COVID-19 Data Dispatch

Category: Testing

  • Seroprevalence, incomplete data in the wake of the Omicron wave

    Seroprevalence, incomplete data in the wake of the Omicron wave

    Almost 60% of Americans had antibodies from a prior COVID-19 case in February 2022, a CDC study found. This rate was even higher among young children and teenagers.

    More than half of Americans have some antibodies from a recent coronavirus infection, according to a new CDC report. The study was published Tuesday in the CDC’s Morbidity and Mortality Weekly Report (MMWR), accompanied by a press conference and other fanfare. To me, this report (and its publicity) reflects the CDC’s current lack of urgency around addressing the pandemic and its continued impacts.

    The CDC regularly surveys COVID-19 antibody levels among the U.S. population, a metric that scientists call seroprevalence. For these surveys, the agency works with commercial laboratories to measure antibody rates from a nationally representative sample of Americans, with updates provided about once a month. The survey specifically looks at a type of antibody that develops in response to infection, not vaccination.

    This most recent iteration of the survey, providing data from February 2022, is particularly notable: the CDC estimates that 58% of Americans had this immune system indicator of a recent COVID-19 infection, immediately after the nation’s massive Omicron wave. Not all of these people got COVID-19 during the Omicron wave, though, since some of these antibodies stem from earlier infections.

    Other notable findings include:

    • National seroprevalence increased from 34% in December 2021 to 58% in February 2022—the largest jump recorded in this survey—reflecting the Omicron wave’s impact.
    • Children and teenagers had the highest antibody levels. For the 12 to 17 age group, seroprevalence went from 46% in December 2021 to 74% in February 2022.

    There are some major caveats to this study, though, including:

    • The imprecise nature of this antibody measurement. The type of antibody measured in this seroprevalence survey “stays positive for at least two years after infection,” CDC scientist Dr. Kristie Clarke said on the agency’s press call.
    • Antibodies wane at different rates and levels for different people, so it’s unclear to what extent this 58% finding actually reflects the share of Americans who have gotten COVID-19 since spring 2020.
    • Plus, some people infected by the coronavirus never seroconvert, meaning that they don’t develop antibodies at all (and thus wouldn’t show up in this study).
    • While we know that the COVID-19 antibodies identified in this study confer some protection against new infections, it’s unclear how long that protection lasts or how it might hold up against new variants.

    To me, this study (and the CDC’s choice to promote it with one of the agency’s infrequent press calls) exemplifies the Biden administration’s COVID-19 response right now.

    As I listened to the press call, the CDC’s interpretation of this study was clear: more than half of Americans have some protection against COVID-19 from a prior infection, and many of those people also have protection against vaccination. Much of that protection applies specifically to Omicron and will likely help us avoid a crisis from BA.2, so it gives the U.S. additional reason to relax safety measures, the CDC suggested.

    (Worth noting: the CDC still recommends vaccination and booster shots for anyone who had a previous coronavirus infection, including children. But that message is not getting across right now, as evidenced by our low booster shot uptake.)

    When you ask for more specifics on that “protection” from prior infections, though, the CDC isn’t able to provide much information. Again, we don’t know how long the protection lasts or how it holds up against other variants. And we have no idea how many people had mild or asymptomatic COVID-19 cases, then did not seroconvert.

    The CDC’s press call also failed to mention Long COVID, which is a risk from any COVID-19 case—no matter how mild. Some Long COVID researchers have also suggested that lack of seroconversion, or even a prior infection in general, may increase a patient’s future risk for prolonged symptoms the next time they get infected.

    And, of course, the CDC report also exemplifies our current lack of surveillance. How many of those Omicron infections between December and February were actually caught by PCR testing and reported to the CDC? A small fraction. At the press call. Dr. Clarke mentioned an upcoming CDC study that estimates how many infections go uncounted for every one reported case:

    In the Omicron period, we found that over that time period, the infection to case ratio was the highest that it’s been, at over three estimated infections per reported case. And that varied by region, so depending on which US census region the estimates were, you know, the ratios were higher or lower.

    Surely that ratio is getting even higher now. To me, this forthcoming study, combined with the seroprevalence report, is a reminder that the cases we see in our datasets and dashboards are a very incomplete picture of actual coronavirus transmission in the U.S. And yet the CDC is using this incomplete picture to suggest we all relax, take our masks off, and forget about the pandemic.

    More federal data

  • The CDC’s isolation guidance is not based on data

    The CDC’s isolation guidance is not based on data

    A study published in the CDC’s own journal indicated that about half of people infected with Omicron are still contagious 5-10 days after their isolation period starts. Chart via CDC MMWR.

    Maybe it’s because I’m a twenty-something living in the Northeast, but: quite a few of my friends have gotten COVID-19 in the last couple of weeks. The number of messages and social media posts I’m seeing about positive rapid tests isn’t at the level it was during the Omicron surge, but it’s notable enough to inspire today’s review of the CDC’s isolation guidance.

    Remember how, in December, the CDC changed its recommendations for people who’d tested positive for COVID-19 to isolating for only five days instead of ten? And a bunch of experts were like, “Wait a second, I’m not sure if that’s sound science?” Well, studies since this guidance was changed have shown that, actually, a lot of people with COVID-19 are still contagious after five days. Yet the CDC has not revised its guidance at all.

    (Also, to make sure we’re clear on the terms: isolation means avoiding all other human beings because you know that you have a contagious disease and don’t want to infect others. Quarantine means avoiding other humans because you might have the disease, due to close contact with someone who does or another reason for suspicion.)

    The current CDC guidance still says that, if you test positive: “Stay home for 5 days and isolate from others in your home.” Yet, in recent weeks, I’ve had a couple of friends ask me: “Hey, so it’s been five days, but… I’m not sure I’m ready to rejoin society. Should I take a rapid test or something?”

