On Vaccines 3: Fear as Fission

Author's Note: I request you remember the ground rules for this post: I am taking it as a given that vaccines are effective and safe. If you would like to dispute this fact, then please review this obnoxiously hyperlinked sentence.

Part 1: Hear no Measles, See no Measles, Think no Measles
Part 2: Risky Feelings

In this series, I am attempting to answer the question, “What keeps people from vaccinating their children?”  We are discussing vaccination behavior as a balance between the perceived risks and benefits of the choice.  In the first post of the series we discussed how we perceive benefits.  In the previous post we discussed the cognitive mechanisms of risk assessment as related to the affect heuristic.  Here we shall continue the discussion of risk and finish by exploring its relationship to our humanity.    

Loss Aversion

            People dislike losses more than they like gains⁴.  Psychologists call this loss aversion.  Many studies have demonstrated this tendency.  When participants were asked how much they would have to win to participate in a coin toss where they would lose $50 if they called it wrongly, people set the average price around $100³.  That is, a participant had to win twice as much as she could lose to make it “worth her while.”  Loss of what you already own is much more painful than a gain is pleasurable.

Loss aversion affects more than financial gambles.  When a dad weighs the risks of vaccines, a prominent fact in his mind is that his child is healthy now.  Suppose he consents to give his son the MMR vaccine at 1 year of age, and then at 3 years he’s diagnosed with autism.  If dad convinces himself of a causal relationship between the two events, then the sense of loss he would feel from his complicity seems unbearable.  If he compares that potential loss of his child’s good health against the potential benefits of preventing vaccine-preventable diseases (VPDs), uncertain and far in the future as they are, then loss aversion can be the key factor in his decision.   

Regret

            Related to loss aversion, people prefer sins of omission (bad outcomes from inaction) compared to sins of commission (bad outcomes from action)⁴.  Consider two investors.  The first owns stock in company A but considers selling it to buying company B.  He decides not to buy, and later learns he would have gained $2,000 if he had purchased B.  The second owns stock in B and then sells it for stock in A.  He also learns he would have $2,000 had he not sold B.  Which investor regrets his decision more?  When asked this questions, 92% of respondents say the second³.  Regret is a powerful emotional manifestation of loss, and our minds feel more regret when our actions are the cause of a bad outcome. 

            Now consider again the dad who is already uncertain about vaccines and is weighing the risks.  He is the one who has to consent to the shot.  His action of consenting directly results in the vaccination of his child.  If his child develops autism and he believes vaccines were the cause, then he will profoundly regret giving his consent, as he believes he is part of the cause.  However, suppose he decides not to vaccinate.  This is the inactive option since his child is naturally unvaccinated.   If his child catches severe pneumococcal pneumonia, it is easy to imagine that he will more readily forgive himself for his inaction and experience less regret.  In the calculus of choice, regret is an important variable.

Reference Frames

Both loss aversion and regret depend on your frame of reference.  What is the default position against which losses and gains are compared? If asked who is happier, someone with $2000 who wins $500 or someone with $3000 who loses $500, the obvious answer is the first person, by a lot³.  If satisfaction only depended on money, then they should be equally satisfied as they have the same amount at the end of the transaction.  However, we see that the starting point matters; it determines whether a final state is a loss or a gain.  If a third person initially had $2000, was given $1000 briefly and then had $500 taken away, we intuitively see this person is less unhappy (and probably quite happy) compared to the second person above.  This third person, although briefly owning $3000, feels little ownership of it – her frame of reference is starting with $2000 and thus her final state is a net gain of $500.

             Applying this logic to vaccines, we can see that the above examples of loss aversion and regret depend on the reference frame of the unvaccinated child with his current state of health.  From here, perceived injury by a vaccine is viewed as a loss of health and prevention of a VPD is a gain.  Now imagine a society with a robust public health program that has swayed public opinion heavily in favor of vaccines.  Society as a whole views the state of being vaccinated as the default state, where a child’s health should be.  Now when a parent considers not vaccinating, that is an act of commission, as it is their choice (choosing not to) which caused the deviation from the reference state⁴.  Similarly, he will feel less regret if he believes his child suffers an injury from the vaccine.  He was merely doing his fatherly duty of maintaining the reference health state.  Unfortunately for many people, the vaccinated child is not their reference state and loss aversion and regret come to bear negatively on their decisions. 

