Opinion: How Taiwan beat COVID-19 – new study reveals clues to its success
Posted 16 April, 2021
Taiwan has been (opens in a new window)widely applauded for its management of the pandemic, with one of the lowest per capita COVID-19 rates in the world and life on the island largely returning to normal.
Just (opens in a new window)11 people have died from COVID-19 in Taiwan since the pandemic began, an impressive feat considering they never went into lockdown.
At the start of the pandemic, Taiwan was considered high-risk for COVID-19 due to its proximity to mainland China and the frequent travel that takes place between the two.
With a history of SARS in 2003, which was not considered to be (opens in a new window)handled particularly well, the Taiwanese government (opens in a new window)acted quickly to close its borders this time around. It set up a (opens in a new window)Central Epidemic Command Centre on January 20 2020 to (opens in a new window)coordinate cooperation across different government ministries and agencies, and (opens in a new window)between government and businesses.
A (opens in a new window)new study in the Journal of the American Medical Association has examined further just why Taiwan did so well at conquering COVID-19. The study’s authors, from a range of health institutes and hospitals in Taiwan and the US, compared the estimated effectiveness of two types of COVID-19 policy in the early months of the pandemic: case-based and population-based measures.
Case-based measures include the detection of infected people through testing, isolation of positive cases, contact tracing and 14-day quarantining of close contacts. The population-based measures included face mask policies, personal hygiene and social distancing.
The effects of these policies were quantified by estimating the effective reproduction number (R number).
The (opens in a new window)R number is a way of rating an infectious disease’s ability to spread – it represents the average number of people that one infected person will pass a virus onto. An R number of greater than 1 means the virus will continue to spread and outbreaks will continue. An R number below 1 means that case numbers (opens in a new window)will start to reduce.
While previous studies elsewhere have simulated hypothetical scenarios, this paper combined transmission modelling with detailed real data to estimate effectiveness.
The authors collected data on 158 cases between January 10 and June 1 2020 from the Taiwan Centers for Disease Control, and all cases were confirmed by PCR testing. The data related to locally acquired cases, confirmed clusters, and imported cases in people who entered Taiwan before March 21 2020.
They then compared the outcomes they found in Taiwan with an estimated R number of 2.5, based on the (opens in a new window)estimated equivalent number in mainland China at the beginning of its COVID-19 outbreak.
The winning combination
The study found that the case-based policies alone, like contact tracing and quarantining, could lower the R number from 2.5 to 1.53. Quarantine contributed the most to lowering the R number.
Case-based interventions could not substantially prevent transmission from one person to another, but could reduce transmission onwards from those secondary cases to a third or fourth person, as long as close contacts quarantined.
Population-based policies like social distancing and face masks, meanwhile, reduced the R number from 2.5 to 1.3.
The authors concluded that it was the combination of case-based and population-based policies, along with widespread adherence, that led to Taiwan’s success in containing COVID. Combining both approaches led to an R number estimated using two different methods to be 0.82 and as low as 0.62. They also found that considerable population-based policies were needed to achieve containment even though the number of circulating infections was small.
Neither approach would have been sufficient alone, even with an effective public health system and sophisticated contact tracing.
What does this mean for other parts of the world?
Acknowledging that all models make assumptions, and this analysis is no different, this paper does confirm that the full suite of public health measures we have been using fairly consistently across the world – to varying degrees of length and stringency – have been necessary. Though it’s worth noting that the results in the study reflects a time when (opens in a new window)new variants with greater transmissibility were not a problem.
The authors assumed that testing and isolation occurred simultaneously. This was the case in Taiwan, but not elsewhere, for example England, where (opens in a new window)delays between testing, results and isolation diminish the effectiveness of case-based measures.
Taiwan has the ability to control the introduction of new cases through border control, and the authors acknowledge the findings of this study may not be fully applicable elsewhere. This is the reason the authors focused on the effectiveness of case-based and population-based interventions on local transmission, rather than on border controls on the number of introductions of COVID-19.
The authors conclude that intensive contact tracing is not possible when public health systems are overwhelmed. This never happened in Taiwan due to the success of its strategies, but it did, for example, take place in (opens in a new window)Ireland in January 2021, which experienced a damaging (opens in a new window)third wave of COVID-19.
This paper also found similar results for seven-day and 14-day quarantine and suggest that the quarantine period could be shortened. This is being considered by some countries, (opens in a new window)including the USA, but it has not been introduced on a widespread basis to date.
We already knew there was much to be learned from Taiwan’s success in preventing COVID-19 from taking hold. Now, as vaccines roll out and new variants emerge, we have more information about the comparative and combined contributions of public health measures.
By (opens in a new window)Patricia Fitzpatrick, Full Professor of Epidemiology & Biomedical Statistics, (opens in a new window)University College Dublin
This article is republished from (opens in a new window)The Conversation under a Creative Commons license. Read the (opens in a new window)original article.
