“Nature” paper praises China’s anti-epidemic model shows no intervention in infected people or more than seven million

As an emerging infectious disease, there are no specific drugs and vaccines available worldwide. As a result, countries have mainly used non-drug interventions (non-pharmaceuticals, NPIs) to contain outbreaks. China’s outbreak containment strategy based on non-drug interventions is effective, but how effective are these measures? How well is it imposed? To answer these questions, quantitative assessment is required.

On May 4th, local time, the top academic journal, Nature, published research by chinese, Anglo-American and multinational research teams. The study found that the three major non-drug interventions used in China not only curbed the development of the new corona epidemic in China, but also won a time window for the world. Without a strong non-drug intervention , the study said , the number of new crown cases in China could increase 67-fold to more than 7 million people .

The above research from the University of Southampton, Fudan University, Wuhan CDC, Harvard University, Johns Hopkins University and other research teams.

Using epidemiological and anonymous human activity data, the researchers created a model framework that used daily travel networks to simulate outbreaks and interventions across China. The model estimates that by 29 February 2020, there will be a total of 114,325 COVID-19 cases in mainland China (quartile range 76776-164576).

Without non-drug interventions (NPI), coVID-19 cases in mainland China could increase 67 times (44-94 times in the quartile range). This means that without intervention, there will be nearly 7.66 million new crown infections in mainland China.

The study also found that:

Early detection and isolation of cases can prevent more infections than travel restrictions and reduced human-to-human contact. However, only a combination of non-drug interventions can achieve the strongest and fastest deterrent effect.

The removal of travel restrictions from 17 February 2020 does not appear to have led to an increase in cases nationwide if social alienation measures can be maintained. The findings help to improve understanding of non-drug interventions on COVID-19 and inform responses around the world.

Three categories of non-drug interventions

As an emerging infectious disease, it is unrealistic to find effective drug interventions within a few months. The current limited medical resources are not able to treat all cases. Therefore, non-drug interventions are an important part of public health response to the outbreak.

These measures include the isolation of the sick, the tracing of contacts, the isolation of contacts, travel restrictions, the closure of schools and workplaces, the cancellation of mass gatherings, etc. These measures are aimed at reducing transmission, thereby delaying the peak of the epidemic and reducing the size of the peak outbreak, buying time for the recovery of the health care system and time for vaccine and drug development.

In order to curb the spread of COVID-19 in China and reduce the scale of its outbreak, China’s non-drug interventions (NPI) include three broad categories.

The first is intercity travel restrictions, which have curbed the further spread of the virus during the Lunar New Year holidays. The city closure measures for cities around Wuhan and Hubei Province were implemented on January 23rd, two days before the Spring Festival (CNY). Since the Spring Festival, other provinces across the country have also imposed travel restrictions.

The second category of non-drug interventions is early identification and isolation of cases, including improved screening, identification, diagnosis, isolation and reporting of tracing of contacts for suspected and confirmed cases. Governments across China have stepped up routine checks and quarantines for travelers in Hubei province in an effort to detect COVID-19 infections as soon as possible. In particular, improvements in detection and diagnosis have reduced the average interval from the onset of symptoms to laboratory diagnosis from 12 days at the beginning of the outbreak to three days in early February.

The third category of non-drug interventions is the implementation of contact restrictions and social alienation measures and the strengthening of personal preventive measures, such as hand washing, to reduce the risk of exposure at the community level. As part of its policy of social alienation, China encourages people to stay at home as much as possible, cancel or postpone large public events and mass gatherings, and close libraries, museums and workplaces. In addition, the school holidays have been extended, the end date of the Spring Festival holiday in Hubei Province was extended from January 30 to March 10, and the end date of the Spring Festival holiday in many other provinces was extended to February 9.

Despite the high economic and social costs, China’s nationwide non-drug interventions have led to a rapid decline in the number of new cases. Previous studies have explored the effects of Wuhan blockades, travel restrictions, airport inspections and case isolation tracking in curbing the spread of the virus. However, there is a lack of comprehensive and quantitative analysis of the effectiveness and timing of different non-drug interventions in China.

Based on COVID-19 epidemiological data and historical and near real-time anonymous population activity data, the researchers developed a random susceptibility-exposure-infection-removal (SEIR) model of travel networks to simulate the spread of new coronal outbreaks in 340 mainland Chinese cities since December 1, 2019. Through the model, the researchers conducted a pre- and for-comparison analysis to quantify the impact of three major non-drug interventions in China. The model also assesses the risk of COVID-19 transmission since the lifting of travel restrictions on 17 February 2020.

Simulating and rebuilding the spread of COVID-19

How can the epidemiological parameters of the early stage of China’s new crown epidemic be valued before extensive interventions are implemented? The researchers used epidemiological parameters estimated at the beginning of the outbreak in Wuhan.

The researchers measured the effectiveness of the three major non-drug interventions by obtaining data on population movements between and within cities, as well as the time interval between onset and reporting cases across the country, using Baidu location services for mobile phone users. Since the intervention, there have been significant changes in population mobility and exposure in China, and the timeliness of case reporting has improved.

