Mixed public health messages and misunderstandings of risk haven’t helped.
At the beginning of the coronavirus outbreak in the US, most Covid-19 cases seemed to be in older folks, who were getting very sick — an artifact of who was getting tested at the time.
But now, especially in states where Covid-19 cases have been rising in June and July, the median age has been dropping.
“We are seeing it in a much younger group, of 20- to 40-year-olds,” compared to the earlier stage of the pandemic, says Thomas Tsai, a health policy expert and surgeon at the Harvard T. H. Chan School of Public Health.
In Florida, where the coronavirus has infected more than 311,000 residents, the median age of positive cases in March was 65. As of July 17, the median age of new cases from the previous 14 days was 39.5.
Related
In Arizona, 61 percent of Covid-19 cases are in people under the age of 45. In Texas’s two largest counties, Harris (home to Houston) and Dallas, about half of the new cases have been in people under 40.
Other states with worrying rises in case numbers are seeing a similar trend. In California, which just reclosed bars and indoor dining, as of July 15, people ages 18 to 34 made up the largest proportion of new cases (24.3 percent), with 35-to-49-year-olds as the second-largest group (19.3 percent of new cases). Even Virginia, where cases have been relatively stable, has seen an increase in new cases among people in their 20s.
Nationwide, “the average age of people getting infected is now a decade and a half younger than it was a few months ago,” Dr. Anthony Fauci, director of the National Institute of Allergy and Infectious Diseases, said in a July 6 press briefing.
Younger people, on average, are less likely to become severely ill and die of Covid-19 — although many do. They can also spread it to older people who are much more vulnerable to severe infection and death.
This seems to already be happening, with assisted living facility cases climbing in Houston and Phoenix, as well as in Florida now. “We first see it in the community, and then we see it in the residents and staff, and then you see the deaths,” David Grabowski, a health care policy expert at Harvard Medical School, told the Wall Street Journal.
Check out this helpful visualization of the ages of new cases in Florida. With many more cases in younger adults (orange band), we are closely tracking spillover into older adults (red and gray bands). Credit: @jasonsalemihttps://t.co/TYinXKTQa1pic.twitter.com/UQgsBq5GS7
So even if you’re a young person who gets infected and don’t get very sick, “you’re part of the propagation of the pandemic,” as Fauci put it.
So how did the coronavirus suddenly start spreading to so many young people?
We’ve known that bars and other crowded indoor spaces are hubs of spread in the new hot spots. But there are likely many other factors at play.
THINK LIKE AN EPIDEMIOLOGIST:What does it mean that the median age of new cases is dropping in some areas? I see three possible explanations, not all good. A thread on how to distinguish between them. 1/10(Figure h/t @ScottGottliebMD) pic.twitter.com/Y6m45qoBL6
Some of these factors include mixed public health messaging and premature reopenings, young people skipping precautions because they perceive a lower risk of getting very sick, people returning to jobs, and more testing of people without symptoms. Let’s unpack some of these, which can help us better understand — and possibly slow — this new trend.
Getting a clear picture of the age demographics of new Covid-19 cases is not as simple as you’d think. States are on their own to decide whom they group together in their statistics, which can make the definition of “young” very different from place to place — and result in skewed numbers.
Arizona, for example, has lumped together people ages 20 to 44, which shows a huge number of cases. But this group is also the largest span of years of the state measures, compared to other groups (such as the 55-to-64 age range, which spans just nine years).(These age groupings might be a legacy from the earlier days of the pandemic, when we were trying to better understand the illness’s severity in older adults.)
Arizona also displays cumulative cases since the beginning of the pandemic for each age group, whereas California highlights the number of new cases by age group for each week. This difference makes it harder to compare the age ranges involved in spreading the virus currently and historically.
To make things even more challenging, most states are also primarily reporting the overall number of cases per age group — rather than the percentage of that demographic’s population. (We can see from states that are providing this information that rates among different age groups can be very different than overall case counts might lead you to believe. For example, Massachusetts’s largest raw number of cases have been in people ages 50 to 59, but by percentage, the rate has been far greater in people over 69.)
Another way to get a sense of how prevalent the virus is in different age groups is to look at the rate of tests coming back positive. This is still not a perfect metric (and depends a lot on who — people who are symptomatic, high-risk individuals, those going back to work, etc. — is being tested). But it does help provide a sense of proportionality so we can do a rough comparison of how frequently people in different age groups are likely to come back with a positive Covid-19 test.
And US Centers for Disease Control and Prevention data can differ from what states are reporting. For example, a recent Bloomberg report found that the CDC was likely underreporting the rate of Covid-19 infections in people under the age of 18 compared to what many states were showing.
Without more uniform — and, in some cases, more granular — data, it can be difficult to pin down precisely which age groups might be most prone to be spreading the virus in different places and what behaviors would be best to target to change.
