Coronavirus : Learn to stop worrying
Want to learn more? Well, allow me to recommend three books that will help you make sense of numbers and statistics, and the "evidence-based" decisions those quantities are supposed to inform. I had nothing to do with writing these books, I recommend th
Briana Mezuk, Ph.D :
Practically daily, the news brings us some pretty unnerving headlines concerning the coronavirus (COVID-19) outbreak. Today my news feed informed me that “U.S. evacuates Diamond Princess cruise passengers; 40 Americans on-board test positive for coronavirus,” that “Taiwan confirms first coronavirus death,” and that “More than 3,000 Britons tested” for the virus. I sigh and scroll past these nerve-wracking, insomnia-inducing click-baits.
Why am I so dismissive of these headlines? Because none of them include any reference to one of the most important concepts in epidemiology: the population at risk. Forty Americans on a cruise ship tested positive? How many Americans (in total) were there on the ship? 41? I’m concerned. 2,000? That means that 98 percent of Americans did not test positive. One death due to the virus in Taiwan, a country with 23.8 million people? That place has an average of 472 deaths every day. There are 66.4 million people in the UK, and 3,000 of them (0.005 percent) were tested for the virus? Snoozefest. Give me a headline about the latest celebrity to get a face tattoo, that is likely to have more of an impact on my life than the numbers these stories are talking about.
Let me give you some information you can actually use to address the ways that we are all deceived by numbers-whether those numbers originate from news headlines, reading about the likelihood of experiencing a side-effect from some new medication, or deciding to finally get that uncomfortable scope that your doctor wants you to have to screen for colon cancer (PSA: Please do this. Colon cancer is one of the most treatable cancers if diagnosed at an early stage. Trust me, I know).
The point of this post is to emphasize that numbers and quantities do not automatically “inform” good decision-making; in fact, sometimes they obscure good decisions or introduce uncertainty into situations where it is unhelpful for taking action. As the examples about coronavirus illustrate, one of the most important concepts to understand is that you should never (a term I hesitate to use, but it is accurate here) interpret a number in isolation. Forty people sounds like a lot when we are talking about hosting a dinner party, but large cruise ships have a capacity of over 2,000 people. Understanding those numbers in terms of the size of the population at risk, the denominator, puts them in context. That context is key to informing the amount of weight you should give them. It is one of the most foundational concepts in epidemiology.
Want to learn more? Well, allow me to recommend three books that will help you make sense of numbers and statistics, and the “evidence-based” decisions those quantities are supposed to inform. I had nothing to do with writing these books, I recommend them simply because they are books that I’ve read (and re-read) and learned a lot from. They are not “textbooks,” they are written in an entertaining and accessible, yet extremely informative, manner. I recommend them to my public health students, and now I’ll recommend them to you.
Intuitive Biostatistics: A nonmathematical guide to statistical thinking by Harvey Motulsky
This book threads the needle of being both accessible to novice quantitative readers and able to pique and keep the interest of more experienced ones. I re-read this book every few years to remind myself of the ways that my first-pass interpretations of quantitative relationships are often “almost exactly right, but not quite…” This book helps me do my job better, both as a researcher and as a teacher.
Thinking, fast and slow by Daniel Kahneman
This book has been widely reviewed, and so I won’t reiterate those comments here. Simply put, this lovechild of psychology and economics is worth your time: It illustrates how experimental studies reveal our cognitive biases, and why our “intuition” is so often illogical and can lead us astray. If you are still on the fence about whether you can commit to it (it is a bit of a tome), then listen to Kahneman discuss the main ideas of this book on the podcast On Being first.
Finally, Factfulness by Hans Rosling
This is an entertaining, rigorous, but fast read about many of the same topics covered in the first two recommendations, but with a tongue-in-cheekiness that the Swedes excel at. Check out his TEDTalk first, and then order the book.
The best way to protect yourself from COVID-19 (or influenza, which is far more likely to affect you this winter) is to wash your hands regularly and thoroughly. When you’re done scrubbing up, take a break from scrolling through your news feed, visit your local library or bookstore, and learn about all the ways that we (mis)use math in real life. And then slowly shake your head in existential disappointment when the next denominator-free headline comes across your screen, you armchair epidemiologist.
