COVID-19 Data Dispatch

Author: Betsy Ladyzhets

  • National numbers, Jan. 17

    National numbers, Jan. 17

    In the past week (January 10 through 16), the U.S. reported about 1.5 million new cases, according to the COVID Tracking Project. This amounts to:

    • An average of 220,000 new cases each day
    • 469 total new cases for every 100,000 Americans
    • 1 in 213 Americans getting diagnosed with COVID-19 in the past week
    Nationwide COVID-19 metrics published in the COVID Tracking Project’s daily update on January 16. Hospitalizations appear to be leveling off, but deaths are still over 3,000 eachday.

    Last week, America also saw:

    • 126,000 people now hospitalized with COVID-19 (38 for every 100,000 people)
    • 23,200 new COVID-19 deaths (7.1 for every 100,000 people)

    Two weeks out from New Years (and the subsequent reporting weirdness), cases seem to be stabilizing, somewhat. But “stabilizing,” at this point in the pandemic, still means ridiculous numbers. 220,000 new cases each day! That’s like the population of Baton Rouge, Louisiana getting diagnosed with COVID-19 every day.

    Meanwhile, COVID-19 deaths are at their highest point in the pandemic, averaging 3,300 deaths per day.

    One positive note in this week’s numbers is that hospitalizations appear to be leveling off. But, as the COVID Tracking Project’s weekly update points out, the hospitalization picture is far different across different parts of the country. And with the more contagious B.1.1.7 variant getting detected in more and more states, the worst point of the pandemic for the U.S. may be still to come.

    more national numbers

  • COVID source shout-out: Kentucky

    At a time when state dashboards have become increasingly crowded with new information—or expanded onto five different GIS pages—I am comforted by the consistency of Kentucky’s COVID-19 reporting.

    The state has posted daily COVID-19 reports since the spring, including all the most important metrics in one place. If you’re looking for total cases, ICU patients, county-level statistics, or demographic data, you can find it all in this one PDF. The report’s formatting has changed over the past few months, but its Cntrl+F ease has not.

    Also, Kentucky started reporting race and ethnicity figures in whole numbers instead of percents recently!  Thanks, Kentucky!

  • Featured sources, Jan. 10

    This week’s featured sources are all about hospitalizations and treatments. See the full CDD source list here.

    • Hospital facilities visualization by the COVID Tracking Project: Last month, the Department of Health and Human Services (HHS) released an extensive dataset showing how COVID-19 patients are impacting hospitals at the individual facility level. (See my Dec. 13 post for more information on this dataset.) The COVID Tracking Project has produced an interactive visualization from this dataset, allowing users to zoom in to individual facilities or search for hospitals in a particular city or ZIP code. I contributed some copy to this page.
    • Therapeutics distribution (from HHS): The HHS is posting a list of locations that have received monoclonal antibody therapies, for the purpose of treating COVID-19. Bamlanivimab, one such therapy, received EUA from the FDA in early November. The HHS page notes that this is not a complete list: “Although monoclonal antibody therapeutic treatments have been shipped nationwide, shipment locations are displayed for those States that have opted to have their locations displayed on this public website.”
    • Hospital discharge summaries (from the Healthcare Cost and Utilization Project): This project, under the HHS umbrella, posts time series data on U.S. hospital patients. The site recently posted summaries on patients from April to June 2020, including datasets specific to COVID-19, flu, and other viral respiratory infections. As epidemiologist Jason Salemi explains in a summary Twitter thread, the data doesn’t provide new information but may be useful for a researcher looking to dig into spring and summer hospitalization trends.
    All featured sources

  • Was the Capitol invasion a superspreader event?

    Like everyone else, I spent Wednesday afternoon watching rioters attack the nation’s Capitol. I was horrified by the violence and the ease with which these extremists took over a seat of government, of course, but a couple of hours in, another question arose: did this coup spread COVID-19?

    The rioters came to Washington D.C. from across the country. They invaded an indoor space in massive numbers. They pushed legislators, political staff, and many others to hide in small offices for hours. They inspired heated conversations. And, of course, none of them wore masks. These are all perfect conditions for what scientists call a superspreading event—a single gathering that causes a lot of infections.

