Slowing the spread through restricted movement measures
Setting the context
Most countries across the world have imposed restricted movement measures in response to the COVID-19 pandemic, recognizing that the fast-spreading virus does not observe national borders, let alone provincial or local boundaries. While some countries have completely closed borders to non-citizens and non-residents, others have kept borders partially open for some categories of travellers. Regardless of who can enter or leave a country, 91 per cent of the global population is currently living with some form of travel restriction. Where economies can afford it, the situation inside countries is no different; movement has been severely curtailed.
According to the WHO, travel restrictions implemented early on can delay the onset of COVID-19 cases by providing countries with essential time to prepare to manage the pandemic. Travel restrictions can delay the spread of the virus by as much as a few weeks and can be effective when supplemented with stringent lockdown measures.
This section is divided into two key topics:
- The impact of border closures on the spread of COVID-19
- The relationship between the stringency of lockdown measures and a country’s ability to recover from COVID-19
Key findings from the analysis
Can closing international borders reduce new COVID-19 cases significantly?
Travel advisories have evolved since the first known COVID-19 case in Wuhan, China in December 2019. Countries have calibrated their approach to restricted movement according to fast-changing scenarios locally and internationally. Based on the interval between the date of the first reported COVID-19 case in a country and the date of international border closures, timing can be divided into three categories: early, on-time, and late.
According to the WHO, countries that close borders right at the start of a pandemic have seen the highest impact on reducing the spread of COVID-19. How do countries fare that close borders as cases rise or much after the peak has passed?
This section includes a historic capture of how countries performed in relation to the closure of international borders during the early stages of the pandemic. Since the publication date, many countries have reopened borders and we continue to see varying results in response.
|Key Finding 1: The impact of border closures on new cases is not immediate. Regardless of when countries close borders, it takes time for new cases to start declining as a result of fewer imported cases. Closing borders is an effective strategy to buy time while governments implement in-country measures to prevent community transmission.|
The literature suggests that the effect of travel restrictions is complex to assess because restrictions are usually imposed alongside other countermeasures. Recent research by Kucharski et al (2020) and Chinazzi M, Davis JT, Ajelli M, et al (2020) has concluded that travel restrictions delayed the overall epidemic progression by a few days/weeks at a national level, with greater impact at the international level due to the reduction of ‘import cases’. To illustrate, the travel restrictions to and from Wuhan delayed the overall contamination by three to five days in Mainland China but had greater effect at an international level where case importations were reduced by nearly 80 per cent in the period up to mid-February 2020. This is supported by related research on influenza (Kucharski et al 2020), which concluded that travel restrictions delay (e.g. by three to four weeks when 90 per cent of air travel is restricted in affected countries, or by two months if even more restrictive measures are introduced) but do not prevent pandemics. As described by the WHO previously, strict travel restrictions (that could interfere with traffic of essential professionals and goods) may be justified only at the beginning of an outbreak, as they will enable countries sufficient time (even if only a few days) to rapidly implement effective preparedness measures.
We considered three countries, Uganda, Germany and Vietnam, each of which has either closed borders completely or partially. The interval between the first reported case and the date of the border closure among the top 20 countries in the GCI can help categorise country response to border closures in three key groups:
- Early: If the country closed borders before the onset of COVID-19 cases or within 29 days of the first reported case
- On-time: If the country closed borders between 29 and 58 days after the first reported case
- Late: If the country closed borders between 58 and 87 days after the first reported case
All three saw a dip in new cases within a few days or weeks of enforcement of border controls, but we wanted to evaluate the comparative impact of the timing of these border closures on the trend of cases.
In Uganda, international borders closed early – even before the emergence of the first reported COVID-19 case in the country (see Figure 35). On 18 March, Uganda started by restricting non-essential travel from 16 countries, including the UK and the US, and requiring a 14-day self-quarantine for people coming from such countries. However, after the first case of coronavirus was confirmed on 23 March, the government of Uganda decided to close all borders while cargo- and humanitarian-related services remained open. Such a fast reaction might have allowed Uganda to reduce imported cases and to delay the virus outbreak for some days – five days with zero reported cases – but did not eliminate cases altogether. Once the first case was reported, new cases increased every few days, but then dropped to zero in April (twice) and May (four times). The first drop to zero cases happened 10 days after international borders closed.
In Germany, the government decided to close the borders on 16 March with a more flexible approach for flights coming from EU member states. International borders were closed for all entries from outside the Schengen area. For flights coming from the Schengen area, several exceptions applied:
- Citizens from EU member states and countries associated with Schengen and their family members who enter Germany in order to travel to their country of origin or residence
- Third country nationals with a long-term residence in an EU Member State
- Third country nationals with essential functions or needs (i.e. healthcare workers, transport personnel employed in the movement of goods and other essential sectors, diplomats, etc)
With the above measures in place, Germany (which closed borders ‘on time’) achieved zero new cases by 1May, which was around six weeks after international border closure (see Figure 36). However, cases started dipping significantly four days after closure.
