Participatory Mathematical Modeling Approach for Policymaking during the First Year of the COVID-19 Crisis, Jordan

Abstract

We engaged in a participatory modeling approach with health sector stakeholders in Jordan to support government decision-making regarding implementing public health measures to mitigate COVID-19 disease burden. We considered the effect of 4 physical distancing strategies on reducing COVID-19 transmission and mortality in Jordan during March 2020-January 2021: no physical distancing; intermittent physical distancing where all but essential services are closed once a week; intermittent physical distancing where all but essential services are closed twice a week; and a permanent physical distancing intervention. Modeling showed that the fourth strategy would be most effective in reducing cases and deaths; however, this approach was only marginally beneficial to reducing COVID-19 disease compared with an intermittently enforced physical distancing intervention. Scenario-based model influenced policy-making and the evolution of the pandemic in Jordan confirmed the forecasting provided by the modeling exercise and helped confirm the effectiveness of the policy adopted by the government of Jordan.

Keywords: COVID-19; Jordan; SARS-CoV-2; coronavirus disease; curfew; decision-making; modeling; respiratory infections; severe acute respiratory syndrome coronavirus 2; vaccine-preventable diseases; viruses; zoonoses.

Figure 1

Epidemiologic indicators and PHSMs in a COVID-19 modeling study, Jordan, March 2020–January 2021. A) Timeline of implemented PHSMs. Colors indicate individual PHSMs; level of shading represents the coverage of each intervention in the timeline, ranging from 0% to 100%. B) Estimated R_t, calculated using the EpiEstem package in R (https://CRAN.R-project.org/package=EpiEstim), which presents the number of new case-patients infected by an average case-patient at time t. Green shading indicates 95% CI. C) Daily incidence and mortality rates for COVID-19 in Jordan. PHSM, public health and social measure; Rt, effective reproduction number.

Figure 2

Model-predicted reported number of COVID-19 cases and cumulative number of associated deaths under scenarios A, B, C, and the baseline scenario in a COVID-19 modeling study, Jordan, March 2020–January 2021. Scenario A assumes the entire population, excepting essential services, will physically distance themselves for 24 hours every Friday while reverting to their usual behavior on the other days of the week. Scenario B assumes the population will physically distance themselves for the entire weekend (Friday and Saturday) while reverting to their usual behavior throughout the week. Scenario C assumes the entire population, except for essential services, will physically distance themselves for the entire week while never reverting to their usual behavior. Baseline scenario assumes no government intervention and half the population instinctively physically distancing themselves to avoid infection. Common to each scenario are 2 parameters used to define the extent of the physical distancing intervention: coverage, which refers to the percentage of the population following physical distancing regulations, and adherence, which refers to the extent to which individual persons follow those guidelines. On days when the interventions are not enforced, simulations assume 80% adherence and 50% coverage of the population practice physical distancing, while on days when the interventions are enforced it is assumed that 80% adherence and 90% coverage of the population physically distance themselves.

Figure 3

Model-predicted heat map showing percentage reduction in COVID-19 incidence (top row) and deaths (bottom row) in a COVID-19 modeling study in Jordan under 3 different scenarios (A, B, and C), relative to the baseline scenario, aggregated for the period November 2020–January 31, 2021. Dark blue corresponds to nearly 100% reduction in incidence and cases relative to the baseline scenario; dark red corresponds to 0% reduction. Scenario A assumes the entire population, excepting essential services, will physically distance themselves for 24 hours every Friday while reverting to their usual behavior on the other days of the week. Scenario B assumes the population will physically distance themselves for the entire weekend (Friday and Saturday) while reverting to their usual behavior throughout the week. Scenario C assumes the entire population, except for essential services, will physically distance themselves for the entire week while never reverting to their usual behavior. Baseline scenario assumes no government intervention and half the population instinctively physically distances themselves to avoid infection. Common to each scenario are 2 parameters used to define the extent of the physical distancing intervention: coverage, which refers to the percentage of the population following physical distancing regulations, and adherence, which refers to the extent to which individual persons follow those guidelines. The coverage parameter was varied between values of 50% and 100% (presented on the horizontal axis of each heat map) on the days when the physical distancing intervention was enforced. On respective days when the interventions were not enforced, simulations assume the coverage was constant at 50%. The adherence parameter varied between 0% and 100% (presented on the vertical axis of each heat map), remaining constant throughout each simulation.

Figure 4

Comparison of COVID-19 incidence (A) and cumulative deaths daily (B) under model scenario A compared with reported data in a COVID-19 modeling study, Jordan, March 2020–January 2021. Scenario A assumes the entire population, excepting essential services, will physically distance themselves for 24 hours every Friday while reverting to their usual behavior on the other days of the week. The scenario is defined by 2 key parameters: coverage and adherence. On days when the physical distancing intervention was enforced, the simulation assumes 60% of the population is following physical distancing regulations (coverage) and that those persons spend 80% of their time adhering to the intervention (adherence).

Figure 5

Percentage changes in mean mobility among the population, Jordan, February 2020–January 2021, including around retail and recreational facilities, grocery and pharmacy stores, parks, and transit locations. Google mobility data are used as a proxy for the population’s coverage and adherence to COVID-19–related physical distancing interventions.