    (The number can vary, based on how you define a superspreading event; for more background, see this post from November.)

    My concerns were quickly echoed by many other COVID-19 scientists and journalists:

    The very next day, Apoorva Mandavilli published a story asking just this question in the New York Times. She quotes epidemiologists who point out that the event was ripe for superspreading among both rioters and Capitol Hill politicians. Many legislators were stuck together in small rooms, having arguments, while some of the Republican representatives refused to wear masks. POLITICO got a video of several Republicans refusing masks in a crowded safe room.

    By Friday, five Congressmembers had tested positive for COVID-19 in a week. It’s true, many of these legislators received vaccines in the first stage of the U.S. rollout in late December. But it takes several weeks for a vaccine to confer immunity, and we still don’t have strong evidence as to whether the Pfizer and Moderna vaccines prevent the coronavirus from spreading to other people. (They likely do, to some extent, but the evidence mainly shows that these vaccines prevent COVID-19 disease.)

    Just this morning, Punchbowl News’ Jake Sherman reported that the attending physician for Congress sent a note to all legislators and staff, warning them that “people in the safe room during the riots may have been exposed to the coronavirus.” I will be carefully watching for more reports of legislators testing positive in the coming weeks. From our nation’s previous experience with COVID-19 outbreaks at the White House, it seems unlikely that the federal government will systematically track these cases—though the incoming administration may change this. 

    As for the rioters themselves, while the events of January 6 may well have been superspreading, we likely will never know the true extent of this day’s impact. As I’ve written previously, we identify superspreading events through contact tracing, the practice of calling up patients to quiz them on their activities and help identify others who may have gotten sick. When case numbers go up—as they are now—it becomes harder to call up every new patient. One county in Michigan is so understaffed right now, it’s telling COVID-19-positive residents to contact trace themselves.

    But even if contact tracing were widely available in the communities to which those rioters are going home, can you really imagine them answering a phone call from a public health official? Much less admitting to an act of treason and risking arrest? No, these so-called patriots likely won’t even get tested in the first place.

    It would take rigorous scientific study to actually tie the Capitol riot to COVID-19 spread to the homes of the rioters. (That said, if you see a study like that in the months to come: please send it my way.)

    Finally, I have to acknowledge one more impact of the riot on D.C. at large: vaccine appointments were canceled after 4 PM that day. One of the most heinous aspects of that riot, to me, was how it pulled our collective attention away from the pandemic, precisely at a time when our collective health needs that attention most.

  • Vaccine confusion abounds—and this is the easy stage

    Vaccine confusion abounds—and this is the easy stage

    !function(){“use strict”;window.addEventListener(“message”,(function(a){if(void 0!==a.data[“datawrapper-height”])for(var e in a.data[“datawrapper-height”]){var t=document.getElementById(“datawrapper-chart-“+e)||document.querySelector(“iframe[src*=’”+e+”‘]”);t&&(t.style.height=a.data[“datawrapper-height”][e]+”px”)}}))}();

    It’s been one month since the Pfizer/BioNTech vaccine was authorized for use in the U.S. Since then, about 22 million Pfizer and Moderna doses have been distributed—and at least 6.7 million of those have actually made it into people’s arms, according to the CDC. (The CDC is not yet tracking second doses.)

    Despite the federal government’s intense push to get vaccines through safety trials, that “last mile” step—from the Pfizer and Moderna factories to people’s arms—has been under-planned and underfunded. In the past month, we’ve been shocked by news stories ranging from a Wisconsin medical employee “intentionally removing” doses from a refrigerator to a local journalist in Florida individually helping over 150 seniors register for vaccination appointments. 

    State public health departments, already overwhelmed from ten months of running every other aspect of pandemic response, needed more money and resources to simultaneously coordinate millions of vaccinations and communicate their importance. The needed money didn’t come until this month, and recommendations from the federal government have left a lot of room for interpretation—leaving state and local health agencies scrambling.

    And this first month was supposed to be the easy part! As The Atlantic’s Sarah Zhang explains, early U.S. vaccination efforts were aimed at easy-to-reach people: those in hospitals, nursing homes, and other long-term care facilities. In these locations, it’s easy to quickly identify the most vulnerable patients and get them registered for vaccination appointments. The next groups of eligible Americans will not be so easy to reach. Doctors’ offices, pharmacies, and many other businesses will need to figure out vaccine logistics while also ramping up campaigns to convince people to even get vaccinated in the first place.

