Fox: Nearly Half of Americans Think Flu Shot Can Make You Sick

Colorized transmission electron micrograph of
Avian influenza A H5N1 viruses (seen in gold)
Image from Cynthia Goldsmith at the CDC

Is it just me or does there need to be a "[sic]" in the title of today's post? I'm pretty sure it should read "...Think the Flu Shot..." but either way, today's article comes from Fox News. It shares the findings of a study published earlier this week in the journal Vaccine about how sharing information with patients affects their willingness to get their annual flu shot.

According to Fox, researchers found that 40% of adult Americans believe that the flu vaccine can give you the flu, and that explaining to patients that this is a myth is not as effective as researchers expected it to be. The researchers polled 1,000 people about their perception of the dangers of flu shots, then they gave some people information about flu shots, some information about the flu, or no new information (control). In addition to the finding regarding belief that the flu shot could give you the flu, they found that the participants who had initially believed the flu shot to be the most dangerous were actually more convinced of their position after reading the information about the flu shot's safety. The flu kills many people every year, and the annual vaccine is the best protection we've been able to develop. So then the question is, what should a health care professional say to some one who is worried about the safety of the flu shot?

In common use, "flu" tends to mean any type of moderately severe viral infection. This is probably related to the idea of flu-like symptoms. Most people who get sick with flu-like symptoms (fever, chills, body ache, head ache, fatigue, sometimes cough, sore throat, and runny or stuffy nose) and then get better on their own in a few days declare "I had the flu!" Probably not. As I talked about Monday, flu-like symptoms are actually the symptoms of your body fighting an infection. Chills and body aches are caused by the fever, which your body is using to slow down the growth of the infection. Most infections (viral, bacterial, or any other type) that effect humans grow most efficiently at normal body temperature (37C or 98F), your body bumps up the internal thermostat to stress the infection and slow it's growth. This stresses your cells too and makes you feel lousy. The respiratory symptoms are caused by cell death in the sinuses or throat, this is common with many infections because many infections are caught by breathing in airborne pathogens, so the respiratory track is the first place they attack. Some examples of common "flues" that are not the flu include sever colds (rhinovirus), enteroviruses (including polio), adenovirus, coronavirus, rubelavirus (less common with modern vaccination against mumps and rubella), RSV (respiratory syncytial virus), and some bacterial infections. Generally there are no tests for these different viruses, and the advice you'd get from a doctor is the same: rest and chicken soup. Antibiotics don't kill viruses so there's nothing to do but let your immune system handle it.

The real flu is caused by influenza viruses. These viruses are all members of the Orthomyxoviridae family, which contains 6 genera of viruses. Because of the broadness of this grouping, influenza viruses are split into three groups based on genus: A, B, and C. Most true flues are type A, some are type B, and type C is the most rare and the least dangerous. Type A is not only the most common, but also the most dangerous.

Here we have to take a break to talk about how viruses replicate. Viruses are not exactly alive in the obvious way cells (and animals and plants) are. A common scientific definition for "alive" is organization, energy use, growth/development, reproduction, adaptation, and response to the environment. Using this definition it's obvious we are alive and so is the cat. With a microscope we can see that bacteria and other single-celled organisms are clearly alive. Viruses are tricky. They are only alive when they have infected a host cell. They can't manage any of the fundamental processes of life without the help of the cell they have infected.

A virus is made of genetic material (DNA or RNA) and a protective shell. The genetic material encodes all the instructions needed to make the virus, but without cellular machinery (such as ribosomes), the virus can't do much. When the virus lands on a cell, it can inject it's genetic material into the cell. The genetic material will then encounter the cellular machinery of the host cell. The host cell's machinery can't always tell viral instructions from legitimate cellular instructions and so begins following the viral instructions. This causes the cell to make new viral shells and new copies of the viral genetic material. These parts then self-assemble inside the host cell. Eventually the cell will wear itself out making new viruses and neglecting it's own chores. When the host cell dies the new viruses escape and are ready to infect new host cell.

