Meeting #17 of the CSA COVID-19 Expert Panel

---

As part of the response to the COVID-19 pandemic, the Chief Science Advisor of Canada created the Expert Panel on COVID-19 to advise her on the latest and most relevant scientific developments. This information assists the CSA in providing current, cross-disciplinary and independent advice to the Prime Minister and government.

---

Overview of discussions

Held by teleconference on May 22, 2020

The following discussion reflects evidence and scientific knowledge up to May 28, 2020

Participating expert panel members:

• Mona Nemer PhD, Chief Science Advisor of Canada (chair)

Disease modelling

• Kamran Khan MD, St Michael’s Hospital
• Caroline Colijn PhD, Simon Fraser University

Risk and behavioural sciences

• Louise Lemyre PhD, University of Ottawa
• Stephen Taylor PhD, University of British Columbia
• Daniel Krewski PhD, University of Ottawa

Biomedical and clinical sciences

• Maziar Divangahi PhD, McGill University
• Deborah Cook MD, McMaster University
• Gary Kobinger PhD, Université Laval
• Joanne Langley MD, Dalhousie University
• Allison McGeer MD, Mount Sinai Hospital
• Samira Mubareka MD, Sunnybrook Research Institute
• Supriya Sharma MD, Health Canada
• Cara Tannenbaum MD, Health Canada

Other

• Masha Cemma PhD, Office of the Chief Science Advisor (support)
• George Enei P. Eng, Office of the Chief Science Advisor (support)
• Vanessa Sung PhD, Office of the Chief Science Advisor (support)

Participating guests:

Canadian Border Services Agency

• Denis Vinette, Vice President, Travellers Branch

Transport Canada

• Michael Keenan, Deputy Minister
• Kevin Brosseau, Assistant Deputy Minister, Safety and Security, Transport Canada

Context

• Currently, the Quarantine Act requires all travellers arriving in Canada to undergo a 14-day quarantine.
• Detailed information is currently being collected from travellers, which includes the place(s) they are visiting, their planned activities, and whether or not they have a suitable self-isolation plan.
• Travellers who do not have access to appropriate self-isolation facilities are invited to stay at federal quarantine sites.

Expert panel deliberations

Criteria and information required for easing travel

Information about countries of origin

• Borders could be differentially managed, for example, considering stably decreasing COVID-19 cases in countries. However, regional heterogeneity/ COVID-19 hotspots, regional diagnostic testing capacity and quality, and potential information lag concerning infection rates must also be considered.

Travel corridors (or coronavirus “bubbles”)

• Travel corridors would allow incoming travellers from countries or regions with similar levels of risk to move freely without mandatory self-isolation upon arrival. Pre-travel measures would remain implemented.
• If implemented, travel corridors should be dynamically updated using evidence of the COVID-19 cases and hotspots both domestically and internationally.

Situation in Canada

• Two key considerations:
  • In a given jurisdiction, how close the COVID-19 situation is to a threshold after which the growth of infections would become exponential.
  • The resilience of the healthcare system in the face of new COVID-19 cases.
• The more a jurisdiction uses its leeway to reopen schools and certain industries, the riskier it becomes to allow travellers to enter.

Information about the traveller

• Information on the occupation, living conditions in Canada and purpose of visit is important, as the risk of an outbreak caused by a leisure traveller is different from a traveller who works or lives in a congregate setting (e.g., agricultural workers, mining or logging workers, meat plant workers, etc.).

Pre-travel measures

Promote responsible behaviour

• Build in flexibility to the travel experience to avoid undesirable behavior.
  • If flights could be changed or cancelled without a penalty, travellers would less likely travel when they are ill.
• Knowledge of random testing at the border (if this is indeed the case) may sway travellers with symptoms to postpone travel plans.

Pre-registration / health declaration (24-48 hours ahead of travel)

• Self-declaration of health status could be incorporated into the check-in portal (similar to declaration of firearms and other goods) as a pre-travel measure.
• Consideration should be given to repercussions for travellers who provide false declarations.

Providing evidence of COVID-19-free status through virus testing

• Testing needs to be easily available to Canadian residents and foreign nationals/travellers across Canadian provinces and territories.
• The closer the test result is to the day of travel, the more valuable it is. A requirement for a test result within 24 h would reduce risk of transmission, but that is difficult and costly to deliver. Results within 48-72 h before travel, together with absence of symptoms, could be considered.
• There is a wide range in the sensitivity of available tests, therefore careful consideration needs to be given before accepting the results. Details of the testing platform used, together with a list of acceptable platforms, could be considered.

