Significant delays were observed before there were eventual resurgences of infection rates after lifting public health distancing measures that had been put in place to control the pandemic of SARS-CoV-2 in countries around the world. Findings from a study of more than 6 months' experience of pandemic control in 131 countries are reported in The Lancet Infectious Diseases (2020; doi: https://doi.org/10.1016/S1473-3099(20)30785-4).
The investigators warned that the absence of any increase in infection rates shortly after lifting specific non-pharmaceutical intervention (NPI) could give a false sense of security unless infection rates remained low well beyond the delay period, the study had found.
"The main take-home message is that the effect of introducing or lifting these public health measures is not immediate," said lead author You Li, PhD, Research Fellow in the Respiratory Viral Epidemiology Group at Usher Institute, University of Edinburgh, Scotland.
Li noted that if a region introduced NPIs they should be aware that the daily number of cases would not drop immediately. "It may take about 1 week to see a drop in the daily number of cases," he said, a finding that was not surprising.
But the study had also been able to quantify significant delays between the lifting of mitigation measures and resulting rebounds in infection rates.
"If a country decided to lift the public health measures, they shouldn't expect an immediate increase in the daily number of cases," he said. "If they didn't see a rise in the daily number of cases within 1 week following the relaxation, that's not a safe signal. They should be aware that the number could go up later-for example, 2 weeks or even 3 weeks after the relaxation," he said.
Relaxing Lockdowns
"At the beginning of the pandemic, many countries decided to introduce a number of public health measures to reduce the spread of SARS-CoV-2," said Li. From May 2020, he noted some of the countries had decided to relax public health measures. "Following that, we have seen a rise in the number of daily cases," he said.
From these data, the study assessed how effective the measures had been and noted the effects of lifting them on disease transmission as indicated by the R number-the reproductive number for disease transmission that needs to be kept below 1.0 to cause infection rates to decline.
"Importantly, what we found was that the effect of introducing and lifting public health measures is not immediate," Li noted. He explained their analysis had used the R number as their principal metric of infectiousness for the virus in each region and under each category of physical distancing mitigation. They calculated the ratio of the value of R before and after any NPI had been applied and then after relaxing it.
The researchers found an increasing trend over time in the R ratio following the relaxation of measures, including school closures, bans on public events, bans on public gatherings of more than 10 people, requirements to stay at home, and internal movement limits.
The change in R was noted for each of the individual NPIs. Increases in R ranged from 11 percent to 25 percent on day 28 following the relaxation of a mitigation measure compared with the last day before relaxation. The study found the increase had been statistically significant only for school reopening (R ratio 1.24) and for lifting bans on public gatherings of more than 10 people (R ratio 1.25).
The investigators observed a median of 8 days following the introduction of an NPI for it to reach 60 percent of its maximum reduction in R. But it took even longer (17 days) after relaxation of a measure to observe 60 percent of the maximum increase in R as infections rebounded in "second waves." This was clearly a danger signal. Policy makers needed to anticipate such delays and not be reassured too soon about the safety of relaxing specific measures.
Combinations of NPIs Most Effective
A control strategy of banning public events and public gatherings of more than 10 people was estimated to have reduced R by a ratio of 0.71 on day 28. This decreased even further to 0.62 on day 28 if measures to close workplaces were added. Adding the restriction of internal movements with regions and countries and requirements to stay at home brought the R value down to less than half its initial value: R ratio of 0.48.
The study concluded that individual NPIs (including school closure, workplace closure, public events ban, bans on gatherings of more than 10 people, requirements to stay at home, and internal movement limits) had clearly been associated with reduced transmission of SARS-CoV-2. This message had been welcome because it clearly opened the door to reducing R below 1.0 and eliminating the virus.
The fact that the effects of introducing and lifting these NPIs had been found to have been delayed by 1-3 weeks (or even longer) could be used as additional evidence to inform policy-maker decisions on the timing of introducing and lifting different NPIs, the researchers concluded.
Li cautioned that the use of the R number needed to be interpreted in the context of its known limitations. For example, the study had taken no account of additional measures, such as the use of face coverings, distancing systems at work and school, transparent screens, and one-way systems-all of which could influence rates of infection.
The big difference between SARS-CoV-2 and the original 2003 epidemic of SARS, had been that COVID-19 had a much higher rate of pre-symptomatic and asymptomatic transmission. "This makes the control of SARS-CoV-2 very difficult, even though both SARS-CoV-1 and 2 have a similar R metric," Li said.
The good news for COVID-19 control was that the study confirmed that policies of combining NPIs had been effective in reducing transmission, said Li. And that was despite the geographical, sociological, and political heterogeneity among the many countries affected by the pandemic-which itself was likely to have played a role in the heterogeneity of outcome patterns across the world.
"What we found from our study was that-across these 131 countries all over the world-what we can see is that these public health measures have been effective in reducing the transmission regardless of which region or which phase the country was [in]."
"On seeing a resurgence in cases, the countries could consider reintroducing these public health measures to control the spread of the virus," said Li. The issue facing community leaders would then choose a combination of measures sufficient to reduce R below 1.0 while maintaining other valued aspects of life.
Targeting Infection Hotspots
When Li was asked about the effectiveness of targeted approaches to imposing and lifting NPIs according to the disease levels in specific regions-the idea of concentrating on applying NPIs in so-called infection "hot spots"-he said this seemed to be a reasonable approach, though his group's analysis had not looked at the subnational level.
"At the beginning of the pandemic, most countries adopted national levels of public health measures," he said. "And that's why now-on seeing within-country variations in terms of the outbreaks-countries are considering more sub-national or regional level measures."
Li's opinion was that this seemed to be more practical and sustainable, since national lockdown, for example, always came with an economic price. "Limiting the measures to only a regional level could help both with the country's development and to protect the people. So, it is overall more practical to do rather than doing a national lockdown," he suggested. He did not comment on the likely sufficiency of such measures.
Peter M. Goodwin is a contributing writer.