disease we’ve achieved that for is smallpox.
Elimination means you no longer have community transmission, but exactly how that’s defined remains to be nutted out. It could be anything from zero tolerance – no community cases at any time – to no community transmission for a month, or a limited number of community cases that link back to an infected traveller.
It’s similar to the process for measles and rubella – Baker is on the regional verification commission for those two diseases. If a country has no cases in the community, or any cases they have are linked to travellers and don’t go on too long, countries are said to have eliminated the disease.
New Zealand will need to agree an elimination definition with other countries if they want to resume international travel, Baker says.
He and other researchers this week called on the Health Ministry to report more details of each new case, to help experts decide whether the country is failing in its elimination goal. A local infection stemming from a returned traveller would indicate a quarantine breach; a case with no known source could indicate widespread community infection; people infected outside a household bubble could be a sign contact tracing wasn’t effective; and infected healthcare workers would suggest faulty infection control.
Auckland University School of Medicine Professor Des Gorman also calls for clearer case reporting.
A common argument supporting the idea New Zealand has broken the chains of community transmission is the fact only 4 per cent of cases come from an unexplained source. But those cases are responsible for another 24 per cent of all cases, so to ignore the other cases is misleading, Gorman says.
‘‘Did the rest of them get infected by magic? Or by some malevolent witch? They’re all community transmissions.’’
Gorman says claiming you’ve eliminated the disease while still reporting new cases is confusing. He prefers the term ‘‘disease control’’.
We’ll know we’ve controlled Covid-19 when we have deep knowledge of community prevalence (the proportion of the population that has been infected), can trace all contacts within 48 hours of diagnosis, and when isolation is foolproof.
Knowing the disease prevalence means understanding the extent of Covid-19’s spread in random samples of society; in those who are more likely to have been exposed – such as people who are out at work or going to supermarkets; and in vulnerable groups.
He questions whether the Health Ministry can really know how much disease has been lurking in the community.
While health authorities have been doing random virus tests, they only pick up people who are infected at that moment in time.
What’s really needed is antibody testing to find out how many people have been exposed to Covid-19, Gorman says.
‘‘It’s overdue . . . The more testing you do, the more confidence you and I can have that prevalence really does approach zero.’’
Most scientists are wary of the tests because they’ve been shown to be inaccurate. An analysis of 14 tests by 50 scientists found only three were consistently reliable.
Gorman says 100 per cent accuracy is not critical if you just want a rough idea of whether there’s virus circulating that you don’t know about. ‘‘If we really have a prevalence that is one in tiddly squat, then a 5 per cent error of tiddly squat doesn’t matter.’’
The ministry has said it had no plans to introduce antibody testing. Asked if that position has changed, it did not respond by deadline.
Baker says antibody testing would be a waste of time and money. While it’s useful in countries where the disease has been rampant, the idea that
2, 5 or 10 per cent of Kiwis might have been infected with no symptoms is ‘‘almost certainly completely rubbish’’. With a ‘‘vanishingly small’’ proportion of the population exposed, false positives would probably exceed real positive cases, he says.
However, with emerging evidence that the disease might spread just by laughing, talking or even breathing, Baker says no-one should expect a return to normal life any time soon.
‘‘I think we have to organise our lives for the foreseeable future on the assumption there’s circulating virus in the population and everyone you meet may be infectious and everything you touch in a public place may be contaminated.’’