    (For a plain-language explanation of this issue that you can send to confused friends and relatives, I highly recommend the latest episode of the Sawbones podcast with Dr. Sydnee McElroy and Justin McElroy.)

    When I updated my vaccine data annotations yesterday, I added notes on how the vaccine rollout is progressing in each state. For the 38 states (and D.C.) now reporting vaccinations, you’ll find two new fields: the state’s vaccination phase (1A, 1B, etc.; 31 states are reporting this) and any prominently featured information on how residents can get vaccinated, such as a registration portal or contact information for local public health departments (at least 12 states are doing this).

    Please note that, while most states do not yet have state-wide vaccine registration portals, many local public health departments are setting up such portals at the regional and county level. I highly recommend searching for your local public health agency to see what they have available. Also, New Mexico, which has a registration portal but no vaccine data dashboard, is not included in the annotations.

    State data availability (as of Jan. 9)

    • 39 jurisdictions are reporting some form of COVID-19 vaccination data on a dedicated page or dashboard
    • 16 states are reporting race and ethnicity of vaccinated residents
    • 20 states are reporting age of vaccinated residents
    • 17 states are reporting gender or sex of vaccinated residents
    • 20 states are reporting vaccinations by county or a similar local jurisdiction
    • 31 states are reporting their vaccination phase (1A, 1B, etc.)
    • 12 states are prominently featuring information on how residents can get vaccinated, such as a registration portal or contact information for local public health departments

    More vaccine data news

    • Jurisdictions with new vaccine dashboards or pages include: Arkansas, Arizona, California, Washington D.C., Kansas, Nebraska, and South Carolina.
    • The CDC’s vaccination data are now available for download, via a table beneath the interactive dashboard. The agency updated its state-by-state data every weekday this past week—an improvement from the past two holiday weeks. A time series isn’t yet available, though.
    • Bloomberg’s vaccine tracker now has time series for both individual states and several countries which have begun administering vaccines. The states currently leading the pack for vaccinations per capita in the U.S. are West Virginia, the Dakotas, and Maine.
    • KFF has updated its COVID-19 Vaccination Monitor with polling data on vaccine hesitancy in rural America. Compared to urban and suburban residents, the foundation found, rural residents are significantly more hesitant. 31% of the rural residents sampled said they would “definitely get” a vaccine, compared to over 40% in other categories. Rural residents are also more likely to say they’re “not worried” that they or someone in their family will get sick with COVID-19.
    • NPR’s Selena Simmons-Duffin and Pien Huang surveyed experts to determine several major ways the U.S. could “jump start its sluggish vaccine rollout.” These include: more money for state and local health departments, more vaccine types (hopefully some easier-to-transport brands), massive administration sites, more regular supplies from the federal government, and public awareness campaigns.
    • The Trump administration is speeding up at least one thing: a plan to help pharmacies administer COVID-19 vaccines. According to POLITICO’s Rachel Roubein, almost 40,000 pharmacies are involved in the federal program, including those part of the Costco, Rite Aid, and Walmart chains. Pharmacies which are already used to administering flu vaccines each year—and already have huge patient databases—are strong candidates for the next phase of vaccine rollout.  
    • After some classic infighting from Governor Andrew Cuomo and Mayor Bill de Blasio, New York state is moving to Phase 1B—meaning seniors and essential workers will start to see vaccinations. However, as City Councilmember Mark Levine pointed out on Twitter, the city has: “One website for H+H sites, another for DOHMH sites, another for Costco. For community clinics, 7 have their own different websites, 4 require calling, and 1 is by email.” (I want to get vaccinated at Costco, personally, if the opportunity arises.)
    • A Twitter thread from KFF Senior Vice President Jen Kates points out more of the methods states and counties are using to get residents signed up for vaccination appointments. They range from the expected online portals to SurveyMonkey and Eventbrite.

