Di Digital Symptom Che hecker an and Trac acin ing Tool l fo - - PowerPoint PPT Presentation

di digital symptom che hecker an and trac acin ing tool l
SMART_READER_LITE
LIVE PREVIEW

Di Digital Symptom Che hecker an and Trac acin ing Tool l fo - - PowerPoint PPT Presentation

Di Digital Symptom Che hecker an and Trac acin ing Tool l fo for SA BASA COVID-19 Project Con Context A review of the health and economic projections for South Africa indicates the impact on both is very significant and while immediate


slide-1
SLIDE 1

Di Digital Symptom Che hecker an and Trac acin ing Tool l fo for SA

BASA COVID-19 Project

slide-2
SLIDE 2

Con Context

  • A review of the health and economic projections for South Africa indicates the impact on both is very significant and while immediate

steps have to be taken to open the economy, additional steps are required to ensure worker safety and to limit the spread of the virus

  • Current COVID – 19 scenarios (from the Actuarial Society of SA) indicate that most of the SA population will be infected

eventually, resulting in 45,000-90,000 deaths (an annual increase of 10-20%).

  • Economic impact of remaining in a relatively high level of lockdown (phased lowering based on health projections by province)

could result in GVA loss of ~R0.8-1.0tn over a 12 month period or almost 20% reduction in GDP. To avoid this outcome, we are expecting a much faster opening up of the economy.

  • While lockdown has lowered the growth rate of COVID-19 to an R0 of ~1.4 (estimated), this growth rate will still result in close to above
  • utcomes (albeit lower infection and mortality). Many countries have lowered R0 below 1, effectively suppressing the virus, resulting

in an accelerated return to a post-COVID-19 health and economic scenarios. Improved hygiene, social distancing and use of PPE (personal protective equipment), etc. all contribute to lowering transmission. Yet re-opening from lock-down and winter season will increase transmission rates. So more is needed to lower transmission.

  • The most effective way of lowering transmission is to identify and quarantine/isolate infected and potentially exposed individuals as

quickly as possible, which also has by far the lowest financial and social cost. Research* indicates COVID-19 spreading takes place predominantly in pre-symptomatic (45%) and symptomatic (40%) states, with lower transmission from environmental (10%) and asymptomatic (5%) vectors. “You can control the epidemic if you can isolate 60% of patients as soon as they have symptoms, and trace

  • ver 50% of their contacts instantaneously and before they infect anybody.” A 30% reduction (50%* 60%) in the current R0 of ~1.4 in

SA would be sufficient to stop the growth at this point in time, while a higher success rate would lead to suppression if maintained.

  • Today we will propose the use of Apps and other digital channel technology to achieve this, which has been the way the countries

who have beaten the virus have done it. The role of the employer in adopting technology to ensure safety of the workplace for their employees, customers and other parties is critical to mitigate the risk introduced by covid-19.

​* Science 2020/04/09/science.abb6936 “Quantifying SARS-COV-2 transmission suggests epidemic control with digital contact tracing”. ​Source: StatsSA, Actuarial Society of South Africa – COVID outcomes 27 April 2020, FRB Epidemiological modelling; FRB and BSA Economic modelling

slide-3
SLIDE 3

Th The short-te term economic cost is massive - ~65% % drop in point-of

  • f-sa

sale spe spend nd measur sured d dur during ng Level 5 lockdo down n

200 400 600 800 1,000 1,200

  • 1. Lead Monday
  • 2. Lead Tuesday
  • 3. Lead Wednesday
  • 4. Lead Thursday
  • 5. Lead Friday
  • 6. Lead Saturday
  • 8. Lead Monday - Phase 1
  • 9. Lead Tuesday
  • 10. Lead Wednesday
  • 11. Lead Thursday
  • 12. Lead Friday
  • 13. Lead Saturday
  • 14. Lead Monday
  • 15. Lead Tuesday - Phase 2
  • 16. Lead Wednesday
  • 17. Lead Thursday
  • 18. Lockdown Friday
  • 19. Lockdown Saturday
  • 20. Lockdown Monday
  • 21. Lockdown Tuesday
  • 22. Lockdown Wednesday
  • 23. Lockdown Thursday
  • 24. Lockdown Friday
  • 25. Lockdown Saturday
  • 26. Lockdown Monday
  • 27. Lockdown Tuesday
  • 28. Lockdown Wednesday
  • 29. Lockdown Thursday