Measures at the border

Temperature screening

• This method has not been shown to be reliable for detecting COVID-19 and other respiratory diseases. Many unrelated conditions can lead to an increase in temperature such as arthritis, sunburn, and anxiety among others.

Virus testing

• Consideration may be given to a pooled-sample approach to mass screening of flight passengers.
  • This method has good predictive value and conserves resources.
  • It could be an effective way to monitor but the repeat assays needed with individual samples to confirm an infected person require a large amount of swabs. The rate of false negatives using this approach is not well-known. A sample collection methodology that does not require a health care worker (e.g., spitting or self oral swabbing) would need to be developed.

• Random testing
  • Knowing that one may be randomly tested for COVID-19 could act as a deterrent to falsifying information about health conditions.
  • Logistically, this might be challenging to organize at land crossings, but feasible at airports and, perhaps, ports.
  • PCR testing could be done in parallel with serological testing for more information.

• Testing before, during and after travel
  • Testing could be given (1) within 3 days before travel, (2) upon arrival, and (3) after 2-4 days of self-quarantine. This is based on the knowledge that people are typically infectious 2-3 days before symptom onset and around 7 days thereafter. This approach would reduce the number of days required for quarantine. Considerations for this approach:
    • For travellers with low pre-test probability (e.g., asymptomatic travellers from countries with low COVID-19 activity), the positive predictive value of the test is low.
    • For travellers with high pre-test probability (e.g., travellers from countries with high COVID-19 activity), it may be more efficient to leave border restriction in place.
    • It is possible for a traveller with very recent exposure to COVID-19 to have an early infection that is not yet detectable by PCR testing. Incubation and transmission periods can vary.
    • Viral transmission can happen during travel and the likelihood increases for indirect routes.

Health information

• Provide incoming travellers with information on potential symptoms of COVID-19 and what they should do if symptoms appear (i.e., how to get tested for COVID-19 or interact with the healthcare system).

Immunity passports/ immunity certificates

• Natural immunity to SARS-CoV-2 virus -- how robust it is and how long it lasts -- is not yet understood, so immunity passports are not presently helpful.
• The requirement for confirmation of vaccination to enter certain countries (e.g., Yellow Fever) exists, so conceptually this approach is feasible if it integrates into the existing vaccination record systems. However, requiring confirmation of vaccination for travel when vaccines are not widely available leads to equity issues. Once vaccines are broadly available, immunity certificates may become relevant.
• Another way is to funnel this through existing paid processes such as Nexus or visa services. Once the science of SARS-CoV-2 immunity is better understood, short-term (e.g., 3 months) individual immunity could be certified through accredited laboratories.

Post-arrival measures

Screening

• Measures could be differentially and proportionally applied based on:
  • COVID-19 activity in the country of origin, where either the border is kept closed to countries with higher COVID-19 cases (hence, higher pre-test probability) or citizens from these countries will be required to undergo more rigorous measures.
    • Information may lag between countries, affecting this assessment.
  • Intended activities or living circumstances in Canada, where working or living in congregate settings (e.g., meat-packing, mining, oil well) or living in a vulnerable community (e.g., Northern community or population with underlying conditions and/or limited heath care services) will require more rigorous measures.
• Depending on a traveller’s risk profile, interventions could include a combination of testing, self-monitoring, mandatory quarantine, a recommendation to wear facemasks, and post-arrival tracking and communication.
• Any approach should take into account Canada’s ability to manage increased number of COVID-19 cases.
• Risk stratification of countries of origin could streamline the process by focussing on the countries from which Canada receives the highest volume of travellers.
  • A limitation to this approach is the delay of information sharing with respect to regional change in disease spread.
  • Bovine spongiform encephalopathy (mad cow disease) is a historical example where countries were categorized based on risk level^{Footnote 1}.
  • This approach would require the collection of a significant amount of information from travellers.

