Log in to MyACC
Timing and Adherence to Social Distancing Measures in COVID-19
Quick Takes
- An agent-based simulation model incorporating transmissibility parameters of SARS-CoV-2 in addition to population-related parameters can help predict case trends for SARS-CoV-2.
- Social distancing measures and their early implementation are effective in reducing transmissibility of SARS-CoV-2.
- The impact of social distancing measures depends on regional characteristics such as population density.
Study Questions:
Can an agent-based simulation model help assess the effectiveness of social distancing measures?
Methods:
The authors used an agent-based model—a class of computational models that can simulate the actions and interactions of autonomous agents such as humans to simulate severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) transmission dynamics. The model represents social network and interactions among persons in a region, considering local population demographic characteristics, population density, the daily number of contacts in the absence of social distancing measures, and adherence to social distancing measures. Major assumptions include that all persons are susceptible to coronavirus disease 2019 (COVID-19), no vaccine is available, and there is no pre-existing immunity. Inputs for the various parameters were derived from estimates of peer-reviewed literature, and data from three different regions: Dane County, Wisconsin, Milwaukee, and New York City (NYC). Adherence to social distancing measures was assessed using cellphone mobility data. The authors simulated the effects of adherence to social distancing, timing of implementation of social distancing, and lastly, timing of easing of social distancing measures.
Results:
Adherence to social distancing has a substantial effect on the cumulative number of SARS-CoV-2 cases. For example, no social distancing other than closing schools compared to 50% adherence increased the total number of cases from 56,433 to 487,501 in NYC in just 26 days. Timeliness of implementation of social distancing measures also impacts cases greatly: earlier implementation by 1 week could have reduced the number of cases in NYC by 80%, whereas a 1-week delay could have increased the number of cases by sevenfold. The effect of implementing social distancing measures depends highly on the region because each has different levels of adherence and transmission. For example, delayed implementation by 1 week in Dane County could increase the numbers by 36%, while in NYC, cases would have increased by 539%, due to the population density. Similarly, earlier easing of measures and decrease in adherence by 15% could have doubled the number of cases from June 8–July 31.
Conclusions:
The timing of implementing and easing social distancing measures has major effects on the number of SARS-CoV-2 cases, of which the magnitude depends on regional specificities.
Perspective:
The authors provide convincing evidence through agent-based modeling that social distancing measures represent an effective approach to mitigate increases in SARS-CoV-2. The findings show that the impact of measures is highly dependent on regional characteristics. While these findings are intuitive, the modeling can serve to guide policy makers in the management of the resurgence. There are important caveats that the modeling does not account for, notably the potential negative impact of aggressive social distancing measures such as lockdowns on other health and economic outcomes. Moreover, while case numbers have increased, health outcomes related to COVID-19 have shifted dramatically, with a significant reduction in disease severity and in-hospital mortality. Case numbers should not be the sole guiding metric in policy decisions.
Clinical Topics: Cardiovascular Care Team, COVID-19 Hub, Prevention
Keywords: Cell Phone, Coronavirus, COVID-19, Immunity, Outcome Assessment, Health Care, Population Density, Primary Prevention, severe acute respiratory syndrome coronavirus 2, Psychological Distance, Systems Analysis
< Back to Listings