Millions

Daily Card Issuing Turnover

Groceries General Retail Dining Out & Entertainment Tourism Fuel Automotive

15 March - State of Disaster declared 23 March - Lockdown announced 27 March - Lockdown effective

3

slide-4
SLIDE 4

4 50,000 100,000 150,000 200,000 250,000 300,000 350,000 400,000 Feb-20 Mar-20 Apr-20 May-20 Jun-20 Jul-20 Aug-20 Sep-20 Oct-20 Nov-20 Dec-20 Jan-21 Feb-21 Mar-21

Economic Scenario by Province

Northern Cape Free State North-West Mpumalanga Eastern Cape Limpopo Western Cape KwaZulu-Natal Gauteng

Th The projected mediu ium-te term economic impact is massive, even unde under a pr projected d lowered d lockdo down n sc scena nario

​351 ​264 ​377 ​308 ​278 ​210 ​-7% ​-44% ​-26% ​-18%

Apr 20 Jun 20 Aug 20 Oct 20 Dec 20 Eastern Cape L5 L4 L4 L3 L1 Free State L5 L3 L3 L2 L1 Gauteng L5 L3 L4 L3 L1 KwaZulu-Natal L5 L4 L4 L3 L1 Limpopo L5 L3 L3 L2 L1 Mpumalanga L5 L3 L3 L2 L1 Northern Cape L5 L3 L2 L2 L1 North-West L5 L3 L2 L2 L1 Western Cape L5 L4 L5 L4 L2

  • ​Potential impact on GVA

​In R billion ​GVA change in % ​-40 to -49%

  • 30 to -39%
  • 20 to -29%
  • 10 to -19%

0 to -9%

​Change in GVA per province

​In % compared to Feb 20 ​Apr 20 Jun 20 Aug 20 Oct 20 Dec 20

​Assumptions

​Source: IHS SA 2019 data, FirstRand high frequency economic measurements and economic assumptions, Government published lock-down measures, Epidemiological scenarios

​ESTIMATE IF LOCKDOWN ENDURES ​-30%

slide-5
SLIDE 5

Al All four ur AS ASSA A sc scena narios s pa paint t a bl bleak out utcome with th the the majority ty of SA A po popul pulati tion n be being ng infected

1 10 100 1,000 10,000 100,000 1,000,000 10,000,000 100,000,000 3/5/20 4/5/20 5/5/20 6/5/20 7/5/20 8/5/20 9/5/20 10/5/20 11/5/20 12/5/20

Scenario 1 - Base case R0 =3, 1.8 (post lockdown); 75% asymptomatic

Susceptible Exposed Recovered Dead Cumulative Detected Cumulative Infections 1 10 100 1,000 10,000 100,000 1,000,000 10,000,000 100,000,000 3/5/20 4/5/20 5/5/20 6/5/20 7/5/20 8/5/20 9/5/20 10/5/20 11/5/20 12/5/20

Scenario 2 - R0 =2.6, 1.6 (post lockdown); 50% asymptomatic

Susceptible Exposed Recovered Dead Cumulative Detected Cumulative Infections

​Source: Actuarial Society of South Africa – COVID outcomes 27 April 2020

slide-6
SLIDE 6

Al All four ur AS ASSA A sc scena narios s pa paint t a bl bleak out utcome with th the the majority ty of SA A po popul pulati tion n be being ng infected

1 10 100 1,000 10,000 100,000 1,000,000 10,000,000 100,000,000 3/5/20 4/5/20 5/5/20 6/5/20 7/5/20 8/5/20 9/5/20 10/5/20 11/5/20 12/5/20