    Yes. The answer is yes. Let’s unpack this.

    Studies indicating contagiousness after five days

    As this NPR article on isolating with Omicron points out, the CDC guidance was “largely based on data from prior variants.” At the time of this five-day recommendation, in late December, scientists were still learning about how Omicron compared to Delta, Alpha, and so on, particularly examining the mechanisms for its faster spread and lower severity.

    But now, almost four months later, we know more about Omicron. This version of the coronavirus, research suggests, is more capable of multiplying in the upper respiratory tract than other variants. People infected with Omicron are able to spread the virus within a shorter time compared to past strains, and they are able to spread it for a higher number of days—even if their symptoms are mild.

    One study that demonstrates this pattern is a preprint describing Omicron infections among National Basketball Association (NBA) players, compared to cases earlier in 2021. Researchers at Harvard’s and Yale’s public health schools, along with other collaborators, compared 97 Omicron cases to 107 Delta cases. NBA players are a great study subject for this type of research, because their association mandates frequent testing (including multiple tests over the course of a player’s infection).

    The big finding: five days after their Omicron infections started, about half of the basketball players were still testing positive with a PCR test—and showing significant viral load, indicating contagiousness. 25% were still contagious on day six, and 13% were still contagious on day seven. These patients also saw less of a consistent pattern in the time it took to reach their peak contagiousness than the players infected with Delta.

    From the NPR article:

    “For some people with omicron, it happens very, very fast. They turn positive and then they hit their peak very quickly. For others, it takes many days” – up to eight or even 10 days after turning positive, says the study’s senior author, Dr. Yonatan Grad, an associate professor of immunology and infectious diseases at the Harvard T.H. Chan School of Public Health.

    While this NBA study is a preprint, other research has backed up its findings. One study from Japan, shared as a “preliminary report” in January, found that people infected with Omicron had the highest levels of viral RNA—indicating their highest levels of contagiousness—between three and six days after their symptoms started. The researchers saw a “marked decrease” in viral RNA only after ten days.

    Another preprint, from researchers at the University of Chicago (and antigen test proponent Michael Mina), examined Omicron infections among healthcare workers at the university medical center. Out of 309 rapid antigen tests performed on 260 healthcare workers, 134 (or about 43%) were positive results received five to ten days after these workers started experiencing symptoms.

    The highest test positivity rate for these workers, according to the study, was “among HCW returning for their first test on day 6 (58%).” In other words, more than half of the workers were still infectious six days after their infection began, even though the CDC guidance would’ve allowed them to return to work.

    Later in February, a study in the CDC’s Morbidity and Mortality Weekly Report (MMWR)—or, the CDC’s own journal—shared similar results. The report, authored by CDC researchers and practitioners at a healthcare system in rural Alaska, looked at antigen test results from hundreds of infections reported to this health system during the Omicron wave.

    The main finding: between five and nine days after patients were diagnosed with COVID-19, 54% (396 out of 729 patients) tested positive on rapid antigen tests. “Antigen tests might be a useful tool to guide recommendations for isolation after SARS-CoV-2 infection,” the authors wrote.

    Following this, an early March preprint from researchers at Massachusetts General Hospital, MIT, Harvard, and other collaborators analyzed infections among 56 people during the Delta and Omicron waves. This study used viral cultures to examine contagiousness directly, rather than simply looking at test results.

    Like past research, this study found that over half of patients (with both Omicron and Delta) were still contagious five days into their infections. About one-fourth were still contagious at day eight.

    Guidance for people testing positive

    All of the above studies suggest similar conclusions: about half of people infected with Omicron will still be contagious five days after their positive test results or the start of their symptoms, despite what the CDC’s guidance says. If you get infected with BA.2 in the coming surge, the best way to figure out whether you’re contagious after day five is by taking a rapid antigen test.

    In fact, for the highest accuracy (and peace of mind), I’d recommend taking two antigen tests, two days in a row. If both are negative, then you’re probably good to return to society—but maybe don’t travel to visit an elderly relative just yet.

    This two-rapid-test guidance comes from the U.K. Health Security Agency, which recommended in December that Brits could isolate for seven days instead of ten if they tested negative on days six and seven of their isolation. (The U.K.’s guidance has since become more lenient, but this is still a good rule for reference—more based in science than the CDC’s guidance.)

    What else should you do if you test positive? Here are a few recommendations that I’ve been giving friends and family:

    • Be prepared to isolate for a week or two, even if you may be able to leave isolation after a shorter period (with rapid tests).
    • After leaving isolation, wear a good mask (i.e. an N95 or KN95) in all public spaces.
    • Look into treatment options near you. The HHS has a database of publicly available COVID-19 therapeutics, while some localities (like New York City) have set up free delivery systems for these drugs.
    • There’s also the HHS Test to Treat program, which allows people to get tested for COVID-19 and receive treatment in one pharmacy visit. This program has faced a pretty uneven rollout so far, though.
    • Rest as much as possible, even if you have mild symptoms; patient advocates and researchers say that this reduces risk for developing Long COVID.

    More testing data

  • Why Utah’s innovative school COVID-19 testing program failed

    Why Utah’s innovative school COVID-19 testing program failed

    In fall 2021, testing events at Utah public schools failed to decrease coronavirus transmission.

    My latest story with the Documenting COVID-19 project is an investigation into Utah’s school COVID-19 testing program, in collaboration with the Salt Lake Tribune.

    As longtime readers know, I have done a lot of reporting on school COVID-19 testing programs. I find these efforts to routinely test K-12 students fascinating, in part because of the unique potential for collaboration between school districts, health departments, and other community institutions—and also because of the immense challenges that arise when schools are asked to become health providers in a way we never would’ve considered before the pandemic.

