Farr's law question from concerned citizen

[jmcmurry]

From a friend " Question to please address (maybe you already have, i haven't had time to thoroughly read your site yet)....

Farr's Law. A paramedic friend in SC (who seems to be too much in the "everything is gonna be fine" camp) mentioned it.

Seems to my layperson mind that there isn't enough data to assume that Covid will follow Farr's timeline......

.....but I'm asking you, how does / doesn't Farr's apply to Covid?

I'll look to your site/twitter for answer, if you have time to address it. "

[macoca]

From a quick Google search:

• Farr's law was formulated by Dr. William Farr in 1840 indicates that Epidemic outbreaks follow a bell-curve pattern.
• Almost all epidemics follow this pattern, including a sharp increase in cases at the beginning of the cycle, a peak, and then a decline in cases and a return to baseline.
• The pattern exists because human beings modify their behavior to avoid getting sick, and those who do get sick seek medical attention.
• China, which had its first cases in December, peaked in late January or early February, and now is in the decline of the spread phase because of modified behavior (quarantines, social distancing, medical care, etc).
• The US is about 45-50 days behind where China sits, still in the 'up' phase of the Bell Curve.
• Farr's law indicates that US Peak in cases should be in April/early May, and then a decline through June/July. Source: https://www.glennbeck.com/radio/coronavirus-update-march-9th

On the other hand, from A History of Epidemiologic Methods and Concepts by Alfredo Morabia we can see that Farr stated that

The probability of dying constantly decreases in acute disease; as the deaths take place at an earlier period than the recoveries

and

Prevalence = Incidence x Average duration of disease

From the same book I can extract:

By using Pearson's method of curve-fitting, Brownlee eventually obtained a curve that effectively described certain symmetrical epidemics, but he could not obtain any function which accounted satisfactorily for the marked asymmetry characteristic of many epidemics.

I haven't found any paper where Farr actually stated that "Epidemic outbreaks follow a bell-curve pattern". The only Farr laws that I found were related to the propagation of the Cholera, and are related to altitude and water drinking habits. It seems that this law attributed to Farr isn't actually his.

[steesdale]

Here is an in-depth description of some of the history on the concept - Brownlee resurrected the concept more recently.

John Brownlee and the Measurement of Infectiousness: An Historical Study in Epidemic Theory on JSTOR https://www.jstor.org/stable/2982487?newaccount=true&read-now=1&socuuid=bce7b82f-36b4-48ac-be57-a2ee73423dfb&socplat=twitter#page_scan_tab_contents

Also, there are many factors that affect the epi-curve as well.

• The rate of contacts in the host population
• The probability of infection being transmitted during contact
• The duration of infectiousness.

Specifically, a main driver for the peak and bell-shaped cure is that the # of susceptible individuals goes down as individuals either get better, die, or change their behavior otherwise to interrupt contact (isolation) or decrease the probability of infection (wash hands, wear mask) never be a contact, etc...

[steesdale]

re: Seems to my layperson mind that there isn't enough data to assume that Covid will follow Farr's timeline......

No one knows the duration or exact shape the epi-curve for COVID will take. It's also possible to have multiple waves or curves back to back to back.

The take-home message of the site is that we want to elongate the time of the curve as best possible and to decrease the magnitude or height of the peak as well. This is to ensure the health system isnt overcome with critically ill patients.