PDA Pandemic Series Update – 4 July 2020

In this pandemic update, we bring to your attention some dashboards and the data they present and some apps created using opensource code. Policy decisions are often made by Governments using data and models. The UK easing of lockdown is being based on the R rate being below 1 and a slowdown in community transmission (with notable regional variations).

The dashboard, “The World in Data” is a collaboration between the University of Oxford and “The Global Change Data Lab”.  The dashboard makes visualisations straightforward and none of the data is hidden behind a paywall.

An example of a chart that may help us to understand the impact of the easing of lockdown is “The Government Response Stringency Index”. This chart shows the number and strictness of measures, countries have been eased and then tightened to respond to emerging situations. Many other charts can be created.

“The Government Response Stringency Index” is presented in a different way by the FT. This dashboard has some interactive features but most of the data is there as an informatic rather than to help you create bespoke visuals or other such outputs.

The newspaper has made most COVID-19 related news and information freely available and their collection of data is regularly updated.

However, it is notable that the published data is not complete as it only reflects Pillar 1 testing and thus may not reflect current spikes in infection rates. Pillar 2 data is not routinely published by the Government. An example of a graph created with incomplete data shows how the spike in Leicester is not captured by using these limited datasets.

An interesting bespoke healthcare app for modelling purposes has been created using open source code. This model maps out healthcare capacity using various assumptions about capacity (including ventilators).

The app shows the extent of assumptions for any underlying model and thus it is imperative that any model should be fully transparent and be able to withstand robust interrogation.

This app is a simple model which explicitly states that it is not for policymaking but merely for educational purposes. However, it does highlight the possibility of small teams being able to utilise open code to create interesting functional apps.

Governments globally made initial policies based on data about infection rates or hospitalisations. As this hard data showed signs that the spread of COVID-19 was slowing, policy makers would increasingly have to rely on trying to understand public behaviours as lockdown eased to ensure that numbers testing positive for COVID-19 did not increase.

A tracker that mapped a whole array of behaviours globally has been created at the Institute of Global Health Innovation, a partnership between Imperial College and YouGov.

An example of one mapped behaviour, avoiding going out, shows that as the pandemic has progressed, and certain restricting measures eased, the public has felt much more confident in going out in general. The outcome of this confidence may have an impact on the transmission rate if appropriate mitigating measures (masks etc) are not observed. The example below shows how this behaviour can be visualised in this tracker.