    Utah’s program caught my eye last year when I was reporting a story for Science News on the hurdles schools faced in setting up COVID-19 testing. This state was an early pioneer of Test to Stay, a strategy in which students must test negative to attend school after a potential exposure rather than going through a (potentially unnecessary) quarantine.

    In Utah’s version of Test to Stay, once 1% of students tested positive for the virus, the entire school would go through a testing event. Students who tested negative could keep attending school without interruption, while those who tested positive (or those who refused to participate) could quarantine. The Utah health department tested out this program in the 2020-2021 school year, and it was so successful that a CDC MMWR boasted it had “saved over 100,000 days of in-person instruction.”

    After that successful test, Utah’s state legislature codified the program into law for the 2021-2022 school year. But Test to Stay crashed and burned this past fall, even before the Omicron variant overwhelmed Utah’s test supplies.

    Here’s why the program failed, according to our investigation:

    • When putting Test to Stay into law, the Utah state legislature doubled the threshold for school cases that would trigger a testing event, from 1% to 2% of the student body. (Or from 15 to 30 students at smaller schools with under 1,500 students.) This higher threshold allowed COVID-19 to spread more widely before testing events took place, leading to higher case numbers when students were finally tested.
    • Utah’s lawmakers also banned schools from requiring masks in fall 2021, leading to more transmission. Experts said the original program was intended to be paired with masks and other safety measures; it was not able to stand on its own.
    • In the 2020-2021 school year, Test to Stay was paired with a second program called Test to Play: mandatory testing every two weeks for students on sports teams and in other extracurriculars. Without this regular testing in fall 2021, Utah schools had less capacity to identify school cases outside of voluntary and symptomatic tests—so it took longer for schools to reach the Test to Stay threshold.
    • The Utah health department allowed individual schools and districts to request rapid tests for additional surveillance testing. Some administrators requested thousands of tests and made them regularly available to students and staff; others were entirely uninterested and did not encourage testing at their schools.
    • Testing in schools has become increasingly polarized in recent months, like all other COVID-19 safety measures. One school administrator told me that he faced some vocal parents who felt “that their rights were being trampled on” by the testing program. Without high numbers of students opting in to get tested, testing programs are inherently less successful.

    Even though the CDC endorsed Test to Stay as part of its official school COVID-19 guidance last December—citing Utah’s program as a key example—its future in the state is now uncertain. State lawmakers paused the program during the Omicron surge in January and have yet to revive it. At the same time, lawmakers have made it even harder for Utah schools to make their own decisions around safety measures.

    What school districts and health departments should actually be doing, experts told me, is stock up on rapid tests now so that they’re ready to do mass testing in future surges. It’s unlikely that the Omicron wave will be our last, much as some Utah Republicans might want to pretend that’s the case.

    You can read my full story at MuckRock’s site here (in a slightly longer version) or at the Salt Lake Tribune here (in a slightly shorter version). And the documents underlying this investigation are available on the Documenting COVID-19 site here.

    More K-12 reporting

  • Five more things, February 27

    Five additional news items from this week:

    • The CDC is not publicly releasing a lot of its COVID-19 data. Last weekend, New York Times reporter Apoorva Mandavilli broke the news that the CDC has withheld a lot of its COVID-19 data from the public, including information on breakthrough cases, demographic data, and wastewater data. This news was honestly not surprising to me because it follows a pattern: the CDC doesn’t like to share information unless it can control the interpretations. But I appreciated the conversation brought on by this article, with public health experts saying they’d rather have imperfect data than a complete data void. (I agree!)
    • BA.2 is definitely more transmissible than the original Omicron strain, but it does not appear to be significantly more severe or more capable of evading vaccines. Two recent posts, one in the New York Times COVID-19 updates page and one from Your Local Epidemiologist, share some updates on what scientists have learned about BA.2 in the past couple of weeks. In the U.S. and other countries with BA.2, this sublineage doesn’t seem to be causing a major rise in cases—at least so far.
    • New CDC study shows the utility of rapid testing out of isolation. More than half of patients infected with the coronavirus tested positive on rapid antigen tests between five and nine days after their initial diagnosis or symptom onset, a new CDC report found. The report includes over 700 patients at a rural healthcare network in Alaska. These findings suggest that rapid testing out of isolation is a good way to avoid transmitting the virus to others, if one has the tests available.
    • January saw record-high coronavirus infections in hospitals. POLITICO reporters analyzed hospitalization data from the Department of Health and Human Services (HHS), finding that: “More than 3,000 hospitalized patients each week in January had caught Covid sometime during their stay, more than any point of the pandemic.” This high number demonstrates Omicron’s high capacity to infect other people.
    • Hong Kong’s surge shows the value of vaccinations. Hong Kong has been a global leader in keeping COVID-19 cases low throughout the pandemic, yet Omicron has tested this territory’s strategy—causing record cases and overwhelming hospitals. One major issue for Hong Kong has been low vaccination rates, particularly among the elderly, as people did not see the need to get vaccinated when cases in the territory were practically nonexistent.
  • States treating COVID-19 as “endemic” is leading to shifts in data collection and reporting

    States treating COVID-19 as “endemic” is leading to shifts in data collection and reporting

    Screenshot from the California SMARTER plan. This week, California became the first state to officially shift to an endemic strategy for dealing with COVID-19.

    Last week, I discussed the recent trend in states ending mask requirements in public schools, businesses, and other settings, by providing readers with some suggestions for encouraging safety during this push to “open everything” (that wasn’t already open). This week, more states are dropping safety measures; for example, Washington governor Jay Inslee announced that the state’s indoor mask mandate will end on March 21, though this change is also contingent on a low level of COVID-19 hospital admissions.

