Here's what an analysis of arrivals testing in Hong Kong tells us about chances of COVID-19 cases on-board your flight

I decided to study this as part of a research paper I am doing for my IB (International Baccalaureate) degree at the American Embassy School, New Delhi, using data sourced from the Hong Kong government, airport, news reports, and Fightradar24.  The data covered approximately 5400 passenger flights that arrived in Hong Kong International Airport (HKIA) in October and November 2020, from 36 countries on 52 different airlines and from 70 originating cities.  Over this period, a total of 253 passengers tested positive on entry into the region.

In addition to testing all arriving passengers, Hong Kong required passengers arriving from or transiting through what they classified as “high-risk countries”, namely Bangladesh, Belgium, Canada, Ecuador, Ethiopia, France, Germany, India, Indonesia, Kazakhstan, Nepal, Pakistan, Philippines, Romania, Russia, South Africa, Turkey, the United Kingdom, the United States, and Ukraine, to take RT-PCR tests within 72 hours prior to departure as well in the originating countries, and board only after the airline checked for the negative test result. Passengers from these countries are then required to take another test after landing.

The findings were very interesting. The analysis showed that every flight from Nepal and India carried COVID-positive cases, while half the fights from Turkey and a quarter of the flights from Germany and Qatar had positive cases. On average, a flight from Nepal had 10 positive cases, and from India, 4, despite passengers producing COVID-negative test results on departure. (See Table 1 for the 10 worst performing countries)

While the table above shows the 10 worst performers, several countries – including Taiwan, China, and Australia – had zero infected originating passengers. These countries are known to have contained COVID-19 more effectively.

Oddly, the US had just 4 infected passengers across 119 flights, performing much better than one might expect.  However, this is likely a result of fewer passengers on each flight going through more reliable screening facilities available in the USA.

In terms of originating cities, Kathmandu plus all four Indian originating cities had at least one infected passenger on every flight. On average, flights from Mumbai and Chennai had 5 positive cases each, Delhi 4, and the sole Hyderabad flight had 1. Once again, this was after passengers submitted a COVID negative test result on departure. (See Table 2 below for the 20 worst performing cities.)

In terms of worst-performing airlines, Nepal Airlines, along with the three Indian airlines that operated flights to Hong Kong (Air India, Vistara, and SpiceJet), had infected passengers on all flights. Royal Flight Russia also earned this dubious distinction. Nepal Airlines had an average of 10 infected passengers per flight, Air India 5.5, Vistara 2.5, and Royal Flight and SpiceJet 1 each. Nepal Airlines and Air India also got banned by Hong Kong 3 times each in the Oct-Nov 2020 period, and Vistara once, for breaching acceptable thresholds of infected passengers on their flights.

It should be noted that Qatar Airways and SWISS (the 9^th and 11^th worst performer overall respectively) also managed to get banned once each during this period. Each ban was for 14 days (see Table 3 below).

What can we take away from all this?

Unsurprisingly, many of the worst performers came from some of the countries that are worst hit by COVID in absolute terms. Intriguingly, most of the countries, cities, and airlines that rank highest on the list also had their originating passengers tested within 72 hours of departure and tested negative. This suggests:

1. Testing may not be very thorough or standardized in these countries, OR


2. Test results are falsified (this is a possibility; Dubai recently blacklisted several labs in India after seeing a pattern of negative test results from those labs for positive cases) OR


3. People caught the virus in the 72 hours after testing, or on the flight (less likely, as it typically takes 5 days for the virus to be detectable in tests)

Extrapolating the findings and relating it to air travel in India, the key learning is, there is a high likelihood of COVID-positive passengers, likely asymptomatic, on India flights, given there is no pre-flight testing requirement for domestic flights (and noting again the India-Hong Kong cases were found despite the fact all departing passengers from India were supposedly pre-tested and COVID-free).

This should not unduly alarm readers: modern aircraft cabins and air circulation and ventilation systems are safe and will not spread viruses. However one can still directly breathe in air from infected passengers nearby, just like on the ground, therefore one must not let one’s guard down on flights and at airports – masks must always be properly worn covering both nose and mouth, and other COVID-safety norms must be followed at all times. As long as these precautions are taken, flying should be safe.

The author is Shiv Kapoor, a Grade 12 student at the American Embassy School, New Delhi. Shiv would like to acknowledge Danny Lee of the South China Morning Post for his assistance in obtaining raw data for the study.