This process means that each parent virus that successfully infects a cell will have hundreds of daughter viruses (why is it "daughter" instead of "son"? Probably sexism, but that's the scientific convention). This high rate of reproduction along with the error rate of the host's machinery (many of the error catching/correcting mechanisms the host has won't get used on the viruses) mean that there is a very high rate of mutation. The simplicity of viruses means that there are fewer wrong answers, and thus the viruses evolve very quickly. It also means that the more common a viral infection is (the more host cells attacked) the faster it evolves. Also, if two similar viruses co-infect a host, they can swap genetic material. This is another way viruses can evolve.

Back to the flu, each type of flu is further categorized by two of the proteins that make up the protective shell, this is why we talk about H1N1 or H5N8. As flu viruses evolve their H and N designations can change. This is important because our immune system recognizes flu viruses as something we need to destroy based on these coat proteins; if they change, our bodies will no longer recognize the virus as dangerous and ignore it until it's already made us sick. This is why you can get the flu many times in your life. The memory of your immune system is also why you can only get chicken pox once (chicken pox doesn't evolve as fast as flu, the specifics of why are beyond the scope of today's post. Short answer: different type of genetic material) and why vaccines work.

Vaccines work by exposing your body to new viruses so that your immune system can "remember" them and fight them off quickly if you are ever exposed. This will prevent you from feeling sick. Vaccines can be live or dead. If they are live, the scientists make a custom virus that has a coat protein like the dangerous virus but lacks the genetic material to be a big trouble maker. When you are vaccinated with this type of vaccine the weak viruses will attack a few cells and get the immune system's attention. Your immune system will clear the virus with no trouble, you won't ever feel more than the tiniest bit sick, and an immune memory for the coat proteins will be made. This memory will protect you if you're ever exposed to the real virus. The other main vaccine type is made of dead viruses or chopped up bits of dead virus coats. Killing a virus involves destroying it's genetic material so that it can not infect any cells. Or, scientists can make bacteria produce the coat proteins, and use purified coat proteins alone as a vaccine. In either case, adjuctvants are added to the empty virus shells to attract the immune system. When this type of vaccine is given, no cells are attacked, the adjuctvant attracts the immune system to attack the empty shells. This also results in immune memory that will protect you from a real infection. This second type of vaccine is safer for people who are immune compromised as no cells are ever infected. In a healthy person both types are safe, and some research indicates that live vaccines are a little bit better, as the natural infection process works better than tricking the immune system with chemical adjuctvants.

This brings us back to flu vaccines. We've established that vaccines work by showing your immune system what a virus looks like before you're first infection. But that will only work if the viral shell does not change too much from the virus you were vaccinated against. For slow evolving viruses, vaccination works really well. For fast evolving viruses the vaccines are less effective, as in the time between vaccine development and you're exposure to the real virus, the virus may have evolved such that your immune system won't recognize it as the same virus. You'll still get sick. We've also established that influenzas are fast evolving viruses (so is HIV). This makes vaccinating against flu difficult. The current solution is to track the virus and it's mutations and then make new vaccines every year based on the new viral shells. Each year the vaccine is made up of scientists' three best guesses based on what they saw last year (remember, these guys and gals do this for a living, their guesses are good). If you get this year's flu vaccine you will (for the rest of your life) be immune to the three strains of flu in the vaccine. However, you could still get a different flu. Or you could get one of those other flu-like viruses. Getting a flu vaccine isn't a guarantee that you won't spend a few miserable days sick under your blankie this winter, but it does improve your chances. Especially if you work in a school or hospital. And if you're immune compromised, really young, or really old, that improved chance could be worth your life. Influenza can be very dangerous in these groups, over the last 30 years the CDC estimates flu caused about 12,000 annual deaths. If you're a normal healthy adult, the life you could save could be an at-risk person you would have sneezed on. This is why hospital employees are required to be vaccinated.