    At the same time, some states are also making major shifts in the ways they collect and report COVID-19 data. State public health departments are essentially moving to monitor COVID-19 more like the way they monitor the flu: as a disease that can pose a serious public health threat and deserves some attention, but does not entirely dictate how people live their lives.

    You may have seen this shift discussed as a movement to treat COVID-19 as “endemic.” An endemic disease, from an epidemiologist’s standpoint, is one that’s controlled at an acceptable level—it hasn’t been completely eradicated, but the levels of cases, hospitalizations, and deaths are generally deemed as levels that can continue without major public health measures. For more on the topic, I recommend this post from epidemiologist Ellie Murray (whom I’ve quoted on this topic before).

    We can argue—and many COVID-19 experts on Twitter are arguing—about whether this is the appropriate time to shift into endemic mode. Still, regardless of individual opinions, state public health departments are starting to make this shift, and I think it’s worthwhile to discuss how they’re doing it, particularly when it comes to data.

    Here’s a brief roundup of four states that are shifting their COVID-19 data collection and reporting.

    California

    California made headlines this week for being the first state to officially shift into “endemic” policy for dealing with COVID-19. State officials have drafted a plan called “SMARTER”—which stands for Shots, Masks, Awareness, Readiness, Testing, Education, and Rx (treatment). I took a look at the plan, which reporters from NBC Bay Area kindly shared publicly on DocumentCloud.

    Here are a few data-related highlights:

    • State officials will “focus on hospital numbers” to gauge how California should react to potential new variants that may be more infectious or more capable of causing severe disease.
    • Unlike some other states, California is maintaining testing capacity going forward, including an expansion of community testing sites and ongoing procurement of at-home antigen tests for public schools, long-term care facilities, and other institutions.
    • Throughout the pandemic, California has invested in genomic sequencing for COVID-19 cases, as well as a statewide modeling tool that compiles several different forecasts. These surveillance tools will be further expanded to respond to COVID-19 and other infectious disease outbreaks.
    • California also intends to “build a robust, regionally based wastewater surveillance and genome sequencing network” that can provide early warnings about new outbreaks.
    • The plan includes a focus on equity: California leaders will monitor testing, cases, and other metrics in minority communities so that resources can be provided to address disparities if needed.

    Missouri

    Missouri started its shift to “endemic” in December, as the governor declared an end to the state’s public health emergency around COVID-19—even though cases were at their highest-ever level in the state. Now, the Missouri health department is preparing to change its data reporting accordingly, my colleague Derek Kravitz and I reported in the Missouri Independent this week. (The Independent, a nonprofit newsroom focused on Missouri’s state government, is a long-time collaborator of the Documenting COVID-19 project, where I work part-time.)

    Here are the planned data changes highlighted in our story:

    Case investigations and contact tracing, where local health departments’ staffers reach out to people exposed to the virus in workplace or other public settings, will cease, unless a new, more transmissive or deadly variant emerges;

    Daily reports on COVID-19 cases and deaths by the state health department will be replaced by aggregate weekly reports. In some cases, metro health departments, including those in St. Louis and Springfield, will likely continue collecting and disseminating daily reports but the state will stop its reporting;

    Positivity rates will be phased out, as they are already difficult to interpret, with many Americans having switched from PCR tests to at-home antigen tests. Most people don’t report their results to local health departments. Missouri officials in January said they were prepared to be a “trend setter” in eliminating positivity rate reporting.

    Hospitalization data will become even more important, with state health officials hoping to make reporting more timely;

    Wastewater surveillance will become a more relied-on data point for public health officials, as a way to spot COVID-19 early in its life cycle and identify potential hot spots. Missouri is a leader in wastewater surveillance, as the state has the highest number of collection sites reported on a new CDC dashboard.

    Iowa

    A couple of weeks ago, I called out the state of Iowa for decommissioning its two COVID-19 dashboards, one dedicated to vaccination data and one for other major metrics. (I’m still bummed out about this, to be honest! Iowa had one of my favorite/most chaotic dashboards to check as a COVID Tracking Project data entry volunteer.)

    The change actually occurred this week: the old link to Iowa’s vaccination dashboard now goes to a 404 page, and all Iowa COVID-19 data are now consolidated in a single “COVID-19 reporting” page on the overall Iowa health department website.

    Here’s a bit more information on Iowa’s data shift, from a press release by the state’s governor:

    • Rather than reporting daily COVID-19 case numbers, vaccinations, and other data, Iowa is now providing weekly updates. The new, pared-down dashboard includes positive tests and death numbers over time, case and vaccination rates by county, and some demographic data.
    • For more frequent COVID-19 reporting, the Iowa dashboard now directs residents to federal data sources. Iowa is still reporting daily to the federal government, as all states are required to do.
    • The state health department “will continue to review and analyze COVID-19 and other public health data daily,” Governor Kim Reynolds said. But some teams focused on the COVID-19 response will return to pre-pandemic responsibilities.
    • This reporting change is intended to align with “existing reporting standards for other respiratory viruses,” Gov. Reynolds said.
    • Iowa is focusing on at-home tests with a program called “Test Iowa at Home,” in which residents can request to have a test kit sent to their homes for free. (It was unclear to me, from browsing the website, whether these are rapid antigen tests or PCR tests.)  The state health department processes these tests and collects data from the program.

