This picture from this article quickly became very popular. Keep in mind that I'm not arguing with the obvious idea that some serious measures should be taken to mitigate the effects of the pandemic. Not even I'm arguing that exactly measures mentioned in the article should be taken.

I still don't get the plot however. Say, the data on the plot is about total number of infected - so after the maximum there're no new cases. Say, in N days we know for sure either infected person will die or will get cured. That will mean that in N days the plot will asymptotically come to zero. It's not obvious (to me) from the plot why in that case second scenario will take longer.

If after the maximum there are newly infected persons than in theory if tail is long enough the number of totally infected persons than in theory if tail is long enough the total number of people infected in second scenario can be bigger in total - it just that it won't be bigger at any particular moment in time. Since mortality rate we consider equal in both scenarios - more people will die.

So, it's obvious that I'm misinterpreting this plot being not able to choose one of two interpretations and not seeing a valid third alternative. Hence the question.

enter image description here

  • 9
    As a mathematician, I would say: Yes that picture is informative. Height of the graph shows number of currently active cases. The two graphs have roughly the same area, which means in the two scenarios the total number of people who do catch it is roughly the same. The difference is, the blue version is slower, so the maximum is lower. This means in the blue version the healthcare system can cope better. The cases that need medical intervention (as with all the other cases) are more spread out in time.
    – GEdgar
    Mar 12 '20 at 23:15
  • 1
    @shabunc It is the English word "peak" you were after - "pique" is french for irritation, resentment or arrousal but is used in English. Mar 13 '20 at 0:13
  • 7
    @shabunc ... in the situation I described, there will be the same number of serious case in both scenarios: cases that would be fatal without treatment. But since treatment will prevent death in many cases, what makes a difference is how many can receive treatment at the same time; thus there are fewer fatalities in the blue scenario.
    – GEdgar
    Mar 13 '20 at 0:21
  • 4
    I'm voting to close this question as off-topic because it isn't a question that needs applied scientific scepticism. There is nothing in doubt.
    – Oddthinking
    Mar 13 '20 at 1:20
  • 1
    @Oddthinking - Your thinking is a bit odd here. I've run across multiple people who initially did not understand the rationale of flattening the curve, even after seeing one or more of the many news articles about the concept. That there are many articles on this topic in highly reputable sources makes the question pass the notability threshold. That the claim has a scientific basis makes the question pass the scientific skepticism threshold. Voting to reopen. Mar 13 '20 at 13:51

This graph was initially used in the 2007 CDC report, "Interim Pre-Pandemic Planning Guidance: Community Strategy for Pandemic Influenza Mitigation in the United States". Ctrl+F for "Figure 1".

Here is their explanation for what the graph means:

  1. Delay outbreak peak
  2. Decompress peak burden on hospitals/infrastructure
  3. Diminish overall cases and health impacts

With further explanation:

The three major goals of mitigating a community-wide epidemic through NPIs are 1) delay the exponential increase in incident cases and shift the epidemic curve to the right in order to “buy time” for production and distribution of a well-matched pandemic strain vaccine, 2) decrease the epidemic peak, and 3) reduce the total number of incident cases and, thus, reduce morbidity and mortality in the community

Interestingly, there is an actual scientifically measured outbreak that closely resembles this graph, which is also being heavily linked online. It is a comparison of the Philadelphia and St. Louis response to the 1918 "Spanish" flu outbreak.

enter image description here

From the source article:

The first cases of disease among civilians in Philadelphia were reported on September 17, 1918, but authorities downplayed their significance and allowed large public gatherings, notably a city-wide parade on September 28, 1918 [sadly, a parade for returning WW1 veterans! --me], to continue. School closures, bans on public gatherings, and other social distancing interventions were not implemented until October 3, when disease spread had already begun to overwhelm local medical and public health resources. In contrast, the first cases of disease among civilians in St. Louis were reported on October 5, and authorities moved rapidly to introduce a broad series of measures designed to promote social distancing, implementing these on October 7. The difference in response times between the two cities (≈14 days, when measured from the first reported cases) represents approximately three to five doubling times for an influenza epidemic.

Hatchett et al. "Public health interventions and epidemic intensity during the 1918 influenza pandemic" PNAS May 1, 2007

Note that this second graph measures deaths versus the illnesses in the first graph.

  • I'm aware of Phil/St.Louis plot but the thing it that it's actually quite different in the sense that in this case it's more obvious that mortality in second case is lower. Thank you for this answer anyway!!!
    – shabunc
    Mar 13 '20 at 0:45
  • also can not help but notice that in original plot (in Economist and in the article you've provided) there's no "Healthcare system capacity" line which actually does not adds anything to the plot from information standpoint.
    – shabunc
    Mar 13 '20 at 0:49
  • 5
    @shabunc actually it does: it says that the capacity for medical treatment is not infinite. Suppose your local hospital has 10 ventilators, all already in use. What's the mortality for the 11th person to show up, who desperately needs to be put on a ventilator? If the hospital had an infinite number of ventilators it wouldn't be a problem. The line shows the threshold where folks who technically might be saved will start dying simply due to lack of resources. Mar 13 '20 at 3:28

Not the answer you're looking for? Browse other questions tagged .