Overestimating Small Probabilities

            In general, humans are terrible with statistics, especially with small probabilities.  Research consistently demonstrates that we overestimate the frequency of rare events.  One group of researchers showed that people consistently overestimate the percentage of death from rare causes⁷. This is particularly true when the events are vivid, as other researchers have shown that people judge death by tornado to be much more frequent than death due to boring, everyday asthma, even though asthma kills 20 times as many people³.  Consider the fear the 9/11 terrorist attacks caused compared to the fear people felt on that same day towards a common, yet much more dangerous, activity such as driving.   Granted, many factors contribute to the fear caused by terrorism, but among them is the fear that you will be the victim of an attack.  This fear is irrational compared to the risk we face everyday by getting into a car. 

            Similar to overestimation, we overweight the importance of small probabilities.  How much would you pay to increase your chance of wining $1000 from 0% to 5%?  What about to increase your chances from 5% to 10%? The latter bet is double the chance to win compared to the former, so would you pay twice as much? Studies demonstrate that you most likely would not; you would pay proportionally more to go from 0 to 5% than from 5 to 10%³.  There is substantial psychological benefit when a person moves from no possibility of winning to any possibility of winning, what researchers call the possibility effect.  There is proportionally less benefit when a person moves from a small chance of winning to a slightly less small chance, and people reflect this preference in the price they are willing to pay for such a gamble.

            Now adverse reactions (AR) to vaccines do happen. All of them are rare; the bad ones, such as severe allergic reactions, are exquisitely rare⁸. Note that the possibility unfortunately does exist.  Given the above discussion, we know that the undecided parent will see the AR rates as much more frequent than they truly are, particularly because they are so vivid.  For the same reason, he will also overweigh the specter of ARs in his risk calculation.

             

Availability Cascade

            We can use these observations about risk assessment to explore a powerful, culture-changing process that strongly influences our response to vaccines, the availability cascade (AC). Availability refers to the degree of familiarity we feel towards a topic.  We feel familiar towards things we’ve been more exposed to.  If a topic is more available in our minds, then we will tend to overestimate its frequency, feel like we know more about it than we actually do, and be more likely to use what we know to draw further conclusions³. 

An AC, as Kahneman defines it, is “a self-sustaining chain of events, which may start from media reports of a relatively minor event [author’s addition: or Twitter messages or Facebook statuses] and lead up to public panic and large-scale government action³”.  This is not a belly-flopping Corgi video that you laugh about at work the next day; this is a piece of news that changes your daily life.  This is a viral idea that becomes a plague. Broadly speaking, this happens when a story catches the attention of a subset of people who become concerned.  The media reports the concern as a news story itself, which reinforces the concern in more people. The cycle repeats until the issue gains so much momentum, it has to be addressed on a large scale.  I find it interesting that it’s called a cascade, which may evoke the gentle image of a waterfall in many people’s minds.  However, I don’t think image does the idea justice. I see the process as more powerful and chaotic, more akin to the violent cascade of radiation in nuclear fission.

            Let’s use MMR causing autism as an example.  In 1998, a researcher makes up a bunch of stuff, writes a paper, submits it to a respected medical journal and it gets published (and later retracted). This is, unfortunately, not too unusual an event. The news media prints an article about the paper.  A portion of the population that was already suspicious of vaccines hears about the article and starts to make a ruckus.  A pretty person with a loud voice lends her celebrity to their concerns.  The controversy between nerdy scientists with their pocket protectors and taped glasses who say vaccines are safe and the supermodel who just knows for certain that vaccines caused her child’s autism is too juicy for the news media to ignore.  So the coverage intensifies, adding undue legitimacy to the controversy.  The anti-vaccination arguments begin to bloom and spread like a mushroom cloud –“toxins!, mercury!, overwhelmed immune systems!, government cover-ups!”  With the force of a bomb, they impact parents at their deepest fear, the safety of their child.  There the radioactive ideas remain, sizzling with intense doubt.   This doubt ignites our risk assessment strategies.  The affect heuristic, loss aversion, fear of regret, overestimating and overweighting small probabilities all burn brightly in the concerned parent’s mind.  The process is repeated millions of times, in the minds of anyone who hears such arguments.  Governments, doctors and scientists respond with a slew of non-made-up studies which state that the MMR vaccine does not cause autism.  Unfortunately, the facts do not fix the fears because they do not survive the cognitive processes of risk assessment.  So the cascade continues, its explosions echoing through history, its shockwaves reverberating into the future.