    South Carolina

    A Tweet from South Carolina data expert Philip Nelson alerted me to this one: not only is South Carolina shifting from daily to weekly data reports, the state is essentially ending all reporting of COVID-19 cases. This is paired with a gradual shutdown of testing sites in the state.

    https://twitter.com/blu3r4d0n/status/1494679355628347397

    Here’s more info on South Carolina’s shift, based on a press release from the state health department:

    • South Carolina’s health department will stop reporting daily COVID-19 case counts on March 15.
    • The agency will continue to report COVID-19 hospitalizations and deaths as important indicators of disease severity, but these will switch to a weekly update schedule rather than daily.
    • The shift away from case reporting aligns with a greater focus on rapid at-home tests, which South Carolina’s health department says are “not reportable.” (While it’s true that the vast majority of rapid at-home test results are not reported, some jurisdictions, like D.C., allow residents to self-report their results!)
    • South Carolina’s health department is planning to gradually shut down almost all public PCR testing sites in the state throughout the month of March. According to the department, these sites have seen “a significant decrease in demand” due to increased availability for rapid tests.
    • The department is also discouraging regular testing for asymptomatic South Carolina residents, saying that individuals who are currently symptomatic or have a close contact who tested positive should be prioritized.

    More news on this topic

    • The CDC continues adding wastewater collection sites to its new dashboard. Two weeks ago, I wrote that only ten states had ten or more sites included on the dashboard; since then, three additional states have crossed that threshold: Illinois, Washington, and West Virginia. But the dashboard is still empty for the majority of states, indicating a lack of this important surveillance tool in much of the country.
    • For an upcoming story, I recently interviewed Lauren Ancel Meyers, a modeling expert at the University of Texas at Austin and lead author on this fascinating paper about using hospital admissions and mobility data for pandemic surveillance. Meyers has considered cases to be a messy indicator throughout the pandemic, she told me. She finds hospital admissions to be more useful, as this metric will directly show how many people are seeking healthcare due to their COVID-19 symptoms.
    • Another interesting paper, published in Nature this week, describes the use of machine learning models to drive COVID-19 testing at a university. The models could “predict which students were at elevated risk and should be tested,” the researchers write; students tested because of the models tended to be tested more quickly and were more likely to test positive than those identified through manual contact tracing or general surveillance. Such modeling could be used to augment the type of random sampling that Natalie Dean described in a recent article, shared in last week’s issue.

    Are there any other states shifting their data reporting for an endemic COVID-19 state that I’ve missed? Email me or comment below and let me know!

    More on state data

  • Three more things, January 30

    A couple of additional news items for this week:

    • Two House Democrats called on the CDC to release more Long COVID data. This week, Rep. Ayanna Pressley (from Massachusetts) and Rep. Don Beyer (from Virginia) sent the CDC a letter insisting that the agency report estimates of Long COVID infection numbers, including demographic breakdowns by race, gender, and age. “Collecting and publishing robust, disaggregated demographic data will help us better understand this illness and ensure that we are targeting lifesaving resources to those who need them most,” said Rep. Pressley in a statement to the Washington Post. While studies that may, theoretically, help provide such data are in the works via the National Institutes of Health’s RECOVER consortium, the consortium has yet to release any results. Long COVID continues to represent one of the biggest COVID-19 data gaps in the U.S.
    • We don’t know yet whether cannabis can treat COVID-19, despite promising early studies. Recent studies have shown that CBD, along with other products containing marijuana and hemp, has some capacity to block coronavirus spread in the body in lab-grown cells and in mice. The studies were quickly turned into sensationalist headlines, even though it’s too early to say whether these products could actually be used to treat COVID-19. An excellent STAT News article by Nicholas Florko and Andrew Joseph describes the studies and their limitations, as well as how these early reports of COVID-19 treatment potential are “adding to the FDA’s existing CBD headache” when it comes to regulating these products.
    • Have you received your free at-home rapid tests from the USPS yet? Last week, I described the federal government’s effort to distribute at-home rapid tests to Americans free of charge, along with the equity issues that have come with this initiative so far. This week, I saw some reports on social media indicating that people have started receiving their tests! Have you gotten your tests yet? If you have, I would love to hear from you—in absence of formal data from the USPS, maybe we can do some informal data collection on test shipping times within the COVID-19 Data Dispatch community.

    Note: this title and format are inspired by Rob Meyer’s Weekly Planet newsletter.

  • FAQ: A refresher on test positivity rates

    FAQ: A refresher on test positivity rates

    Test positivity trends for New York City, calculated and reported by the city health agency. Chart retrieved on January 23.

    I’ve recently been getting a lot of questions about test positivity rates, both from COVID-19 Data Dispatch readers and from friends outside this project, which reminded me of just how confusing this pandemic metric can be. So, here’s a brief FAQ post about test positivity; if you have more questions, shoot me an email!

    What is a test positivity rate?

    A test positivity rate is calculated through simple division: the number of positive tests counted in a particular region or setting during a particular period of time, over the number of total tests (positive and negative) conducted during that same period.

    Where do test positivity rates come from?

    While the test positivity rate calculation may seem simple, matching together the right numbers for that numerator and denominator can get pretty tricky. This is because, at the federal level as well as at most state and local health departments, positive tests and total tests are reported through different systems.

    Positive tests—also known, more simply, as cases—are prioritized for reporting. This is because public health departments need to know how many cases they are currently dealing with for contact tracing, potential hospital utilization in the coming weeks, and other crucial health system reasons. If a health department is pressed for time during a surge or coming back from a holiday break, it will analyze and report out case data before going through total test data. Similarly, many labs report their positive tests to health agencies separately from (and earlier than) total tests.

    As a result, simply dividing the new cases reported on a particular day over the new tests reported that day often won’t give you an accurate test positivity figure. Instead, the data analysts that calculate these rates typically match up the dates that tests were conducted. So, instead of dividing “all cases reported on Tuesday” over “all tests reported on Tuesday,” you’d divide “all tests conducted on Tuesday that returned positive results” over “total tests conducted on Tuesday.” This calculation provides a more accurate picture of test positivity.

    Also, different states and localities might report tests using different units, like “tests conducted,” “people tested,” and “testing encounters”—making it difficult to compare test positivity rates across states. This was a larger problem earlier in the pandemic; I recommend reading this excellent COVID Tracking Project analysis post for more info on the issue.

