Science, Policy and Pandemics: Applying Statistical Methods

17 April 2020


Reported by Kate McNeil, CSaP Communications Coordinator

As part of CSaP's podcast series Science, Policy and Pandemics, we asked: what can statistical methods tell us about burden of disease?

Listen to the discussion here:

Produced in partnership with Cambridge Infectious Diseases and the Cambridge Immunology Network, CSaP's Science and Policy Podcast's series on science, policy and pandemics aims to answer questions about our understanding of the current pandemic, including the epidemiology, on what basis governments are making current decisions, how much confidence we can have in the knowledge models are producing, and how to manage the uncertainties involved in the present crisis.

In our fourth episode, Dr Rob Doubleday spoke with Dr Daniela De Angelis, Professor of Statistical Science for Health at the University of Cambridge to discuss applying statistical methods to epidemiology, disease transmission, and how we're using models to understand the burden on the NHS posed by covid-19.

Dr De Angelis' work focuses on synthesizing routinely collected data with the goal of developing models and building our understandings of the dynamics of infectious diseases. She explained that epidemiological models are mathematical expressions which are built, constructed and put together to approximate a phenomenon which typically not observed - such as community transmission - with the goal of estimating the burden of an infectious disease. Here, inputting data and parameters based on what we know can help to generate ranges of possible scenarios regarding the trajectory of the pandemic.

These models are adapted regularly, as new information becomes available, and are used to understand how many people are being infected, how the lockdown has changed the course of the virus' spread, and to predict burden on the NHS. At best, she cautions, these models are able to forecast scenarios for roughly a 2-3 week period, as there is a high level of uncertainty regarding both the inputted data, and regarding behavioural responses to recommended health measures - as epidemiological models rely on assumptions of how people interact and transmit the virus.

CSaP's Science and Policy Podcast's special series on Science, Policy and Pandemics is available across all major podcasting platforms, including Spotify, Apple Podcasts, Google Podcasts, Google Play, RadioPublic, Pocket Casts, Podbean, ListenNotes, Acast, Player.FM, Podcast Addict, and Castbox.


Cover Photo by CDC on Unsplash

Reported by Kate McNeil, CSaP Communications Coordinator