Vietnam adopted a longer process, gradually restricting international travel throughout March 2020. The government began by imposing entry restrictions on travellers from Iran and Italy and then suspending new visas for nationals from the Schengen area and UK during the first and second weeks of March. A week later, the government suspended visa issuance for all foreigners over COVID-19 concerns, leading to the complete closure of all international borders on 21 March. Although Vietnam deployed early actions to restrict visa issuance, borders were effectively closed ‘late’. Zero new cases were achieved by 8 May, which was around 2.5 weeks after international borders closed (see Figure 37). The number of cases started dipping significantly two days after border closure, perhaps caused by a reduction in imported cases.
Countries in each category ended up with significantly fewer new cases over time, but closing borders as early as possible may provide more time for governments to adopt measures which prevent community transmission. A reduction in imported cases will slow down the number of new daily cases. As shown above, regardless of timing for imposing international travel restrictions, all countries eventually experienced a downturn in new cases, with occasional drops to zero new cases within a month and a half of restricting international non-essential travel.
|Key Finding 2: Even if countries close borders late, it may help to flatten the curve alongside other curtailment measures.|
China is an example of a country that imposed international travel restrictions (in March 2020) long after the first cases had appeared in December 2019. In late January 2020, China began with in-country mobility restrictions in Wuhan, the epicentre of the COVID-19 transmission then, and the neighbouring cities of Huanggang and Ezhou. Eventually all the remaining cities in the province of Hubei were added. These restrictions involved suspension of all public transportation including buses, railways, flights and ferry services. Moreover, the residents of Wuhan were not allowed to leave the city without permission from the authorities. Severe lockdown measures were implemented allowing families to send one person for grocery shopping every two days. In Wuhan and other ‘high-risk areas’, people were permanently housebound.
Effective lockdown measures were possible thanks to strict regulation and enforcement, community efforts and monitoring technology. Lockdown relied upon individual communities to self-enforce social distancing and travel bans. Entrances to communities were shut down, leaving only one entry point for residents with valid IDs and face masks. The introduction of mobile apps, ID readers, and temperature checks enhanced with digitally enabled detection and prevention systems, all contributed to the monitoring of community mobility and the close monitoring of quarantined and treated patients.
Based on these measures, China created an environment designed to fight community transmission from imported cases at a time when other countries in Europe were already presenting high numbers of confirmed cases.
As with many peer countries, Cambodia´s core response was based on case detection, isolation and contact tracing. Schools were closed in mid-March and large entertainment and religious gatherings were prohibited starting from early March. The Prime Minister cancelled the Khmer New Year holiday scheduled for mid-April – one of the largest celebrations of the year amongst Khmer people.
Cambodia imposed travel restrictions on 30March with 107 accumulated cases, almost two months after the first case in late January 2020. These measures included suspending the visa exemption policy and the issuance of tourist visas, e-visas and visas on arrival for all foreigners for one month. Moreover, any foreign individual faced a 14-day mandatory quarantine. This condition was then changed to require an updated medical certificate demonstrating that the individual was not infected and proof that the individual held at least USD 50 thousand for medical coverage. Strict quarantine measures were put in place after imported cases from foreign cruises and tourist groups boosted daily cases in March 2020. Quarantine was also applied for national garment workers or any company employees taking leave during Khmer New Year. According to the Phnom Penh Capital City Hall, an estimated 15,726 garment workers employed in 672 factories returned to Phnom Penh from the provinces. Although Cambodia closed borders when daily cases were already at their peak, the country did manage to reduce the spread of imported cases through strict mobility measures and quarantine for all international and domestic travelers.
Cambodia and China have managed improvement from their peak number of cases as of 17 May. Although critical, closing international borders is unlikely to be the main driving force in reducing COVID-19 cases. Other protective and precautionary measures restricting in-country mobility at the community level are vital too.
Can lockdown measures help countries recover quickly from COVID-19?
Some countries have developed and enforced strict lockdown measures, while others have been more relaxed. The strictness of lockdown measures may be linked to countries’ income levels. Depending on where a country stands on the income spectrum, the stringency of lockdowns might have a different impact on the ability to recover from COVID-19.
|Key Finding 3: Lockdowns are consistently associated with improvements in recovery of COVID-19 cases.|
Countries have adopted various types of social mobility restrictions with diverse levels of stringency. Evidence suggests that the number of new cases has dropped regardless of level of lockdown stringency. The stringency of lockdowns can be measured by various indices, of which the Oxford Stringency Index is the most comprehensive. Based on this index, countries can be divided into four categories: very strict, strict, moderate, and relaxed. Among the top 20 countries in the GCI, all countries fall in the first three categories.