The researchers estimate that as of February 29, 2020, there were 114,325 COVID-19 cases in mainland China (quartile spacing 76776 – 164576), of which Hubei Province accounted for 85%.

Before the Spring Festival, China’s new cases showed an exponential increase, but after non-drug interventions, the peak of the epidemic arrived quickly, during the Spring Festival holiday, which took about a week. The model’s predictions of epidemics and peaks are highly consistent with the data reported on the date of onset. The model’s daily estimates are also highly correlated with cross-time and cross-region reporting data.

As of 29 February 2020, there was a significant overall correlation between the estimated number of cases in the model and the number of cases reported in the provinces (p?lt;0.001, R2 s 0.86), and the sensitivity of the model (91 per cent, 280/308) and specificity (69 per cent, 22/32) were also high when predicting that there were or no new coronal cases in a city.

Quantify the impact of different non-drug interventions

Model projections show that without non-drug interventions, the number of COVID-19 cases in China will be much higher than the current actual number as of February 29. Among them, Wuhan Increased 51 times (33 times -71 times in the quartile range) and the number of cases in other cities in Hubei Province increased by 92 times (58 times -133 times in the quartile range). The number of urban cases in other provinces will increase 125 times (77-180 times in the quartile range).

However, the apparent effects of different interventions vary.

The blockade of Wuhan did not completely prevent the spread of the virus from Wuhan, because the blockade was imposed at a relatively low time for the Spring Festival to return home. However, if intercity travel restrictions are not enforced, cities and provinces outside Wuhan will see more new cases from Wuhan, and the geographical scope of the impact will be extended to china’s remote western regions.

In general, early detection and isolation of cases can prevent infection more quickly and substantively (5x to 2.6x) than a reduction in human-to-human exposure and social alienation nationwide. However, without reduced intervention in human-to-human contact, the epidemic across the region will multiply in the long run. As a result, integrated non-drug interventions have the strongest and fastest deterrent effect on the COVID-19 outbreak, with the peak of the outbreak reaching about a week after the implementation of the integrated non-drug intervention.

Assess the timing of intervention: How different will it be to move forward or one, two, or three weeks ahead?

If intervention sits one, two or three weeks earlier, the number of cases in China can be reduced by 66% (50-82%), 86% (81-90% in the quartile range), and 95% (93-97% in the quartile range).

The geographical extent of the affected areas will also be reduced from 308 to 192, 130 and 61 cities, respectively.

However, if non-drug interventions are carried out one, two, or three weeks later than the actual implementation time, the cases may be increased three times (2-4 times in the quartile range), 7 times (quartile range 5-10) or 18 times (quartile range 11-26).

Assessing the lifting of travel restrictions

After the intervention, outbreaks outside Hubei Province were contained at a low level in early March (10 new cases per day, excluding imported cases), while it could take up to four weeks for Hubei provinces to reach the same level as other provinces.

However, if the level of contact levels returns to normal levels in previous years, the removal of travel restrictions from 17 February may lead to a further increase in the number of cases. As a result, models suggest that by the end of April, even to a limited extent, maintaining social distance (for example, a 25 per cent reduction in exposure on average) will help ensure that outbreaks in AREAS such as Wuhan’s COVID-19 epicenters are contained.

The model is sensitive to the number of basic infections (R0), and if R0 is high, the epidemic peak swells and occurs later, and it takes longer to contain the outbreak. Sensitivity analysis also shows that the model can effectively measure the relative change of intervention effect under different epidemiological parameters and transmission sensitivity.

The researchers’ findings suggest that co-imposed non-drug intervention facilities have greatly curbed the spread of the new coronal outbreak in China. Early implementation may significantly reduce the outbreak level and geographical scope. Conversely, delayed responses can lead to larger outbreaks.

China’s strong, multi-pronged containment has averted a worse situation and slowed the spread of the new coronavirus around the world. China offers a window of time and opportunity to curb the spread of COVID-19 in other parts of the world and mitigate its impact.

The researchers’ results raise three key points. First, population movements and close contact play an important role in the spread of the new crown. Since the blockade of Wuhan is in the late part of the Spring Festival, travel restrictions and can completely prevent the spread of the virus from Wuhan, but it did prevent more cases from being exported from Wuhan to the wider region.

Second, the importance and impact of the three non-drug interventions are different. Better detection and isolation of cases and social alienation may have a greater impact on outbreak control than travel restrictions. Interventions of social alienation have reduced people’s exposure to travellers from outbreak centres. This may be particularly useful in curbing the spread of the virus to the wider community and reduce the risk of asymptomatic or mild infection.

Third, given that work is resumed in China, China should consider retaining at least some non-drug interventions to ensure control of the outbreak and avoid another outbreak. For example, given cases imported from other countries and asymptomatic or subclinical infections found in China, early case identification and isolation should continue, which may also help prevent or delay the second wave of outbreaks.