Tracking and communications

• Current update on ArriveCAN app^{Footnote 2}:
  • The app was first developed as a way to digitize the traditional paper process of border entry.
  • The next iteration of the app will include a functionality for travellers to complete health check-ins, as well as the ability to push social and health messaging to encourage compliance. The next iteration is scheduled for the end of June.
  • The use of geolocation data is not under consideration due to privacy concerns.
  • Provinces have also expressed interest in the information collected from the app.
  • Discussions are ongoing about how to increase user engagement for the app, while consideration is given to people who do not own a mobile device or do not have access to the Internet.
  • The app use, which is currently optional, could be mandatory under the Quarantine Act.
• CBSA’s ArriveCAN app could be an effective mechanism to tie together traveller health declarations with the ability to interact with travellers post-arrival.
  • Travellers could input daily updates on their condition and symptoms.
  • Leveraging behavioural science could help to optimize user engagement.
  • Information could be shared on COVID-19 symptoms and how to interact with Canadian healthcare system if symptoms are developed.
• Tracing or tracking of international travellers and citizens for 14 days upon arrival in Canada. Considerations for this approach:
  • Ability of current technology (i.e., cell phones) to provide precision is needed.
  • Social acceptability and potential privacy/ethical issues with the use of geolocation.
  • Some people do not own mobile devices.
  • Legal framework.

Reducing exposure risk during travel

• Update from Transport Canada on measures at the airports:
  • Transport Canada is in the process of developing overarching guidance for airports, with recognition that airports are configured differently and vary in passenger volumes.
  • Most airports have instituted frequent sanitation and hygiene practices, including sanitation measures at the security check.
  • Airports are attempting to reduce crowding via lineups at checkpoints (e.g., security, customs) to maintain physical distancing. This is feasible now because of low passenger volume, but will become problematic once travel restrictions are lifted.
  • The following questions need to be addressed:
    • How can exposure risk be reduced in light of the space limitations in airport halls?
    • How can large volumes of passengers be moved through checkpoints safely?
    • Can the required distance between passengers (i.e., 2 meters) be lowered by limiting exposure time in conjunction with mandatory facemask use?
    • What processes could be added to enable some degree of crowding that would be acceptable for a short and controlled period of time?
• Considerations for reducing the distance between passengers from 2 meters to 1 meter:
  • Risk of exposure is stable over time, so reducing time of exposure does not necessarily reduce risk – depending on the physical arrangement of the lineup (straight vs serpentine line).
  • The World Health Organization recommends maintaining a 1-meter distance. The risks associated with a 1-meter distance compared to a 2-meter distance could be modeled, but the results would be difficult to generalize because they depend on variables such as the heating, ventilation and air conditioning systems (HVAC).
• To reduce passenger exposure upon disembarking:
  • Airports may choose to limit passenger mixing by managing queues and baggage claim by flight and country of origin, rather than by time of arrival.
  • There are conflicting perspectives on whether a separate line for travellers with the verified immunity would be beneficial. On the one hand, this would accelerate the process of going through security and reduce both the duration and length of the line; but on the other hand, individuals with verified immunity may serve as buffers between susceptible individuals, thereby limiting potential spread of the virus.
  • Passengers accept the risk of traveling in close proximity of each other by virtue of travelling. An alternative to lineups at customs could have customs officers at the entrance of planes, calling passengers up one by one. Officers could also board the planes to collect information or scan passports.
  • Airlines could keep spare empty rows in case they need to isolate passengers who become symptomatic mid-flight.
• There is a concern that implementation of “immunity certificates” based on immunity due to prior exposure (once verified) could incentivize people to intentionally get infected.
• Simulation models using anonymized people paired with computational fluid dynamics may be a useful tool for determining transmission risks associated with travel.
• Canadian COVID-19 Genomics Network (CanCOGeN) could prioritize viral genomic sequencing of returning travellers. The goal would be to follow new introductions of the virus and track potential genetic mutations as it travels.

Annex A – Literature scan on the use of temperature checks and other border measures

