Estimating the failure risk of quarantine systems for preventing COVID-19 outbreaks in Australia and New Zealand

Introduction

New Zealand and Australian states have successfully eliminated community transmission of the pandemic virus 'severe acute respiratory syndrome coronavirus 2' (SARS-CoV-2),¹ albeit with occasional outbreaks from imported cases that have typically been quickly brought under control. These two countries have primarily used hotel-based quarantine for citizens returning to their countries during the pandemic period, with 14 days of quarantine combined with polymerase chain reaction (PCR) testing and mask use in any shared spaces (eg, common exercise areas used in New Zealand, but not in most Australian states).

Converting hotels for quarantine purposes has the advantage of making use of a resource that would otherwise be underused during a pandemic, given declines in international tourism. However, the major disadvantage of hotel-based quarantine is that it is likely to be less effective than purpose-built quarantine facilities owing to shared spaces and lack of safe ventilation (as per World Health Organization advice on air flow²). Moreover, the consequences of leakage of the virus out of quarantine (eg, through facility workers) may be more severe given higher population density in urban settings where the hotels are located. Given these issues, we aimed to estimate the failure risk of quarantine systems in New Zealand and Australia in terms of the spread of 'coronavirus disease 2019' (COVID-19) infection into the community.

As of 13 June 2021, the rolling 7-day average number of COVID-19 vaccine doses administrated per 100 people was 0.46 in Australia and 0.31 in New Zealand.³ However, this was counted as single vaccine doses and does not equal the total number of people vaccinated (eg, the Pfizer/BioNTech vaccine which is currently used in New Zealand requires two doses).³ The majority of border workers in Australia and New Zealand have been vaccinated (eg, in New Zealand over 56,000 doses had been administered to border workers as of 28 March 2021,⁴ and all hotel quarantine workers in Victoria who have face-to-face contact with returned travellers received their first dose of the vaccine by the first week in April 2021⁵).

Methods

We defined a quarantine system failure as where a border/health worker or person in the community with a link to the quarantine/isolation system, became infected with SARS-CoV-2. This definition included people infected in hospital from cases who had been transferred from a quarantine facility (as such cases were still in the 14-day quarantine process), but did not include pandemic virus transmission between returnees within the quarantine facilities.

We searched official websites in both countries, and for the eight states and territories in Australia, to identify outbreaks and border control failures associated with quarantine systems (searches conducted between 6 January and 23 June 2021). Where an outbreak source was uncertain (eg, the Auckland, New Zealand, August 2020 outbreak) we used the best available evidence to classify it as a quarantine failure or not. The decision to label an incident as a quarantine system failure was confirmed by all co-authors. We used two denominators: a) the estimated number of travellers who went through quarantine facilities during the 2020 year (data were for the period starting 1 April 2020 for Australia and 17 June 2020 for New Zealand) up to 15 June 2021; and b) the number of SARS-CoV-2 positive people who went through these facilities in this same time period. The unit of analyses were New Zealand, the eight Australian states and territories, and both countries combined.

For New Zealand, we used official data on both travellers going through the quarantine system⁶ along with official (Ministry of Health) data on SARS-CoV-2 positive cases,⁷ although there are some discrepancies in the information about when regular testing began in Managed Isolation and Quarantine (MIQ) facilities. For Australia we used overseas arrival data⁸ and health data.^{9, 10}

Results

The collated data for quarantine system failures are shown in Table 1 (available in PDF), with specific details of each event in the Appendix (Table A1 - available in PDF). In Australia, 22 failures were identified, one resultant outbreak caused over 800 deaths (Victoria’s second wave) and with eight lockdowns linked to quarantine system failures. In New Zealand, there were ten failures, with one causing an outbreak with three deaths, and also a lockdown.

Given our estimates of the number of travellers processed via quarantine systems (Table 1 - available in PDF), the overall risks for both countries combined were one failure per 20,156 travellers, and one failure per 163 SARS-CoV-2 positive cases in quarantine. The combined data can also be interpreted as quarantine system failures leading to a lockdown response per 71,665 travellers; and approximately one death from COVID-19 per 803 travellers (using the 800 deaths estimate from Australia and the three deaths from New Zealand – although this figure is largely driven by the second wave in Victoria and is unlikely generalizable forward in time).