    How do you know a test positivity figure is reliable?

    As I explained in a recent post about the John Hopkins University (JHU) dashboard, the test positivity rates that appear on national dashboards often are not reliable because they fail to take these timing issues into account. A dashboard like JHU’s, which automatically scrapes data from state health agencies, does not have the backend information about the dates tests were conducted needed to calculate accurate positivity rates.

    JHU recently changed its test positivity calculations to better address differing testing units across states. Still, as the team behind this dashboard explains in a blog post, a lack of standardization across how states report their testing data makes it difficult to calculate positivity rates that can be accurately compared between jurisdictions.

    For that reason, I tend to trust test positivity rates calculated by individual state and local health agencies over those calculated by large, aggregating dashboards. For example, the NYC health department reports its own test positivity rate and does so with a three-day lag, in order to allow time for matching testing dates to case dates.

    In addition, I would be wary of test positivity rates that are calculated for a longer period than one or two weeks. Test positivity, as a metric, is meant to be an indicator of the current situation in a state, region, or a specific setting like a university campus; when reported for a longer period (like a month) or cumulatively, this metric doesn’t tell you anything useful.

    If you’re looking for a national test positivity rate source, the HHS’s Community Profile Reports include these figures for states, counties, metro areas—albeit with some reporting delays and gaps in certain states.

    How do you interpret test positivity rate data?

    I find this explanation from the COVID Tracking Project very helpful:

    Test positivity can help us understand whether an area is doing enough tests to find its COVID-19 infections. The metric is widely used by local, state, and federal agencies to roughly gauge how well disease mitigation efforts are going. Put simply, when test positivity is high, it’s likely that not enough tests are being done and that most tests that are done are performed on symptomatic people. Both of these factors—insufficient testing and only testing people who feel sick—make it very likely that many cases are going undetected.

    What would we consider a “high” test positivity rate? The CDC threshold here is over 10%; such a positivity rate means that one in ten tests conducted are returning positive results, indicating a lot of symptomatic people are getting tested for COVID-19 and a lot of cases are going undetected. A region with a positivity rate over 10% should step up its testing efforts and encourage asymptomatic people to get tested for surveillance purposes.

    On the other end of the spectrum, 3% and 5% are commonly used as thresholds for low test positivity. The specific number might depend on an institution’s testing capacity; at a business that regularly tests all of its workers and is already looking for asymptomatic cases, a test positivity over 2% might already be cause for concern.

    Generally, though, if this number is under 5%, it’s a good indicator that the region or setting has high enough test capacity to identify asymptomatic cases—and the majority of cases are being caught.

  • CovidTests.gov early rollout raises equity concerns; where’s the data?

    CovidTests.gov early rollout raises equity concerns; where’s the data?

    The federal government’s policies aimed at helping Americans get free rapid tests are insufficient for many households including people of color. Graphic via KHN.

    This week, the U.S. government unveiled a new website where Americans can get free at-home COVID-19 tests. The site is hosted by the U.S. Postal Service (USPS)—which will also distribute the tests—and it’s been lauded for its straightforward navigation and ability to handle a high level of traffic, both of which are unusual with government sites.

    On Tuesday, the site went live early in “beta test” form before its formal launch on Wednesday. Within hours of it going live, public health experts were already raising equity concerns about the free test distribution program. To address these concerns, the federal government should release data on where the free tests go—including breakdowns by state, county, ZIP code, race and ethnicity, the tests’ delivery dates, and more.

    As the link to the testing order site was shared widely on social media, one thing quickly became clear: people who lived in high-density settings were at a disadvantage. Americans in traditional apartment buildings, houses split into multiple living spaces, dormitories, and other multi-unit dwellings attempted to order tests—only to get an error message stating someone at their address had ordered tests already.

    The USPS ordering page is set up to allow just one test order per address, to prevent people from abusing the free test program. But, despite having literally every address in the U.S. on file, the USPS apparently failed to account for many apartment buildings. Some apartment-dwellers were able to get around this issue by placing their apartment number on the first address line, removing “Apt” from the address, or otherwise adjusting how they filled out the form, but these tricks didn’t work for everyone.

    I myself ordered the free tests before I learned about these issues on Twitter; I later sheepishly texted the groupchat for my Brooklyn, seven-unit apartment building, preemptively apologizing in case I’d fucked up my neighbors’ chances of obtaining free tests. (Luckily, my building seemed to be unaffected by the USPS issue—one of my neighbors responded saying that she was able to order the tests without a problem.)

    This issue “stems from buildings not being registered as multi-unit complexes and affected only a ‘small percentage of orders,’” the USPS said in a statement to POLITICO. And people facing this issue as they order tests can file a service request with USPS or call the agency at 1-800-ASK-USPS, according to KHN.

    Still, a “small percentage of orders” could add up to millions of people living in multi-unit housing who were unable to obtain free tests, or would have to share just four tests among an apartment building’s worth of residents. Without more precise data, it’s hard to understand the scope of this problem.

    All the Twitter discourse about apartment buildings obscures another group that shouldn’t have to share a small number of tests among many people: large households. The USPS is sending just four tests in each order—not four testing kits, four individual tests. That’s not enough for a family of four to test themselves according to FDA recommendations (i.e. twice within two days) after a potential exposure; it’s certainly not enough for large families including five or more people.

    And minority communities are more likely to include such large households. According to a Kaiser Family Foundation analysis of Census data: “More than a third of Hispanic Americans plus about a quarter of Asian and Black Americans live in households with at least five residents…Only 17% of white Americans live in these larger groups.”

    Households in West coast states are also more likely to include five or more residents, according to a similar analysis from the University of North Carolina Chapel Hill’s Carolina Demography center. States with the highest shares of five or more resident households are: Utah (18.8%), California (13.7%), Hawaii (13.5%), Idaho (13.2%), and Alaska (12.9%). On the other hand, in some East coast states, under 7% of households include five or more residents.