    Related posts

    • Sources and updates, November 12
      Sources and updates for the week of November 12 include new vaccination data, a rapid test receiving FDA approval, treatment guidelines, and more.
    • How is the CDC tracking the latest round of COVID-19 vaccines?
      Following the end of the federal public health emergency in May, the CDC has lost its authority to collect vaccination data from all state and local health agencies that keep immunization records. As a result, the CDC is no longer providing comprehensive vaccination numbers on its COVID-19 dashboards. But we still have some information about this year’s vaccination campaign, thanks to continued CDC efforts as well as reporting by other health agencies and research organizations.
    • Sources and updates, October 8
      Sources and updates for the week of October 8 include new papers about booster shot uptake, at-home tests, and Long COVID symptoms.
    • COVID source shout-out: Novavax’s booster is now available
      This week, the FDA authorized Novavax’s updated COVID-19 vaccine. Here’s why some people are excited to get Novavax’s vaccine this fall, as opposed to Pfizer’s or Moderna’s.
  • National numbers, Jan. 10

    National numbers, Jan. 10

    In the past week (January 3 through 9), the U.S. reported about 1.7 million new cases, according to the COVID Tracking Project. This amounts to:

    • An average of 240,000 new cases each day
    • 447 total new cases for every 100,000 Americans
    • 1 in 196 Americans getting diagnosed with COVID-19 in the past week
    Nationwide COVID-19 metrics published in the COVID Tracking Project’s daily update on January 9. More than 3,000 Americans are now dying of COVID-19 every day.

    Already, in 2021, America has reported 2.1 million new COVID-19 cases. That’s 31 times the number of cases South Korea has reported in the entire pandemic. (Remember, the two nations had their first cases on the same day back in January 2020.)

    Last week, America also saw:

    • 131,000 people now hospitalized with COVID-19 (40 for every 100,000 people)
    • 21,700 new COVID-19 deaths (7 for every 100,000 people)

    The nation is now recording an average of 3,000 deaths every day, more than the number of lives lost on September 11, 2001. Yet cases are still rising—the COVID Tracking Project reported a record 310,000 on January 8—and hospitals continue to fill with patients.

    Some of the cases and deaths added to national counts this week were likely reported late, making up for holiday dips over the winter holidays. (See previous issues for more on this phenomenon.) But many weren’t. 

    “Things will get worse as we get into January,” Dr. Fauci said in an interview with NPR this week.

    more national numbers

  • School data update, Jan. 3

    Many school districts across the nation will once again open for in-person instruction later this month. But data on how COVID-19 spreads in schools remain inadequate.

    At the request of one of my readers, I’ve updated my annotations of state K-12 data reporting, first published on December 6. The annotations are posted on a new resource page, which also includes notes on the four major national sources for COVID-19 school data.  I’ll be updating this page every two weeks.

    Here’s how the state data stand, as of January 1:

    • 34 states and the District of Columbia are reporting COVID-19 cases in K-12 schools, in some form
    • 7 states are reporting incomplete data on school outbreaks or cases in school-aged children
    • 20 states are separating out school case counts by students and staff
    • 5 states are reporting deaths linked to school outbreaks
    • 1 state is reporting COVID-19 tests conducted for school students and staff (New York)
    • 2 states are reporting in-person enrollment (New York and Texas)
    K-12 data annotations

    Related posts

    • COVID-19 school data remain sporadic
      On November 18, New York City mayor Bill de Blasio announced that the city’s schools would close until further notice. The NYC schools discrepancy is indicative of an American education system that is still not collecting adequate data on how COVID-19 is impacting classrooms—much less using these data in a consistent manner.

  • Vaccination data update, Jan. 3

    Despite the holidays, several more states began reporting vaccination data in the past week. The Centers for Disease Control and Prevention (CDC) also made a huge update: this national dashboard is now posting vaccination counts at the state level.

