A study finds that “supertasters” might be protected from the pandemic coronavirus.

This week, straight from the, that’s-just-crazy-enough-to-be-true file – a study that suggests that perception of bitter taste can predict whether you’ll catch COVID and how bad your disease will be.

And no – before you go there – this is NOT a study that says that the degree to which you lose taste AFTER you get COVID tells you how sick you’ll get.  This study, appearing in JAMA Network Open looked at people before they got the disease.

OK here’s what you need to know.

First off, there are certain genetic traits that change how you perceive certain tastes. You may have heard that some people have a mutation that makes delicious cilantro taste like horrible soap. My condolences to these poor souls who can never truly enjoy guacamole.

But that’s not the only taste-linked gene. The gene of interest in this study is called T2R38. It is a bitter taste receptor, and there are essentially two flavors of the gene, one called PAV and one called AVI.

If you inherit two copies of PAV, certain chemicals like propothiouracil taste obnoxiously bitter to you. If you inherit two copies of AVI – you don’t taste it at all. And if you have one copy of each, you’re somewhere in the middle.

What does this have to do with COVID?

Well, bitter taste receptors are special because they don’t just live on the tongue. T2R38 is expressed throughout the respiratory tract and, when it is activated, causes cells in those areas to release nitric oxide. That nitric oxide causes the cilium in your nose to beat faster – getting the offensive particles out of there.

Nitric oxide also has some in vitro activity against original SARS , for what it’s worth.

The idea is that, in general, bitter is a sign that we shouldn’t eat something. And what’s good for the tongue is good for the respiratory passages as well. These receptors may form part of the innate immune system. There is at least some biologic plausibility that dysfunction of these receptors might increase the risk of COVID.

Researchers took 1,935 people who had been exposed to COVID – they were mostly healthcare workesr - and tested their ability to taste a bitter chemical. About a quarter were found to be “supertasters” – they had a strong reaction. A quarter had no reaction. And the rest were somewhere in the middle.

Researchers ensured the participants didn’t currently have COVID (via PCR) or a history of COVID (via antibody testing).  Then they waited. Of the 1,935 patients, 266 eventually came down with COVID-19.

Exposure? Ability to taste the bitter chemical.  Outcome: COVID-19 and severity of COVID-19. Straightforward study.

And the findings were pretty stark. Just 3% of supertasters came down with COVID-19, compared to 29% of non-tasters. 9% of non-tasters were hospitalized, compared to zero supertasters. Of those who did get sick, symptom duration was 5 days among supertasters, 24 days among non-tasters.

If this is all true, it would make your ability to taste propothiouracil one of the strongest predictors of COVID-19 risk. But is it true?

A couple things to worry about here.

First, your ability to taste declines with age – this was demonstrated in the study – non-tasters tended to be a bit older.

The researchers adjusted for age and sex in their models, but there are plenty of other things that might affect your ability to taste (smoking, I’m looking at you), and also increase your risk of COVID-19. Maybe the taste receptors have nothing to do with it.

The other thing to remember is that all these folks were exposed to COVID-19. True, they had a negative PCR test at the time of taste testing – but could an infection have been brewing? Enough, perhaps, to attenuate taste a bit, but not to meet the PCR cycle threshold?

And, just to say it – some of the authors of this study apparently have a company that is trying to market a set of taste tests to predict health.

The best way to get rid of confounders would be to genotype people. 52 patients in the cohort were actually genotyped for the T2R38 receptor. The genotype matched the taste test in 49 out of 52 patients. The 3 who didn’t match, according to an e-mail I got from the lead author, were genetically supertasters but were older – so they didn’t perform as well on the taste test.

The other big missing piece? Documentation that SARS-CoV-2 interacts with T2R38 in any way. Like – just show me some protein epitope stimulates this receptor in vitro I don’t care. I scoured the literature for this but couldn’t find any evidence one way or the other – if I missed something, let me know in the comments.

Oh – by the way – I’ve tried to figure out how you can test your own T2R38 receptor since most of us don’t have propthiouracil lying around and enjoy our thyroids intact thank you very much. Some data suggests supertasters at this gene locus find cruciferous vegetables pretty bitter – so if you, like me, think Brussels sprouts are the worst, well, for this pandemic at least, that could be pretty sweet.

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