Outline
■Outlook for Contact Rate Parameter（pp.4-12）
■Based on past correlations between various human traffic Measures and the contact rate parameter (after adjusting the economic activity level) in the Fujii-Nakata model, we calculated "the assumed contact rate parameter when the human traffic would return to the level of pre-Covid crisis"
■Outlook for Infection and the Economy（pp.13-21）
■Assuming that the state of emergency will be lifted in mid-October and that human traffic and the economic activity will be gradually restored to "pre-Covid crisis level" over a six-month period
■Analysis of the number of new infections at the time of undeclaration divided into the following two scenarios
■Basic Scenario: about 4,000 people
■Alternative Scenario: about 7,000 people
Important Points
■Large uncertainty in the contact rate parameter when human traffic and economic activity return to pre-Covid crisis
■Therefore, large uncertainty exists even in each of the scenarios considered
■Note that the number of basic reproduction consistent with the steady-state value of the contact rate parameter obtained from the human traffic Measures is lower than the number of basic reproduction obtained from "the number of basic reproduction*2 based on overseas data at the outset of Covid crisis (considering Delta/Alpha variants)"
■In both scenarios, a major wave of infection is likely to occur at the end of this year and the beginning of the new year
■At peak period, the number of new infections per day could exceed 20,000
■In both scenarios, the number of seriously ill patients and inpatients is likely to remain at a level that exceeds the current number of secured hospital beds until next spring
■At peak, it may double or triple current capacity of medical system
■If the COVID-19 medical system can be strengthened to properly care for patients who are seriously ill or whose condition suddenly changes at home (overnight), the average number of deaths per day is likely to remain below 20
■With vaccination of the elderly completed about 90%, and with gradual progress in vaccination of people aged 40-64
■Without proper cares, the number of deaths will increase significantly
Analytical Framework
■3 Time-Series models
■Linear, Nonlinear (Exponential), Nonlinear (SIR-Based)
■Objective variable: Time-Varying Contact Rate Parameter in the past Fujii-Nakata model
■Adjusted for the change in Alpha variant ratio, the change in Delta variant ratio, and vaccination rate
■Explanatory variable: Various human traffic Measures and temperatures
Human traffic Measures considered
Google Mobility Index
Relative infectivity of Delta variant to the conventional variant (1.9 times)
Google Mobility Index
Relative infectivity of Delta variant to the conventional variant (1.7 times and 2.1 times)
Voluntary restraint rate
Relative infectivity of Delta variant to the conventional variant (1.9 times)
Voluntary restraint rate
Relative infectivity of Delta variant to the conventional variant (1.7 times and 2.1 times)
Number of people staying in downtown areas
Relative infectivity of Delta variant to the conventional variant (1.9 times)
Number of people staying in downtown areas
Relative infectivity of Delta variant to the conventional variant (1.7 times and 2.1 times)
About Simulations
■In this simulation, economic activities are incorporated into the SIRD model, and the risk of spread of infectious diseases and economic losses are analyzed simultaneously.
■The following parameters and predicted values were used in the simulation. Parameters are estimated before applying the model.
■Parameters
・Relationship between people flow and GDP
・Delta mutation rate (percentage of patients infected with delta variants among those positive)
・Severity rate (percentage of patients with severe symptoms among those who tested positive)
・fatality rate (percentage of deaths among those who tested positive)
・Hospitalization rate (percentage of patients with positive results who are hospitalized)
■Predicted values
・Changes in the number of newly infected patients per day (average number of infected patients from Monday to Sunday)
・Number of vaccinations per day
■Model Details
Fujii and Nakata, “Covid-19 and Output in Japan",
https://covid19outputjapan.github.io/JP/files/FujiiNakata_Covid19.pdf
Assumption about future contact rate parameter
(About 4,000 newly infected patients per day at the time of the lifting of emergency declaration in mid-October)
β：Infection rate without consideration of people's economic activities
βtilde：Infection rate in consideration of people's economic activities
* The solid vertical black line represents actual values
Assumption about Percentage of Delta Variant (common to all scenarios)
* The solid vertical black line represents actual values
Serious illness rate (common to all scenarios)
*As vaccination progresses, the severity rate will remain at a lower level
*Criterias for severely ill patient
・metropolitan standard: Patients on mechanical ventilation or ECMO
・national standard: Patients requiring intensive care unit (ICU) management, ventilator management, or extracorporeal cardiopulmonary support (ECMO) management
* The solid vertical black line represents actual values
Hospitalization rate assumption (common to all scenarios)
*It is assumed that the hospitalization rate will be lower than before due to the effects of vaccination, etc.
* The solid vertical black line represents actual values
Fatality rate assumption (common to all scenarios)
* The solid vertical black line represents actual values
About 4,000 newly infected patients per day at the time of the lifting of emergency declaration in mid-October
*Assumptions other than the parameters shown on pages 15 to 19 are given on page 22.
Note 1: Thin lines are optimistic and pessimistic cases of contact rate, fatality rate, serious illness rate, and hospitalization rate parameters
Note 2: The number of deaths is in the case of all patients are cared properly through the strengthening of the medical system; otherwise, the number could increase significantly
About 7,000 newly infected patients per day at the time of the lifting of emergency declaration in mid-October
*Assumptions other than the parameters shown on pages 15 to 19 are given on page 22.
Note 1: Thin lines are optimistic and pessimistic cases of contact rate, fatality rate, serious illness rate, and hospitalization rate parameters
Note 2: The number of deaths is in the case of all patients are cared properly through the strengthening of the medical system; otherwise, the number could increase significantly
Details
■Fujii and Nakata (2021)： Covid-19 and Output in Japan
■https://covid19outputjapan.github.io/JP/, https://covid19outputjapan.github.io/JP/resources.html
■Assuming that after the state of emergency is lifted, human traffic and the economic activity will be recovered to pre-Covid crisis level over a six-month period
■On the left, the black vertical solid lines (middle and right) are the current time (4th week of August and 5th week of September). The two thin lines are the optimistic and pessimistic cases of the steady-state values of contact rate parameter
■The bottom-right panel is the cumulative number of deaths in one year (including deaths to date) and economic losses over the next year. The middle-right panel is the monthly GDP in the region (details of the estimation method will be explained in a paper to be published in the near future).
■It is assumed that the Alpha variant is 1.3 times more infectious than the conventional variant and that serious illness/fatality rate is 1.4 times higher. It is assumed that the Delta variant is 1.5 times more infectious than the Alpha variant and that serious illness/fatality rate is 1.1 times higher. The ratio of the Delta variant is assumed to currently be 90%
■It is assumed that the number of new infections will rise and then decline over the next 2 or 3 weeks
■An increased fatality rate due to the tightening of the medical system is not taken into account here
■Considering the season, the contact rate parameter is 1.1 times the steady-state value for the winter peak and 0.9 times for the summer Trough
■The vaccination pace is expected to be 1.2 million doses per day until the end of August (national equivalent, of which 700,000 doses are for the elderly) and it is assumed that the vaccination pace will gradually decline thereafter. It is assumed that 90% of the elderly wish to be vaccinated and 70% of the non-elderly wish to be vaccinated (80% of those who wish to be vaccinated are assumed to be aged 15 or older, and those under 15 are assumed to be unvaccinated). Vaccine efficacy is consistent with "SPI-M-O: Summary of further modelling of easing restrictions – Roadmap Step 4, 9 July 2021". See Fujii and Nakata (2021) for details of the model
