There are several answers to this question. One was overheard the other day among a bunch of well educated people oriented towards science who were taking a break from their job.
Person 1: “So, how effective is the seasonal flu shot?”
Person 2: “I heard about 1%. If you get the flu shot, you’ll have a 1% difference in if you get the flu.”
Person 3: “That’s crazy. I don’t know where you are getting your data from. It can’t be 1%, but I admit I don’t know what the actual answer is, but it can’t be that.”
Persons 4 through 6: “Well, if YOU don’t know, and HE says 1%, I’m going with the 1%. Too much trouble to get a flu shot anyway.”
Person 3: “Wait, wait! That’s crazy! That makes no sense!” as persons 1, 2, 4, 5 and 6 are filing out of the break room to go back to work. “You can’t leave thinking that 1% is correct! It can’t be correct!!!”
Person 5: [Over her shoulder on the way out the door.] “Hey, why don’t you go ask your husband. He’s a Scienceblogger, right?”
[This is a repost in honor of Get Your Flu Vaccination Week. Which is now. Did you get your vaccination yet?]
Another answer is found by checking with the CDC web site. They say that in persons under 65 who are otherwise not at risk, the flu vaccine reduces the chance of getting influenza by about 70 to 90%
But, if you go to one of the authoritative science blog sites like Effect Measure, you get a somewhat different story:
CDC wants us to get vaccinated for flu every year. … They want us to get vaccinated because they think the vaccine works … citing figures that the vaccine is 58% effective or 91% or effective or some other number, depending on what group is being talked about … This post is not about contradicting CDC, since I mostly agree that flu vaccination programs are sound public health. It is about clarifying some things that are glossed over when CDC talks … bla bla bla bla bla
and so on and so forth and I’m pretty sure I know less than when I started after reading that post. (… in a good way … it is an excellent post.)
It turns out that the issue is quite complicated, but I think I can explain it. But you have to be ready to change how you think about a couple of things.
First, you have to understand that at some level, a vaccine only works on people who get the flu. The vaccine does not shield you like a force field on Star Trek so the virus does not get at you. No. The virus gets at you, gets in you, and is inside your body …. so in a sense, you’ve “got” the flu, and then your immune system fights it off hopefully before you become symptomatic. Or maybe it fights it off enough that you get sick but not too sick. Or maybe it fights it off enough that you end up not spreading it to someone else. The vaccine sets things up so that your immune system is prepared to do this much more effectively than if you had not had the vaccine. (Oh, and by the way, does that person who did not get the flu from you, because you had the vaccine and thus did not pass it on, count when counting percentage of effectiveness of the vaccine?)
So you can see that it is fairly complicated. How to count the flu as an infection or the vaccine as an anti-flu agent is a matter of infection, symptom, virulence, reinfection, and so on.
There is another level of complication: There is not one flu. The so called “seasonal flu” is, in a given year, several different strains of influenza virus. The so called “flu vaccine” is a mixture of vaccines for each of several strains that are believed to be the most likely to circulate. Some of these vaccines may actually give cross immunity to other strains, or they may not. Generally not much, it would seem. Usually, the strains that are addressed by the vaccine are the very ones that are circulating, but some years, a strain or two circulates above minimal levels that was not included in the mix. A very high number for infection rates for a non-vaccine strain might be 20 or 30%, but that is rare.
Here’s the question one really wants to know the answer to: Imagine two alternative universes. In one there is seasonal flu but no vaccine, in the other there is a vaccine. Within both of those universes we pick out a large number of people matched up in both … the dopplegangers if you will … and vaccinate all of them in the second universe. Then we ask, Of these paired-up people, how many got the flu in the first, no-vaccine universe, vs. how many got the flu in the second, vaccine-capable universe?
If we picked 10,000 people to vaccinate, and in the non-vaccine universe, 1,000 of them got the flu badly enough to be as sick as a dog or worse (like, they died) and in the vaccine-capable universe, of the 1,000 dopplegangers only 200 got the flu to this level, we could say that the flu vaccine is 80% effective. I’m pretty comfortable saying that this 80% is a useful, meaningful number.
But the reality is more complex. In truth, the vaccine might be 100% effective for one strain, 50% for another, and there might be a rare strain circulating that is not addressed in the vaccine, but the prevalence of each strain vs. differential effectiveness resulted in the observed apparent overall effectiveness of the strategy of vaccination of 80%. In truth, 78 of those people who did not get the flu didn’t get it not because the vaccine worked but because they were never exposed to the flu in the vaccine capable universe because the person who was going to give them the flu did not get it themselves. In truth of those 800 people who “didn’t get” the flu, 123 of them did get it, and did get sick, but not too sick and they didn’t think it was the flu even though it was. In truth, 18 people in the first universe were counted as having the flu but it was a false positive. And so on.
In conclusion, no, it is not 1%. It is closer to 70 to 90% just like the CDC says, on average, for people not otherwise at risk who are under 65.
If you don’t get vaccinated because you think vaccines are dangerous or you think you can avoid the flu by taking vitamins then you are a moron. If you do’t get vaccinated because you think the flu vaccine is only 1% effective then you are … an entirely different kind of moron. Either way, get a shot and don’t be a moron. The person you give the flu to because you didn’t get the shot may be one of those that dies. Wouldn’t that be smart. Not.
UPDATED (in relation to a comment posted below): Effectiveness of the shot varies across various demographic and health related variables such as age. It will have less of an effect in very young children simply because vaccines rely on a person’s immune system, and young children have less effective immune systems. One study showed that kids between 6 months 5 years had half the incidents of flu if they were vaccinated. The risks of the flu to children under 2 are much higher than for others, so even if the vaccine is less effective for them, it is more essential that they get it. The only group for which the vaccine is not recommended is age 0 to 6 months. Kids that age don’t have enough of an adaptive immune system for a vaccine to work. If you have an infant under 6 months old, you need to avoid exposure to the flu.