Africa COVID Dashboard

Overview

Data Clinic collaborated with the COVID-19 in Africa Data Science team to develop an open source interactive multilingual dashboard that facilitates access to their nuanced insights of the trajectory of COVID-19 in Africa and supports decision-makers across the continent.

The dashboard, written using a React-based frontend and a NestJS API, displays infection, recovery, and death statistics for each African country along with forecasts based on the model produced by the COVID-19 in Africa Data Science Initiative.

Repo structure

The inital repo is setup as a yarn + lerna based monorepo. It has two packages

• frontend : A create-react-app template using typescript
• api : A nestjs server for developing the API, this is also written in typescript

In addition to these javascript packages there are two other directories

• data: which should contain any rawor proccessed data we need for the project
• scripts: which should contain any scripts we are using to clean and transform data. For example any ETL should live in this folder.

These are just placeholders and we might swich these out for another type of server / frontend framework as the project progresses.

Running the application

First install the dependencies using

yarn install

From the root dir and then run

yarn start

To get the servers running. The frontend will be served at http://localhost:3000 and the API will be hosted at http://localhost:4000

Data

The majority of data that the team is using comes from the COVID-19 Data Repository by the Center for Systems Science and Engineering (CSSE) at Johns Hopkins University, https://github.com/CSSEGISandData/COVID-19.

Population data comes from The World Bank, World Development Indicators, https://datacatalog.worldbank.org/dataset/world-development-indicators.

In addition, we have some model output data that can be found in the data/raw directory.

For an example of the model outputs + some other data files, look at the summary here :

• Model output data summary

Some additional variables come from the Institute of Health Metrics and Evaluation (IHME), Population Reference Bureau

Model specifications

IV Regression Model

model <- ivreg(lncaseload_lastobs ~ lnrchange + lnexpo + lnsdi + lnurban + lnp70p +
                 lnhhsn + lnihr2018 + lnsqualty + lnasthma + lnhiv  | 
                 lnexpo + lnsdi + lnurban + lnp70p +
                 lnhhsn + lnihr2018 + lnsqualty + lnasthma + lnhiv + lntraffic,data= modeldata)

• lnrchange : Change in cases between the first two weeks of the virus ?
• lnexpo : Inital expansion factor of the virus ?
• lnsdi : socio demographic index
• lnurban : Urbanisation of the country
• lnp70p: Age profile ?
• lnhhsn : Average household size
• lnihr2018 : Adherance to international Health regulation
• lnsqualty :health care quality index.
• lnasthma : Prevalance of Asthma (from Institute of Health Metrics and Evaluation (IHME))
• lnhiv : Prevalance of HIV (from Institute of Health Metrics and Evaluation (IHME))
• lntraffic : Incoming air traffic

License

This repo and the dashboard are shared pursuant to the Apache 2.0 License