At the country level, there were 10.5 failures per 1000 SARS-CoV-2 positive cases transiting quarantine in New Zealand (95% confidence interval [CI]: 4.0 to 17.0), compared to 5.2 per 1000 SARS-CoV-2 positive cases in Australia (95%CI: 3.0 to 7.3) – a two-fold difference in risk (relative risk: 2.0, 95%CI: 1.0 to 4.2).

Discussion

This analysis identified 32 failures of quarantine systems in Australia and New Zealand combined (up to 15 June 2021). The relatively higher failure risk per 1000 SARS-CoV-2 positive cases transiting quarantine in New Zealand versus Australia could reflect a lower quality approach in the former, with perhaps some of the difference due to greater detection in New Zealand from more border worker testing over a longer period.

These estimates are subject to chance variations due to the low numbers of failures. These estimates will also probably be an underestimate of all quarantine system failures, as not all of those infected will transmit the virus and start a detectable chain of transmission. Genomes of the first 649 viral isolates collected in New Zealand show that only 19% of virus introductions resulted in ongoing transmission of more than one additional case.¹¹ Therefore, counts of border system failures are sensitive to how they are identified and defined. Indeed, with increased testing (eg, testing of people after leaving quarantine on day 16 as is now common in Australia), we may be detecting failures that previously would have been undetected.

Looking forward, the failure risks per month in New Zealand and Australia may increase, given that the proportion of travellers returning to these countries who are infected is increasing due to global intensification of the pandemic and the increasing infectivity of new SARS-CoV-2 variants.¹² Indeed, there have been several clearly documented cases of spread within quarantine hotels (eg, two instances in Melbourne in February 2021, two instances in Sydney in April 2021, and one in South Australia in May 2021), highlighting the increased risk and evolving situation with more highly infectious variants arriving from overseas.

However, offsetting this trend will be measures such as the vaccination of quarantine workers. In New Zealand, the vaccination of border workers began in February 2021 with the Pfizer/BioNTech vaccine, and as of 11 June 2021 all MIQ workers were fully vaccinated.¹³ However, as of 29 June 2021, over 1600 frontline border workers outside of MIQ (eg, those working at ports or as aircrew) were still unvaccinated.¹⁴ A notable limitation of this study was the lack of publicly available data on the vaccination of quarantine workers in Australia. While quarantine workers and other border staff were prioritised for vaccination as part of the first phase of the national vaccine rollout, there was no information available on how many workers had or had not been vaccinated. As of 15 June, only 3% of Australians were fully vaccinated.¹⁵ Some states claimed to have required all border staff to be vaccinated earlier in 2021 (eg, ^{16, 17}), but investigations of outbreaks in Victoria, Western Australia, Queensland, and New South Wales in June 2021 demonstrated that many border workers, as well as health and aged care workers and residents, had not yet been fully vaccinated.

The full vaccination of frontline border workers could likely have prevented a number of quarantine system failures. However, vaccination does not fully eliminate the risk of contracting SARS-CoV-2, nor does it fully protect against SARS-CoV-2 transmission, although a moderate degree of protection is likely. For example, infection rates were halved for the AstraZeneca vaccine,^{18, 19} reduced by 70% for the Moderna mRNA vaccine as indicated by using swab results for asymptomatic infection plus symptomatic cases,²⁰ and reduced by 95% for the Pfizer/BioNTech mRNA vaccine as indicated by national surveillance data in Israel.²¹ For vaccinated people who are infected, primate study evidence suggests (consistent with expectation) that the infectivity is decreased in peak and duration,²² further protecting the border.

Furthermore, the level of testing of quarantine workers has been increasing (eg,²³; which will find some failures before they have a chance to establish as an outbreak in the community). There have been other improvements in the quarantine systems over time (eg, improved security, introduction of mask wearing within quarantine settings, reduction in shared spaces, improved personal protective equipment (PPE) used by workers, and other procedures as detailed in both countries^{24, 25}).