A comparison of countries in the ‘very strict’ and ‘moderate’ categories shows how both approaches can be linked to either sustaining a positive or improving the recovery path. Denmark was the first country in Europe to implement lockdown measures. New Zealand and Greece also fall under the ‘very strict’ category and have shown remarkable recovery. Both Denmark and New Zealand requested all public servants to stay at home as well as private sector employees unless working on essential services. Education services for primary, secondary and university students were cancelled and transitioned to virtual schooling. All places for social/religious gatherings were closed. Assemblies of more than 10 people became illegal with associated fines of DKK 150 in Denmark. Greece adopted similar restrictions with additional measures to limit mobility, only allowing its citizens to leave home for seven specific reasons such as going to a pharmacy, food store or bank, exercising outdoors briefly and walking pets. The recovery index in Denmark, as shown in Figure 40, jumped 27.2 points from its lowest point since the onset of COVID-19, followed by New Zealand (25.8 points) and Greece (18.4 points). Recent international literature also supports the positive impact of large-scale lockdown measures and school closures in decreasing COVID-19 cases and has arguably prevented 3.1 million deaths in Europe.
Although recovery index performance in South Korea and Taiwan is much more modest, with a jump of 10.5 points in Taiwan and zero points in South Korea, this might imply that even ‘moderate’ lockdowns seem to support recovery paths over time if they are coupled with other strategies to prevent contagion. Both countries have been widely known for implementing mass contact tracing and testing policies enabled by technological solutions which potentially offset the need for strict lockdown measures.
Key Finding 4: Countries with large-scale efficient contact tracing, testing and isolation measures can potentially afford less strict mobility restriction measures while controlling health risks.
The government of South Korea decided early on not to impose draconian quarantine and lockdown measures. It focused instead on mass testing and contact tracing. South Korea is a compelling case for using technology-based solutions for contact tracing, testing and isolation measures. South Korea used cell phones, closed-circuit television (CCTV) and bank transactions to collect data on the movements of COVID-19-infected people and shared information via text message with people living nearby. Moreover, it put in place widespread drive-through testing facilities where examination and temperature checks were done with drivers able to remain inside the car.
Quarantine and isolation measures were put in place for infected people in government shelters. High-risk patients with underlying illnesses were prioritised for hospitalization while others with moderate symptoms were placed in repurposed corporate training facilities and public spaces to receive medical support. Recovered patients and those with minimal symptoms were sent home and requested to self-quarantine for two weeks, receiving twice-daily calls from monitoring teams to check on compliance and review symptoms. All individuals under self-quarantine were required to download an app that alerts authorities about non-compliance, leading to a fine of up to USD 2,500.
In April 2020, after the number of daily confirmed cases fell into single digits, the government decided to relax some mobility restrictions in stores, restaurants, gyms, schools, bars, religious services and national parks. Such intervention did not lead to an increase in the number of daily cases (there were under fifty new daily cases in April, and under forty in May); but it did enable the reactivation of economic and social activity. South Korea leveraged the use of big data to reduce the cost of contact tracing. However, questions around information privacy and episodes of data leaks have created controversy and fears of social stigma.
Like South Korea, Taiwan adopted an alternative approach deploying a policy of maximum health information transparency and supervised self-discipline without resorting to strict lockdown measures. The government of Taiwan established a centralised epidemic command centre (CECC) to provide information and perform detailed surveillance of the movements of infected people. All confirmed cases with COVID-19 symptoms reported to public and private hospitals and were immediately reported to the CECC. The CECC traced patients’ recent movements and identified their footprint in public spaces such as supermarkets and restaurants. Text messages were sent to warn other people who might have been near the patients and encourage compliance with social distancing rules. The government also established an ‘electronic fence’ system, allowing them to monitor any infected person leaving their designated quarantine area.
|Key Finding 5: Gradual relaxation in lockdown measures helps sustain recovery paths.|
Countries that have eased the strictness of lockdown step by step and gradually have observed sustained improvement in the recovery index (see Figures 45 and 46 below).
Effective Practice Case Study: Australia’s four phase roadmap Lockdown measures and gradual reopening of society and economy
The lockdown measures: Like many other countries, Australia gradually adopted strict lockdown measures to reduce the risk of community contamination. In early March, government started imposing self-isolation for all international travellers for 14 days and restricting social gatherings larger than 100 people. The government then closed international borders for all non-essential travel both for non-citizens and residents. It imposed a major round of closures of non-essential services including all pubs, clubs, cafes, restaurants and gyms nationwide, excluding only takeaway services. By late March, travel restrictions were in place for international and domestic flights, gatherings were limited to two people, people over age 70 were advised to self-isolate and government advised all people to stay at home.