Scenario 3 - R0 =3.0, 1.5 (post lockdown); 75% asymptomatic

Susceptible Exposed Recovered Dead Cumulative Detected Cumulative Infections 1 10 100 1,000 10,000 100,000 1,000,000 10,000,000 100,000,000 3/5/20 4/5/20 5/5/20 6/5/20 7/5/20 8/5/20 9/5/20 10/5/20 11/5/20 12/5/20

Scenario 4 - R0 =3.0, 2.1 (post lockdown); 75% asymptomatic

Susceptible Exposed Recovered Dead Cumulative Detected Cumulative Infections

​Source: Actuarial Society of South Africa – COVID outcomes 27 April 2020

slide-7
SLIDE 7

50 100 150 200 250 300 350 400 450

New cases (ex incoming/international) Step-up scenario

CO COVID-19 o 19 outcomes a are a almost b binary y – if if suppres essed ed, healt ealth outcomes mes ar are e ben enig ign, if if le let run, the e bulk lk of f the e popula latio ion will ill be e in infec ected ed*

R0:

viral reproductive number > 1: broad sustained infection < 1: suppression

United States Australia South African R0

​* SEIR models predict infection growth until Herd Immunity achieved ​Source: FirstRand Covid-19 team parameter estimation model – 3 May 2020

Initial R0 of 2.7 pre lockdown in line with international experience Published case data indicated R0 below 1 after lock-down However, last 3 weeks consistently indicate higher R0 0f ~1.4

slide-8
SLIDE 8

Th The P Prop

  • pos
  • sal
  • Symptom checking and rapid tracing. Most countries who suppressed the COVID-19 spread used technology to enable this, with China, South

Korea, Taiwan and Singapore being particularly effective (barring some mishaps of course). The key to the use of technology is widespread adoption on secure, authenticated and trusted platforms, which is why leveraging bank infrastructure is extremely compelling in SA. Bank digital solutions can cover ~20m* bank customers using all digital channels (App, internet and USSD), i.e. more than 95% of formally employed

  • workers. Solutions will also allow individuals to register using only their cellphone number (i.e. they don’t need a bank account). Data cost is

always a concern, but the bank Apps generally provide free data (i.e. the bank pays the telco provider directly for the zero rated use of the App

  • n any phone).
  • OHS regulations gazetted on 29 April by the Minister of Employment and Labour, requires an employer to “screen any worker, at the time they

report for work”. The regulations also stipulate a number of steps to be taken in terms of risk awareness, steps to take when an employee is diagnosed and administrative support for contact tracing, etc. All of these steps can most effectively be addressed by the use of digital information technology.

  • Data ownership and privacy: At this stage it is not envisaged that legislative changes are required to run the symptom checker as data is
  • btained with opt-in consent of staff/customers and only de-identified data is shared with the DoH to improve symptom screening. For co-

location, de-identified tokens are shared between participants. The reporting of positive infection cases and exposed individuals will follow established DoH protocols and legislated requirements and such information would be owned by the DoH. It is envisaged that future developments will include sharing of detailed ‘hotspot’ data by the DoH/NICD for participants to further reduce risk for their employees &

  • customers. All collection, storage, transmission and dissemination of data will be managed according to strict legislative, data privacy and ethical

controls.

  • Governance. A governance structure is being put in place to provide oversight over the development and management of the symptom checker.

It is proposed that representation including the DoH, NICD, CSIR, Wits Health Consortium, BASA, B4SA and PPGI (Public Private Growth Initiative) will be required

  • Employer requirements. This aspect has not been developed, but is under consideration as clearly the employer needs to know if any staff have

been infected and in the case of secondary exposure, such staff would need to be isolated, put on medical leave etc.

