Inside Story

The circuit-breakers

Are short, sharp lockdowns working?

Catherine Bennett 15 February 2021 1294 words

Victorian premier Daniel Andrews talks to journalists on Saturday, the first day of the state’s “short, sharp circuit-breaker.” Erik Anderson/AAP Image

Australia no longer needs to use lockdowns as a primary way of managing large numbers of Covid-19 cases. We had hoped we wouldn’t even need them to deal with outbreaks after the virus breaches our international borders. But for now, at least, the new, more infectious variants have put paid to that hope.

Now we are in the age of “circuit-breaker” lockdowns. In contrast to the extended lockdowns of waves one and two, which were designed to snuff out invisible chains of community transmission, these short, sharp shutdowns are all about buying time to enable the public health response to get ahead of the virus.

Long lockdowns work by keeping the average number of close contacts low across the whole community, thwarting the virus’s attempts to move through the population. They usually run for six to eight weeks, holding infected people in relative isolation long enough to ensure that entire households are no longer infectious.

Circuit-breakers, on the other hand, don’t need to be in place for even one full incubation period to be effective. They are generally used in one of two ways: to slow an established outbreak or — as has happened four times recently in Australia — to help close down a new outbreak.

If they come early in an outbreak, circuit-breakers can minimise the risk of further transmission while health authorities chase down cases and their contacts and use mass testing to map how far the virus has penetrated. They don’t go on long enough to cause the virus to die out; instead, they keep the majority of the population in something akin to quarantine until everyone who might have been exposed is identified and isolated.

The first of these lockdowns came in Adelaide in November, when a new case appeared to have contracted the virus via a very casual exposure while picking up a pizza. The fear was that the infection might signal a superspreader event, but further investigation revealed he had been exposed more intensively by working at the venue with a known case. When it became clear the outbreak was contained, the six-day lockdown was truncated to three days.

Brisbane came next. A three-day lockdown was declared on 8 January after a cleaner working in hotel quarantine became infected with what genomic testing revealed to be one of the new infectious variants, B.1.1.7. Given the impact of the variant during Britain’s uncontrolled second wave, authorities were worried that it might be more difficult to contain.

As it transpired, we were lucky. Despite being out and about for a few days, the cleaner was not especially infectious and only passed the virus on to her partner. The rapidly implemented precautionary lockdown was, thankfully, not needed. But that also meant contact tracing was not tested in those circumstances, so we don’t know whether it alone might have been sufficient to contain the potential outbreak even if the index case had been more infectious.

Next came Perth. Four days after announcing a hotel worker had tested positive for the same B.1.1.7 variant, the WA government sent the city and parts of the state’s southwest into a five-day circuit-breaker lockdown at the end of January. Once again, the concern over the potential impact of this strain on outbreak containment led to this significant step.

Luck was once again on our side. No secondary transmission had occurred, not even to close contacts. This is typical of the virus — 20 per cent or so pass it on; most do not. Once again, the circuit-breaker proved to be an unneeded precaution. Even though they hadn’t needed to deal with wider community transmission, the Brisbane and Perth circuit-breakers were labelled a “success.”

But these events will provide rich information to determine whether the lockdowns added necessary additional containment, or whether contact tracing alone was coping. We will be able to determine whether all contacts were in quarantine before the lockdown came into effect, for example, or before they could have been infectious.

We now have more detailed data from Britain on the increased infectivity of B.1.1.7. Public Health England reported that 14.7 per cent of close contacts of cases became infected compared with 11 per cent for previous variants. This 34 per cent acceleration in new cases has proved devastating in countries where the variant now dominates, and demonstrates why we must not let this (or other) mutant variants escape into the community in Australia.

The good news is that even with these higher infection rates, we can contain outbreaks if we catch them early. If ten cases each have ten close contacts, we might see fifteen secondary cases on average, instead of eleven. That is manageable with contact tracing. Yet when Victoria experienced its latest crossover into the community from hotel quarantine, again involving the B.1.1.7 variant, the entire state was placed into a five-day lockdown.

In this latest event, still playing out as I write, other factors had to be considered. Four staff were infected in quick succession in hotel quarantine, along with two returned travellers sent home before the outbreak was detected. In other words, six people were potentially seeding community outbreaks simultaneously: six more chances of a highly infectious case within one generation of spread. No wonder authorities were nervous.

As it stands, four of those initial six cases didn’t pass on the virus. Of the other two, one infected a spouse only and the other, unknowingly infectious, attended a large family event that led to a second-generation of spread of seven further cases. This dinner occurred before the outbreak was apparent, and six full days before lockdown started.

That family dinner is now the central concern. All attendees have been identified, tested and isolated, but the seven new cases resulting from this exposure event were potentially infectious for up to four days before they were identified as close contacts of a known case. If none of their close contacts tests positive, we can relax; the risk that they may have passed the virus on to more casual contacts is then significantly less.

Will the circuit-breaker lockdown contain transmission in Victoria? It couldn’t prevent the secondary community transmission because the exposure occurred too early in the chain of events — before the hotel worker who seeded the outbreak was recognised to be a case, and before the infected person’s last workplace test yielded a negative result.

That person must have been becoming infectious at the time of the family dinner, but their fluctuating viral load produced a very weak positive result a day later, insufficient to register as a positive result the first time the sample was tested. The event was not recognised as an exposure site for another five days.

That will be the true test of whether we need circuit breakers. We can learn from this latest public health response, as well as the previous ones, whether we can rely on contact tracing even in the face of more transmissible variants in a more complex outbreak. If lockdown does help quash further spread — and we will know that soon enough — we can then use this information to inform future decisions on when circuit breakers should be used, and what they need to involve.

Building on the “success” of the previous, untested lockdowns — lockdowns that have progressively been extended in both duration and geography — is dangerous. We risk believing they are an essential tool when in fact they had nothing to do with the outcomes in Queensland and Western Australia. The next few days will add the Victorian experience to the mix, and even if this lockdown ends up being redundant, we will have learned more valuable lessons. •

The publication of this article was supported by a grant from the Judith Neilson Institute for Journalism and Ideas.