The categorical imperative is a nice heuristic to determine whether some behavior will benefit the greater good, but it's also nothing more than that. There is nothing inherently 'ethically dubious' about deviating from a rule of thumb, firstly because the heuristic returns false positives and secondly because you may disagree whether something is a 'greater good'.
If you want to argue that this deviation from this rule of thumb is 'ethically dubious', you'll have to argue exactly that and not argue some blanket claim that everyone should disagree with.
The categorical imperative is not the only or best rule, but it is a good rule of thumb. It's a good test in this case because it clearly demonstrates the folly of going unvaccinated-- if everyone relies on "herd immunity" there will be no herd immunity.
It matters what we say. The mixed messages out there already are dangerous. Kottke put it best in TFA:
So get your kids (and yourselves) vaccinated and save them & their playmates from this whooping cough bullshit, which is actually killing actual kids and not, you know, magically infecting them with autism.
As long as only a small percentage rely on "herd immunity", we're fine. If only those who actually have severe reactions to vaccines rely on "herd immunity", we're fine. Even if a few nutjobs fear autism and a few others simply don't get around to it, we're fine.
The problem is that too many people are relying on it for foolish reasons (fear of autism) and not weighing the risks appropriately. Going unvaccinated without extremely good reason, in the present circumstance, is foolish, but don't overgeneralize.
As an entire country, you are correct (for many diseases, especially ones who have vaccines in the 85% effectiveness ranges). The range of the outbreaks will largely be limited to those 15% who didn't get immunity once vaccinated, and the remainder of those who aren't vaccinated due to their ignorance.
However for small groups, they GREATLY increase the chance of the disease hitting your population. By small groups, I mean circles of friends, co-workers, schoolmates, nurserymates, etc. You should IMMEDIATELY kick people from those groups, as they increase your and your children's chances of getting infected quite a bit. (Remember, you and your children have a 15-30% of the vaccine you DID take to not work).
Remember, you're not looking at just 15-30% from one disease. They have 0% immunity to EVERY disease.
Something people overlook is that the slippery slope fallacy is only a fallacy in a logical context. People aren't uniformly logical entities, and slippery-slope arguments are not always fallacious when applied to human actions.
At some point in any analysis of human behavior and ethics, philosophy is no longer the right tool for the job. You have to turn to psychology.
There are a bunch of other actions that most people would consider perfectly ethical that fail this test too. Is going childless ethical? If everyone did that, we'd all die out. How about going to your favorite hamburger joint? If everyone did that, the line would be miles long, and nobody would get served.
The problems work in the other way too: there're some actions that people do consider unethical where there are no problems if everyone did it. Consider cheating (by a constant percentage) on your tax returns. If everyone did it, the government would just raise tax rates to compensate, everyone would end up paying the same taxes anyway, and nobody would be worse off (except perhaps having to throw an additional calculation in). Yet few people would say that cheating on your taxes is ethical.
Why is going childless considered okay, yet killing your infant considered abhorrent? They have the same effect on the world - in both case, there's no kid. They have the same effect on the world if everyone does them - in both cases, we all die out. And yet one is considered fine while the other is highly unethical.