​* Estimate from the major bank’s published digital channel users, including App, online banking and ussd.

slide-9
SLIDE 9

COVID19

Covid-19 Functionality

  • List Testing stations with option to get directions
  • Report my Status (If you declare you’ve tested

positive this will trigger the notification to people you’ve been in contact with and advise them to self-isolate and get tested if they start displaying symptoms, this is an anonymous notifications – people will not know the identity

  • f the person who tested positive)
  • Stats - View COVID19 stats for South Africa

(Infected, Deaths etc. per Province)

  • Ability to Share App – Share this App
  • Symptom Tracker – sending daily reminders to

staff to capture any symptoms they may be experiencing after being in contact with someone who's tested positive

  • Temperature Screening – allows user to capture

current temperature

*120*321#

slide-10
SLIDE 10

Covid-19 Tracer Applet on Launcher

Existing App users

Once a customer has opted-in, their device will start to detect other devices (with App installed) that it comes into close contact with using BLE technology. Both customers need to have opted-in to this service for their phones to be able to detect each other. Information screens to tell users what this is and benefit of it Screens to explain and ask user to enable permissions

slide-11
SLIDE 11

Available to non-FNB users

The only difference in this flow is the screen to capture a new users cellphone number and OTP verification

slide-12
SLIDE 12

COVID19 SA Infection Statistics

slide-13
SLIDE 13

Report your Status

As soon as you state that you tested positive we trigger a message to all anonymized ID’s saved on your phone for relevant risk period

slide-14
SLIDE 14

App Push Notification

The below is the notification a user will receive if they’ve been in close contact with someone that has confirmed to be COVID19 positive

slide-15
SLIDE 15

Temperature Screening

Functionality to capture users temperature and upload as well as retrieve history

slide-16
SLIDE 16

Check My Symptoms

slide-17
SLIDE 17

Check My Symptoms

slide-18
SLIDE 18

Testing Stations

slide-19
SLIDE 19

Mobi

slide-20
SLIDE 20

USSD – Check my symptoms

slide-21
SLIDE 21

USSD – Check my symptoms

slide-22
SLIDE 22

USSD – Symptom Results

slide-23
SLIDE 23

USSD – Log my temperature

slide-24
SLIDE 24

Now Singapore is working to share its code with the world. "We are working around the clock to finalise our protocol reference documents and reference implementation, to open source what we have built, so that others may deploy their

  • wn flavours of TraceTogether - each implementing the

BlueTrace protocol," says the BlueTrace Manifesto

Singapore to open-source national Coronavirus encounter-tracing app and the Bluetooth research behind it

slide-25
SLIDE 25

Current Architecture

Service UUID: <App Unique ID> Advertising ID: <Unique Anonymous ID> Service UUID: <App Unique ID> Advertising ID: <Unique Anonymous ID> Details shared with 2 devices in close proximity

FNB App Backend

DB with Advertising ID to Customer

  • link. This is used to notify people that

have been in close contact with a person that has been reported COVID-19 positive

Advertising ID DB Notifications Service

Notification Registration 1 3 2

Our Bluetooth technology works similarly to TraceTogether – FNB’s ‘Service UUID’s’ picks up the other FNB Apps. Once this occurs and all parameters are met the device copies the users ‘Advertising ID’ (Anonymized ID) and vice versa. Once a user enters a positive test status all ID’s saved on their device are triggered and Push Notification alerting them of close proximity to person who tested positive is sent out

slide-26
SLIDE 26

Future Architecture

Details shared with 2 devices in close proximity

FNB App Backend

DB with Advertising ID to Customer’s

  • link. This is used to notify people that

have been in close contact with a person that has been reported COVID-19 positive

Advertising ID DB Notifications Service

Notification Registration 1 3 2 Notify Third Parties Based on Organization’s ID https://bluetrace.io/

Third Party Organization’s App

Third Party Organization’s Backend

Independent Aggregator

Advertising ID DB Notifications Service

3

Advertising ID: <Unique Anonymous ID’s Exchanged>

FNB willing to share the technology for Bluetooth co- location and assist with design of central infrastructure as well as security/ privacy safeguards

slide-27
SLIDE 27

Data output example

1 person 2 weeks Pilot group 2 weeks

Individual Anonymized ID’s saved