There are so many headlines flying around at the moment about the coronavirus that it’s sometimes hard to keep up. From the daily updates on numbers of cases, to the endless speculation about exactly whose fault the latest crisis is, it’s increasingly impossible to get to everything. But despite that, there’s a recent media storm that stuck out, because it just seemed so very odd and unlikely.
You see, according to news from across the world, being tall makes you more likely to catch coronavirus. People over 6 foot in height, apparently, are in dire danger of breathing in particles carrying COVID-19, which puts them at a disadvantage, based on a new scientific study.
Now, there is some speculation about risk factors for COVID-19. There are lots of methods of transmission, and a huge amount of debate about exactly how big a role aerosols and bigger droplets play. But one thing that has never really come up before is the idea that being tall would necessarily put you at a higher risk of disease.
Fortunately for the tallies out there, the headlines were wrong. There’s no evidence that being tall makes you more likely to catch coronavirus after all.
The study that all these headlines are referring to is a new preprint that looked at work-related hazards for COVID-19 infection. They surveyed 2,000 people in the US and UK — split evenly between the countries — and asked them a bunch of questions about their work habits and whether they’d had COVID-19. The authors then correlated the various answers with people’s risk of having had COVID-19, and made a bunch of conclusions about what might be driving risk of infection.
One of the things that the authors found in their statistical model was that, if you split people into two groups — over and under 6 foot tall — those who were over 6ft were more likely to report catching COVID-19 in the past. In fact, they had more than double the odds of catching the disease, which sounds pretty terrifying to anyone breathing the rarefied air of the Rather Large.
Except, there are a few pretty huge caveats that make it quite a bit less likely that tall people are actually at risk from coronavirus.
In fact, it’s hard to conclude anything about height and COVID-19 from this study at all.
This is a problem if your study design splits people up by height.
We can see this pretty easily when we look at some proportions. In the US, only about 11% of men and 1% of women are taller than 6ft. In the total population, that works out to be very roughly 7% of people who are this tall.
In the study, 17% of their entire sample — which was by design 50% male/female — reported being taller than 6ft. Even if we assume that’s mostly men, it would mean that 25–30% of all men in the study were >6ft tall. Unless the study exclusively picked basketball players, it’s unlikely that this is a true representation of the height of the participants.
Even if these numbers are correct, it throws a lot of doubt onto the sampling methodology used in the study. This is perhaps not surprising, because the way that the authors got their sample was by paying a marketing firm to recruit people online. The participants were “paid a small amount for completing the survey which took about 5 minutes on average to complete”, which sounds like most online surveys. Now, there are ways to correct such surveys, but they are still notoriously problematic and often give you biased results that may be entirely useless.
If nothing else, it’s also pretty much impossible to trust self-report when it comes to COVID-19 diagnosis. Without some sort of verification (i.e. test results), the people who said “yes” to the question “have you been diagnosed with COVID-19?” could be lying, wrong, or simply misunderstand the question.
On top of this, the study had other issues. These are a bit technical, so I’ll italicize for those who just want to skip past.
The statistical analysis was potentially problematic for the sample they had. More specifically, they used a logistic model with 21 covariates in a sample including only 106 diagnosed cases of COVID-19. The results show how this is a bit of a mess, with the odds ratio for >6ft women, for example, being 9.09 (3.74–22.11). They also committed a classic epidemiological blunder, called the Table 2 Fallacy, which involves over-interpreting the results from a single statistical model as causal links. In other words, it might not be that being tall causes COVID-19, it might simply be that the statistical model wasn’t calibrated to tell us much about height anyway. It’s also questionable whether dichotomizing — splitting people into 2 groups — is the right thing to do with height at all. Height is not a dichotomous variable, so by splitting it up you are losing quite a bit of information and potentially biasing your results. Also, as with most dichotomization, it’s entirely arbitratry — would you have seen the same ratio if you’d dichotomized at 5'11"? What about 5'10"?
You can see these issues pretty easily when you look at other things that the authors found. For example, being in a trade union increased your risk of COVID-19 by more than four times. Using cash to pay doubled your risk in the UK, but halved it in the US. Owning a car reduced your risk of infection by more than 50%. None of these things themselves are causing COVID-19 — trade unions aren’t secretly spreading the disease — they are just vague correlations that sprung out of the statistical model and survey used in this research.
All of this makes it almost impossible to know if the results of this survey mean anything at all. Without any verification of the COVID-19 diagnoses, and an apparent bias in height reporting, all we can really say is that there seems to be some correlation between people saying they’ve had coronavirus and fudging their height up a few centimeters. Anything more is just speculation.
Tall People Are Probably Fine
Ultimately, this study tells us almost nothing about whether being tall is an issue for catching COVID-19, but it’s hard to see why it would be in the first place. There’s some vague idea that having your head slightly higher off the ground might expose you to more aerosols, but this is completely unsubstantiated and also a bit unlikely. If nothing else, many of the biggest super-spreading events for coronavirus have happened to people who were mostly sitting down, equalizing the differences in height.
While we may see further research in the future that shows tall people are at a higher risk — anything’s possible — it seems pretty unlikely. The scary headlines may have been startling, but at this point there really is no good reason to believe that tall people are more at risk from COVID-19 than anyone else.
Giants everywhere can breathe a sigh of relief.