If they're recruiting 10,000 people then over a year, assuming it's a perfect slice of the population, you'd expect around 65 to die. Or roughly 5 a month.If you take a test group of a thousand people, and ask them to test-taste cups of tea, it's possible that one or more of them will die in the following month.
As a precaution, you might perhaps call a halt to your tea-tasting trials until you find out the cause of death.
Does that alarm you?
I'd guess they've got to trim out the very unwell, but you'd want to include a good sample of older people as part of the tests.