dcyphr | Differential Effects of Intervention Timing on COVID-19 Spread in the United States


It is important to understand how non-medical interventions impact COVID-19 transmission, especially to plan future measures to combat the ongoing pandemic. In this study, the researchers used data from American counties of reported infections and deaths along with data about how people move from place to place and a mathematical transmission model. Combining these three things, the researchers quantified changes in COVID-19 transmission in the U.S from March 2020 to May 2020. They found that the basic reproduction number (R0), which is the number of susceptible people who will be infected by one infected person, was reduced in major metropolitan areas when they used social distancing and other control measures. Mathematical simulations of what would’ve happened if these measures had been implemented 1-2 weeks earlier showed that a substantial number of deaths and infections would have been avoided. Specifically, 56.5% of reported infections and 54.0% of reported deaths would have been avoided if interventions were implemented 1 week earlier. The researchers also examined what would happen if there were delays in re-implementing social distancing after relaxing control measures. They found that a longer response time would result in a stronger rebound of infections and death. The results of the study show the importance of early intervention and against response in controlling COVID-19.


The researchers wanted to understand how non-medical interventions, such as social distancing, impacted the transmission of COVID-19. They also wanted to see what would’ve happened if these interventions were implemented 1-2 weeks earlier. Finally, they wanted to see what would happen if re-implementation of social distancing was delayed after current control measures are relaxed.


COVID-19 has quickly spread in the United States after its first case in January 2020. Beginning in mid-March, control measures enforcing social distancing were implemented across the U.S. In other countries, these measures successfully controlled the spread of COVID-19. However, in the U.S the efficacy of these control measures has been less evident. It is important that changes in COVID-19 transmission by implementation of non-medical control measures be quantified in the U.S, so that the effects of earlier interventions on cases and deaths can be evaluated.


The researchers used human mobility data (data about how people travel across the country) to inform a dynamic metapopulation model. This model represented SARS-CoV-2 transmission in 3142 U.S counties, and simulates documented and undocumented infections with separate transmission rates. They defined a separate transmission rate for counties with greater than 400 cumulative cases to reflect heterogeneity in transmission rates between counties. They also included the following parameters: ascertainment rate (the fraction of infections documented as confirmed cases), the latency period (the time that passes between being exposed to SARS-CoV-2 and having symptoms of COVID-19), the average duration of infection, and a parameter for how travel impacts transmission.


In six metropolitan areas (New York, New Orleans, Los Angeles, Chicago, Boston, and Miami), the basic reproduction number decreased after March 15th after social-distancing politics and practices were implemented, and continue to decrease. The ascertainment rate slowly increased in all six areas after April 5. These results indicate that non-medical interventions adopted in the U.S effectively reduced COVID-19 transmission in metropolitan areas.

During the initial growth of a pandemic, infections increase exponentially. This means that a fast response and early intervention are important to limit infections and deaths. The researchers quantified the effect of earlier intervention in the U.S on infections and deaths. They performed two simulations in which the interventions were pushed back 1 and 2 weeks. The researchers found that if measures were implemented 1 week earlier, the U.S would have avoided 56.5% of cases and 54.0% of deaths nationwide. In New York, the pandemic epicenter of the U.S, 83.3% of cases and 85.6% of deaths would have been avoided. These percentages increased if implementations happened 2 weeks earlier. These dramatic reductions show the need for early response to the COVID-19 pandemic.

Rapid response is also essential to avoiding rebounds in infections and deaths in locations that are reopening. The researchers quantified the effects of response time after reopening on infections and deaths using more simulations. They assume that control measures are relaxed beginning May 4, 2020 in all counties, resulting in an increased basic reproductive number. They used a response time of 2 and 3 weeks with a 25% weekly reduction in transmission rates. They found that for both 2 and 3 weeks, the number of daily confirmed cases declined for almost 2 weeks after easing control measures. This decreasing trend, along with a lab between actual infection and confirming infection, communicates a false sense of security that COVID-19 is under control. However, the researchers found a large rebound in cases and deaths peaking in early- and mid-June even after the implementation of control measures. A one-week delay in re-implementation of these measures resulted in 32,379 additional deaths by July 1, 2020.


Modeling of disease transmission is based on hypothetical assumptions, and does not reflect the real-world complexities of implementing public health interventions. However, countries with early responses to the viruses have seen a sharp decline in infections and deaths. This study found that if control measures were implemented in the U.S at a similar time, a dramatic decrease in infections and deaths could have been observed. The findings emphasize the need for continued caution after reopenings in various areas of the country. It is vital to balance the economic need for reopening and the health consequences it could cause. Many countries have achieved this balance. The strategies in these countries could be used to guide politics in the U.S, including broader testing and contact tracing.