In spite of the value of annual flu vaccines, vaccination rates for flu are low: only about 33%. In contrast the vaccination rate for children born in 2013 (in the midst of the current anti-vax nonsense movement) was over 90% for MMR, TDaP, Polio, Hib, Hep B, and chicken pox (CDC). A vaccination rate of 90% or better is ideal, because at that rate herd immunity can protect non-immunizable individuals (for example, people with immune deficiency or certain cancers can't be vaccinated; it won't do any good, their immune systems can't learn). With everyone protected the viruses will go extinct. We were able to do this with small pox and we're getting close with polio. So, why aren't more people getting vaccinated for the flu and what can health professionals do? This is what today's research paper seeks address.

So what did the authors of the Vaccine paper find out about adult's resistance to flu vaccination?

 The data for this study was collected via polling. A sample of 1000 adults was collected from the 2012 Cooperative Congressional Election Survey, a survey that mainly looked at political attitudes. This sample was matched and weighted to the overall US population (i.e. according to the US Census Bureau our country is 17% Latino, so 170 Latino people were polled, 14% of people are over 65 so 140 people over 65 were polled, etc). After an initial polling about the participants beliefs regarding flu and flu vaccine, the 1000 participants were randomly split into three groups. One group received information debunking the myth that people can contract flu from the vaccine (Correction group). The second group of respondents received information about the risks posed by influenza, a standard type of pro-vaccination information that was included to provide a comparison measure of messaging effectiveness (Danger group). Finally, a third group of respondents received no additional information about the flu or flu vaccines prior to answering the outcome questions (the control group).

The initial polling showed that 24% of respondents were "very concerned" (13%) or "extremely concerned" (11%) about the side effects of the flu vaccine. The researchers labeled these participants "high concern" and everyone else as "low concern". They tracked these two sub groups' responses to information independently.

After getting the participants in the danger and correction groups to do their assigned reading, all the individuals were polled again. This time researchers were interested in misperceptions about the flu vaccine, beliefs about flu vaccine safety, and intention to get vaccinated. Each of these were measured with one question. "True/False: You can get the flu from the flu vaccine", "Just based on what you know, how safe do you believe the seasonal flu vaccine, meaning the flu vaccine available every year, is generally for most people to take?", and "How likely are you to get a flu vaccination this year?"

Researchers wanted to poll the participants one last time a few months after the first round of polling to see if opinions had been changed in a lasting way by the danger or correction information. However, there was non-random drop out from the study; a particularly large number of the "high concern" participants declined to take the last poll.

First we'll look at the interaction between information and belief that the flu vaccine can give you the flu (it can't). Before being given any information 32% of participants thought that was very false, 25% somewhat false, 31% somewhat true, and 12% very true. For the control group, 38% of the low concern members of that group thought it was somewhat or very true that the vaccine could give you the flu, while 70% of the high concern members thought it was true. After being given Correction information 30% of the low concern members of that group still thought it was somewhat or very true that the vaccine could give you the flu, while 50% of the high concern members still thought it was true. And after being given the Danger information 42% of the low concern members of that group still thought it was somewhat or very true that the vaccine could give you the flu, while 70% of the high concern members still thought it was true. The Correction information does statistically significantly change people's belief that the flu vaccine causes flu (p-value < 0.05). In fact, it seems to have a larger effect on high concern participants than low concern; this seems very promising. The danger information seems to have have no effect on the high concern participants and a statistically insignificant negative effect on the low concern group.

Next we'll look at the interaction between information and the belief that the flu vaccine is safe. Before being given any information 44% of participants thought that the vaccine was very safe, 40% somewhat safe, 12% somewhat dangerous, and 4% very dangerous. For the control group, 10% of the low concern members of that group thought the flu vaccine was somewhat or very dangerous, while 50% of the high concern members thought it was dangerous. After being given Correction information, 10% of the low concern members of that group thought the flu vaccine was somewhat or very dangerous, while 35% of the high concern members thought it was dangerous. After being given Danger information, 10% of the low concern members of that group thought the flu vaccine was somewhat or very dangerous, while 40% of the high concern members thought it was dangerous. From this we see that information does not change the low concern participants opinion on the safety of the vaccine, likely because they already believe it safe. However, both types of information have a positive effect on the high concern members' beliefs, but the effect is not statistically significant. In conjunction with the change in beliefs about the flu vaccine causing the flu it seems that providing people with the Correction information is beneficial. But let's look at the last question; it's the important one.