Another risk reduction practice would be using better or purpose-built facilities in rural locations as these have less risk from close contacts in central business district hotels and within-building spread from poor ventilation systems. To date, there have been no failures at the Howard Springs facility outside of Darwin, and the success of the facility was cited in the announcement in June 2021 of the construction of a new purpose-built quarantine facility in Victoria.²⁶ Other infection prevention and control measures (eg, PPE) will remain important in all quarantine facilities.

Limitations of our analysis include residual uncertainty around the cause of some outbreaks (eg, the Auckland one in August 2020), and imprecision with denominator data on traveller numbers for Australia (eg, some travellers were moved between states on domestic flights which is not captured in the official data we used). Additionally, case numbers are constantly changing, due to the number of reclassifications caused by false positives and duplications. We also did not assess the change in the quarantine system failures over time due to the relatively small number of failure events. The risk of system failures is probably highly dynamic on a month-by-month basis as traveller volumes, infection rates, and quarantine processes change, and as vaccination rates among border workers and in the wider community increase.

To substantially reduce the risk of SARS-CoV-2 incursion out of quarantine (until such time as enough of the population is vaccinated), the most obvious action is to reduce arrivals, or even suspend arrivals, from high infection locations (as New Zealand and Australia temporarily did for travel from India and other high risk countries in April 2021²⁷). Beyond this, there are a range of other potential improvements in ongoing arrangements and processes as detailed in Table 2 (available in PDF). Furthermore, the start of quarantine-free travel between Australia and New Zealand (also known as a "green zone") in April 2021 provides an opportunity to benchmark COVID-19 border control policies and practices, identify potential improvements in both countries, and harmonise best practices across the region. The green zone further intertwines the biosecurity status of both nations and it is therefore even more important to lower the risk of border failures that could disrupt such travel. This shift from a one-size-fits-all strategy to a risk-based approach to border management can be summarised as a 'traffic light' approach.²⁸

Conclusions

In summary, Australia and New Zealand have had 32 COVID-19 identified failures arising from quarantine systems up to 15 June 2021. Quarantine system failures can be costly in terms of lives and economic impacts such as lockdowns. Ongoing improvements or alternatives to hotel-based quarantine are required.

Ethics approvals

There was no requirement for ethics review as the study only utilized publicly available anonymized data.

Competing interests

No relevant disclosures.

References:

