Ta'anit 19

Ta'anit 19a ~ On the Definition of a Pandemic

What, exactly, is a pandemic? Perhaps you think this is an easy question, because, after all, we have been living through a pandemic for almost two years. So surely we must all know what we mean by the term. So before we go on, write down your definition of a pandemic (and don’t Google it first). All done? Let’s see how you did.

The World Health Organization and Its Definitions

In its discussions of what constitutes a pandemic (in this case, of influenza,) the World Health Organization went through various iterations:

1999

There was no real definition in use. According to Peter Doshi, the closest the WHO came to was this:

At unpredictable intervals, however, novel influenza viruses emerge with a key surface antigen (the haemagglutinin) of a totally different sub-type from strains circulating the year before. This phenomenon is called “antigenic shift”. If such viruses have the potential to spread readily from person-to-person, then more widespread and severe epidemics may occur, usually to a similar extent in every country within a few months to a year, resulting in a pandemic.

2005

The WHO still didn’t really have a definition, but said that a pandemic will be said to have begun when a new influenza virus subtype is declared to have reached Phase 6. Phase 6 is defined as “Increased and sustained transmission in the general population.

2009

Ibn 2009 the WHO stated that “Phase 6, the pandemic phase, is characterized by community level outbreaks in at least one other country in a different [second] WHO region in addition to the criteria defined in Phase 5. Designation of this phase will indicate that a global pandemic is under way.”

These definitions have important consequences, as I pointed out in my book on the history of influenza. Most people think of a pandemic as a disease that spreads and kills thousands of people. That description is echoed in the WHO’s official definition of the word as an infectious disease that causes “enormous numbers of deaths and illness.” But in talking about the 2009 outbreak, the WHO used a more academic and narrow definition that focused only on how much diesease was out there (called its prevalence,) not severity. After this was pointed out by an astute CNN reporter, a WHO spokeswoman announced that the organization had erred in using the more apocalyptic definition. “It was a mistake, and we apologize for the confusion,” she said, noting that the word painted “a rather bleak picture and could be very scary.” Quite so.

Some other definitions of a pandemic

Here is a comparison of the WHO and the CDC definitions of a (n influenza) pandemic, taken from Doshi’s helpful (though now dated) 2011 paper The elusive definition of pandemic influenza. As you can see, they are different.

From Doshi, P. The elusive definition of pandemic influenza. Bull World Health Organ. 2011;89:532–538 | doi:10.2471/BLT.11.086173

More recently, the CDC has defined a pandemic as an event in which a disease spreads across several countries and affects a large number of people. Fine, but how many countries exactly? It is not defined. Writing on a patient information page in the influential Journal of the American Medical Association, Dana Grennan defined a pandemic as an epidemic that spreads globally - and an epidemic is an outbreak that spreads over a larger geographical area than an “outbreak.” As you will note, none of these definitions address how deadly the disease is. In fact, by this standard you can have a pandemic of a completely mild and clinically unimportant disease, though no-one would really pay it any attention.

It is ironic that part of the recent problem with pandemic terminology arose not because of inherent vagueness but because of well-meaning attempts to eliminate ambiguities.
— D. Morens, G. Folkers and A. Fauci The Journal of Infectious Diseases 2009 Vol. 200 Issue 7 Pages 1018-1021.

The definition of a Pandemic in the Talmud

All of this is by way of introducing today’s page of Talmud, which takes a stab at defining a pandemic:

תענית יט, א

אֵיזֶהוּ דֶּבֶר? עִיר הַמּוֹצִיאָה חֲמֵשׁ מֵאוֹת רַגְלִי, וְיָצְאוּ מִמֶּנָּה שְׁלֹשָׁה מֵתִים בִּשְׁלֹשָׁה יָמִים זֶה אַחַר זֶה — הֲרֵי זֶה דֶּבֶר, פָּחוֹת מִכָּאן — אֵין זֶה דֶּבֶר

What is considered a plague? If a city that sends out five hundred infantrymen, [i.e., it has a population of five hundred able-bodied men, and three dead are taken out of it on three consecutive days,] this is a plague of pestilence, which requires fasting and crying out. [If the death rate is lower than that, this is not pestilence.]