    Here are the notable updates:

    • I launched a vaccination data page on the CDD site which includes annotations on ten major national sources and every state’s vaccination reporting. I’ll be updating it weekly—the most recent update was yesterday.
    • Five states have started regularly reporting vaccination data since December 27: Alabama, Alaska, Indiana, Mississippi, and Wisconsin. 32 total states are now reporting these data; 15 states are reporting race and ethnicity of vaccinated residents. See more details on the resource page.
    • On December 30, the CDC started reporting state-level vaccination data on its national COVID-19 dashboard. For every state, the CDC is reporting total vaccine doses distributed and total people who have received their first dose. The dashboard also includes national counts—both for the U.S. as a whole and for long-term care facilities. Data are not yet available for download. According to the most recent update (yesterday, January 2), 4.2 million Americans have received their first dose.
    • Drew Armstrong, the Bloomberg reporter who runs the publication’s vaccination dashboard, posted a vaccine data user guide on Twitter. While the Tweet thread primarily describes the methodology and design choices behind Bloomberg’s dashboard, it also provides useful context for vaccination data overall. Two notable details: all vaccination data lag (the CDC’s data lag by about 50 hours, according to Armstrong), and Bloomberg is working on making the underlying data behind their dashboard public.
    • Benjy Renton halted updates for the “Doses Administered” tracker on his Vaccine Allocation Dashboard. As the CDC is now providing standardized state counts—and Renton is a one-person tracking operation—he’s switching to focus on analyzing vaccination trends and accessibility.  
    • Distribution delays: Operation Warp Speed promised that, if the Pfizer and Moderna vaccines received Emergency Use Authorization from the FDA, 20 million Americans would get vaccine doses by the end of 2020. That clearly didn’t happen. What went wrong? To answer that question, I recommend two articles: this STAT News story and this CNN story. Both articles suggest that a lag in data reporting may be one reason why the current vaccination counts look so low. Still, there’s a big difference between 4.2 million and 20 million.
    • Vaccination and the new COVID-19 strain: As the B.1.1.7 coronavirus variant, identified in the U.K., becomes an increasingly ominous threat to America’s COVID-19 containment, vaccination becomes increasingly urgent. Zeynep Tufekci’s latest piece in The Atlantic explains the issue. One piece that stuck out to me: the U.S. doesn’t have good genomic surveillance—or, a system to systematically sequence the virus genomes for people infected with SARS-CoV-2—which makes us less equipped to see where the new strain is actually spreading. As Tufekci puts it: “we are flying without a map.”
    • One dose or two? Scientists and public health leaders have been debating changing our vaccination protocol. Should the U.S. stick to the script, so to speak, and reserve enough vaccine doses that everyone who receives one dose can receive a second in the prescribed time window? Or should we give as many people first doses as we can, accepting that some may not get a second dose for months—or at all? (The U.K. opted for the latter earlier this week.) University of Washington professor Carl Bergstrom has compiled some Twitter threads that explain the debate. Dr. Fauci said on Friday that the U.S. will stick to the official two-dose regimen, but the scientific discourse will likely continue.

    Related posts

    • Sources and updates, November 12
      Sources and updates for the week of November 12 include new vaccination data, a rapid test receiving FDA approval, treatment guidelines, and more.
    • How is the CDC tracking the latest round of COVID-19 vaccines?
      Following the end of the federal public health emergency in May, the CDC has lost its authority to collect vaccination data from all state and local health agencies that keep immunization records. As a result, the CDC is no longer providing comprehensive vaccination numbers on its COVID-19 dashboards. But we still have some information about this year’s vaccination campaign, thanks to continued CDC efforts as well as reporting by other health agencies and research organizations.
    • Sources and updates, October 8
      Sources and updates for the week of October 8 include new papers about booster shot uptake, at-home tests, and Long COVID symptoms.
    • COVID source shout-out: Novavax’s booster is now available
      This week, the FDA authorized Novavax’s updated COVID-19 vaccine. Here’s why some people are excited to get Novavax’s vaccine this fall, as opposed to Pfizer’s or Moderna’s.
  • Your guide to choosing a COVID-19 data source

    Your guide to choosing a COVID-19 data source

    In preparing for this re-launch, I asked a few of my readers what they liked about the COVID-19 Data Dispatch and how it could better serve them. One common answer was that the publication has helped readers navigate the landscape of COVID-19 data sources, and pick the best source for a given story.

    The first two resources pages I’ve produced take this service to the next level.

    First: The Featured Source List is an upgraded version of the Google spreadsheet I’ve been using to keep track of data sources featured in the newsletter since July. You can use the table to search, sort, and filter all 82 featured sources by their names and categories. The little green plus icons toggle expanded views, with more details on every source. Much friendlier than a spreadsheet!  (Though, if you want to see the raw spreadsheet, it’s still accessible here.)

