Explained: Why quarantine works better than airport screening

Karishma Mehrotra
Explained: Why quarantine works better than airport screening

National Guard personnel who carry out bio screening on arriving passengers walk around the Luis Muñoz Marin Airport in an effort to detect and isolate the new coronavirus, in Carolina, Puerto Rico, Tuesday, March 17, 2020. (AP Photo: Carlos Giusti)

On Monday, the Indian Council of Medical Research (ICMR) published research from February 27 that attempts to “identify rational intervention strategies that might work towards control of the outbreak in India” of COVID-19. The research compares the effectiveness and feasibility of two approaches — attempting to contain an outbreak at the border, and quarantining symptomatic cases within the country.

What is the key finding?

The research makes a case for post-travel tracking rather than border containment. It uses mathematical modelling to show that spending resources on quarantining symptomatic cases can achieve a meaningful impact on the disease burden (assuming an “optimistic” scenario), rather than attempting to achieve infeasible levels of containment at the borders.

At the same time, it also accounts for the inevitability that an outburst of cases would make lab confirmations impractical. Therefore, the paper proposes “symptomatic surveillance” to be included with quarantine measures.

The government has focused on random sampling of patients with severe symptoms and quarantining positive cases. In the initial weeks of rising cases in India, asymptomatic travellers were not tested.

How does the paper interpret the feasibility of border containment?

The researchers found that if India screened all symptomatic airport arrivals from China, the epidemic would occur in 45 to 47.7 days. If all asymptomatic arrivals from China were screened, India would need to identify at least 75% of asymptomatic infected arrivals in order to achieve an “appreciable” delay in the outbreak. If 90% were identified, the delay would be 20 days.

The argument is that there would be little impact from addressing only symptomatic cases, but also that covering the necessary asymptomatic infections (almost all of them) is “practically infeasible”. Further, there is no accurate, rapid test to achieve the required detection levels, the paper notes, citing other studies to show that thermal screening can miss at least 46% of infections. The only way to achieve the needed detection levels, in fact, may be isolation of all arrivals from the specified airports.

Also read | Air travel suspended in India: What impact will it have on travellers, airlines?

“Any containment strategy focused on symptomatic infections, no matter how comprehensively tends to be negated by the asymptomatic infections that escape detection and can go on to cause onward transmission in the community... Resources may be better spent on the mitigation of infection in the community,” the paper says.

Explained: Why quarantine works better than airport screening

Click to enlarge

And how does it model the effectiveness of quarantining symptomatic cases?

The researchers built their model with two scenarios. The optimistic scenario assumes that each infected person transmits the virus to 1.5 other people (known as R0 or reproduction number) and that asymptomatic infections do not infect others. The pessimistic scenario assumes each infection transmits to four other people, and that asymptomatic cases are half as infectious as symptomatic cases.

A different study by scientists at the Institute of Mathematical Sciences in Chennai shows that India currently has a R0 around 1.7, much lower than other badly-hit countries. Research is beginning to show that asymptomatic cases do transmit the infection, but it’s still unclear by how much.

📢 Express Explained is now on Telegram. Click here to join our channel (@ieexplained) and stay updated with the latest

What are the findings from the ICMR paper’s models?

The models show that once community transmission occurs, the epidemic’s peak and duration can be greatly affected by quarantining symptomatic cases — but only in the optimistic scenario. This relates to a concept called “flattening the curve” — spreading the number of cases over time. This may not decrease the total number of cases, but would lighten the load on the healthcare system at any point of time.

The paper’s hypothetical model found that quarantining 50% of symptomatic cases within three days of their symptoms would reduce overall cases by 62% and the peak number of cases by 89% in an optimistic scenario. The intervention would reduce overall cases by 2% and the peak number by 8% in a pessimistic scenario.

Explained: Why quarantine works better than airport screening

India coronavirus lockdown: The paper proposes “symptomatic surveillance” to be included with quarantine measures. (Representational Image)

Which regions does the model factor in?

The model looked at Delhi, Mumbai, Kolkata, and Bengaluru, assuming that these metros would be the initial focal points of transmission. The paper uses another paper’s global risk assessment to deduce that the probability of an infected air traveller coming to India as the final destination is 0.209%, with the highest relative import risk in Delhi (0.064%) followed by Mumbai, Kolkata, Bengaluru, Chennai, Hyderabad, and Kochi.

Don't miss from Explained | Coronavirus lockdown: How prepared is India to meet the supply requirements of essential items?

The graph of Delhi’s outbreak is significantly higher than that of other cities because of these inputs. In fact, Maharashtra has more cases than Delhi, and Kerala is a significant focal point that the paper seems to have missed.

For Delhi, intervention leads to a projected decrease of cases from 150 per 10,000 to less than 25 per 10,000 at the epidemic’s peak in an optimistic scenario. In a pessimistic scenario, intervention reduces cases from over 1,000 in every 10,000 to under 1,000 in every 10,000. (see graphs left)

What are the other takeaways?

The paper cites other research that estimates eight to 10 severe and 40-50 non-severe COVID-19 cases for every death. Another paper looking at the Diamond Princess cruise ship estimates 26% of the entire population to get infected, and one in 450 infected individuals to die. The ICMR paper deduced that 5% of the infected will need intensive care, and half of those will need mechanical ventilation.

What are the limitations of the model?

The model assumes that cases are only coming from certain regions in China. We now know that many cases in India have actually come from the Middle East and the UK. It inputs only 500 daily arrivals from COVID-19-affected regions in China. Before travel from China was closed, 3,565 passengers were arriving from there every day.

The model also ignores urban-rural migration. It makes estimates about rail and road travel between cities. Other parameters about COVID-19 are admittedly not clear, such as the average duration of the infection, the incubation period, and the fatality rate.

Here’s a quick Coronavirus guide from Express Explained to keep you updated: Are smokers at high risk form coronavirus? | Can Vitamin-C prevent or cure coronavirus infection? | What exactly is community spread of coronavirus? | How long can the Covid-19 virus survive on a surface? | Amid the lockdown, what is allowed, what is prohibited?