The Problem with the Word ‘Mild’

One of the public health communication problems surrounding COVID-19 has been the use of the word ‘mild’ to describe infections. This appears to have been interpreted as, at least by those who aren’t health professionals, a few days of fever and the ‘ughs’, followed by recovery. But when you look at the data for those who have tested positive for COVID-19, it’s clear this is a very serious infection:

Screen Shot 2020-03-24 at 9.12.51 AM

When you look at these numbers while simultaneously seeing how some people are reacting, it seems to me that lots of people don’t realize how bad the median (or near median) outcome is.

Now, I realize a lot of people are going to say that the denominator is too small. People aren’t getting tested when they have symptoms due to the lack of tests–though it’s worth noting that a fair number of people who are symptomatically COVID-19 turn out to have something else (e.g., influenza, mono, other viral or bacterial pneumonia). Also, there are probably some asymptomatic or very mildly symptomatic people who have been exposed, and will never be counted*.

At the same time, we need to realize that the hospital admission rates also are deflated: a lot of people who would wind up in the hospital, at least for a short stay, are being told to wait it out at home, until it gets bad. Things are not normal for the numerator either.

What scares me about COVID-19 isn’t the prospect of death–and I realize that I might be willfully oblivious and also that I’m lucky to not be in a high risk group–but hospitalization. I really don’t want to be in the hospital, or worse, the ICU (are there any healthy people who do?). It’s why I’m a bit of a nut about washing your damn hands, and why I get a flu shot every year. I don’t want to get pneumonia or need breathing assistance (again, who does?).

We need to start describing the outcomes more rigorously, so people take this more seriously.

*Which, in the absence of an antibody-based test, makes controlling COVID-19 harder.

This entry was posted in COVID-19. Bookmark the permalink.

2 Responses to The Problem with the Word ‘Mild’

  1. r. clayton says:

    I don’t know how to read this chart. If the range high-end is the number of* persons hospitalized, in ICU or dead among all persons hospitalized, in ICU or dead, then why isn’t the hospitalization range high-end for 0-19 age group 100%? None of them have died, none are in ICU, so they all must be hospitalized.

    * I’m guessing that should be percentage of according to the column header, otherwise how do you get fractional people?

  2. ElHongo says:

    There’s basically 2 flavors of “mild”. The actual mild cases – up to light flu grade – are about 30-40% depending on age and appear separate only in listings that distinguish a “moderate” category also.

    Most often one sees only a distinction between “mild” and “severe” tho – the notorious “80% mild cases”, usually -, but this defines “mild” as “anything that isn’t severe or critical” and lumps “moderate” with “mild”.

    In the latter scheme, “mild” cases are anything to severe flu grade and above, half of which requiring hospitalization (at least in theory; if such people are hospitalized depends on your local government; then again, most cases of moderate to bad flu aren’t either) and potentially oxygen masks and a ICU bed at the ready for the worst days, but never becoming life-threatening in and by themselves.

    “Severe” cases are latently life-threatenung, while in the “critical” cases there is prolonged acute risk of fatality.

    Or at least that’s the best I could make of the vast morass or third-hand data I’ve seen since January.

    Another nugget: much is made of mortality affecting mainly age 60+ and being excessive for this age group while apparently low for all others.
    But AFAICT this is almost entirely a triage artifact. One could tehcnically save many more of those aged 60+, but this would come at the cost of grossly higher fatalities at lower ages.
    It’s ugly and stark and plain simple when you think it through; hence most people are in denial about it, yet barely anyone who read Garrett (1994) is in any way surprised.

Comments are closed.