Practically daily, the news brings us some pretty unnerving headlines concerning the coronavirus (COVID-19) outbreak. Today my news feed informed me that “U.S. evacuates Diamond Princess cruise passengers; 40 Americans on-board test positive for coronavirus,” that “Taiwan confirms first coronavirus death,” and that “More than 3,000 Britons tested” for the virus. I sigh and scroll past these nerve-wracking, insomnia-inducing click-baits.
Why am I so dismissive of these headlines? Because none of them include any reference to one of the most important concepts in epidemiology: the population at risk. Forty Americans on a cruise ship tested positive? How many Americans (in total) were there on the ship? 41? I’m concerned. 2,000? That means that 98 percent of Americans did not test positive. One death due to the virus in Taiwan, a country with 23.8 million people? That place has an average of 472 deaths every day. There are 66.4 million people in the UK, and 3,000 of them (0.005 percent) were tested for the virus? Snoozefest. Give me a headline about the latest celebrity to get a face tattoo, that is likely to have more of an impact on my life than the numbers these stories are talking about.
Let me give you some information you can actually use to address the ways that we are all deceived by numbers-whether those numbers originate from news headlines, reading about the likelihood of experiencing a side-effect from some new medication, or deciding to finally get that uncomfortable scope that your doctor wants you to have to screen for colon cancer (PSA: Please do this. Colon cancer is one of the most treatable cancers if diagnosed at an early stage. Trust me, I know).
The point of this post is to emphasize that numbers and quantities do not automatically “inform” good decision-making; in fact, sometimes they obscure good decisions or introduce uncertainty into situations where it is unhelpful for taking action. As the examples about coronavirus illustrate, one of the most important concepts to understand is that you should never (a term I hesitate to use, but it is accurate here) interpret a number in isolation. Forty people sounds like a lot when we are talking about hosting a dinner party, but large cruise ships have a capacity of over 2,000 people. Understanding those numbers in terms of the size of the population at risk, the denominator, puts them in context. That context is key to informing the amount of weight you should give them. It is one of the most foundational concepts in epidemiology.
Want to learn more? Well, allow me to recommend three books that will help you make sense of numbers and statistics, and the “evidence-based” decisions those quantities are supposed to inform. I had nothing to do with writing these books, I recommend them simply because they are books that I’ve read (and re-read) and learned a lot from. They are not “textbooks,” they are written in an entertaining and accessible, yet extremely informative, manner. I recommend them to my public health students, and now I’ll recommend them to you.
Intuitive Biostatistics: A nonmathematical guide to statistical thinking by Harvey Motulsky
This book threads the needle of being both accessible to novice quantitative readers and able to pique and keep the interest of more experienced ones. I re-read this book every few years to remind myself of the ways that my first-pass interpretations of quantitative relationships are often “almost exactly right, but not quite…” This book helps me do my job better, both as a researcher and as a teacher.
Thinking, fast and slow by Daniel Kahneman
This book has been widely reviewed, and so I won’t reiterate those comments here. Simply put, this lovechild of psychology and economics is worth your time: It illustrates how experimental studies reveal our cognitive biases, and why our “intuition” is so often illogical and can lead us astray. If you are still on the fence about whether you can commit to it (it is a bit of a tome), then listen to Kahneman discuss the main ideas of this book on the podcast On Being first.
Finally, Factfulness by Hans Rosling
This is an entertaining, rigorous, but fast read about many of the same topics covered in the first two recommendations, but with a tongue-in-cheekiness that the Swedes excel at. Check out his TEDTalk first, and then order the book.
The best way to protect yourself from COVID-19 (or influenza, which is far more likely to affect you this winter) is to wash your hands regularly and thoroughly. When you’re done scrubbing up, take a break from scrolling through your news feed, visit your local library or bookstore, and learn about all the ways that we (mis)use math in real life. And then slowly shake your head in existential disappointment when the next denominator-free headline comes across your screen, you armchair epidemiologist.
(Briana Mezuk, Ph.D., is Associate Professor of Epidemiology and Co-Director of the Center for Social Epidemiology and Population Health at the University of Michigan’s School of Public Health).