September 06, 2020
Dr Rawiri Taonui | Covid-Maori | Searching for the Source of the Auckland OutBreak | 6 September 2020
Dr Rawiri Taonui | Covid-Maori | Searching for the Source of the Auckland OutBreak | 6 September 2020
This column examines six scenarios for the source of the Auckland OutBreak. Searching for the source of the 11 August Covid-19 Auckland Outbreak is an important part of halting the spread of the virus. It is also important for understanding how to minimise future outbreaks. Integral to this is understanding the parameters and limitations of swab tests, genome sequencing and serology tests.
THE SEQUENCE OF INFECTION
The Earliest Case
The South Auckland community cluster began with a worker at an AmeriCold storage facility in Mt Wellington. The worker became sick on 31 July. While it can be longer or shorter, five to six days is the usual incubation period for Covid-19, the time between exposure to the virus and symptom onset. This means the worker was infected around 25/26 July. This worker’s infection is the earliest in the cluster. All other known infections are thought to emanate from this person.
The Source Case
The AmeriCold worker is not the source case of the outbreak. They were infected by someone else we can call the ‘source case’. This person was themselves infected around 19/20 July then passed the virus to the AmeriCold worker.
The source case may have been part of a chain or multi-generation sequence of cases before the infection passed to the Americold worker. Each generation, the passage of infection from one case to another, is usually 5-6 days.
A chain of infection can remain one person to another for six to seven generations (30 to 40 days). If that is the case here, the origin of the sequence could extend back to the beginning of July or even mid to late June. However, by the time chains reach 10 to 12 generations multiple simultaneous cases appear. This increases the chance of detection because more likely than not someone has become symptomatic and seen a doctor, or they are found through random testing.
UNDERSTANDING TESTING
Swab Testing
Three types of testing are being used in the battle against Covid-19. Swab testing has an accuracy of about 80% meaning that up to 20 to 30% can produce ‘a false negative’.
False negatives can relate to the timing of the test. If a sample was taken too soon after someone was exposed to Covid-19, there may not be enough of the virus in the nose or throat to be detected. Similar can happen at the other end of the illness. A person feels better and has less virus left from their resolving illness, and the test misses that altogether.
Human error is another cause of false negatives, for example, if the swab is not inserted far enough into the pharynx (the mucous membrane behind the nose and mouth), then the sample may not be adequate.
The double testing regime New Zealand conducts on day 3 and 12 in MIQ (Managed Isolation and Quarantine) reduces the risk of a false negative to around 4%. The 14-day isolation provides insurance against that because it is thought that this period covers first infection, sickness, and recovery.
Some countries think the risk period is longer. China, for instance, requires 21 days isolation for those entering the country, including entry conditional on evidence of a negative test from a Chinese government-approved testing provider in the country of origin (through one of their 286 embassies and diplomatic offices), 14 days isolation in a facility and 7 days self-isolation.
The government has made much of the more 305,000 tests conducted since the outbreak began on 11 August. However, swab testing, while integral in detecting new cases and mapping the perimeter of the outbreak, has limits in detecting early cases because 99% of those tested two weeks after recovering will test negative. The person who infected the AmeriCold case would have recovered around the end of July and now test negative.
Genome Sequencing
The Institute of Environmental Science and Research (ESR) is genome sequencing (tracing genetic versions of the virus) on all cases that return positive for Covid-19 by comparing them with a genome database of previous cases.
The ESR has identified a BTriple1 (or B.1.1.1) strain of Covid-19 at the centre of the Auckland OutBreak. Most frequently documented in Britain, Australia and to a lesser extent in Peru, Uruguay, South Africa, Switzerland, and Bangladesh. Otago University scientist Dr Jemma Geoghegan has said that the strain in New Zealand is most closely related to England.
Swab testing will only show whether someone is positive or negative. It will not identify the strain of the disease. Genome sequencing will only follow a positive test. Given the probability that the source transmission occurred more than a month ago, the source case will test negative and therefore never be sequenced.
