The Sift Here’s what we’re Sifting today

COVID-19 rebounds in some places

by Lynde Langdon
Posted 6/25/20, 04:58 pm

As many as 20 million Americans have contracted the coronavirus, according to U.S Centers for Disease Control and Prevention officials on Thursday. Though the official count of confirmed U.S. cases is about 2.4 million, “our best estimate right now is that for every case that was reported, there actually are 10 more infections,” CDC Director Dr. Robert Redfield said on a call with reporters.

What does that mean for public health? The estimates show the COVID-19 pandemic remains a serious threat to the United States. Hospitalizations and the percentage of positive tests are rising in many parts of the country, while 12 states showed recent increases in the number of daily deaths. States such as Washington, California, Nevada, North Carolina, and Virginia have mandated people wear face coverings in public. But the country’s preparedness to handle outbreaks of the virus has drastically improved since March, Redfield noted.

Dig deeper: Read Harvest Prude’s report in The Stew about Congress’ review of the White House coronavirus response so far.

Read more from The Sift Sign up for The Sift email
Lynde Langdon

Lynde is a WORLD Digital's managing editor. She is a graduate of World Journalism Institute, the Missouri School of Journalism, and the University of Missouri-St. Louis. Lynde resides with her family in Wichita, Kansas. Follow Lynde on Twitter @lmlangdon.

Read more from this writer


You must be a WORLD Member and logged in to the website to comment.
  • AlanE
    Posted: Fri, 06/26/2020 08:13 am

     “our best estimate right now is that for every case that was reported, there actually are 10 more infections,”

    In light of that, here are some of the most important questions to be asking:

    1. Are we seeing more cases because there are more cases, because we're better targeting the individuals to test, or some mixture thereof? If the latter, can we be more precise about how much belongs to each part?

    2. Are any changes from previous rates taking place in the rate of hospitalization of those identified as infected? 

    3. Are there any changes happening in the infection demographic?

    4. What role is people converging into air conditioned buildings producing in the spike of cases?

    If we don't know the answers to these questions, any changes in policy are pure guesswork. We need to know a whole lot more than how many tests, how many positives, and how many deaths. Unfortunately, though, we can't seem to move beyond those metrics.

  • RC
    Posted: Fri, 06/26/2020 02:59 pm

    Alan E  - While you are asking some interesting questions they are impossible to answer because they require past data that would have been gathered and was not, so you can not measure curret data to past datra that was not gathered.

    You are making a huge assumption that detailed questions have been asked and answered recorded and tracked, as this has been going on from the beginning. It has not.  Even when questions have been asked, answers recorded and tracked, it has been extremely uneven all across the country. Then everybody does not answer because many people think they should not respond due to medical privacy. Remember HIPA.     

    No one has or is being targeted for testing. Testing has been expanding over time because the test were not even available early on across the country.  So trying to measure anything around testing over time will give you flawed results. The hospitalization and infection rates are hard to measure because a lot of people come who are infected and not sick enough to be hospitalized yet or ever. They may come back later if they get sick enough, or they could go to another medical facility (and the first one never gets notified) which then screws up the whole reporting process.

    Gathering in an A/C enclosed building question is interesting, but how do you gather the data on that? Even if you ask an infected person you will need to ask a string of questions to qualify that answer and likely the person will not know, remember or have noticed.  When was it? Where was it? How long were you there? How many other people were in there when you were there? How were the other people protected? etc.

    The most critical knowledge about the unreported 10 to 1 infection rate, is that the 10 never really got sick or had to be hospitalized.  So the virus hardly makes sick, and is not deadly, to about 90% of the population. The problem is that we still do not know enough to really tell if you are in the higher risk 10% group or 90% no risk group. We know some of the risk factors but not all of them, yet.                     

  • AlanE
    Posted: Fri, 06/26/2020 07:19 pm


    Actually, the data is extractable. I do not believe that testing is targeted in the most usual sense of the word, but people who have been in contact with those who have tested positive are definitely urged to be tested. That is a type of targeting. We're doing better at that than we were. So, there's good reason to believe the groups we're testing now are less random in nature than the groups we were checking, say, two months ago. So far, nobody is factoring that into the analysis--at least not explicitly so.

    It is definitely possible to track hospitalization rates of those who have tested positive.

    Population demographics? Same there. My state publishes tables by age and by ethnicity. State departments of health are definitely tracking at least that much. Given that, it's worth knowing if those metrics are changing. I suspect there are other metrics they're tracking as well.

    If you're savvy with data, you can get a very good idea if air conditioning is a problem. You compare the demographics of people who spend higher amounts of time in air conditioning with those who spend lower amounts of time in air conditioning (job type is one very good way to do this). If the demographics of the positive group better match the air conditioning group, you have a useful piece of information. Moreover, if not a month ago certainly now, anybody who tests positive should be asked to evaluate how much time a day they spend in air conditioning with large groups of people. It's reasonably important we get a handle on this situation and the question is really not very invasive.