1. Baker M, Wilson N, Blakely T. Elimination may be the optimal response strategy for covid-19 and other emerging pandemic diseases. BMJ 2020; 371: m4907. doi: 4910.1136/bmj.m4907.
2. World Health Organization. Considerations for quarantine of contacts of COVID-19 cases: Interim guidance, 2020. https://www.who.int/publications/i/item/considerations-for-quarantine-of-individuals-in-the-context-of-containment-for-coronavirus-disease-(covid-19) (accessed Aug 2021)
3. Ritchie H, Ortiz-Ospina E, Beltekian D, et al. New Zealand: Coronavirus pandemic country profile published online at OurWorldInData.org 2021 (updated Apr 2021).  https://ourworldindata.org/coronavirus/country/new-zealand?country=NZL~AUS (accessed Apr 2021).
4. New Zealand Ministry of Health. COVID-19: Vaccine data, 2021 (updated Apr 2021). https://www.health.govt.nz/our-work/diseases-and-conditions/covid-19-novel-coronavirus/covid-19-data-and-statistics/covid-19-vaccine-data (accessed Apr 2021).
5. Ilanbey S, Dexter R, Estcourt D. All Victorian frontline hotel quarantine workers to be vaccinated by Thursday. The Age (Victoria/Coronavirus pandemic) 2021; 5 Apr. https://www.theage.com.au/national/victoria/no-new-local-covid-19-cases-as-calls-grow-for-mass-vaccination-centres-to-be-fast-tracked-20210405-p57ghc.html (accessed Apr 2021).
6. Ministry of Business Innovation & Employment. Managed isolation and quarantine data. New Zealand, 2021. https://www.mbie.govt.nz/business-and-employment/economic-development/covid-19-data-resources/managed-isolation-and-quarantine-data/ (accessed Jun 2021).
7. Ministry of Health. COVID-19: Case demographics 2021 (updated Jun 2021). https://www.health.govt.nz/our-work/diseases-and-conditions/covid-19-novel-coronavirus/covid-19-data-and-statistics/covid-19-case-demographics (accessed Jun 2021).
8. Australian Bureau of Statistics. Overseas Travel Statistics, Provisional (Released 16 June 2021). https://www.abs.gov.au/statistics/industry/tourism-and-transport/overseas-travel-statistics-provisional/latest-release (accessed Jun 2021).
9. Australian Department of Health. Coronavirus (COVID-19) common operating picture, 2021. https://www.health.gov.au/resources/collections/coronavirus-covid-19-common-operating-picture (accessed Jun 2021).
10. O’Brien J, Monteiro C, Rezny M, et al. COVID-19 in Australia, 2021 (updated Apr 2021). https://www.theage.com.au/national/victoria/three-new-covid-19-cases-linked-to-holiday-inn-cluster-emerge-20210219-p573wq.html (accessed Apr 2021).
11. Geoghegan JL, Ren X, Storey M, et al. Genomic epidemiology reveals transmission patterns and dynamics of SARS-CoV-2 in Aotearoa New Zealand. Nat Commun 2020; 11: 6351-6357.
12. COVID research updates: 7 January — Evidence grows of a new coronavirus variant’s swift spread. Nature 2021, 8 Jan. https://www.nature.com/articles/d41586-020-00502-w (accessed Apr 2021).
13. Ministry of Business, Innovation & Employment. All MIQ workers fully vaccinated [press release]. 2021; 11 Jun. https://www.miq.govt.nz/about/news/all-miq-workers-fully-vaccinated/ (accessed Jul 2021).
14. Cheng, D. Covid 19 coronavirus: 1600 border workers still unvaccinated; 160 MIQ workers overdue for testing. New Zealand Herald (Politics) 2021; 29 Jun. https://www.nzherald.co.nz/nz/covid-19-coronavirus-1600-border-workers-still-unvaccinated-160-miq-workers-overdue-for-testing/XMNZGQVT3ZO2EOO7I3L5VYCBAY/ (accessed Jul 2021).
15. Australian Department of Health. COVID-19 vaccine rollout update - 15 June 2021. 2021; 15 Jun. https://www.health.gov.au/sites/default/files/documents/2021/06/covid-19-vaccine-rollout-update-15-june-2021.pdf (accessed Jul 2021).
16. Ilanbey S, Dexter R, Estcourt D. All Victorian frontline hotel quarantine workers to be vaccinated Thursday. The Age (National/Victoria/Coronavirus pandemic) 2021; 5 Apr. https://www.theage.com.au/national/victoria/no-new-local-covid-19-cases-as-calls-grow-for-mass-vaccination-centres-to-be-fast-tracked-20210405-p57ghc.html (accessed Jul 2021).
