A new study published on preprinted servers medRxiv* in July 2020 revealed the important nature of time as a factor in controlling mortality rates in the current COVID-19 pandemic. The background of this paper is the outbreak in Switzerland, where around 30,000 people were infected with acute respiratory syndrome coronavirus 2 (SARS-CoV-2), and more than 1,800 died from late February to early May 2020. This study shows that the introduction of NPI in March 2020 prevents thousands of virus-related deaths in Switzerland.
Study: Time is crucial: containment of the SARS-CoV-2 epidemic in Switzerland from February to May 2020. Image credit: Roger Gantner / Shutterstock
Plague Course in Switzerland
The first case in this country came from a region adjacent to the area affected by Italy, but the disease quickly spread to the small country. In response to the outbreak, a series of non-pharmaceutical interventions (NPI) were quickly introduced, including the closure of elementary schools, shops selling non-essential items, and restaurants, and limiting the gathering of more than five people on March 20, 2020.
The epidemic showed a decline after the introduction of these measures in March. On May 10, 2020, restrictions were lifted. The paper by researchers at the University of Bern aims to show how the timing of NPI implementation affects the course of the epidemic and its results.
New model number of infections, hospitalized patients, intensive care unit (ICU) occupancy, and deaths during the SARS-CoV-2 epidemic in Switzerland. Solid lines show the maximum likelihood estimates of the model and shaded area according to the 95% prediction interval. The model fits the data shown as white circles. The number of reported infections (gray box) is shown for comparison. The vertical dashed line shows the strengthening of social distance on March 17, 2020.
Potential Effects of Previous or Next NPI
The researchers used modeling to estimate the effects of various scenarios on the Swiss epidemic. That is, if the NPI has been applied a week before or a week later, how has the incidence of infection, the number of cases treated at the hospital, and mortality, changed?
This study is based on a population-based transmission model and includes a gradual introduction of the NPI. Previous studies have shown that this causes a sharp reduction in the amount of effective reproduction (Re) in Switzerland.
Reduction of effective reproduction number of Re during the SARS-CoV-2 epidemic in Switzerland. Solid lines show the maximum likelihood estimates of the model and shaded area according to the 95% compatibility interval. We assume that the introduction of NPI sequentially results in a reduction in sigmoid transmission rates for about 2 weeks. The vertical dashed line shows the strengthening of social distance on March 17, 2020.
Using public data about the daily number of confirmed cases, hospitalization, admission to the intensive care unit (ICU), and virus-related deaths, the researchers found that initially, infections rose exponentially.
The researchers also estimate that the total number of infections is about ten times the number of cases diagnosed and confirmed initially. Then, the diagnosed cases account for around 20% of the total, but finally, after all the BOP is present, it drops again to around 10% of the total. The total estimate, therefore, amounts to around 264,000 cases.
Using a simulation of the previous NPI introduction, the model shows that around 2,000 deaths will occur if the NPI continues to negatively affect the number of deaths after May 10. If they had been implemented a week before, the model concluded that the peak number of hospitalizations would be much lower, at ~ 440, as would be the peak of ICU occupancy and death, respectively at ~ 88 and 400.
If the NPI is introduced a week later, these numbers will increase substantially, with a peak inpatient number of around 10,200, peak ICU occupancy of 1,900, and deaths of more than 8,000 each.
Doubling Time Triples with NPI
Current studies show that the effective reproduction rate (Re) fell from about 2.6 at the beginning to 0.64, which corresponds to a doubling time of about 3 days and 9 days at the start of the epidemic and post-lockdown phase, respectively. This underlines the urgency of the initial NPI in containing the spread of the virus.
The researchers commented, “One week of exponential increase in new infections during the spread of the initial epidemic requires 3.1 weeks of ‘locking’ to reduce the number of infections to the same level.”
Implications of this Study
This is the first study to analyze the possible effects of counterfactual scenarios regarding changes in the timing of NPI implementation. This finding is partly supported by lower mortality in Austria, which reported its first few cases at the same time as Switzerland, and who also fell victim to the spread from neighboring Lombardy, in Italy. However, the initial and strict introduction of the NPI within one week of the first case led to a reduction in the number of deaths.
In addition, the researchers showed that the high peak ICU occupancy rate of nearly 2,000 that could potentially result from a one-week delay in the NPI might flood the national health system, because the effective capacity of the ICU in Switzerland was only around 1,275 in April 2020.
In summary, this study concludes, “Our study illustrates that time is the most important when it comes to outbreak responses. We point out that while previous implementation of the NPI in Switzerland would result in a much lower number of cases and deaths being treated in hospitals, the steps taken succeeded in preventing the much higher morbidity and mortality burden associated with SARS-CoV-2. “
Because one week of exponential virus spread requires three weeks of locking to bring the number of daily cases back to their original level, the delay not only increases the number of deaths but extends the locking period. Therefore, the adoption of a stringent NPI will reduce the social and economic burden of these measures. Second, there are significant differences in the number of infections in various cantons, and this might indicate that new local outbreaks might be limited to separate areas.
* Important Notification
medRxiv publishing initial scientific reports that are not reviewed by colleagues and, therefore, should not be considered conclusive, guiding clinical practice / health-related behaviors, or treated as pre-existing information.