The Facts do not fix the Fears

            If there is main point for healthcare providers it is this: facts are not communication.  Too often, doctors sit with patients and tell them the facts of their illness and how to treat it.  The patient returns 1 month later having done nothing the doctor suggested and we blame it on non-compliance.  Each fact a patient hears has to navigate the gauntlet of our cognitive machinery.  If we do not communicate with an appreciation for people’s loss aversion, heuristics and biases, or terribleness with probabilities, then we are not communicating with a human brain, we are speaking to ourselves.   The only way to do this is both unfortunately cliché and extremely powerful: listen.  By listening, we can identity the patient’s cognitive process and speak not only to their intellectual gaps, but also to their doubts and fears.

            If there is a main point for anybody with an opinion about vaccines it is this: we all use these same cognitive mechanisms, therefore we are all at risk of the same thinking errors.  By extension, we are all wrong about a lot of things a lot of the time. There are members of our species that came up with scientology, believe that Reptilian overlords run the government, and think the Twilight movies are actual artistic achievements, and now this species is debating the merits of vaccines.  It should surprise no one that there are people who conclude wrong things, nor should it surprise you when you are one of those people.  If we understand this, we can approach debates with humility, understanding we are not so different than our opponents. After all, it’s certain that one of your beliefs appears to someone as anti-vaccination beliefs appear to you.  Now don’t get me wrong, this is a serious issue, and we are ultimately debating over the health and safety of our kids.  But in order to convince someone why they are wrong we have to first understand why they are wrong.  We have to understand the quirky shenanigans of cognition and humbly acknowledge that no matter which side of the debate, we all have similar brains that are quite fallible and, therefore, quite human. 

References

1.         Smith PJ, Humiston SG, Marcuse EK, et al. Parental Delay or Refusal of Vaccine Doses, Childhood Vaccination Coverage at 24 Months of Age, and the Health Belief Model. Public Health Reports. 2011;126(Suppl 2):135-146.

2.         Rottenstreich Y, Hsee CK. Money, kisses, and electric shocks: On the affective psychology of risk. Psychological Science. 2001;12(3):185-190.

3.         Kahneman D. Thinking, fast and slow. Macmillan; 2011.

4.         Cappelen A, Mæstad O, Tungodden B. Demand for Childhood Vaccination – Insights from Behavioral Economics. Forum for Development Studies. 2010;37(3):349-364.

5.         Slovic P, Finucane ML, Peters E, MacGregor DG. Risk as analysis and risk as feelings: Some thoughts about affect, reason, risk, and rationality. Risk analysis. 2004;24(2):311-322.

6.         Finucane ML, Alhakami A, Slovic P, Johnson SM. The affect heuristic in judgments of risks and benefits. Journal of behavioral decision making. 2000;13(1):1-17.

7.         Lichtenstein S, Slovic P, Fischhoff B, Layman M, Combs B. Judged frequency of lethal events. Journal of experimental psychology: human learning and memory. 1978;4(6):551.

8.         Medicine Io. Adverse Effects of Vaccines: Evidence and Causality National Academies Press 2012.

On Vaccines 2: Risky Feelings

Author's Note: I request you remember the ground rules for this post: I am taking it as a given that vaccines are effective and safe. If you would like to dispute this fact, then please review this obnoxiously hyperlinked sentence.

What keeps people from vaccinating their kids against potentially deadly vaccine-preventable diseases (VPDs)? In the last post we discussed how we make medical decisions.  A classical model of decision-making is that a person measures the risks and benefits and makes a choice based on which is greater. While true in a limited sense, this theory does not tell the whole story.  Last time we examined how a human brain evaluates the benefits of vaccinating.  Here we will discuss the hugely important topic of risk assessment and its role in judgment and decisions.

