Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

First 100 days of pandemic; an interplay of pharmaceutical, behavioral and digital interventions -- A study using agent based modeling

Gauri Gupta, Ritvik Kapila, Ayush Chopra, Ramesh Raskar

TL;DR

This paper demonstrates that agent-based models can capture the nuanced interplay of pharmaceutical, behavioral, and digital interventions during a pandemic, using a 100k-agent Kings County, WA population over 180 days. It introduces a vectorized ABM framework with four interventions—Testing, Self-quarantine, Vaccination, and Contact Tracing (including digital and manual variants)—and evaluates their standalone and combined effects. Key findings show that the initial 100 days largely shape the pandemic's trajectory, and that integrating behavioral and digital strategies with vaccination significantly reduces infections and delays the peak, often with superior cost-effectiveness to vaccination alone. The work provides a practical decision-support framework with a configurable pipeline and real-world data, highlighting policy implications for rapid, multi-pronged pandemic responses.

Abstract

Pandemics, notably the recent COVID-19 outbreak, have impacted both public health and the global economy. A profound understanding of disease progression and efficient response strategies is thus needed to prepare for potential future outbreaks. In this paper, we emphasize the potential of Agent-Based Models (ABM) in capturing complex infection dynamics and understanding the impact of interventions. We simulate realistic pharmaceutical, behavioral, and digital interventions that mirror challenges in real-world policy adoption and suggest a holistic combination of these interventions for pandemic response. Using these simulations, we study the trends of emergent behavior on a large-scale population based on real-world socio-demographic and geo-census data from Kings County in Washington. Our analysis reveals the pivotal role of the initial 100 days in dictating a pandemic's course, emphasizing the importance of quick decision-making and efficient policy development. Further, we highlight that investing in behavioral and digital interventions can reduce the burden on pharmaceutical interventions by reducing the total number of infections and hospitalizations, and by delaying the pandemic's peak. We also infer that allocating the same amount of dollars towards extensive testing with contact tracing and self-quarantine offers greater cost efficiency compared to spending the entire budget on vaccinations.

First 100 days of pandemic; an interplay of pharmaceutical, behavioral and digital interventions -- A study using agent based modeling

TL;DR

This paper demonstrates that agent-based models can capture the nuanced interplay of pharmaceutical, behavioral, and digital interventions during a pandemic, using a 100k-agent Kings County, WA population over 180 days. It introduces a vectorized ABM framework with four interventions—Testing, Self-quarantine, Vaccination, and Contact Tracing (including digital and manual variants)—and evaluates their standalone and combined effects. Key findings show that the initial 100 days largely shape the pandemic's trajectory, and that integrating behavioral and digital strategies with vaccination significantly reduces infections and delays the peak, often with superior cost-effectiveness to vaccination alone. The work provides a practical decision-support framework with a configurable pipeline and real-world data, highlighting policy implications for rapid, multi-pronged pandemic responses.

Abstract

Pandemics, notably the recent COVID-19 outbreak, have impacted both public health and the global economy. A profound understanding of disease progression and efficient response strategies is thus needed to prepare for potential future outbreaks. In this paper, we emphasize the potential of Agent-Based Models (ABM) in capturing complex infection dynamics and understanding the impact of interventions. We simulate realistic pharmaceutical, behavioral, and digital interventions that mirror challenges in real-world policy adoption and suggest a holistic combination of these interventions for pandemic response. Using these simulations, we study the trends of emergent behavior on a large-scale population based on real-world socio-demographic and geo-census data from Kings County in Washington. Our analysis reveals the pivotal role of the initial 100 days in dictating a pandemic's course, emphasizing the importance of quick decision-making and efficient policy development. Further, we highlight that investing in behavioral and digital interventions can reduce the burden on pharmaceutical interventions by reducing the total number of infections and hospitalizations, and by delaying the pandemic's peak. We also infer that allocating the same amount of dollars towards extensive testing with contact tracing and self-quarantine offers greater cost efficiency compared to spending the entire budget on vaccinations.
Paper Structure (19 sections, 12 figures, 4 tables)

This paper contains 19 sections, 12 figures, 4 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Method
  4. Testing
  5. Self-quarantine(SQ)
  6. Vaccination(VACC)
  7. Contact Tracing (CT)
  8. Results
  9. Analysis of individual impact of different interventions
  10. Age stratification of infections for different interventions
  11. Cost analysis of individual interventions
  12. Coupled effect of pharmaceutical, behavioral, and digital interventions
  13. Geographical spread
  14. Discussion
  15. Conclusion
  16. ...and 4 more sections

Figures (12)

  • Figure 1: Implementation of different interventions - Testing, Self-quarantine, Vaccination, and Contact Tracing. (1) Infection spreads through the interaction of infected with susceptible agents, and the states of the agents are then updated based on disease progression. (2) Upon experiencing symptoms, exposed agents get themselves tested (3a) If tested positive, agents undergo self-quarantine with compliance. A quarantined agent then engages in no further interactions until the quarantine period ends. The interaction graph of quarantine agents is thus an isolated point (3b) Agents that have not tested positive or are not quarantined get vaccinated. Vaccination reduces the susceptibility of an agent to infection risk (3c) In case of contact tracing: interactions of the positively tested agents (that own app in case of DCT) from the previous interaction graphs of past days are tracked; (4c) exposure notifications are sent to the possibly exposed tracked agents (that own the app in case of DCT); (5c) notified agents then opt for self-quarantine. (Last) After simulating for N days, the aggregate statistics of the agent states are computed. Agent states here are: susceptible (S), exposed (E), infected (I), recovered (R), mortal (M), and vaccinated (V)
  • Figure 2: Comparison of Digital vs. Manual Contact Tracing: Digital tracing requires app ownership for both interacting agents but can effectively track unknown or random interactions, while manual tracing captures household and occupational contacts but may miss random interactions
  • Figure 3: Comparative analysis of the individual impact of different interventions on pandemic progression; No Interventions (NI), Self-Quarantine (SQ), Vaccination (VACC), and Contact Tracing (CT). (a) Peak hospitalizations showcase the strain on healthcare under each scenario, with notable stress in the NI and SQ cases. The dotted line represents the hospital bed availability for Kings County, Washington (b) Daily new infection rates highlight the efficacy of interventions, with CT significantly lowering the infection rate. (c) Cumulative infections over time reveal the pervasive nature of the pandemic in the absence of effective measures and a substantial reduction in total infections under VACC, SQ, and CT.
  • Figure 4: Age-stratified cumulative infections in Kings County, Washington, illustrating the impact of contact tracing (CT), self-quarantine (SQ), and vaccination (VACC) intervention scenarios on different age groups.
  • Figure 5: Comparison of costs associated with different intervention strategies. The figure shows contact tracing (CT) is the most cost-effective over both self-quarantine (SQ) and vaccination (VACC); excluding $0 cost for no-intervention (NI)
  • ...and 7 more figures