    Second: The Data Source Finder tool tells you exactly where to find the data you need for a given story.  (Or for a Facebook post, or an argument with your friend, and so forth.)  The tool includes detailed annotations on 16 data sources which I consider the primary COVID-19 sources in the U.S.

    Here’s how to use it. You start out by selecting the geographic scale on which you’d like to see data (global, U.S. states, counties, or cities), then choose the type of metric you’re looking for. The tool will return your options, including each dataset’s available metrics, methodologies, update schedule, download links, and more.

    It’s essentially an interactive flowchart, aimed to make it easy to compare and contrast sources for reporters on deadline and students engaged in Twitter debates alike. You can also find the full set of annotations linked on the page.

    While I compiled the annotations, the interactive tool was coded in Twine by my girlfriend, Laura Berry.  Your membership fees will help me buy Laura a nice dinner to thank her for her work.

    Support the COVID Data Dispatch
  • Support the COVID-19 Data Dispatch

    Support the COVID-19 Data Dispatch

    For the past five months, I’ve produced this publication for free. It’s been an act of service to my fellow COVID-19 reporters, public health communicators, and readers who simply want to understand the pandemic a bit better.

    The newsletter will continue to be free, as will many of the COVID-19 data resources I publish. But in tandem with this new site, I’m launching a membership program. 

    This program will enable COVID-19 communicators to connect more directly with each other, as well as to provide feedback that will shape what I cover.  It’ll also help me cover my own costs, which have grown significantly as I moved platforms.

    I already talked about my technical reasons for moving from Substack to a full-fledged website. I have another big reason for setting up a site, though: I’m planning to keep the CDD going beyond this pandemic. Its name might change later in 2021 or 2022, but my mission will stay the same—building accessibility and accountability for public health data in the United States.

    This publication won’t end when COVID-19 does. But even that idea, COVID-19 “ending,” feels tenuous to me. Maybe you feel that way, too. Maybe you’ve been reading articles like Ed Yong’s “Where Year Two of the Pandemic Will Take Us” or Maryn McKenna’s “2021 Will Be a Lot Like 2020,” that unpack how far we still need to go before life returns to some semblance of normalcy. Maybe you realize that America’s recovery from the pandemic won’t be so simple as 70% of the population getting vaccinated. Maybe you feel haunted by the structural inequities that COVID-19 revealed in our healthcare system and beyond, and you know you could never write enough stories or donate to enough mutual aid funds to make up the gap.

    Covering COVID-19, I’ve realized, is not just about this virus.  It’s about making sure we’re ready for the next public health crisis.  And we do that not just by growing our scientific capability but by prioritizing the public in public health.  To change the systems in which we live, we need to understand them—and we need to bring our communities along with us.

    If you feel this way, too, join me!  Help me build a network that will be ready to cover this pandemic and the next one.

    Support the COVID Data Dispatch

    And now, the technical details.  Here are the benefits of membership:

    • Community: Join a Slack server where COVID-19 reporters and communicators share resources and advice.
    • Resources: Exclusive cleaned datasets, visualizations, and other tools to assist you in your work.
    • Shape the Dispatch: Your priorities and needs will shape what the CDD covers and which new resources are produced.
    • Accessibility: Keep the CDD free for all its readers! Support accountability for public health data!

    The recommended membership fee is $10/month.

    But I understand that the pandemic is a difficult time for financial commitments. As such, I’m also offering pay-what-you-will pricing, starting at $2/month. There’s no difference in benefits between the two price tiers.

    In the interest of transparency, I’ve published my major costs here. To break even, I would need 120 members to join at the recommended $10/month tier.

    I also want to call attention to the second line on that costs page: Intern’s research and writing time. That’s right—this is going from a one-person publication to a two-person publication!  My friend (and current Barnard junior) Sarah Braner has agreed to join me as an intern for their spring semester.  You’ll learn more about them next week.

    As I am extremely against unpaid internships, my top financial priority right now is paying Braner’s salary. That shakes out to 18 members joining at $10/month.

    If you’re not ready to commit to membership right now, you can still support the publication with a one-time donation on Ko-fi.