Evidence for Behavioral Autorepression in Covid-19 Epidemiological Dynamics

Area of Science:

• Epidemiology
• Mathematical Modeling
• Behavioral Science

Background:

• Epidemiological forecasting models often omit human behavioral responses due to a lack of empirical evidence.
• Behavioral autorepression, where population behavior self-regulates epidemic spread, has been hypothesized but not empirically demonstrated.
• This omission can lead to significant inaccuracies in epidemic forecasting and intervention planning.

Purpose of the Study:

• To provide empirical evidence for behavioral autorepression during the COVID-19 pandemic.
• To quantify the strength of behavioral autorepression across different populations.
• To assess the impact of behavioral autorepression on epidemic forecasting and the efficacy of non-pharmaceutical interventions.

Main Methods:

• Analysis of early COVID-19 dynamics across 96 countries, 50 US states, and 562 US counties.
• Identification of behavioral autorepression signatures through mobility-death correlations, fluctuation-magnitude analysis, and infection wave dynamics.
• Incorporation of behavioral autorepression into conventional epidemiological models for forecasting and analysis.

Main Results:

• Evidence for behavioral autorepression was identified across multiple scales using distinct analytical approaches.
• Accurate forecasting of COVID-19 mortality required the inclusion of behavioral autorepression in models.
• The strength of behavioral autorepression was found to influence the effectiveness of non-pharmaceutical interventions.

Conclusions:

• Behavioral autorepression is a significant factor in regulating epidemic dynamics, as evidenced by COVID-19 data.
• Incorporating behavioral autorepression into models substantially improves epidemic forecasting accuracy.
• Understanding and quantifying behavioral autorepression can lead to more effective public health interventions and tailored strategies for future epidemics.