Effective pandemic policy design through feedback does not need accurate predictions.

Area of Science:

• Epidemiology
• Control Theory
• Public Health

Background:

• The COVID-19 pandemic caused significant health and socioeconomic disruptions.
• Public health interventions have historically managed epidemic curves, acting as feedback systems.
• Existing pandemic projections often lack explicit integration of decision-making feedback loops.

Purpose of the Study:

• To formalize and demonstrate the incorporation of feedback control principles into pandemic management.
• To shift focus from purely predictive modeling to designing effective interventions.
• To illustrate how feedback control can enhance pandemic projections and decision-making.

Main Methods:

• Developed an epidemiological model for COVID-19.
• Incorporated a feedback control design linking case numbers/hospitalizations to policy restrictions.
• Simulated pandemic scenarios with uncertainty in key parameters (e.g., R0, time to hospital admission).

Main Results:

• Demonstrated robust pandemic control using feedback responsive to hospital cases, even with high uncertainty in R0 and admission times.
• Shorter feedback delays and response to case counts reduced cumulative cases and intervention levels.
• Feedback control proved robust to changes in public compliance, variants, and vaccination programs.

Conclusions:

• Coupling disease propagation models with decision-making processes reduces pandemic impact.
• Feedback control offers a more realistic approach to public health decision-making than open-loop projections.
• Feedback principles enable effective control strategies under uncertainty, promoting timely, smaller interventions.