• Stephen Grice

Scenario Modelling using the tool for recent New Zealand conditions

Updated: Apr 16

A modelling tool demonstrates what-if scenarios using New Zealand input variables and initial conditions. The results are presented. There is a brief discussion of findings is given.


You can find the modelling tool here and the scenario input files used in the blog post are included in this zip file here. Download the zip file and extract the files out of the zip file and save them to your file system somewhere easily accessible to use in the modelling tool. Once you have loaded an input file you can change the mitigation periods and levels to perform your own scenario simulations. We recommend only using up to date browsers with this software.


Caution must be exercised in drawing conclusions from this data as future scenarios described are speculative and the model is only as good as its input values. The input values are described in the assumptions section below with qualifying notes. Note, however, the input values are copied from, derived and extended from the Hendy et al model published on 26 March 2020.


Scenario 1 - "Let it rip - no mitigation"

The first scenario simulates the disease profile if COVID-19 Alert Levels and social distance measures had not been applied in New Zealand. Load the let-it-rip input file and click "Run simulation" button. The results are as follows:



Scenario 2 - "Flatten the curve"

The second scenario simulates what the disease profile looks like today. The simulation applies COVID-19 Alert Level 2 between 24 March 2020 and 26 March 2020, then shifts to COVID-19 Level 4 on 26 March 2020, and assumes COVID-19 Level 0 (no mitigation) from 1 May 2020. Load the flatten-the-curve input file and click "Run simulation" button. The results are as follows:



Scenario 3 - "Eliminate" The last "Eliminate" scenario is the same as the "Flatten the curve" scenario except that COVID-19 Level 2 mitigation is applied from 1 May 2020. Load the eliminate input file and click "Run simulation" button. The results are as follows:




Discussion

The modelling tool allows to you to apply COVID-19 Alert levels to the Hendy model developed for New Zealand using the publicly available data. The data is current as of 11 April 2020 and is provided to develop predictive what-if scenarios and planning strategies.


The 3 simple conclusions from the scenarios are as follows:

  1. The "let it rip" scenario would have been extremely costly both in terms of lives lost and economic damage. The Government's decision to implement COVID-19 Level 4 avoided this outcome.

  2. The implementation of "Flatten the curve" strategy shifts the time taken for the infection to peak and slightly decreases the effects of the pandemic. But the ultimate outcome still has high costs albeit time shifted.

  3. The "Eliminate" strategy further increases the time taken for infection to peak and decreases the amount of the peak compared to "Flatten the curve" scenario. If the R0 value (beta) for COVID-19 Level 2 is > 1 then the consequences will require a further elevated COVID-19 Level 4 period later in the year if we are to avoid a second wave of contagion.


Assumptions

All the assumptions are taken from the Hendy Model specification.

The R0 values are a function of COVID-19 Alert levels are as follows:

R0(Level 0) = 2.5, R0(Level 1) = 2.0, R0(Level 2) = 1.5, R0(Level 3) = 1.1, R0(Level 4) = 0.5. Levels 0 and 4 are derived from the Hendy Model 26 March 2020. Levels 1, 2 and 3 are arbitrary values that are evenly spread in the infectious range R0 > 1 to R0 < R0(Level 4).

The remaining population initial conditions are taken from the New Zealand Government Ministry of Health website on COVID-19 current data.

The total population for New Zealand was obtained from Statistics New Zealand.



225 views

©2020 by CloseAssociate Covid  |  Terms & Conditions Privacy Policy