What’s This Curve We’re Flattening?

Epidemiology basics and COVID-19

Image for post
Esther Kim@k_thosandCarl T. Bergstrom@CT_Bergstrom / CC BYWikimedia Commons

In the age of COVID-19, everyone keeps talking about flattening or bending the curve. But what the heck kind of curve are they talking about?

Epidemiology looks at the patterns of diseases across populations. This is a fundamental part of public health. Epidemiological curves, also known as epidemic curves or epi curves, are used to visually display data about a condition.

The image above is a generic epi curve. In this particular example, the horizontal axis represents time, and the vertical axis represents the number of cases. An actual epi curve would give more specific information, such as dates, active or resolved cases, etc.

The dashed horizontal line represents healthcare capacity. The healthcare system is capable of handling a finite number of patients at a given point in time, and that limit is stable unless resources are added to or removed from the system to change the capacity.

The large peak represents uncontrolled transmission. The number of cases goes up so sharply because, without effective measures in place to control the spread of a highly contagious illness, each person who gets sick infects 10 other people (the 10 in this example is a totally arbitrary number), and each of them infects 10 other people, who in turn each infect 10 more, etc., etc, and a lot of people get sick very quickly.

Where the shit really hits the fan is when the number of cases far exceeds the flatline healthcare capacity. More people are getting sick and dying than hospitals can handle.

On the other side of this large peak is a sharp drop in the number of cases, because by that point most people will have gotten sick and either died or recovered and developed at least some degree of immunity.

By implementing control measures to “flatten the curve”, you get the second peak shown in the above graph. There’s a more gradual increase in cases, and the number of cases at any point in time doesn’t greatly exceed the healthcare capacity.

Image for post
RCraig09 / CC BY-SAWikimedia Commons

This fancy little GIF from Wikipedia shows how the curve can be flattened with mitigation measures to decrease transmission. It also shows increasing the healthcare capacity, but that’s not necessarily easy to do. In the case of COVID-19, the production of more ventilators allows for increased capacity to deliver critical care. However, healthcare personnel numbers can only increase so much, since you need to get trained and qualified bodies from somwhere.

Image for post
RCraig09 / CC BY-SAWikimedia Commons

As many areas are talking about “reopening,” that’s where this fancy GIF is waving a figurative caution flag. This graph doesn’t show a clear line for healthcare capacity, but the green represents capacity and the pink/red area exceeds capacity.

Just because the curve of the initial surge has been flattened doesn’t mean the number of illness cases can’t flare up again. If there aren’t enough mitigation strategies in place to keep the number of new cases down, more people will start to get sick, and each of them will get 10 more people sick, etc., etc.

Image for post
U.S. Centers for Disease Control and Prevention

The first three graphs weren’t COVID-19 specific, but this one from the CDC is, and it includes actual data. The horizontal axis shows time, with specific dates. The vertical axis shows the cumulative number of deaths in the United States (so, for example, the number of deaths May would include all deaths from March and April).

This is a national forecast; forecasts for individual states are also available from the CDC here. The CDC page notes that this information was last updated on May 6, so it doesn’t incorporate any changes in either case numbers or mitigation measures that have happened since then.

The data points connected by solid lines up until just after May 1 represent actual data. Beyond that is the forecasted deaths based on epidemiological modelling. This kind of modelling is very complex, and since no one knows the future, models need to be based on certain assumptions about different factors that will affect how many people the virus kills. The curve on the left breaks it down so the projections of each model are shown and labelled in different colours, while on the right the various models are combined. The models come from a number of sources; for example, the ones labelled CU-20 come from Columbia University.

Just like a weather forecast, the ability to make accurate predictions decreases the further into the future that you’re looking. That, combined with the different assumptions made in different models around things like social contacts and speed of restrictions lifting and businesses reopening, ends up giving a sort of fan shape to the projections.

The take-home, though, is that even in the best-case scenario projections, it looks like more people are going to die. In the worst-case scenario projections, a whole lot more people will die.

I wonder if the people who’ve been protesting because they want haircuts, or because they think restrictions are communist measures that take away their freedom, are keen to volunteer their parents, grandparents, and immunocompromised family members to be among the potentially 70K+ more people who are going to die if restrictions are lifted too much, too quickly. Priorities, right?

Originally published at https://mentalhealthathome.org on May 11, 2020.

Mental health blogger | Former MH nurse | Living with depression | Author of 3 books, latest is Managing the Depression Puzzle | mentalhealthathome.org

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store