A new study reminds us that people who never had COVID can have symptoms that look like long COVID.

I’ve mentioned it before – there is a problem with Long COVID. Without a definitive diagnostic test, we have to rely on other markers – symptoms, diagnoses, medication prescriptions to understand the epidemiology of the condition. But with so many potential long covid symptoms – from shortness of breath to palpitations, brain fog to depression – it is inevitable that we may believe someone has long covid, when in fact they would have developed those symptoms regardless of prior covid infection.

It is for this reason that different studies estimate dramatically different risks of long covid – ranging from a few percentage points to as high as 50%.

But a new study – looking at long covid in kids – is instructive. It’s how this research needs to be done.

The study appears in JAMA Pediatrics and I’m going to start with how the authors could have presented this data. They didn’t, but I think it’s important to understand how we can get such drastically different estimates of long covid in kids or adults.

This is a huge cohort study leveraging the electronic health record data of more than half a million kids from around the US.

Of the 59,893 kids in the study who tested positive for COVID (this is before October 2021, so pre the home testing boom), 41.9% - almost half, had a symptom, sign, or medication suggestive of long covid in the 28 to 180 days after that positive COVID test. The headlines write themselves.

But this is not, thankfully, how the study was conducted. The authors recognize that with so many potential manifestations of long covid (they code 121 syndrome clusters and 30 medication clusters), kids without covid are bound to experience those outcomes as well.  You need a control group.

The authors used a logical one: kids tested for COVID, whose test was negative.

Entry into the retrospective study occurred at the time of a COVID test, regardless of the result.  The authors then looked to see which of those 121 diagnoses or 30 medications showed up in the medical record 28 to 180 days after the test.

The surprising finding? 38.2% of the kids with a negative COVID test had findings consistent with long COVID in the EHR. To be clear, that is less than the 41.9% of kids who had COVID – statistically significantly less – but it’s still a lot. What this tells us is that long covid likely occurs in kids, but also that kids get sick for plenty of non-COVID reasons. After adjustment for the fact that kids who get COVID had more comorbidities in general than kids who didn’t, the authors estimate that the risk of long covid is about 3.7% in children. That’s important – but not catastrophic.  But just to reiterate the central finding here, almost half of kids with COVID will have some symptom in the 1-6months after their diagnosis. But so will kids without COVID. Kids get sick.

Where this study shines though is finding those specific symptoms and signs that are uniquely prevalent in kids with COVID. These are the things that, when we think about long covid, we should really be looking for.

I’m highlighting some of the interesting ones here. Kids with COVID were about twice as likely as kids without COVID to have loss of taste or smell after recovery. They were about 50% more likely to have hair loss – we should note that these are both conditions that are pretty darn rare in kids in general.

Kids with COVID were three times more likely to be diagnosed with myocarditis 1-6 months later, and three times more likely to develop acute respiratory distress syndrome. Concerningly, they were 60% more likely to require treatment for a mental health condition. These were all highly statistically significant findings.

Who is most at risk for long covid? The authors found that kids with complex chronic conditions were more than 3 times as likely to develop long COVID than other kids.  By age, 5-11 year olds had the lowest risk. Encouragingly, the risk seems to be decreasing as the pandemic winds on.

What we can take from this large, epidemiologic study is some reassurance that long term sequela from COVID are rare in children. But it is clear that the long covid syndrome exists, and that, for some small number of kids, it can have quite profound implications.

A version of this commentary first appeared on Medscape.com.