Updated on May 28, 2020

Evidence on the use of temperature checks for screening COVID-19

• An investigation by CNN in Feb found that after 1 month of temperature screening at 11 US airports, and 30,000 passengers screened by the CDC, not a single case of COVID-19 was caught by airport temperature checks (Feb 20, 2020)^{Footnote 3}.
• Effectiveness of airport screening at detecting travellers infected with novel coronavirus (2019-nCoV) (Eurosurveillance Feb 6, 2020)^{Footnote 4}.
  • Modeling finds that temperature checks will fail to detect cases nearly half (46/100) the time.
  • Authors estimate that syndromic screening at airports to prevent infected travellers from entering is only achievable if the rate of asymptomatic infections that are transmissible is negligible, screening sensitivity is almost perfect, and the incubation period is short.
• Thermal image scanning for influenza border screening: results of an airport screening study^{Footnote 5}.
  • Infrared thermal imaging scanners performed moderately well in detecting fever but the proportion of influenza-infected travellers who were febrile was low – thermal scanners were not much better than chance at identifying infected travellers.
  • Authors suggest that thermal screening is unlikely to be effective at preventing entry of the virus into a country.
• International travels and fever screening during epidemics: a literature review on the effectiveness and potential use of non-contact infrared thermometers^{Footnote 6}.
  • The evidence does not support high efficacy of temperature screening to detect symptomatic individuals. If fever prevalence is fixed at 1% in all studies to allow comparisons, the derived positive predictive value of non-contact infrared thermometers varied from 3.5% to 65.4% and the negative predictive value was >=99%.
• Thermal image scanners to detect fever in airline passengers, Vancouver and Toronto, 2003^{Footnote 7}.
  • Thermal imaging scanners were deployed for both arriving and departing passengers from mid-May to mid-December 2003. An elevated temperature was detected in 1435 people (0.031%). No person was assessed as having SARS.
  • Note thermal image scanners were not put into operation in Canadian airports until after the last traveller with SARS entered Canada on April 1. Of persons who were exposed to SARS in Canada and then travelled and became ill elsewhere, none was symptomatic when departing this country.
• SARS in Hong Kong: From experience to action^{Footnote 8}.
  • 36.3 million people underwent screening by thermal scanner. 1921 (0.00053%) were febrile, 40 of these people were admitted to hospital but none were diagnosed with SARS.

Additional literature on border entry screening

• Effectiveness of interventions targeting air travelers for delaying local outbreaks of SARS-CoV-2^{Footnote 9}.
  • A mathematical model assessed the efficacy of three syndromic screening intervention scenarios: 1) no screening, 2) exit-only, and 3) exit-and-entry screening.
  • Exit and entry screening in combination with traveller sensitization can delay a local outbreak by 8 days (50% CI: 3-14 days) when the rate of importation is 1 infected traveller per week at time of introduction.
  • If there are 10 infected travellers per week at the time when this is introduced, the outbreak is delayed by only 2 days (50% CI: <1-5 days). Also, the additional benefit of entry screening is small if exit screening is effective as the combination of only exit screening and traveller sensitization can delay an outbreak by 7 days (50% CI: 2-13 days).
• Imported novel coronavirus infections: Observation on active and passive case detection in Thailand^{Footnote 10}.
  • Active screening for COVID-19 was conducted at Thailand international airport, and 21,374 travelers were screened. 4 imported cases of COVID-19 were detected at health screening points at immigration posts at international airports.
• Estimated effectiveness of traveller screening to prevent international spread of novel coronavirus (2019-nCoV)^{Footnote 11}.
  • Under best-case assumptions, authors estimate that screening will miss more than half of infected travellers.
• Evaluation of border entry screening for infectious diseases in humans^{Footnote 12}.
  • Based on modeling and reports on the H1N1 and SARS experiences, authors conclude that border screening is not effective at detecting infectious persons. Moreover, border screening has high opportunity costs, financially and in terms of use of scarce public health staff resources during a time of high need.
  • Outbreak-associated communications for travelers at border entry points, together with effective communication with clinicians and more effective disease control measures in the community, may be a more effective approach to international control of communicable diseases.

Annex B. Border measures: travel bubbles and COVID-19 testing and tracing at international airports

Updated on May 28, 2020

Travel bubbles

The idea is to reopen borders between countries with comparable epidemiological situations and where sufficient capabilities are in place in terms of hospitals, testing, surveillance, and contact tracing.

EU (EC’s guidelines and recommendations on safe resumption of travel)^{Footnote 13}.

• On May 13, the European Commission recommended a phased approach to the reopening of border. This means that EU countries with “similar overall risk profiles” on the pandemic will open to tourists from each other’s countries.
• Baltic nations Lithuania, Estonia, and Latvia opened their borders to each other on May 15, creating a so-called Baltic bubble. Anyone entering from outside the zone must still self-quarantine for 14 days. Anyone who has not travelled outside the Baltic states in the past two weeks, is not infected, and has not been in contact with somebody who has tested positive may travel freely to the other nations.
• Austria has announced plans to open borders to visitors from “safe countries” like Germany and the Czech Republic.
• Greece has proposed travel corridors with Cyprus, Germany, Austria, and non-EU member Israel.
• The UK and France are reportedly in talks to establish a travel corridor.

