It seems to me you picked out a really bad example to prove your point. Didn't it occur to you that flu vaccinations work, and that they prevent deaths, hospitalizations, doctor visits, and finally, not the least thing, the misery of the flu is avoided by most people who get a flu shot. How can more people getting flu shots be a bad thing?
For efficacy claim, show some double blind placebo controlled trial of the flue vaccine. Otherwise you are mearly spinning wishful tales about some imaginary efficiacy. If something works, people will flock to it without needing push and one wouldn't need "strategies" to "incentivi$e" administrators and doctors.
I guess we see what we want to see. Plug this into pubmed:
("vaccines"[MeSH Terms] OR "vaccines"[All Fields] OR "vaccine"[All Fields]) AND (Controlled Clinical Trial[ptyp] AND "humans"[MeSH Terms] AND "influenza"[All Fields])
You should get 260 papers.
Pick your strongest paper and if its paywalled, supply the pdf. For efficacy, you need randomized double blind trial with at least one replication by independent research group (rather than just manufacturer's claims; these are always fraud riddled). The trials should cover several seasons, rather than just one where by some luck they happened to hit the right combination.
In order to claim they are beneficial to take (rather than to pass), there should be net health benefit in vaccinated group (fewer deaths or any other health problems from all causes, not just some cherry picked pinhole view metrics, such as counting antibodies in such and such tissue, or some narrow health outcome,... etc). Otherwise a "therapy: consisting of "shooting intervention group with a bullet in the head" would have 100% efficiacy against any diesease, since the treated group would surely be safe from the target disease. Hence, I want to see the "net benefit" from patient perspective (that's who needs to decide whether it is good for them), not from some conveniently filtered pinhole view.
For safety, you need the same, except done long term (like 10-15 years) study. If you pick short enough observation window, nearly anything can be made to appear safe.
I would be especially interested to see study with pregnant women (they are pushing hard the flue vaccines for them), with all health effects on fetus and later infants followed up.
Edited by nightlight, 20 January 2013 - 10:40 PM.