The Experiential and Analytical Systems

Figure 1: The Experiential and Analytical Systems.  

(Slovic et al., 2004)

            Since risk assessment is a mental process, we need to discuss the structure of our mind before we continue.  In recent decades, cognitive psychologists have described two different, parallel systems that we use to make decisions.  The first system is the Experiential system (ES)⁵ which utilizes associations, intuitions, and emotions to make decisions.  It is quick to make judgments and loathe to verify them.  It extrapolates from minimal information and integrates the data in the larger context of what is known, felt, and thought (Figure 1).  The second system is the analytical system (AS).  It is logical and methodical in its evaluation of data.  It is slower than ES and responsible for fact checking it.  It is concerned with being correct, not quick.  ES tends to be the process we use more in everyday decisions.  It can be very reliable when the patterns and associations it’s evaluating are familiar.   We see 8 kinds of peanut butter on the grocery store shelf.  ES knows we like the Jif with the red cap and striped label and quickly identifies it.  AS evaluates information more carefully.  Money may be tight, so you look at 2 different HEB coupons - one gives 20 cents off sausage flavored mayo if you buy two Jif cans the other gives two free bags of potting soil if you buy the HEB brand peanut butter – and AS decides which is the best deal. 

           

Thinking Errors

The important thing to understand about both systems is that they each have weaknesses.  To think quickly, ES employs mental shortcuts, “heuristics,” which are prone to make predictable errors in judgment, “biases” (the main subject of Kahneman’s award-winning book Thinking Fast and Slow³ which I shall reference frequently).  In the peanut butter example, you employed a pattern recognition heuristic that said, “red top and striped label = Jif”.  While most likely a reliable conclusion, the fact remains that a knock-off brand could use similar packaging.  To confirm its authenticity you have to read the label. Cheap brands might find that 1 in 10 people do not check the label and exploit this heuristic for their own gain by selling red-topped “Piff”.  We discussed heuristics in the last post while talking about benefits of vaccination.  Vividness, immediacy effect, and affective reasoning all fall in this category.  Heuristics also play an important role in risk evaluation, as we shall see below, and are worth understanding to avoid making erroneous judgments.

            AS’s weakness is that it is lazy.  It is often content to let ES draw the conclusions, especially if the conclusions “feel right,” what psychologists call “cognitive ease”³.  You end up with inferior Piff peanut butter in your basket because AS was content with the available information; the packaging, the size of the jar, the syllables and cadence of the name were consistent with Jif.  Had it been so inclined, it could have made you slow down to read the label to check its accuracy.  It could have gone on to evaluate whether cheaper Piff was a more intelligent purchase than more expensive Jif and avoided the disappointment you will feel at home when your sandwich tastes like cardboard.  In this instance, its contentment with its cognitive ease superseded its desire to be correct. 

            Below we will talk about how ES and AS evaluate risk.  We will talk about ES more because I believe its activity impacts everyday risk evaluation more than AS.  Plus, we are assuming that vaccinations are proven safe and effective.  This is an AS conclusion based on evidence and logic; a conclusion outside the scope of this post (see the linked up disclaimer above.)

Affect Heuristic

            We discussed the affect heuristic in the last post, except I called it “affective reasoning” because we didn’t have the language of heuristics.  Affect, our feelings toward a topic, dramatically alters perceptions of risks and benefits.  Paul Slovic, one of the foremost researchers in this area, demonstrated an inverse relationship between our perceptions of risks and benefits based on our affect: positive feelings magnify benefits and diminish risk while negative feelings do the opposite⁵.  Researchers demonstrated this by measuring people’s response to information about either the risks or benefits of nuclear power.  If a person was convinced that the benefits were higher for nuclear power, they downplayed the risks.  If they were told the benefits were low, then the risks were perceived as higher.  The same relationship was true for risk perception – information describing nuclear power as more risky decreased the perceived benefits and information describing nuclear power as less risky increased the perceived benefits⁶.  