Pacific

• Australia and New Zealand have committed to a trans-Tasman travel bubble when it is safe.

Country/region-specific approaches to testing and tracking

Austria - Vienna International Airport – Testing & quarantine

• All travelers entering Austria must either present a medical certificate that confirms a negative PCR test results (not older than 4 days), otherwise they are required to begin a 14-day quarantine.
• Arriving travelers who have Austrian citizenship, or have a valid residence permit, without a medical certificate can get a PCR COVID-19 test at the airport, the results of which are available within 3-6 hours^{Footnote 14}. If test results are negative, they can bypass self-quarantine. Exiting travelers can use their negative test results as a certificate of health to enter their destination country, depending on their entry requirements.
• If tests results are positive, the passengers and public authorities are notified.
• The test is a private service which costs €190 ($208) for passengers either coming or leaving Vienna, and is by appointment only.

Japan - Tokyo Narita International Airport – Testing

• Officials have introduced on-site COVID-19 testing for arriving passengers from widely infected countries like Italy and the U.S., who are required to undergo a virus check.
• The test can take anywhere from six hours to two days for the test results to come in. Arriving travelers who undergo the test aren’t allowed to use public transportation, so unless they arrange for someone to pick them up, they have to stay in the airport to await test results.
• Tokyo Narita has arranged some special accommodations for travelers who must wait in its baggage claim area. Those accommodations consist of special temporary “beds” made of heavy-duty cardboard, along with a mattress and a quilt. The system probably works for now because the number of arriving international passengers is currently so small.

Hong Kong - Hong Kong International Airport – Testing & tracing

• Mandatory COVID-19 testing for all travelers arriving at its international airport (first to implement mandatory testing)
• Passengers coming in from countries that are considered high-risk will be given tests that can produce results in about eight hours, and they’ll have to wait out that time at a temporary testing facility in an exhibition center near the airport.
• Those coming from lower-risk places will get a different test that may take days to produce results. They won’t have to stay at the airport, but they’ll have to wear a tracking wristband that officials can check on to make sure they self-isolate.

South Korea - Incheon International Airport – Testing & tracing

• Walk-through testing facilities for travellers with no symptoms.
• Up to 2000 people can be tested daily at 16 open-air testing booths installed in 5 locations at the airport’s 2 terminals. Roughly that means testing 1 person every 5 min.
• Those with symptoms, regardless of nationality, are separated and tested at quarantine facilities set up inside the airport.
• Among incoming travelers from Europe without symptoms, Korean citizens and foreign nationals with long-term visas are required to get tested within three days upon arrival and self-quarantine for 14 days. Asymptomatic travelers from the US don’t have to be tested, but are required to self-quarantine for the two-week period.
• Foreigners will be deported if they breach the rules. Adherence is monitored with a mobile application. No one can enter Korea without downloading the app.

Taiwan - Taoyuan International Airport – Tracing

• Relies on monitored, mandatory quarantine system, not testing.
• Upon arrival in Taiwan, travelers are required to fill out a paper or online form to state their personal information and accommodation plans for quarantine, which could be either one’s home or a government-assigned hotel. An official double-checks information given and takes down the individual’s phone number, dialing it on the spot to ensure that it is correct.
• Taiwan also created a health questionnaire that passengers accessed by scanning a QR code with their phones while still on planes as they were arriving.
• Depending on their health status, they were sent a text message providing them with a health declaration pass to fast-track through immigration. Anyone at a higher risk of infection was urged to go into self-isolation at home and tracked via their cellphone to make sure they stayed there. False reporting of health information is a fineable offence.
• Those passing through the health inspection are required to leave the airport either by driving home, being picked up by family, or by use of a special quarantine taxi.
• Returnees would receive a special care package and two phone calls per day during the quarantine. They are reminded that their phone is being tracked, and instructed to report any symptoms.
• The mobile phone tracking system automatically alerts authorities when a quarantined person leaves their address. Police respond within 15 minutes when someone triggers an alert.

UAE - Dubai International Airport – Testing

• Emirates Airline working with Dubai Health Authority to take passenger blood samples at the airport before departure. Results can be obtained within 10 min.