The last question measures the interaction between information and intention to get vaccinated for flu in the upcoming season (2012-2013, so two years ago from our perspective). Before being given any information 38% of participants were very likely to get the vaccine, 10% somewhat likely, 7% slightly likely, 5% slightly unlikely, 7% somewhat unlikely, and 33% very very unlikely. This is a strongly bimodal distribution; very few people are undecided. For the control group, 55% of the low concern members of that group were likely to get the vaccine, while 46% of the high concern members were. After being given Correction information, 60% of the low concern members of that group were likely to get the vaccine, while 48% of the high concern members were. After being given Danger information, 60% of the low concern members of that group were likely to get the vaccine, while 28% of the high concern members were. Information does not statistically significantly change the likelihood of vaccination for the low concern group. For the high concern participants, only correction information had a statically significant effect, but that effect was negative. High concern participants were statistically significantly less likely to get vaccinated than the controls who saw no information! (p-value < 0.05)

Based on these findings researchers conclude that the myth-debunking Correction information did decrease the number of people who believed the vaccine was unsafe or caused the flu, but--almost paradoxically--the information made people who were initially concerned about the safety of the vaccine less likely to get vaccinated. This work is consistent with similar research done with parents decided to give their child the MMR vaccine; when parents were initially worried it caused autism, telling them it did not made them less likely to vaccinate their child even though they (apparently) accepted the information as true.

The researchers acknowledge that their use of only one survey question to address each issue of flu vaccine belief represents a weakness of the study. And that other correctional information may have been more effective than the CDC web page information they chose. They believe that further research is needed to understand this illogical connection between vaccine misconceptions and vaccine hesitancy.

What are my conclusions?

Firstly, the statistics in this paper are obnoxiously presented. Things that should be in tables are presented as graphs, and the data wasn't spelled out as well in the results text as I'd have liked to see. That being said, I don't think this represents a critical error in the findings.

It looks like the researchers have run into a quirk of human nature. The more you try to supply proof that someone's beliefs are false, the more they dig their heels in. This is why we don't discuss religion or politics in mixed company (or at least we shouldn't). As of now, no one has been able to come up with a solution for how to counter factually inaccurate ideological beliefs, which is what we are dealing with. The New Yorker has an interesting review of some of the work that's been done on this topic. But given the potential public health implications of the vaccination rate, I'm sure researchers will keep looking. Personally I've always held out hope if I can get some one to let me explain why the facts are what they are (i.e. how does a vaccine give you protection from a virus or how does evolution explain the diversity of life on our planet) and what the mechanisms behind the scenes are, people will see that the facts make sense. They aren't just pulled out of some academic's butt. Maybe then they'll believe the truth is true.

As for the news article, another good one. A little longer than many others, with a fair bit of detail. I thought it did pretty good justice to the somewhat complex and contradictory findings of the research. The article includes the name of the journal (not italicized, tsk tsk Fox editor!) and the name of the lead author. Another A for science journalism. This time for Fox News. I don't know how to feel about that. Mostly glad that the science is right.

References

Cox, R. J., K. A. Brokstad, and P. L. Ogra. "Influenza virus: immunity and vaccination strategies. Comparison of the immune response to inactivated and live, attenuated influenza vaccines." Scandinavian journal of immunology 59.1 (2004): 1-15.

Centers for Disease Control and Prevention (CDC. "Estimates of deaths associated with seasonal influenza---United States, 1976-2007." MMWR. Morbidity and mortality weekly report 59.33 (2010): 1057.

Nyhan, Brendan, and J. Reifler. "Does correcting myths about the flu vaccine work? An experimental evaluation of the effects of corrective information" Vaccine (2014)