In a couple of pages the Talmud will further define the definition we have in the Mishnah:

תענית כא,א

. תָּנוּ רַבָּנַן: עִיר הַמּוֹצִיאָה חֲמֵשׁ מֵאוֹת וְאֶלֶף רַגְלִי, כְּגוֹן כְּפַר עַכּוֹ, וְיָצְאוּ הֵימֶנָּה תִּשְׁעָה מֵתִים בִּשְׁלֹשָׁה יָמִים זֶה אַחַר זֶה — הֲרֵי זֶה דֶּבֶר

The Sages taught: If a city that sends out fifteen hundred infantrymen, i.e., one that has a population of at least fifteen hundred men, e.g., the village of Akko, and nine dead are removed from it on three consecutive days, i.e., three dead per day, this is considered a plague of pestilence.

But pandemics don’t follow linear rules, or recognize neat periods of 24 hours, and the Talmud delves a little further:

בְּיוֹם אֶחָד אוֹ בְּאַרְבָּעָה יָמִים — אֵין זֶה דֶּבֶר. וְעִיר הַמּוֹצִיאָה חֲמֵשׁ מֵאוֹת רַגְלִי, כְּגוֹן כְּפַר עֲמִיקוּ, וְיָצְאוּ הֵימֶנָּה שְׁלֹשָׁה מֵתִים בִּשְׁלֹשָׁה יָמִים זֶה אַחַר זֶה — הֲרֵי זֶה דֶּבֶר

If all nine died on a single day, while none died on the other days, or if the nine died over a period of four days, this is not a plague of pestilence. And a city that sends out five hundred infantrymen, for example, the village of Amiko, and three dead are removed from it on three consecutive days, this is a plague of pestilence.

Now this might sound a little confusing. First we are told that three people need to die each day for three consecutive days for a pandemic to be declared. But then we learn that if all nine die on one day, or over four days rather than three, then there is no pandemic declaration. And then comes this:

בְּיוֹם אֶחָד אוֹ בְּאַרְבָּעָה יָמִים — אֵין זֶה דֶּבֶר

If all three died on one day or over four days, this is not a plague of pestilence.

Rashi explains that if all three die on a single day, or over four days - אין זה דבר, דאקראי בעלמא הוא - this does not meet the definition of a pandemic, but rather it is a chance occurrence. The point here is that for a pandemic to be declared, we need a pattern of disease over a unit of time. The Talmud is trying to find the best pattern over the best unit of time that would make the declaration of a pandemic meaningful. We might argue that a better definition might be found, but as we have seen, even today the definition of a pandemic is elusive and changes frequently.

BONUS CONTENT: Even More on how we measure Pandemic Deaths

In 1854 in London there was a deadly outbreak of cholera. The British physician John Snow determined that it was caused by a contaminated water supply, and, so the famous story goes, when he removed the handle to the pump that supplied the dangerous water, the epidemic ended.

But what exactly was the effect of removing that handle? Well, it depends on how we measure things. Here for example is one way of visualizing that effect:

This and the following images are from From Edwin Tufte, Visual Explanations, pp. 27-37

It’s pretty impressive, right? But now let’s visualize the same data in another way:

Now the effect of removing the handle seems even more impressive. The deaths dropped from about 500 per week to about 100. But all we have done here is to slightly change the way in which the dates are grouped together. In the first chart the x-axis had August 20-26, then August 27-September 2 and so on. In the second chart the x-axis was August 18-24, August 25-31, and so on. Same deaths, different way to display the data. Now let’s take one last look at the same data, but this time the x-axis does not display periods of seven days. Instead, it displays day-by-day.

When the data is displayed in this way, the effect of the removal of the handle from the Broad Street pump seems to disappear entirely, because the cholera epidemic was already waining. In fact, the result of Snow’s intervention will depend on the arbitrary choice of time periods and the way we display the data. And we can also generalize to all pandemics. Any method to count pandemic deaths will be arbitrary, but that does not make it useless. We just have to be clear about why we decided to count the way that we did, and explain that decision. That is true whether you are the World Health Organization, The Centers for Disease Control, of the rabbis of the Talmud.

Talmudology field trip to the Broad Street Pump in London, July 2007.

Print Friendly and PDF