COVID Tracking Project / NIAID / The Atlantic

Editor’s Note: The Atlantic is making vital coverage of the coronavirus available to all readers. Find the collection here.

The United States set a new record for reported cases this week, breaking 500,000 for the first time in the pandemic as the third surge continued to build across nearly every state in the country.

Today, the country recorded 88,452 new cases of COVID-19, its highest single-day total since the pandemic began. Over the past two weeks, 25 states have set a new record for cases in the past two weeks, including 17 states with record highs since last Wednesday.

The country reported a record number of tests, at 8.2 million, but case growth (24 percent) far outpaced test growth (9 percent), as we explained earlier this week. That’s also true for the entire month of October: Forty-seven of the 50 states, along with the District of Columbia, have seen cases rise faster than reported tests since October 1.

These cases are translating into higher numbers of COVID-19 hospitalizations in many states in every region of the country. All but 11 states saw a rise in people hospitalized this week, the largest increases occurring in the upper Midwest and Texas. Although we are not yet close to the hospitalization peaks of almost 60,000 that we observed in the spring and summer, the average number of people hospitalized this week rose to 42,621, a very substantial increase from the lows of about 30,000 that we saw just a month ago.

In some worrying states—North Dakota, South Dakota, and Wisconsin—hospitalization rates are approaching those of the Sun Belt–surge hot spots over the summer. At this point, we see no evidence that any state in the current surge has reached its peak and begun to decline.

COVID-19 deaths have risen, but not nearly as sharply as hospitalizations. The current explosion of cases is only three weeks old, and we know from the data that reported deaths lag behind cases and hospitalizations. According to the Centers for Disease Control and Prevention, the approximate timeline from symptom onset to the reporting of a death to public-health authorities is about three weeks, on average. This lag varies by state, and some states such as Florida take an average of more than six weeks to log COVID-19 deaths. Given the current trends, we should expect deaths to continue to rise in the coming weeks.

More encouraging, as our team explained this week, the fatality rate for hospitalized patients has declined as treatments have improved and younger and healthier people have made up more of those hospitalizations.

It’s important to note, though, that the hardest-hit states are seeing notable increases in deaths. In Wisconsin, the state that led off this surge, deaths rose 56 percent this week and are up 270 percent since the week ending October 1. Deaths have also climbed markedly in the Dakotas to a combined 101 this week, from 11 in the week ending September 3. On a per-capita basis, that is worse than any individual week of deaths in Florida at any time in the pandemic.

With the presidential election next week, the swing states of Michigan, Ohio, Pennsylvania, and Wisconsin would already have been in the spotlight. But they also happen to be in the midst of a major uptick in COVID-19 cases. Beyond the large rises in cases since October 1, hospitalizations are up at least 96 percent in all four states. Ohio and Wisconsin have also surpassed their previous hospitalization records.

If we look at the county-level data from these four states, we see another pattern emerge. In the early days of the pandemic, the outbreaks in these states were concentrated in major cities, especially Detroit and Philadelphia, but this new surge is much more geographically dispersed.

Since September, the racial composition of people who are confirmed to have the coronavirus has also changed remarkably in the three worst-hit states, North Dakota, South Dakota, and Wisconsin. Our COVID Racial Data Tracker shows that the per-capita infection rate in Black, Latinx, and Asian communities has declined, while per-capita infections among white people have risen very slightly. (In states with mostly white populations, even a small rise in per-capita cases among white residents can translate to large increases in total numbers of cases in this population.)

In absolute terms, the disparities remain pronounced: In North Dakota, since the start of the pandemic, one in every 12 Black people has tested positive for COVID-19, compared with one in 29 white people. In South Dakota, one in every 14 Indigenous people has tested positive, compared with one in 26 white people. In Wisconsin, one in every 13 Latinx people has tested positive for COVID-19, compared with one in 30 white people. But recent data suggest that the virus is now circulating in whiter populations outside the major cities.

We’re seeing a dramatic rise in long-term-care-facility cases in many of the states now experiencing surges, an alarming trend given that long-term-care deaths account for roughly 41 percent of all COVID-19 deaths in the U.S. This week saw 17,848 new cases associated with long-term care facilities. Because many states do not separate out resident cases and staff cases, this total includes both.

Wyoming reported a record number of new cases in long-term care facilities for the third week in a row. The state has reported 309 total resident and staff cases since the beginning of the pandemic. About one-quarter of those are from this past week, and half are from the past two weeks.

Ohio reported 791 new cases of COVID-19 among nursing-home residents this week, the highest number in the state since we started tracking cases in late May. And Wisconsin’s long-term care facilities reported their highest number of resident and staff cases since May: 241.

Hospitalizations have risen at least 96 percent in Wisconsin, Michigan, Pennsylvania, and Ohio since the beginning of October.

We continue to be alarmed that Arizona and South Dakota do not report comprehensive long-term-care data. Arizona releases the number of affected facilities but does not provide facility-level case or death numbers. And South Dakota does not release COVID-19 data on long-term care facilities at all.

The bottom line this week is simple: The third surge is in full swing, hospitalizations are rising nationwide, and deaths have begun to increase in Wisconsin, the state that could be seen as the canary in the coal mine of this surge. But it is difficult to predict what might happen from here. Just as the summer surge in the Sun Belt unfolded differently than the spring outbreak in the Northeast, we are seeing distinct patterns emerge at this stage of the pandemic. Cases and hospitalizations are more geographically dispersed and appear to be rising less steeply but over longer periods. And the behavioral shifts brought on by the intense cold arriving in the upper Midwest may accelerate the spread of the virus.

Testing accessibility and in-hospital patient care have improved, so it seems unlikely that we’ll return to the case-fatality rates of the spring. The lower death rates from COVID-19, however, are predicated on stable and functioning health-care systems. As this outbreak continues to grow in so many urban and rural communities, public-health officials have brought back the “Flatten the curve” mantra of the spring. Each fall, hospitals prepare for an increase in patients requiring respiratory support due to influenza season.

What was true in the spring is true now: If hospitals are overwhelmed with COVID-19 patients, they cannot provide the quality of care that has reduced the number of Americans who die from this disease, and from other health emergencies. And local reporting suggests that communities from El Paso to Green Bay are in danger of letting the virus grow beyond the capacity of health-care workers to contain the damage.


This post appears courtesy of The COVID Tracking Project. Artis Curiskis, Alice Goldfarb, Erin Kissane, Júlia Ledur, Alexis Madrigal, Jessica Malaty Rivera, Charlotte Minsky, Kara Oehler, Joanna Pearlstein, Sara Simon, Peter Walker, and Nadia Zonis contributed to this analysis.

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.