Effectiveness of emergency declarations, booster vaccinations, and vaccine passports

Date: 2021.09.14
Researcher: Yukio Ohsawa
Organization: The University of Tokyo (School of Engineering)

Premise of the COVID-19 Room (draft)

Premise of this simulation

Model A: The framework is a lattice model of SEIRS circuit that allows for the analysis of inter-prefectural round trips and international round trips, taking into account the possibility that the movement of people may affect the spread of infection in the destination (Ohsawa and Hayashi 2021: https://arxiv.org/abs/2104.09719 )Refer to the next page for each circuit.

Double-R model with evolved lattice unit of SEIRS circuit (circuit equivalent to one region) of Model A

* Arrows indicate only transitions between initial states (N) and elliptical states as well as serious illness due to infection (including death).

Model A Results: The declaration lifted at the end of September with the slow attenuation of the vaccine’s effectiveness (antibodies decreased by 64% in 200 days). No vaccine passport

The declaration lifted in the middle of October with the slow attenuation of the vaccine’s effectiveness (antibodies decreased by 64% in 200 days). No vaccine passport

The declaration lifted at the end of September with the rapid attenuation of the vaccine’s effectiveness (antibodies decreased by 64% in 100 days). No vaccine passport

The declaration lifted in the middle of October with the rapid attenuation of the vaccine’s effectiveness (antibodies decreased by 64% in 100 days). No vaccine passport

The declaration lifted at the end of September with the slow attenuation of the vaccine’s effectiveness (antibodies decreased by 64% in 200 days). * With vaccine passport

The declaration lifted in the middle of October with the slow attenuation of the vaccine’s effectiveness (antibodies decreased by 64% in 200 days). * With vaccine passport

The declaration lifted at the end of September with the rapid attenuation of the vaccine’s effectiveness (antibodies decreased by 64% in 100 days). With vaccine passport

The declaration lifted in the middle of October with the rapid attenuation of the vaccine’s effectiveness (antibodies decreased by 64% in 100 days). * With vaccine passport

Model (B) of MultiLayer-MultiAgent for SEIR

*1. v1_det = min(v1, v1 - weeks elapsed since first vaccination * v_det_rate / v_det_term)
*2. v2_det = min(v2, v2 - weeks elapsed since second vaccination * v_det_rate / v_det_term)

Model (B) of MultiLayer-MultiAgent for SEIR: setting

・Data
　✔️Artificial synthesis data from Professor Murata of Kansai University
・Infection transmission model
　✔️Adopted SEIRS model after considering multi-agent based vaccine effectiveness.
　✔️Considered the transition of status I from mild to serious illness, and then from serious illness to death
・Vaccination
　✔️Vaccination reduces the transition probability of S/V/V1 nodes to E by -X%
　✔️Vaccine effectiveness declines gradually

Model B Results: Number of newly infected people

Even in a relatively slow vaccine effectiveness attenuation scenario, the number of newly infected people will increase unless the human flow is strictly controlled.

[[Settings for calculation]]
・10,000 households (about 10,000 agents), random number seed 1,
・[W-m0 related] family: fully combine, workplace: W=3, m0=1, school: W=4, m0=2, neighborhood: W=4, m0=2, Travel events: randomly with 2 persons

Model B Results: Cumulative number of infected people

With a virus of the expected infectiousness, it is difficult to control the cumulative number of infected people.

Model B Results: Number of seriously ill patients

If the vaccine prevents serious illness quite effectively, a sharp increase in the number of seriously ill patients can be avoided by reducing human flow by -10% to -25%

Model B Results: Cumulative number of deaths

The pace of the increase in the cumulative number of deaths will be relatively slow if serious illness is avoided (highly effective) by vaccines and human flow is controlled at a certain level.

Summary

(1) It is unlikely to cause a significant difference in the results whether lifting of the state of emergency declaration at the end of September or in the middle of October, but this depends on the speed of decrease of antibodies after two doses of vaccination (if they decrease rapidly, it is better to lift in the middle of October).

(2) The most effective way is a booster vaccination, and it is an absolute requirement (This is the third time that this point has been reported. Collaborative medical-engineering research has also been published at http://medg.jp/mt/?p=10452 , https://news.yahoo.co.jp/articles/a668c27e23f1fb7c58d0aa201be0fd1e10c81fdc). The difference in the spread of infection with and without booster vaccination appears from December, so it is advisable to start from this year. However, as shown in the previous reference, it is also important to equalize the vaccination rate (not humanitarian considerations) in each country around the world in order to increase vaccine effectiveness, so a balance needs to be considered.
(3) Booster vaccinations should be administered based on a physician's judgment of antibody attenuation and other conditions, not based on the proposals of "eight months after the second dose” or similar. It is possible to determine the need for booster vaccinations, not by directly looking at antibody titers, but by judging with the available information such as medical history and use of immunosuppressive drugs (Dr. Morihito Takita, Navitas Clinic).

(4) Regarding passports, there are few businesses that immediately give "preferential treatment to holders," and "they respond to non-holders in the same manner with those under emergency declarations and respond normally to holders," but the effect can still be fully expected.

(5) In conclusion, the combination of vaccine passport and booster vaccination is desirable, but the appropriate solution is to leave the booster vaccination to the discretion of physicians and the pharmaceutical distribution market.

* In the MultiAgent network model, some simulations show only a moderate increase in the number of newly infected people and number of serious illness at the end of the target period, even without booster vaccination. We believe this can be interpreted as a difference in the models and will be discussed in the future.