Optimal control by social distancing in pandemic preparedness: cost prioritization, adherence and fatigue

Developing control principles for future pandemics is crucial for public health systems. Learning from COVID-19 entails understanding the tension between social distancing’s direct health costs and indirect societal-level costs, impacting control options from elimination to suppression and to mitigation. Optimal control is a promising tool for this goal. In this presentation, I’ll discuss open-loop optimal control of a COVID-19 transmission model during a very aggressive outbreak, aiming to identify social distancing interventions that balance direct epidemiological costs with indirect costs. In particular, I will compare the results of a pre-behavioral model, assuming fixed adherence as a free constant, with a more complex one where adherence is endogenized through an evolutionary game equation involving individuals’ fatigue. As for the pre-behavioral model, I will show how optimal social distancing responses vary based on prioritization of indirect costs, adherence, intervention timing, and ”allowed” traveler inflow. In detail, Increasing prioritization of indirect costs shifts the optimal response from elimination to suppression and ultimately to mitigation; furthermore, the ”effective” mitigation zone, where hospitals’ capacity is never overwhelmed, is notably narrow. Ultimately, a delicate balance between adherence and timeliness of governmental response emerges, where low adherence and delayed response unavoidably lead to ineffective mitigation as the sole option. When adherence is endogenized, although the model behaviour obviously becomes richer, cost prioritization remains the key determinant of whether suppression rather just mitigation will be selected as the optimal policy. However, the main implication of including fatigue is that the optimal strategies will always be intermittent. These findings show the importance of optimal control, traditionally absent in public health preparedness, as a tool to inform robust preparedness guidelines.