UCLA experts dispute effectiveness of airport screening in detecting coronavirus
By Annie Yeh
Feb. 20, 2020 1:13 a.m.
Airport screenings may only catch half or fewer travelers infected by the 2019 novel coronavirus, UCLA experts said in research that is still under peer review.
The 2019 novel coronavirus, or COVID-19, was first reported in Wuhan, China, in December. As of Wednesday, there have been more than 75,000 confirmed cases of the virus in 26 counties and more than 2,000 deaths, according to the World Health Organization. The United States has seen 15 confirmed cases of the virus, none of which were caught by airport thermal screenings, according to CNN.
By adapting previously researched models of fever screening used during past outbreaks, such as the 2009 H1N1 flu pandemic and the 2014-2016 Ebola virus outbreak, UCLA experts found that airport screenings may not be effective in preventing this new outbreak. The results from their study are currently available as a preprint on medRxiv, but the research is under peer review.
The term “coronavirus” refers to a family of viruses that vary from causing mild illnesses, such as the common cold, to more severe diseases, such as COVID-19 and severe acute respiratory syndrome, said Ana Gomez, the second author of the study and an ecology and evolutionary biology graduate student at UCLA.
When outbreaks of unknown diseases occur, one of the measures to contain the spread of the infection is to implement airport screenings, said Katelyn Gostic, lead author of the study and a postdoctoral researcher at the University of Chicago. But these measures may not effectively identify all travelers carrying the disease, based on research of previous viral outbreaks, she said.
Before the new coronavirus outbreak, the research team built a model using characteristics of different diseases from outbreaks to try to understand why airport screening is ineffective, said Gostic, a UCLA alumna.
The team updated the model to include characteristics of COVID-19 to determine how likely airport screenings are to detect infected travelers, Gostic said.
These screenings may vary, but they typically consist of a written questionnaire and infrared thermometers to take a traveler’s temperature, Gomez said.
The model and calculations were inputted into an interactive tool that allows users to change certain parameters to get an updated result online of how effective traveler screening will be, Gomez said.
The model is currently available for public use, Gomez added. And this allows users to log in and choose their own inputs to alter the models in real time, Gostic said.
“Our hope is that by updating models so that it’s specific (to) this current virus outbreak, it might be useful to people who are making decisions about how to prevent the outbreak from spreading,” Gostic said.
The research’s overall conclusion was that even best-case scenarios, such as travelers disclosing when they have had contact with people with infections, resulted in screenings missing infected cases more often than detecting them, said James Lloyd-Smith, a senior author of the study and a professor in the department of ecology and evolutionary biology at UCLA.
“And, if you make more realistic or pessimistic assumptions, it’s down to a success rate of 25% or even lower,” Lloyd-Smith added.
One of those assumptions is that a fraction of cases are mild, lacking detectable symptoms, but still capable of transmitting the virus, he said.
Additionally, a large number of cases never show symptoms. This would make it even more difficult to detect travelers using conventional screening methods, Lloyd-Smith said.
“It means that, for one thing, for every one individual who is actually detected by these kinds of screening methods, you should expect there will probably be at least one and maybe two or more other individuals who were not detected out in your community,” Lloyd-Smith said.
However, Gostic said these numbers should not raise alarm bells, because the model operates under the assumption that individuals are infected and traveling with the coronavirus.
“This is just the idea of knowing that having traveler screening doesn’t completely protect your country or your city from a potential infection spreading,” Gomez said.
Additionally, the UCLA study is still preliminary and under peer review. MedRxiv, the archive hosting the research, said the results should not be used to guide medical practice.
It became clearer upon finishing the study that a large portion of individuals will only manifest mild symptoms, Lloyd-Smith said.
“This does not mean that we are criticizing how these programs are being done,” Lloyd-Smith said. “We’re just saying that by the biology and epidemiology of this virus, we don’t expect the programs to be super effective at catching everyone.”
This is also largely due to the fact that the virus will incubate in a patient for a period of time after exposure, during which time the patient will exhibit no symptoms, Gostic said.
“The moment you get exposed to a virus is not the same moment that you’re going to develop a fever and start sneezing and coughing,” Gostic said.
This is why, from a purely scientific perspective, quarantine is effective, because people with the virus will show symptoms by the time the two weeks are over, Gostic said.
Part of the speed of viral transmission is also up to chance, even if the virus is very infectious. If a single infected person is introduced in a new environment, there is a chance the individual will not spark a sustained chain of infections, Gostic said.
This could just be because the person does not interact with others before recognizing they are ill, she added.
“We have tools in place to try to stop the transmission before it starts, even if there are a few sparks,” Gostic said.
The fact that the disease is not spreading uncontrollably in the United States is a testament to the fact that it is possible to break up chains of transmission, Gostic said. This may just depend on people modifying their behaviors when they recognize they are sick, such as by staying home when they are sick, washing hands and maintaining other basic habits to keep healthy, she said.
Although there have been some worrying signs of coronavirus spreading in America, such as patients with no travel history getting infected, there is no current threat to UCLA, Lloyd-Smith said.
“There’s no need for any kind of specific overreaction to this particular virus right now,” Lloyd-Smith said. “Should it establish spread in the U.S., then it’ll be time to look at things more closely. But for the time being, just wash your hands a lot, have good hygiene, and don’t sneeze on your friends.”