By nearly all measures it has been a horrible week, a horrible month (nine days in), and a horrible year. The United States set pandemic records in all three metrics that measure the pandemic’s severity this week, recording a total of 1.4 million new cases and 15,966 deaths. Yesterday, states and territories reported 3,088 deaths from COVID-19—a record no one wanted to see—and the average number of deaths per day this week exceeded 2,000, surpassing the highest average we saw in the spring’s deadly first surge. More than 106,000 people are currently hospitalized with COVID-19.
If the patterns we’ve traced here since spring hold true, the worst is yet to come. Given the rapid increase in the number of new cases, we expect the metrics for hospitalizations and deaths to continue to rise in the coming weeks—especially if in-person gatherings over Thanksgiving led to increased spread of the novel coronavirus, as public health experts warned.
A crucial bit of muddiness remains in the data itself: Some of the reported deaths we are seeing now are likely related to the post-Thanksgiving backlog we’ve written so much about in the past month. What's not exactly clear is how many of these deaths are simply following the case increases that occurred before and during the holiday week. Either way, we expect that deaths will climb steeply given current case and hospitalizations numbers. Test reporting, too, is probably still slightly low because of the holiday effect. Case numbers appear to be largely back online following the holiday reporting disruption, however, which makes sense—historically, case numbers have been quicker to recover from holidays and weekends than tests and deaths.
Case numbers were high across the US this week, with each of the country’s seven most populous states reporting more than 50,000 new cases among their residents. Fourteen states in regions across the US reported a record number of cases this week; that’s on top of the 16 states that hit new-case records last week.
The slowing case rises in Midwestern states that we noted last week has continued: After the Thanksgiving data dip and subsequent rise, new cases appear to be declining in the region. Meanwhile, in the Northeast, South, and West, cases are rising sharply.
After 14 weeks of North or South Dakota posting the highest per-capita case numbers in the country, Rhode Island has taken their place with 1,253 cases per million on the seven-day average. (The below chart, along with most of the others we use in our weekly posts, is available as a mobile-friendly interactive tool in the charts section of our site.)
With the catastrophic rise in cases, communities of color across the country are dealing with the terrible effects of the pandemic. In Rhode Island, where the average new cases per capita is the highest in the country, more than 1 in 8 Latinx people have tested positive for COVID-19, compared to 1 in 31 white people. In both North Dakota and South Dakota, 1 in 8 Black people have tested positive. Some of the highest per-capita statistics are reported for American Indians or Alaska Natives. South Dakota, where the average cases per capita has been among the highest, is reporting that 1 in 7 Native Americans has tested positive for COVID-19. Notably, as other case disparities have decreased in South Dakota, that has not been the case for Native Americans, for whom the weekly new cases reported per capita have been higher than for any other group, or the state overall, for the past three months. In surveying racial and ethnic disparities in other parts of the country, we also note that in Louisiana, 1 in 6 Asian people have tested positive for COVID-19.
US hospitals are still feeling the strain as hospitalizations climb
We’ve frequently written that rising hospitalizations lag behind rising cases, and the same is true when cases fall. Hospitalizations rose in 26 states this week, mostly concentrated in the Northeast, South, and parts of the West. Hospitalization data came through Thanksgiving with no major reporting artifacts, and we take these rising numbers as a strong indication of worsening outbreaks in many of the country’s most populous states. Although the Midwest still has the highest per-capita number of people hospitalized, the actual numbers there are falling; 9 states in that region reported drops in the number of people hospitalized compared with last week.
New Hampshire, Delaware, and Maine reported the largest percentage increases in COVID-19 hospitalizations in the past week, with California and New York right behind them. In California, the already-high number of hospitalized COVID-19 patients rose 23 percent, to 11,965; the state is reporting that statewide, ICUs are 88.4 percent full. Intensive care units in Silicon Valley and parts of the Central Valley are full, and San Francisco, which long boasted of relatively low case levels given the city’s density and population, now says it may run out of hospital beds in the coming week. In Los Angeles County, ICU and overall hospital capacity are running low, and a record 1,719 healthcare workers reported they have COVID-19 as of Sunday, more than double the number reported sick in the county the week before.
Earlier this week we published an analysis comparing hospitalization data collected by The COVID Tracking Project to a dataset from the US Department of Health and Human Services. We found that the federal data on current COVID-19 hospitalizations very, very closely tracks our dataset when we take into account differences in data definitions and how dates are reported. We think this federal metric, which HHS publishes alongside several other potentially useful hospitalization figures that we can’t get from states, is a good candidate to use alongside and eventually in place of the state data we compile.
This week in long-term-care data
The pandemic continues to play out in particularly distressing ways in long-term-care facilities. As our LTC team reported yesterday, a surge in cases in the Northeast and West continues; last week both regions reported their highest numbers of new cases in the past six months. The Midwest and South saw a small downturn in new cases, which is promising, yet the week still saw the nation’s highest number of newly reported cases—51,574—in long-term-care facilities since The COVID Tracking Project began collecting this data in May. And, as we always note, these figures are an undercount, since not every state reports long-term-care data.
Nearly 40 percent of the new cases in the Northeast are coming from Pennsylvania; the state reported more new cases of COVID-19 in LTC facilities than any other state except California, which has more than three times as many residents. Our researchers have been unable to obtain current data about the number of people living in Pennsylvania’s long-term-care facilities. But if numbers have remained relatively constant over the past five years, we can estimate, based on a 2015 figure of residents in LTC facilities, and available COVID-19 data from the state, that nearly 30 percent of people living in Pennsylvania LTC facilities have had COVID-19. And nearly one in five Pennsylvania long-term-care residents who contracted the virus has died.
Alice Goldfarb leads The COVID Tracking Project’s part in The COVID Racial Data Tracker.
Erin Kissane is a co-founder of the COVID Tracking Project, and the project’s managing editor.
Jessica Malaty Rivera has an MS in Emerging Infectious Diseases and is the Science Communication Lead at The COVID Tracking Project.
Charlotte Minsky helps lead the COVID Racial Data Tracker team.
Joanna Pearlstein is an editorial lead at the Covid Tracking Project and a former editor at WIRED and Protocol.
Three of our four topline COVID-19 metrics are still recovering from data interruptions over the Thanksgiving holiday weekend. Meanwhile, hospitalizations are soaring.
The upcoming holiday means that many COVID-19 metrics are going to drop—and then rise—in ways that may trip up unsuspecting observers. Here’s what to watch out for.
Despite a rocky beginning, the current hospitalization and new admissions metrics from the HHS Protect public dataset have stabilized—and they’re now largely harmonious with state-reported hospitalization metrics if we account for differences in data definitions and reporting lag time.