17. Pilat, L. Mandatory vaccine enforced for WA hotel quarantine workers. WAtoday (National/WA/Coronavirus pandemic) 2021; 19 Apr. https://www.watoday.com.au/national/western-australia/mandatory-vaccine-enforced-for-wa-hotel-quarantine-workers-20210419-p57kjt.html (accessed Jul 2021).
18. Voysey M, Costa Clemens S, Madhi SA, et al. Single dose administration and the influence of the timing of the booster dose on immunogenicity and efficacy of ChAdOx1 nCoV-19 (AZD1222) vaccine: a pooled analysis of four randomised trials. Lancet 2021, 397: 881-891.
19. Emary KRW, Golubchik T, Aley PK, et al. Efficacy of ChAdOx1 nCoV-19 (AZD1222) vaccine against SARS-CoV-2 variant of concern 202012/01 (B.1.1.7): an exploratory analysis of a randomised controlled trial. Lancet 2021; 397: 1351-1362.
20. Baden LR, El Sahly HM, Essink B, et al. Efficacy and safety of the mRNA-1273 SARS-CoV-2 vaccine. N Engl J Med 2021; 384: 403-416.
21. Haas EJ, Angulo FJ, McLaughlin JM, et al. Impact and effectiveness of mRNA BNT162b2 vaccine against SARS-CoV-2 infections and COVID-19 cases, hospitalisations, and deaths following a nationwide vaccination campaign in Israel: an observational study using national surveillance data. Lancet 2021; 397:1819-1829.
22. Brouwer PJM, Brinkkemper M, Maisonnasse P, et al. Two-component spike nanoparticle vaccine protects macaques from SARS-CoV-2 infection. Cell 2021; 184: 1188-1200,e1119.
23. Covid-19: Why Australia’s quarantine rules are changing. Radio New Zealand (World) 2021; 9 Jan. https://www.rnz.co.nz/news/world/434278/covid-19-why-australia-s-quarantine-rules-are-changing (accessed Apr 2021).
24. Morrah M. Revealed: The strict new coronavirus rules for New Zealand's COVID-19 Defence Force staff. Newshub 2020; 25 Nov. https://www.newshub.co.nz/home/new-zealand/2020/11/revealed-the-strict-new-coronavirus-rules-for-new-zealand-s-covid-19-defence-force-staff.html (accessed Apr 2021).
25. Premier of Victoria. A stronger quarantine program to protect what we’ve built [press release]. 2020; 30 Nov. https://www.premier.vic.gov.au/stronger-quarantine-program-protect-what-weve-built (accessed Apr 2021).
26. The Victorian government reaches a deal with the Commonwealth over new quarantine hub. ABC News 2021; 4 Jun. https://www.abc.net.au/news/2021-06-04/new-quarantine-hub-to-be-built-in-victoria/100189846 (accessed Jun 2021).
27. Cheng D. Watch: Covid Response Minister Chris Hipkins reveals travel ban for ‘very high risk countries’. NZ Herald 2021; 23 Apr. https://www.nzherald.co.nz/nz/watch-covid-response-minister-chris-hipkins-reveals-travel-ban-for-very-high-risk-countries/ZRZAJIGZ6Q3OBWPX6C2ZB7BLPM/ (accessed Apr 2021).
28. Wilson N, Baker M. New Zealand needs a ‘traffic light’ system to stop COVID-19 creeping in at the border. The Conversation 2020; 4 Nov. https://theconversation.com/new-zealand-needs-a-traffic-light-system-to-stop-covid-19-creeping-in-at-the-border-149262 (accessed Apr 2021).
29. Northern Territory Government. Mandatory supervised quarantine for repatriated Australians 2020. https://coronavirus.nt.gov.au/travel/quarantine/mandatory-supervised-quarantine-for-repatriated-australians (accessed Apr 2021).
30. Swadi T, Geoghegan JL, Devine T, et al. Genomic evidence of in-flight transmission of SARS-CoV-2 despite predeparture testing. Emerg Infect Dis 2021; 27: 687-693.
31. Brooks JT, Beezhold DH, Noti JD, et al. Maximizing fit for cloth and medical procedure masks to improve performance and reduce SARS-CoV-2 transmission and exposure. MMWR Morb Mortal Wkly Rep 2021; 70: 254-257.
32. Sickbert-Bennett EE, Samet JM, Prince SE, et al. Fitted filtration efficiency of double masking during the COVID-19 pandemic. JAMA Intern Med 2021; Apr 16: e212033.
33. Cheng D. Covid 19 coronavirus: Chafing under the rules - 76 bubble breaches in four months. New Zealand Herald (Politics) 2020; 28 Nov. https://www.nzherald.co.nz/nz/covid-19-coronavirus-chafing-under-the-rules-76-bubble-breaches-in-four-months/5WLODXO7S4SBCZY7UWBWKSZOUI/ (accessed Apr 2021).