By Liora Engel-Smith
Four months into the pandemic that led to about 1,400 deaths in North Carolina, sickened thousands more, created massive unemployment and disrupted nearly every aspect of life, the magnitude of coronavirus in the state remains hard to grasp. State officials cited some of those indicators as the deciding factors in halting the state’s reopening process and the passage of a mask-wearing mandate late last month.
To help you make sense of the numbers that led to that decision, we’ve compiled three charts that explain what’s happening with coronavirus in North Carolina.
The rise of coronavirus cases in the state coincides with the easing of stay-at-home restrictions.
Gov. Roy Cooper relaxed some of the stay-at-home restrictions right before Memorial Day weekend. Summer weather, combined with the reopening of hair salons and restaurants for sit-in dining have meant that more people are out and about. During that time, protests over police brutality also increased the number of people in the street.
Because of coronavirus’s long incubation period (as much as 11 days) and then the time it takes for people to see a doctor for their symptoms, case tallies don’t increase immediately after someone is exposed. It can take roughly two weeks to see the impact of loosening restrictions, according to state epidemiologist Zack Moore.
North Carolina Department of Health and Human Services data illustrates this trend. The state’s coronavirus cases have been climbing faster since late May, state data shows. At the beginning of May, the daily case tally and the seven-day average curve — a measure that separates a clear trend from a random, one-off rise — are closer together. But by the end of May, the two lines are farther apart, indicating that cases are climbing faster now.
The governor’s mask order went into effect on June 26 and it’ll take roughly two weeks to begin seeing its impact on cases. Data from states that already put mask orders in place, so far, show that cases decline significantly when people consistently wear masks in public.
For example, New York Governor Andrew Cuomo imposed a mask mandate starting on April 17. Two months later, that state’s number of positive cases has dramatically declined to about one percent of the people tested.
We can tell North Carolina’s coronavirus surge is real because hospitalizations have been climbing.
Since the pandemic began, health officials have made concerted efforts to increase the number of coronavirus tests in the state. To date, more than 1 million tests have been completed, up from roughly 500,000 a month ago. But the rise of coronavirus tests isn’t the only reason North Carolina’s caseload is going up. We know that because coronavirus hospitalizations are going up.
At the beginning of the pandemic, only people who had obvious coronavirus symptoms got tested. As testing capacity increases, more asymptomatic people and those with mild illness are getting diagnosed, but the number of people with complications serious enough to be in the hospital on any given day is continuing to increase, suggesting that coronavirus is spreading through the community.
North Carolina’s COVID-19 hospitalizations dipped briefly in the middle of May near the end of the state’s lockdown. But since the easing of some stay-at-home restrictions in late May, cases have been steadily climbing. Hospitalizations topped 900 every day over the last week in June, for the first time since the pandemic began. A seven-day average, which helps with separating trends from day-to-day fluctuations that aren’t meaningful, is also trending up.
We know coronavirus is spreading because the proportion of positive cases is also increasing.
As the pandemic evolved, the state has expanded testing to include not only people with serious illness but those who may have been exposed but show no symptoms. The proportion of positive cases in the state continues to climb, meaning that the virus is spreading through the community.
While the percent of people who test positive on any given day varies, a seven-day average illustrates the proportion of people who are testing positive — a number that is also trending upward. In mid-May, the percent of positive cases was 7 percent. Roughly a month later, 9 percent of coronavirus tests came back positive.
Not all labs that test for COVID-19 report their negative results to the state, and that has the possibility of distorting this measure, according to Moore, the state epidemiologist. Reporting only positive cases would make the spread look worse than it is. The department gets around this issue by calculating the percent positive only using results from labs that reported both positive and negative tests.
By only including the labs that report both percents of positive and negative results, he said, the department has a more accurate depiction of the spread of coronavirus in the state.
When several key indicators trend upward, spread is the likeliest explanation.
No coronavirus indicator is perfect. Hospitalizations, for example, depend on hospitals consistently reporting their coronavirus cases to the state, but not all of them do. Testing tends to capture only some of the cases, and according to a CDC estimate, the total number of people who have become infected by the virus over the past six months may actually be 10 times higher. The impact of this rise on coronavirus-related deaths in the state remains to be seen because it takes several weeks or more from the onset of coronavirus to hospitalization and recovery or death from the disease.
Even with these limitations, when several indicators trend upward at once, the message is clear: coronavirus is spreading.
Republish our articles for free, online or in print, under a Creative Commons license.
“Not all labs that test for COVID-19 report their negative results to the state, and that has the possibility of distorting this measure,.”