The results: After imposing strict travel and quarantine restrictions throughout March and reaching the highest number of daily cases (497) on 28 March, daily cases began to drop drastically. In April, the average number of new daily cases was 25 (with the highest point being 64 new cases on 14 April). In May, the average number of new daily cases was 15 to 17, with a peak on 14 May with 30 cases.
The reopening plan: As the number of new cases decreased in April, a four-phase roadmap to gradually ease the lockdown measures and resume economic and social activities was put in place. Each phase is separated by three to four weeks and gradually provides greater flexibility regarding social distancing, gatherings and reopening economic activities and travel. This plan has been constantly revised and updated by the authorities to enable preventive action if the rate of new cases increases.
Note: As of 21 July, Australia has witnessed several new outbreaks including a return to lockdown across the Melbourne Metropolitan area in the state of Victoria where over 370 new cases were recorded in one day. This highlights the changing nature of the pandemic and how much we need to continue learning the effect of gradual easing measures taken by different countries.
Source: Website, Department of the Premier and the Cabinet – Government of Western Australia
Recommendations: Interventions for countries to consider during the early stages of an outbreak
Recommendation 1: Close international borders to prevent import of new foreign cases and buy time for adoption of preventive and protective measures.
- Travel restrictions are feasible with a careful risk assessment that takes account of the public health risks and immediate economic impacts. As the situation evolves, room to reconsider such restrictions is recommended.
- A plan for closing international borders can make exceptions to allow essential travel for national citizens and residents and mobility of essential workers and goods.
- For essential travellers, provide clear recommendations and establish protocols for testing, tracking and self-quarantine measures.
- Closing international borders is not enough. International mobility measures work best when combined with in-country measures.
Recommendation 2: Implement and enforce lockdowns; adjust stringency based on capacity to test and isolate.
- Lockdown measures are a highly contextualised policy with no one-size-fits-all recommendation. Countries may want to analyse lockdown policy in light of government’s capacity to effectively test, track, trace and isolate infected cases and prevent the spread through physical interaction. The lower a country’s surveillance and case management capacity, the more stringent its social mobility restrictions must be.
- If required, define a lockdown plan starting with high-risk areas for contagion (places for social or religious gatherings), reinforcing work-from-home policies for non-essential workers and social distancing measures in places allowed to operate. Communicate widely to encourage universal hygiene actions to prevent infection (e.g. washing hands, wearing masks and constant disinfecting).
- Before, during, and after lockdown, governments can take rapid action to increase capacity for contact tracing and isolation through digital apps, widespread mobile use, and technological infrastructure. Nation-wide strict lockdown measures are not sustainable for long periods of time. Governments can develop alternative approaches that enable more targeted actions.
Recommendation 3: Relax lockdowns gradually as the data show that the spread of the disease has been controlled.
- The relaxation of lockdown measures can be based on an analysis of both the health risks and the economic risks of prolonged lockdowns. Lifting a lockdown too early might bring a second wave of infections; lifting it too late will involve larger economic loss.
- Define a clear roadmap that specifies stages, dates and actions to gradually resume social and economic activities. Ensure the roadmap provides clear metrics regarding social distancing, size of gatherings and type of activities allowed to reopen, allowing citizens and business owners to take immediate action.
- Closely monitor the impact of lockdown relaxation on viral spread at the community level. Update the plan and specific actions accordingly.
 More than nine-in-ten people worldwide live in countries with travel restrictions amid COVID-19, PEW Research Center, 1 April 2020
 Updated WHO Recommendations for International Traffic in Relation to COVID-19 Outbreak, World Health Organisation, 29 February 2020
 Travel Restrictions and the Spread of COVID-19 – What Does the Research Say? CATO Institute, 23 March 2020
 Kucharski, A. et al. (2020), “Early dynamics of transmission and control of COVID-19: a mathematical modelling study”, The Lancet Infectious Diseases, Vol. 0/0. Chinazzi M, Davis JT, Ajelli M, et al. The effect of travel restrictions on the spread of the 2019 novel coronavirus (COVID-19) outbreak. Science. 2020
 Updated WHO Recommendations for International Traffic in Relation to COVID-19 Outbreak, World Health Organisation, 29 February 2020
 Flaxman, S., Mishra, S., Gandy, A. et al. Estimating the effects of non-pharmaceutical interventions on COVID-19 in Europe. Nature (2020). https://doi.org/10.1038/s41586-020-2405-7