• ajkay

The Curve is Already Flat

Updated: Apr 12

CDC data suggests COVID-19 was here in Nov 2019

by AJ Kay & Thomas P Seager, PhD

AJ Kay graphs the CDC data showing a mystery flu that infected tens of thousands of Americans per week in late 2019.
AJ Kay graphs the CDC data showing a mysterious flu-like illness that infected tens of thousands of Americans a week in late 2019.

More than half of U.S. states have instituted lockdown measures in response to the spread of the COVID-19 virus. These policies are justified as an effort to “flatten the curve,” a phrase coined by Dr. Howard Markel, a pediatrician, and professor of medical history at the University of Michigan.

The hypothetical COVID curve we are trying to flatten — the time rate at which the number of people needing intensive medical treatment will grow to exceed our capacity to care for them — is based on assumptions about when COVID was brought to the US, the speed of its transmission, and the ability of our healthcare system to accommodate the most severely affected patients. By isolating citizens in their homes, lockdown policies intend to slow the rate of infection, thereby “flattening” the curve, which will allow us to ration healthcare resources over time.

“Flatten the curve!” has become the rallying cry of politicians, public health officials, celebrities, and social media users who believe that, without extreme social distancing measures, the American healthcare system will invariably be overwhelmed resulting in several million unnecessary deaths. The theory goes that if we succeed in flattening the curve, millions of lives will be saved.

It’s important to remember that a flat curve is not one in which no one gets infected. A flat curve is one which, at its peak, does not create enough critically ill patients to overwhelm the health care system.


The model with the most profound impact on public health policy was produced at Imperial College on 16 March 2020. It compared the health outcomes of suppression (i.e., lockdown) with less restrictive policies and predicted that millions in the United States and United Kingdom (2.2 million and 500K respectively) would die from COVID-19 unless aggressive containment measures were instituted immediately.

Since publication, new data has challenged the Imperial College model. Points of contention have centered around hospitalization, fatality, and transmission rates. The lead author of the study has since revised the fatality expectations substantially downward. So there are some adjustments being made to the initial assumptions.

But there’s one data point being used in all of the prevailing models, including Imperial, that no one is talking about.

When did the COVID curve begin?

And it’s an important point. How can we interpret the data we currently have if we don’t know where on the time axis we actually are?

Starting the Curve “Seed date” is a term used in pandemic epidemiology to identify when a disease was first established in a specific location. Knowing that date, or at least an approximate range, allows scientists and data modelers to calculate how far the disease has spread, how fast it’s spreading, and use that information to design effective responses.