This week, NIH Director Francis Collins tweeted that an NIH-funded study found that “teens & young adults who use e-cigarettes...were 5 to 7 times more likely to be diagnosed with COVID-19 than their peers who did not vape.”
Quite a disparity in risk! Perhaps vaping is deadly after all. But not so fast. Let’s look at the actual article.
First, an initial sign of trouble was pointed out by Nick Brown on Twitter: Some of the basic descriptive numbers just aren’t possible. For example, the respondents were asked about their mother’s level of education, and 609 said that their mother had “started college” (but not graduated). Of these 609, 48% were never-users while 52.0% had vaped.
But that’s impossible: 48% of 609 is 292.32, while 52% is 316.68. People exist in integers — 1, 2, 3, etc. — and there’s no such thing as someone who has .32 or .68 of a mother. (If the researchers had rounded the numbers, that still wouldn’t explain the results here.) Similar impossible numbers can be found elsewhere.
As Brown says, this is a warning sign: If researchers can’t do something as simple as calculate the percentage of people in their study who answered a certain way, what can you really trust about their numbers?
Second, the reported risk here doesn’t make sense. If you look at Table 1 of the study, 17.5% of the vapers claimed to have gotten a COVID-19 test, compared to 5.7% of the non-vapers. Then, Table 1 says that 2.3% of vapers had a positive COVID-19 diagnosis, compared to .8% of the non-vapers.
This actually works out to a roughly equal rate of COVID in both groups! Sure, the vapers were more likely to say they had a positive COVID-19 test, but they were also more likely to say they got tested at all. Thus, a roughly equal number of vapers and non-vapers tested positive for COVID-19 if they got tested at all.
Third, do we trust those numbers in the first place? The study claims that it surveyed people via a “survey Web link on gaming sites, social media, customer loyalty portals, and through website intercept recruitment.” For all we know, teens and young adults who vape might have been more likely to say that they got a COVID-19 test or that it was positive.
Fourth, it isn’t believable that 17.5% of vapers had been tested for COVID-19. If 17-18% of people were getting tested, there would have been nearly 58 million tests nationwide at that time, which is about five times more than had actually occurred! This indicates a significant bias in who was responding to the survey.
All told, this doesn’t seem like a reliable study. Vaping might be bad for you, but this study doesn’t show it.