Covid by numbers: making sense of the pandemic with data
Joanna Hindley reviews Covid by Numbers, in which statisticians David Spiegelhalter and Anthony Masters distill how we might “data” our way out of the pandemic.
Book review
Year: 2021
Authors: David Spiegelhalter and Anthony Masters
Rating: Entertaining 4/5 | Informative 5/5 | Inspiring 4/5
Image credit: Penguin Books
The COVID-19 pandemic saw conversations about data thrust into the mainstream as never before. Reproduction numbers (‘R’ numbers) and the relative efficacies of different brands of vaccine were not only discussed in academic journals and at conferences, but in newspaper headlines and ministers’ speeches. During this time, individuals such as Professor David Spiegelhalter and Anthony Masters became familiar voices to many, clearly communicating mathematical concepts and scrutinising the claims made by politicians and in the media. Following from their successful column in The Observer, Covid by Numbers aims to distill some of the key metrics that informed the debate, and the pitfalls that arose when they were misused and misunderstood.
Each short chapter explores a different question relating to the pandemic, mostly contextualised in the United Kingdom. These cover an impressive range—from the nature of the virus and the utility of epidemiologic modelling, to the wider consequences of successive lockdowns. Attention is given to debates that dominated the headlines, such as (incorrect) comparisons between COVID-19 and the flu, as well as issues less frequently discussed, such as the outcomes for patients being discharged after hospitalisation with the virus.
All of this is written in an uncomplicated style with clear and relevant figures. Excessive detail on statistical concepts is avoided, however there is much to be learned here for both the expert and general reader. Crucially, the authors emphasise the imperfect processes inherent in collecting and analysing data. They explain, for example, why it is challenging to measure levels of hospital-acquired coronavirus and how this led to wildly different estimates being quoted in the media. In doing so, they make clear that while statistical science is a powerful tool for understanding the world around us, it has limitations and claims of certainty should always be approached with a healthy degree of scepticism.
Masters and Spiegelhalter also carefully ensure that the data are never separated from the human suffering that each figure represents. “Counts and measurements are stories writ large” they say in their introduction, and this contextualisation is a continuing thread throughout each chapter. Such sensitivity is important: the reader is likely all too aware of the tragedy brought about by this virus. Furthermore, it strengthens their case for greater consideration of data in handling the present and future pandemics. It is easy to view statistics as a subject for dry academic debate. However, emphasising the human journeys that lie behind such numbers makes it hard to dispute their relevance in all our lives.