The possibility? No, not reporting negatives CERTAINLY distorts the measure. If someone makes 50 winning stock trades and does not mention the losing trades, then one’s stock picking prowess can easily be distorted. ALL wins and ALL losses must count. When it comes to COVID, ALL positives and ALL negatives must count. Or else the % positive number is meaningless.
There are other distortions. For example, four times in the last two weeks I have had my forehead scanned to determine if I had a temperature. A high temperature is a very common–one of THE most common–signs of COVID. Four times my temp was 100.3, then those five will be encourage to get a COVID test, since the high temp is an indication of infection. Those 5 are obviously more likely to have it then the 95 with the lower temps.
If only the 5 with a high temp are required to take a test, and not the 95 that do not, then the positive rate is certainly going to be skewed.
The percentage positive can only be a true representation of NC if it is random and not skewed by pre-screening.
Would anyone trust a pollster who asked likely voters to identify who they voted for in 2016, and then counted only those voters who choose Donald Trump (or Hillary Clinton)? Of course not. That said, why would anyone trust a state health official who engages in his or her own sampling bias by making less effort to count the negatives?
Have a nice evening.
Hi Glen, thanks for your comment. Up till now, the state has used only the negatives and positives from labs that reported both in computing the percent positives, meaning that the proportion of positive to negative tests is the best estimate of what you might expect in the entire state. If they included positives from labs that didn’t report negatives, the results would definitely be more skewed. But you’re right, it’s not exactly the same and the actual percent positive day to day may be different.
NCDHHS announced today that it’s requiring all labs to report both positive and negative tests, so from now on, the percent positive calculation will reflect all the tests that are being done in the state. You’re correct though, no measure that we have is exact and every number has a weakness or limitation. But what matters are the trends. Ours are going up, even if we don’t know the exact percent of people who tested positive day-to-day. We will know moving forward, however.
The recent protests for BLM and other causes has certainly caused a rise in positive numbers as well as the non-reporting of negative tests. Many protestors not wearing masks. It is sad that the BBC was the first major news organization to print the truth concerning the protests. Our own national media refuse to give the American people facts if those facts don’t fit their political agenda.
Everybody give Governor Cooper a hand for taking his mask off his face while walking with other protesters. Way to lead by example gov.
The increase in hospitalizations is most concerning. Let’s hope Covid 19 deaths don’t follow the same trajectory.
Elizabeth, thank you for reading the story. Generally, when cases go up, we can expect more hospitalizations and more deaths, though at a later time because it generally takes a couple of weeks for someone to reach the point of complications in the disease trajectory. However, Sec. Cohen has said that our case increase is driven by younger people, and we know that younger people without chronic diseases tend to do better as far as COVID goes. We likely won’t know what the current surge in cases and hospitalizations will do to our death counts for at least another month, but we will keep monitoring the numbers and updating our readers as more information becomes available.
Conveniently missing from the article is the amount of COVID-19 spread by protesters and looters. While it is lawful to peacefully protest, the fact of folks closely packed together and screaming through homemade cloth masks on a near daily basis for weeks on end has certainly spread the virus. Of course this fact has not been discouraged because it does not serve the narrative of much of the media. But it may be the case that looters contributed less to the spread of COVID, given the heat from the flames as a result of the businesses they torched…….
NYC had massive BLM protests for a week and didn’t see any uptick in cases or positivity. The protesters were outside and observational studies to date point to very, very little cluster spread from being outdoors.
Minneapolis, also no uptick in cases or positivity:
Finally, reopenNC didn’t report many cases coming from their protests which were maskless but OUTSIDE
Amazing that amongst your several, and excellent charts of data, with moving averages included, you insist on leaving out the one true ‘telling’ data trail – deaths directly from Covid-19. However you are included in a large group of National Reporters in consistently avoiding this primary indicator of severity. Please include a chart of deaths in future representations.
I would have cemsured, or fired, research engineers for such carelessness – or obvious bias – in my days as R&D Director @ DuPont and given low grades to graduate students as NCSU as Distinguished Professor.
Thanks for reading the story and for commenting. We tally daily death tolls on our landing page and we also note them in Coronavirus Today roundups if you’re interested, but deaths are actually not the most immediate way to detect or understand coronavirus surges. Deaths are what epidemiologists call a lagging indicator, which means it takes them longer to show changes. It makes sense if you think about the course of disease. It can take two weeks for people to show symptoms and a couple more weeks to develop symptoms that require hospitalization if that’s their disease trajectory. Death is the last thing that happens, and because of that, the death numbers we have are more of a reflection of what happened a month or more ago.
On top of that, it’s hard to say what’s going to happen to our death tally. It’s been climbing slowly, but the current rise in cases is driven by younger people, who on the whole tend to do better with COVID (provided they don’t have underlying health conditions).