Modeling serological testing to inform relaxation of social distancing for COVID-19 control

The COVID-19 pandemic necessitated widespread social distancing measures. While effective in controlling initial outbreaks, these measures had significant social and economic costs. The study explores the potential of serological testing as a tool to safely relax social distancing. Serological testing could identify individuals with immunity (seropositive individuals), allowing them to resume more social interactions while minimizing transmission risks. The success of this approach depends on the accuracy of serological tests and the extent to which seropositivity correlates with immunity. The researchers aim to quantify the potential benefits of this strategy in reducing mortality and easing social restrictions, considering different testing frequencies and test specificities.

The paper reviews existing literature on the COVID-19 pandemic, including the impact of social distancing measures, the prevalence of SARS-CoV-2 infection (which significantly exceeded documented cases), and the performance of various serological assays for detecting SARS-CoV-2 antibodies. It highlights the variation in test performance, emphasizing the importance of high specificity (minimizing false positives) to avoid increasing transmission risk when using such tests to inform social distancing policies. The literature also indicates that while reinfections are possible, they are relatively rare, supporting the potential of using serological testing to identify individuals with substantial protection against reinfection. The review emphasizes the need to consider the trade-offs between relaxing social distancing and the risk of increased transmission due to imperfect test specificity.

The study uses a deterministic, compartmental transmission model to simulate the epidemiological consequences of mass serological testing. The model incorporates various factors, including age, gender, susceptibility, infection stages (latent, symptomatic, asymptomatic), and testing rates, which are functions of testing frequency, test sensitivity, and specificity. The model considers interactions within different settings (home, work, school, other) among three age groups: children/young adults (0-20 years), working adults (20-64 years), and elderly (65+ years). The model was calibrated to data from New York City, South Florida, and Washington Puget Sound, accounting for variations in social distancing measures and epidemic trajectories. A Markov Chain Monte Carlo (MCMC) approach was used to fit the model to time series data of deaths and cross-sectional seroprevalence. Simulations were performed to assess the impact of different serological testing strategies, incorporating various testing frequencies and test specificities, coupled with a 'serological shielding' strategy, where seropositive individuals return to normal levels of social interaction while susceptible individuals maintain reduced interaction.

The model predicts that without intervention, the pandemic would have resulted in substantial additional deaths across the three locations. Retrospective simulations indicate that widespread serological testing starting in late 2020, coupled with a serological shielding strategy, would have significantly reduced deaths. The effectiveness of this strategy depended strongly on test specificity and testing frequency. Using a highly specific test (99.8% specificity), monthly testing maximized the benefit, reducing deaths substantially and allowing a significant proportion of the population to safely resume normal activities. However, using a less specific test (90% specificity) reduced the effectiveness of the strategy, highlighting the critical importance of test accuracy. Sensitivity analyses explored the impact of varying other model parameters (e.g., transmissibility of asymptomatic infections, recovery rates), and found the effectiveness of shielding to be sensitive to the relative transmissibility of asymptomatic cases.

The findings demonstrate the potential benefits of using widespread serological testing combined with a shielding strategy to manage the COVID-19 pandemic, particularly in the pre-vaccine era. High test specificity is crucial for success. The results suggest that timely and accurate serological testing could have lessened the severity of the pandemic by enabling a more nuanced approach to social distancing. While mass testing is a resource intensive endeavor, the study notes that testing capacity in the US already grew significantly during the pandemic, suggesting that such a scale-up may be achievable. The study also discusses potential future applications, such as combining serological testing with vaccination status to identify individuals with immunity and safely reduce social distancing. The success of a serological testing strategy is contingent on the availability of highly accurate tests, and careful consideration of ethical implications, especially regarding potential disparities in access to testing.

The study shows that strategically implemented serological testing, alongside a shielding strategy prioritizing interactions between immune and susceptible individuals, could have significantly reduced COVID-19 mortality and facilitated the safe relaxation of social distancing. High test specificity is critical for this strategy’s success. The findings emphasize the potential of combining serological testing with vaccination to further optimize public health responses to current and future outbreaks. Further research could focus on optimizing testing strategies, addressing equity concerns, and considering the evolving landscape of new SARS-CoV-2 variants.

The study relies on several simplifying assumptions, including the immediate onset of detectable antibodies after infection, consistent immunity throughout the simulation period, and the assumption that antibody detection correlates with immunity against reinfection. The model assumes a random allocation of testing. Real-world implementation would likely involve targeted testing, potentially improving efficiency. The accuracy of predicting the trajectory of the pandemic is constrained by the imperfect quantification of social distancing practices. The model assumes that all other variables are constant and ignores any other external factors. These factors need to be addressed when employing the developed model.