Quantitatively assessing early detection strategies for mitigating COVID-19 and future pandemics

The COVID-19 pandemic highlighted the need for improved early warning systems for novel pathogens. While non-pharmaceutical interventions (NPIs) played a role in mitigating the pandemic's impact, the potential of earlier detection to improve mitigation strategies has received limited quantitative investigation. This study addresses this gap by focusing on the potential impact of early detection on pandemic outcomes. Several proposed systems aim to detect novel pathogens earlier than current surveillance methods, including monitoring hospital patients, wastewater, and air travel. These systems involve continuous monitoring using multiplex testing for conserved nucleic acids of known pathogen families. The primary research question is to quantitatively assess how much earlier these systems would detect outbreaks and, consequently, how much they could improve pandemic mitigation. This question carries significant weight given the substantial financial investments proposed for such systems.

Existing literature extensively explores the impact of various NPIs on COVID-19 transmission, including mobility restrictions, school closures, and testing policies. These studies optimized NPI parameters such as testing frequency and quarantine length. However, the focus on early detection strategies remains limited. While early detection theoretically enables intervention at smaller outbreak sizes, thus making mitigation strategies more cost-effective, the quantitative impact remains unclear. This paper seeks to fill this knowledge gap, providing a quantitative assessment of the potential benefits of early detection systems.

The researchers developed and empirically validated a quantitative model to simulate disease spread and detection time for various diseases and detection systems. The model uses a branching process simulation, incorporating epidemiological parameters such as R0 (basic reproduction number), serial interval, dispersion, hospitalization rate, and time to hospitalization. The model simulates three detection systems: hospital monitoring, wastewater monitoring, and air travel monitoring. The model considers different delays in detection for each system, for example, the delay between infection and hospitalization compared to the delay between infection and fecal shedding in wastewater. The detection threshold for individual-based systems (hospital and air travel) is an absolute number of cases, while for community-based systems (wastewater), it's measured in prevalence (cases as a percentage of the population). The model parameters were derived from existing literature and data sources, including studies on COVID-19 epidemiology, wastewater surveillance, and case counts in various states. The model was validated by comparing its predictions to observed COVID-19 detection times in the US, taking into account various delays and underreporting factors. A compact formula was derived to approximate the model's simulations, facilitating its application to a broader range of diseases and detection systems. The formula helps interpret the relationship between detection times, epidemiological parameters, and detection system characteristics.

The model's application to the COVID-19 outbreak in Wuhan showed that hospital monitoring could have detected the outbreak approximately 0.4 weeks earlier, detecting around 2,300 cases compared to the actual 3,400 statistically confirmed cases at the time of discovery. Wastewater monitoring, however, would not have accelerated detection in Wuhan but showed benefits in smaller catchments. Air travel monitoring did not significantly accelerate outbreak detection in most scenarios. The researchers then applied the model to other diseases (mpox, polio, Ebola, and influenza), demonstrating that detection system effectiveness varied with epidemiological parameters. Hospital monitoring outperformed wastewater monitoring for diseases with high hospitalization rates (e.g., Ebola), while the reverse was true for diseases with low hospitalization rates (e.g., polio). Wastewater monitoring performed better in smaller catchments. The study found that early detection systems could detect outbreaks up to 52% smaller (wastewater for polio) or 110 weeks earlier (hospital for HIV/AIDS) than the status quo. A generalized analysis across a wide range of epidemiological parameters confirmed the findings, highlighting that hospital monitoring is best for diseases with higher R0 and shorter times to hospitalization, while wastewater monitoring is advantageous for diseases with longer times to hospitalization and higher fecal shedding probabilities. The mathematical formula derived closely approximated the model simulations, enhancing its usability and allowing for clearer interpretation of results.

The findings demonstrate that the benefits of early detection systems vary greatly depending on the specific disease and the characteristics of the detection system. While some systems may offer substantial advantages for certain diseases, as seen in the significant improvement for HIV/AIDS, the impact on COVID-19 was relatively modest. The study emphasizes that effective early detection requires a rapid and coordinated response to the detected outbreak. The effectiveness of these systems hinges upon timely implementation of pre-determined policies upon detection and factors beyond detection such as resource availability, economic feasibility of interventions, and political considerations. While wastewater monitoring may offer earlier detection in some instances, it lacks the ability to differentiate between mild and severe disease unlike hospital monitoring. Cost-effectiveness and the quality of evidence provided by the system should inform investment decisions.

This study provides a quantitative assessment of the potential impact of various early detection systems for mitigating pandemics. The findings highlight the variable effectiveness of different strategies depending on the epidemiological characteristics of the disease. While early detection can offer significant benefits in some cases, a coordinated and rapid public health response is crucial for effective mitigation. The developed model and derived formula offer valuable tools for future research and policymaking in pandemic preparedness.

The model relies on assumptions about epidemiological parameters and detection system characteristics, which may not perfectly reflect real-world scenarios. The study’s focus is on detection time and does not explicitly incorporate the cost-effectiveness of different detection systems. Further research considering cost-benefit analysis would enrich the understanding of the practical implications of early detection strategies.