    !function(){“use strict”;window.addEventListener(“message”,(function(e){if(void 0!==e.data[“datawrapper-height”]){var t=document.querySelectorAll(“iframe”);for(var a in e.data[“datawrapper-height”])for(var r=0;r<t.length;r++){if(t[r].contentWindow===e.source)t[r].style.height=e.data["datawrapper-height"][a]+"px"}}}))}();

    The USPS test distribution system also gave an advantage to Americans with internet access. At one point on Tuesday afternoon, the USPS order site was drawing more than half of all government website traffic, demonstrating its popularity with internet users—while people without internet were not yet able to order tests.

    As of Friday, those without internet access can order the free tests over the phone, at 1-800-232-0233. This phone line is open daily from 8 AM to midnight Eastern Time, according to NPR, and Americans can order in over 150 languages. The USPS website itself is available in English, Spanish, and Chinese.

    While this phone line is very helpful now, the delay between the website’s release (on Tuesday) and the phone line’s release (on Friday) means that Americans without internet may be behind in the queue for actually receiving their tests. Already, the federal government has said that people who ordered their tests may need to wait for weeks to receive their tests.

    Of course, as analysis from KHN has shown, Americans of color are less likely to have internet access than their white neighbors. 27% of Native Americans, 20% of Black Americans, and 16% of Hispanic Americans have no internet subscription, compared to 12% of white Americans.

    Finally, the USPS test distribution system leaves out one major group of vulnerable Americans: those who don’t have an address at all. Homeless people are particularly vulnerable to COVID-19: many outbreaks have occurred in shelters, and many of these people have health conditions that increase their risk of severe symptoms. The impact of COVID-19 among homeless Americans is not well understood due to a lack of data collection; still, we know enough to indicate free tests should be a priority for this group.

    The White House has said that equity will be a priority for the free rapid test rollout: each day, 20% of test shipments will go to people who live in highly vulnerable communities, as determined by the CDC’s Social Vulnerability Index. This index ranks ZIP codes according to the communities’ ability to recover from adverse health events, based on a number of social, environmental, and economic factors.

    This priority is nice to hear. But without data on the test rollout, it’ll be difficult to evaluate how well the federal government is living up to its promise of equitable test distribution. I’d like to see data on the free test distribution that goes to the same level of detail as the data on our vaccine distribution, if not even more granular.

    The data could include: tests distributed by state, county, and ZIP code; tests distributed to ZIP codes that rank highly on the Social Vulnerability Index; tests distributed by race, ethnicity, age, gender, and household size; dates that tests were ordered and delivered; tests delivered to single- and multi-unit buildings; and more.

    Unlike other COVID-19 metrics that are difficult to collect and report at the federal level, the federal government literally has all of this information already—they’re collecting the address of every person that orders tests! There is no excuse for the government not to make these data public.

    In short: USPS, where is your free rapid test distribution dashboard? I’m waiting.

    https://twitter.com/JuliaRaifman/status/1483542002990002176

  • COVID source callout: JHU positivity rates

    On Friday, a COVID-19 Data Dispatch reader asked for my help in interpreting a wildly high test positivity rate: 544% in Washington, D.C. The source of this rate, she said, was Johns Hopkins University (JHU)’s COVID-19 dashboard.

    Test positivity rates seem simple; they’re calculated by dividing the number of positive tests over the total tests reported in a particular place, over a particular period of time. But these rates can be hard to calculate accurately because positive tests—a.k.a. COVID-19 cases—are often reported on a different time scale from all (positive and negative) tests.

    If a health department is swamped with COVID-19 data—or if it’s coming off of a holiday break—it will prioritize analyzing and reporting the case numbers over other metrics, because case reporting is most important for public health measures like contact tracing. Similarly, some labs might send in positive test results before they send in negative test results. This can lead to something like 100 cases reported on a Monday, but the tests used to find those cases not getting reported until later in the week.

    States and localities that calculate their own positivity rates have systems to account for these time differences, usually by matching up the dates that tests took place. But JHU doesn’t do this, because JHU test positivity rates come from automatic data scrapes and calculations with none of the backend timing information that you’d need to actually determine an accurate positivity rate.

    https://twitter.com/kissane/status/1478019697261957120

    In short, if you see a wildly high test positivity rate sourced from JHU’s dashboard, don’t trust it. Go look at the state, city, or county’s own COVID-19 data, or check the CDC dashboard instead.

    Also: I’d like to write more about test positivity next week, since this is such a confusing metric right now. If you have questions on this topic, send them my way!

    more source spotlights
  • FAQ: Testing and isolation in the time of Omicron

    FAQ: Testing and isolation in the time of Omicron

    After exposure to the coronavirus, someone may test negative on rapid antigen tests for multiple days before their viral load becomes high enough for such a test to detect their infection. Chart by Michael Mina, adapted by the Financial Times.

    As Omicron spreads rapidly through the U.S., this variant is driving record case numbers—and record demand for testing, including both PCR and rapid at-home tests. In other words, it feels harder than ever to get tested for COVID-19, largely because more people currently need a test due to recent exposure to the virus than at any other time during the pandemic.

    Also this week, the CDC changed its guidance for people infected with the coronavirus: rather than isolating for 10 days after a positive test, Americans are now advised to isolate for only five days, if they are asymptomatic. Then, for the following five days, people should wear a mask in all public settings. This guidance change has prompted further discussion (and general confusion) about who needs to get tested for COVID-19, when, and how.

    Here’s a brief FAQ, to help navigate this complicated testing-and-isolation landscape. In addition to the CDC guidance, it’s inspired by a recent question from a reader about testing and isolation following a positive PCR result in her family.

