https://en.wikipedia.org/wiki/Salmonellosis_in_the_United_St...
142,000 cases were reported annually several years ago. Not 1.3 million. So it is indeed an order of magnitude off.
Imagine comparing estimated flu infections vs people that go to the doctor for flu and you'd get similarly ridiculous results.
The 142000 number is from chicken, alone. The 1.3 (or 1.2 at the top of the same wiki page you cite) is for all sources.
There is a epidemiological problem here, of a) reported vs. estmated and b) the basis of both those things. But it isn't the one you thought it was.
If you can find actual reported cases in the US from all sources, that would be great to compare -- otherwise we shouldn't make the comparison at all.
So, for example, if it's the case that only 1/100 people who get salmonella have symptoms severe enough to cause them to go for treatment, then the _estimates_ of salmonella will be 2 orders of magnitude _higher_ than the _reported_ cases.
