One thing I didn’t understand before becoming a parent was the scale of the instinct to keep them safe. I obviously understood that parents had such an instinct, in the same way I understood that parents loved their kids. But after the birth of my sons, now 2 and 4, I learned just how powerful each of those emotions is when considering your child. I love them both more than I could have understood, and the urge to keep them safe is no less potent.

I say this not to cast those without children as incapable of appreciating that parents worry about kids. I say it because, for many parents, it adds a hard-to-appreciate level of complexity to the pandemic. Our 4-year-old was too young for pre-K last year but is now attending, meaning that for the first time in the pandemic we’re trusting someone else to ensure that he doesn’t contract the coronavirus. It’s novel for us, so it’s potent in a way that it probably isn’t for parents who’ve been sending their kids to school for months — but it’s also happening at a moment when the pandemic itself seems to be shifting.

On Thursday, for example, the New York Times reported that there had been a surge in children being hospitalized with covid-19, precisely the sort of thing that an admittedly overwrought parent doesn’t want to hear. But, while that’s true, the broader context of how the pandemic is affecting kids at the moment is at least somewhat reassuring. In short, the risks to kids remain low, even if no parent ever enjoys those harrowing nights of regular temperature checks and feverish, sobbing youngsters.

The Centers for Disease Control and Prevention makes public a large data set of anonymized infection records. It tracks race, age and outcomes by day, allowing us to see the patterns in how the pandemic has unfolded. Unfortunately, it is produced at something of a delay, meaning that the available data takes us only to the beginning of August.

From those figures, though, we learn that the shift that began with the emergence of the delta variant, spurring the fourth surge that started in late June, has only subtly increased the density of children among new cases. There are rises in hospitalizations and deaths, though they occur only at the end of the time series, thanks in part to the fact that hospitalizations and deaths trail new infection data by several weeks, given how diseases unfold. They are also only shown to be increasing at the end because the data are aggregated by the date of the positive test or the date the case was first reported to the CDC, not by date of hospitalization or death.

Is that spike at the tail end of the deaths graph worrisome? It’s not great, of course, but it is again useful to keep it in context. Below, we see the total number of cases, hospitalizations and deaths since March 2020 broken out into four age groups. The two youngest groups make up very few instances of those more severe infections.

The data used by the Times comes not from the CDC but from the Department of Health and Human Services. Beginning in July 2020, that data set includes daily figures on new hospitalizations by state, allowing us to assess the number of children who are entering hospitals and the percent of hospitalizations that number represents.

In short, the picture is the same: Yes, there’s been an alarming rise in hospitalizations among children, but that’s in part because there’s been a big surge in new cases. Children now make up a smaller percent of all hospitalizations than they did in early June or last summer. It is, however, rising.

Of course, we’d expect this to some extent, given that so many more adults are vaccinated against the virus. One of the benefits of being vaccinated is that you’re less likely to see worst-case effects of covid-19. At this point, 93 percent of the group most at risk, those 60 and over, has received at least one dose of a vaccine. Among adults overall, the figure is 75 percent. Among those 12 and up, 73 percent. Among those under 12? Almost none. So while that group has consistently been the least likely to need hospitalization, it’s also the least protected by vaccinations.

We see that there’s a loose inverse correlation between vaccination and hospitalization of children — meaning more vaccinations, fewer hospitalized kids — because there’s a stronger inverse correlation between vaccinations and the number of new infections.

You can see the recent increase in hospitalizations among children when breaking out the data by state. In Florida, Louisiana, Texas and Tennessee — all states hit hard by the delta variant — the cases have spiked. In states less affected by delta (and often better vaccinated), that’s not the case. (In some states, wild fluctuations in numbers due to small totals have been made mostly transparent in order not to disrupt the clarity of the chart.)

But, again, as a percentage of cases since the fourth surge began, the effect is generally subtle.

Again, this data is imperfect, reported by different standards and at different times in different places. But the two broad themes are clear: More children are being infected and hospitalized, but they still make up a small percentage of the total seeing the worst effects.

On Thursday, President Biden announced new policies and incentives aimed at increasing vaccinations. He encouraged parents worried about children under 2 (and therefore too young to get vaccinated) to get the shot themselves.

“Children have four times higher chance of getting hospitalized if they live in a state with low vaccination rates rather than in states with high vaccination rates,” he said. It’s not clear where those figures come from, but they may be low. The data from HHS used elsewhere in this article indicates that the 10 states with the lowest vaccination rates are seeing an average of 14.5 pediatric hospitalizations per million residents at this point, compared with 2.3 in the 10 most vaccinated states.

“For any parent,” Biden said at another point, “it doesn’t matter how low the risk of any illness or accident is when it comes to your child or grandchild. Trust me — I know.”

On that, he’s not wrong.