    What’s the difference between being infected and being contagious?

    As we think about interpreting COVID-19 test results in the Omicron era, it’s key to distinguish between being infected with the coronavirus and being actively contagious.

    • Infected: The virus is present in your body.
    • Contagious: The virus is present in your body at high enough levels that you can potentially spread it to other people.

    In a typical coronavirus infection, it takes a couple of days after you encounter the virus—i.e. breathe the same air as someone who was contagious—for the coronavirus to build up enough presence in your body that tests can begin detecting it. PCR tests can typically detect the virus within one to three days after an infection begins, while rapid tests may take longer.

    How do you use testing to tell if you’re infected and/or contagious?

    Timing is extremely important with coronavirus tests, and has become even more so with Omicron. If you learn about a recent exposure to the virus, you don’t want to get tested immediately after that exposure, since the test would not pick up a potential infection yet. Say you had dinner with a friend on Wednesday, and they tell you on Thursday that they just tested positive; you should wait until Friday or Saturday to get tested with PCR, or until Saturday or Sunday to get tested with a rapid at-home test. (And ideally, you would avoid interacting with other people while you wait to get tested.)

    PCR tests can detect the virus within a couple of days of infection. Rapid tests, which are less precise, generally can’t detect the virus until it’s at high enough levels for someone to be contagious. This can take time—though Omicron may have shortened the window between infection and becoming contagious to just three days, according to some early studies. A new CDC study released this week provides additional evidence here.

    This chart, an adaptation of a figure by rapid test expert Michael Mina published in the Financial Times, shows how someone could potentially test negative on rapid tests for multiple days after a coronavirus exposure, even though they are infected:

    When this person tests positive on a rapid test, the result indicates that they’ve become contagious with the virus. Then, it’s possible that the person may continue testing positive on PCR tests after they stop testing positive with antigen tests, because they are no longer contagious but continue to carry enough virus genetic material that a PCR test can pick it up.

    How do you get ahold of rapid tests, in the first place?

    In order to use rapid tests to tell whether you’re contagious with the coronavirus, you need to get some rapid tests! Here are a couple of suggestions:

    • Order online from Walmart: If you look at this website right now, Walmart will probably say that Abbott BinaxNOW rapid tests are out of stock. But if you leave the page open and refresh often, you may be able to snag some rapid tests right after Walmart restocks (which happens roughly once a day, I think). I like ordering from Walmart because they’re cheaper than other BinaxNOW vendors and ship quickly, usually within a week.
    • Order online from iHealth Labs: iHealth Labs is one rapid test manufacturer that’s grown in popularity recently, as an alternative to BinaxNOW. You can order up to 10 packs (with two tests each) directly from the manufacturer, and report test results in an app. In my experience, though, iHealth Labs is slower to ship than other distributors; an order I placed on December 22 is due to arrive two weeks later, on January 5.
    • Use NowInStock to see availability: This website tracks rapid test availability at a number of websites, including CVS, Walgreens, Walmart, Amazon, and others. It’s helpful to see your options for a number of different tests, but bear in mind that tests sold by third-party vendors (like Amazon) may be less reliable than those sold directly by pharmacies.
    • Follow local news: A lot of city and state governments have recently started making rapid tests available to the public for free, from D.C. libraries to Connecticut towns. I recommend keeping an eye on local news and government websites in your area to look for similar initiatives—or, if your area isn’t making rapid tests available, call your local representative and ask that they do!

    Why did the CDC change its guidance for isolation?

    As I mentioned above, the CDC recently changed its guidance for people who test positive for the coronavirus. If someone has no symptoms five days after their positive test result, they can stop isolating from others—but they need to wear a mask in all public settings.

    According to the CDC, the new guidance is “motivated by science demonstrating that the majority of SARS-CoV-2 transmission occurs early in the course of illness, generally in the 1-2 days prior to onset of symptoms and the 2-3 days after.” In other words, the CDC is saying that people are generally contagious for a few days after their symptoms start. After that, they’re less likely to infect others, so isolation may be less necessary—and good mask-wearing may be sufficient to prevent further coronavirus spread.

    Many experts are attributing the guidance chance to economic needs: as Omicron causes flight cancellations, closed restaurants, and other business disruptions, a shorter isolation period can help people get back to work more quickly. The recent isolation change follows a similar guidance change the previous week, which said healthcare workers could shorten their isolation periods if their facilities were experiencing staffing shortages.

    What are experts saying about the new guidance?

    Much of the commentary is not positive. While the CDC said the new guidance is “motivated by science,” the agency has failed to cite specific studies backing it up—though some such studies exist, as Dr. Katelyn Jetelina discusses in this Your Local Epidemiologist post.

    Generally, it does seem that most people—particularly vaccinated people—are no longer contagious five days after their symptoms start. (Reminder: five days after symptoms start could be seven to nine days into the infection period, since it takes time for the virus to build up in your body and cause symptoms.) But this is by no means guaranteed for everyone, as each person infected with the coronavirus has a unique COVID-19 experience.

    As a result, many experts have said that the CDC should have required negative rapid tests for people to leave isolation after five days. A negative rapid test would indicate that someone is no longer contagious, the argument goes, and they can then go back into the world. In the U.K., two negative rapid test results are required to shorten isolation from ten to seven days.

    However, for everyone in the U.S. to be able to rapid test out of isolation, the country would need a far greater supply of those tests than we currently have available. This Twitter thread, by epidemiologist Matt Ferrari, explains the challenges posed by limited rapid testing:

    Ferrari argues that the CDC guidance makes sense, given the information and resources currently available in the U.S., as well as the fact that simpler rules are easier to follow. Still, I personally would say that, if you have the rapid tests available to test out of isolation, you should.

    More Omicron reporting