When you’re looking at a whole bunch of numbers, as the Monitor does every week while tracking the status of COVID-19 in the state, it’s tempting to focus on the outliers.
An example came last Thursday when the number of new hospitalizations, a data point that was recently added to the tracker, suddenly spiked. Four of them were reported by the state in one day.
That doesn’t sound like many since New Hampshire saw up to 15 hospitalizations a day during the pandemic peak in May. But it stood out as the biggest number since mid-summer; in fact, we hadn’t seen more than one or two new hospitalizations for a full month.
The tracker added new hospitalizations this month to tell whether serious cases of COVID-19, as compared to milder cases, were spreading. State epidemiologist Dr. Benjamin Chan mentioned hospitalizations recently as an important measure of whether we were seeing a new surge.
Hospitalizations is what is called a lagging indicator, meaning that its ups and downs lag behind the ups and downs of another indicator – in this case, the number of new COVID-19 cases. The question is whether that hospitalization spike was the start of a surge in seriousness, lagging behind the increase in new cases that we’ve been seeing for weeks.
The answer: It’s too soon to tell. (You knew that’s what I’d say, right?)
It could just be a coincidence, with a couple of doctors around the state happening to decide that their borderline COVID-19 case were serious enough to hospitalize. That’s why you shouldn’t fixate on a single day’s data but average it out over a period of time – 14 days, in my case – to detect the actual trends.
With that in mind, here are the trends from the past week and how our four metrics are going. Note that you can see three charts about the virus in New Hampshire that I update each weekday on these links: New cases, new hospitalizations, and total deaths.
Goal 1: No sustained increase in number of new hospitalizations related to COVID-19. Have we met this goal? Yes.
Despite that one-day surge discussed above, the two-week average has been hovering around one case every days all month. It isn’t going down, but the numbers are so low that hospitalizations would have to virtually ceases for the average to decline any more.
On a similar note, COVID-related deaths remains very low, less than one every two days.
Goal 2: A two-week drop in new cases. Have we met this goal? No, but that’s not as bad as it sounds.
New cases are the leading indicator for COVID-19 – the first data point that we see as the coronavirus moves in a population.
I average the daily number over two weeks. That average bottomed out Aug. 28 but then rose for the next two weeks. It has been fairly steady over the past week at around 36 cases, double the minimum.
That’s not good, but it indicates that the outbreaks we’re hearing about on some college campuses, public schools and workplaces have not spread into the general population, which is a very good sign.
Goal 3: Fewer than four new cases per 100,000 people each day, or 54 new cases a day. Have we met this goal? Yes.
The average number of new cases has been below 54, the per-capita level commonly used as a sign of widespread infection, since mid-June.
Goal 4: Conducting at least 150 PCR tests per 100,000 people each day, or 2,000 tests per day. Have we met this goal? Yes.
The average number of test results reported each day was about 3,500 last week, thanks to increased testing at schools and colleges.
However, we need to keep testing all over the state for anybody who asks for it. That’s the only way we’ll know what’s actually going on.
Goal 5: A positive rate of PCR tests below 5%, indicating that we’re doing enough testing to get a good handle on how widespread it is. Have we met this goal? Yes.
As measured by the Department of Health and Human Services, the positive rate has been below 3% for most of the summer.
Return to the Concord Monitor