These findings are due to how ES integrates the information, including emotional information, pertaining to the topic “nuclear power”.  You see, ES has a habit of answering questions it is not asked, particularly if it can answer an easier question instead of a harder one³.  The question “is nuclear power safe?” could be answered by evaluating safety data of power plants in a country, disease rates near the plant, security information, etc., but this is information that is either not available to ES or requires lazy AS to compute.  If ES can instead answer the easier question, “how do I feel about nuclear power?” then it can arrive at a conclusion! Its favorite thing! And in general, you feel good about things that have many benefits and little risk and you feel bad about risky things with little benefit.  Therefore, if you feel good about nuclear power, then you will tend to think it is a great technology that does not pose a threat to society.  This relationship holds for quick judgments requiring ES (the same researchers showed this relationship increases under time pressure⁶), so if you’re reading this and thinking that you can see how it is beneficial AND risky, you’ve engaged AS in the discussion and are no longer using the affect heuristic.

      This has compelling implications for vaccine acceptance.  The negative portrayal of vaccines has gained prevalence in our culture.  Even if you love vaccines, you’ve heard the negative propaganda and thought how frightful it would be if it were true.  Vividness comes into play here.  Just as we discussed before that VPDs are not prevalent and therefore not very vivid, so autism, autoimmune disease, crying children, or whatever your vaccine injury of choice are quite prevalent and quite vivid.   Such negative images have a way of clawing into our brains and nesting there.   

Imagine a Well-Meaning Mommy who has heard the scary, negative propaganda about vaccines, but between 3 kids and 8,000 dirty diapers, has not had time to critically evaluate the information.  She comes into my clinic and I say its time for shots! At that moment in her brain, ES’s eyes pop open.  ES looks around suspiciously, thinking it’s heard something about vaccines before, and spots the TV interview with super hot whatsherface who said something about toxins and hurt children.  It also recalls the conversation with a fellow mommy proudly raging against vaccines for whatever reason. ES acknowledges these memories aren't necessarily reliable information; however, what is reliable is that these interactions make her feel uncomfortable toward vaccines.  Mom begins to feel a rising discomfort that she can’t quite put her finger on.  I happily explain to her all the great benefits of preventing VPDs, unaware that such statements sound empty to her compared to the growing dread of shoving a needle into her child and giving her toxins.  She politely declines, saying she’s just not comfortable with vaccines at this moment.  Mildly surprised, I ask what makes her uncomfortable.    With growing confidence she says she feels they’re risky and she’s not sure if her child needs them.  I sit back dejectedly, wishing she had read this blog post so we could discuss the affect heuristic. 

People’s stated reasons for vaccine refusal differ widely, but it is surprising how frequently I hear “I’m not comfortable”, as if your intuitive feeling alone is a good reason to ignore over 100 years of vaccine research.  Such is the power of the affect heuristic - positive and negative feelings not only convince us to make decisions based on little evidence, but also, once decided, create justifications for it.    

More on our cognitive relationship to risk in the next post.

References

1. Smith PJ, Humiston SG, Marcuse EK, et al. Parental Delay or Refusal of Vaccine Doses, Childhood Vaccination Coverage at 24 Months of Age, and the Health Belief Model. Public Health Reports. 2011;126(Suppl 2):135-146.
2. Rottenstreich Y, Hsee CK. Money, kisses, and electric shocks: On the affective psychology of risk. Psychological Science. 2001;12(3):185-190.
3. Kahneman D. Thinking, fast and slow. Macmillan; 2011.
4. Cappelen A, Mæstad O, Tungodden B. Demand for Childhood Vaccination – Insights from Behavioral Economics. Forum for Development Studies. 2010;37(3):349-364.
5. Slovic P, Finucane ML, Peters E, MacGregor DG. Risk as analysis and risk as feelings: Some thoughts about affect, reason, risk, and rationality. Risk analysis. 2004;24(2):311-322.
6. Finucane ML, Alhakami A, Slovic P, Johnson SM. The affect heuristic in judgments of risks and benefits. Journal of behavioral decision making. 2000;13(1):1-17.
7. Lichtenstein S, Slovic P, Fischhoff B, Layman M, Combs B. Judged frequency of lethal events. Journal of experimental psychology: human learning and memory. 1978;4(6):551.
8. Medicine Io. Adverse Effects of Vaccines: Evidence and Causality National Academies Press 2012.