The Impact of Host-Based Early Warning on Disease Outbreaks

Hernandez, Mark; Milechin, Lauren E; Davis, Shakti K; DeLaura, Rich; Claypool, Kajal T; Swiston, Albert

Objective: The detection of communicable pathogens responsible for major outbreaks relies on health care professionals recognition of symptoms manifesting in infectious individuals. Early warning of such communicable diseases before the onset of symptoms could improve both patient care and public health responses. However, the potential impact of such a host-based early warning system on containing the spread of an outbreak and in steering public health response is unknown. Methods: We extend the deterministic SEIR (Susceptible, Exposed, Infectious, Recovered) model to simulate disease outbreak scenarios and to quantify the potential impact of a host-based early warning capability to mitigate pathogen transmission during an outbreak. In particular, we compare and contrast the performance of five different policies: Self-monitoring and reporting (baseline SEIR model), Quarantining the entire population, Quarantine-on-alert (with high sensitivity early warning), Quarantine-on-alert (with high specificity early warning), and Quarantine-on-alert (ideal early warning). We further evaluate these five policy options against four different outbreak scenarios with high or low disease transmission and high or low initial population exposures. Results: For all scenarios, a quarantine-on-alert policy coupled with the near-ideal early warning capability reduces quarantine needs with only a small increase in the number of additional infections. The cost of a highly specific early detection system (i.e., a reduction in false alarms and thus quarantine costs) is an increase in additional infections relative to the near-ideal system. Conversely, a highly sensitive early detection system increases the percentage of the population in quarantine compared to both the ideal and high-specificity early detection system while also reducing the number of additional infections to nearly the numbers seen by quarantining the entire population a priori. Conclusions: Our simulations demonstrate the utility of host-based early warning systems in controlling an outbreak under various outbreak conditions. Our tools also provide a simulation capability for evaluating public health policies enabling quantitative evaluation of their impacts prior to implementation.