If swabs do not produce enough viral protein, then only a partial genome sequence is possible. About 40% of positive tests in MIQ do not produce sufficient RNA material to be sequenced. There are also holes in our database with only 650 of the 1504 cases in the first wave sequenced. The government has said that the BTriple1 strain has ‘not been seen in New Zealand’. David Welch an epidemiological researcher at AUT has said there were two BTRiple1 cases in MIQ in April.
Serology Testing
Serology tests are blood tests that detect antibodies. This will reveal if someone had the disease and has now recovered or whose test was a file negative. They do not identify the strain of the disease because they do not test for the viral proteins.
Serology tests also have a problem producing more ‘false positives’ than true positives by two-to-one. The false leads complicate follow-ups so much that serology tests are better as surveys of a wider population rather than the intimate tracking of a sequence of cases.
LIKELY SOURCE OF THE AUCKLAND OUTBREAK
1. Breach in MIQ
In this scenario, the source case was in MIQ and has passed Covid-19 to a border or MIQ worker and the virus made its way to AmeriCold. The host case might have been a false negative and therefore was not genome sequenced. Alternatively, they may have tested positive but did not produce enough virus to be genome sequenced hence BTriple1 remained invisible. This far on from July, both the source case and the border or MIQ case would show negative on a swab test.
This scenario coincides with an absence in random testing during July, an upward surge in numbers entering MIQ from the last week of July and a concomitant rise in the number of positive cases appearing in MIQ.
2. Air and Maritime Crew Untested at the Border
In this scenario, the virus could have been passed by an Air New Zealand or other foreign passenger or freight aircrew, the majority of whom are not tested upon entering New Zealand.
Similar applies to maritime crew who are required to self-isolate on their vessels but can disembark to unload cargo. There is also no evidence that security around maritime crews has been particularly tight or that any random testing of vessels in port was conducted, their isolation relying on self-recognisance. The AmeriCold cases have links with Tauranga Ports. A source case would now test negative because of time elapsed since July.
If such transmission occurred, it would follow the Rydges MIQ example. The source case could have flown or sailed to New Zealand, walked through customs at an airport or been on a ship in port, infected a border worker, a staff in MIQ, or someone at a wharf. That intermediary passed it to the AmeriCold worker. The intermediary would now test negative.
3. False Negative
A third possibility is a case slipping through MIQ as a false negative. Swab testing can have up to a 20 to 30% error. As above the double-testing New Zealand conducts on day 3 and 12 of isolation reduces this error to 4%. The 14 days of isolation is regarded as a good safeguard.
However, as someone in isolation rather than quarantine, they could have passed it to a MIQ worker from where it made its way to AmeriCold. Given the time elapsed this case would now test negative.
4. June Transmission
It is possible that Covid-19 was passed from an untested source case in MIQ. Up to 1360 people were released from MIQ between 9 and 16 June without being tested.
While the majority were minimal risk because they had completed 14 days isolation, over 50 were released on compassionate grounds without completing their full 14 days. Transmission from this source could have taken five or six generations to when the AmeriCold worker was infected.
The 1360 were followed up and testing conducted, however, based on the last report from the Ministry of Health, 277 were not found, 161 were found and not tested and 97 refused testing. This is important because one of the 50 may have carried the BTriple1 strain.
UNLIKELY SOURCES OF THE AUCKLAND OUTBREAK
Long Incubation
One possible but unlikely scenario relates to studies suggesting the incubation period for Covid-19 can be longer than 14 days. China has reported two cases where the incubation period was 17 days. This is one reason they require 21-days isolation for incoming arrivals (14 days in MIQ and seven in self-isolation). For this to have occurred, the disease would have lain latent and undetected in our MIQ for 14 days only to emerge after the subject left isolation.
Long First Wave Transmission
The BTriple1 lineage has been seen in New Zealand in two cases in MIQ in mid-April. However, transmission from this source is unlikely because this would require both a leak from MIQ and undetected transmission for 12 or more generations over 120 days until late July.
Noho haumaru, stay safe and self-sovereign
Copyright © 2020, UMA Broadcasting Ltd: www.waateanews.com