As our therapy session came to a close, I handed General Public (or “the General” as he liked to be called) a claim receipt for submission to his insurance company. The receipt listed the General’s identifying information along with the diagnosis code 300.29 (American Psychiatric Association, 2013). This code indicated that the diagnosis for this particular therapy session was specific phobia. I did not include the specifying number for his particular phobia because Ebolaphobia was so new that the authors of the ICD-10 coding system had yet to generate a corresponding specification number.
I breathed a large sigh of relief as I shut my office door behind the General. Therapy had grown incredibly difficult over the last few sessions because I too suffered from Ebolaphobia. I couldn’t help but silently feed off of the General’s obsessional content during our sessions. And as I tried to allay his fears, I felt like a cat trying to council a fellow feline on the harmless nature of water.
I reminded myself that even doctors have doctors: after all, the degree does not alleviate the idiosyncrasies of being human. So I settled myself into my chair and prepared myself for my next client, Uncle Sam, with his persistent anxiety regarding his various debts.
As of October 24, 2014, 4,877 people have lost their lives to the Ebola virus (WHO, 2014). One of these victims, Thomas Duncan, was diagnosed and subsequently passed away on American soil. Since Duncan’s passing, media coverage has steadily escalated to the point where one cannot watch the news or pick up a newspaper without being confronted by the Ebola virus.
Of the 9,937 suspected cases (some estimates double this number), 4 have been diagnosed in the United States (WHO, 2014). Thus, using conservative estimates, we see that the diagnostic burden in the United States represent 0.04% of the global diagnostic sum. Just to put this number in perspective, Eastern Equine Encephalitis (EEE), a mosquito-borne illness, affects 33% more Americans than Ebola every year, killing a third of those infected, and causing permanent brain damage in another third; only the final third survive without deficit (Centers for Disease Control, 2010). Despite these grim statistics, EEE is a rare topic for news coverage and has certainly not captured the American public’s imagination the way Ebola has.
So if America is bearing such an infinitesimally small fraction of the global disease burden, why are news outlets everywhere pummeling our collective consciousness with Ebola fears? Why is our collective perceived risk so much greater than the purely statistical risk? To answer these questions we must examine how the human mind assesses risk in general.
I wish to first acknowledge the tragedy and staggering loss of human life that has resulted from this unprecedented outbreak of the Ebola virus. Ebola is undoubtedly a deadly and terrifying pathogen. It is not my intention to trivialize the awful enormity of this illness. Instead, I hope that by examining the psychological construct of risk perception we might be freed from the inhibiting paralysis of terror.
We must act to stop the Ebola virus and halt the mounting loss of life. But action will emerge from a collective mind unclouded by fear and paranoia.
So how does the human brain evaluate risk?
I will use aerophobia, or fear of flying, as an analog for the Ebola virus in America. The current Ebola outbreak as it relates to America is simply too new and of too small a number to generate any meaningful data for discussion. However, the airline industry has decade’s worth of detailed data that we can examine. Furthermore, aerophobia is estimated to affect up to a quarter of the American population with varying degrees of severity, so chances are that it will serve as a relatable allegory for the Ebola virus (Rothbaum et al., 2000).
In the months that followed the tragic events of September 11, 2001, nearly 1.4 million travelers cancelled their upcoming flights, opting instead to drive or abandon their travel plans altogether (Floyd et al., 2004). As a result of the increased motor vehicle traffic in the subsequent months, an estimated additional 2,170 people were killed in automobile accidents (Blalock et al., 2009). The 2,170 deaths were on top of the almost 40,000 motor vehicle deaths observed annually at baseline. So I ask you, what is scarier: flying on a plane or driving in a car?
About 25% of the population reports a moderate to severe fear of flying (Rothbaum et al., 2000). Compare this figure to the estimated 5% of the population who endorse a moderate to severe level of anxiety associated with the operation of a motor vehicle (Taylor et al., 2011).
The per-flight risk of dying in a commercial airline crash between 2000 and 2007 was just 1 in 80 million (Barnett, 2009) while the per-trip risk of dying in an automobile crash, given the average American car trip of 6 miles (National Household Travel Survey, 2009), was 1 in 14.6 million (Insurance Institute for Highway Safety, 2012).
As it turns out, the old adage is actually true: you are about 5 (actually 5.7) times as likely to die in your car on the way to the airport as you are aboard an airplane. And yet, 5 times as many people fear airplanes as fear cars.
A 5-fold increase in risk may not seem like much, but consider how much more often we drive than we fly. The average American travels by air about once a year, while he or she will tend to take about 3 trips in his or her car every day (National Household Travel Survey, 2009). If we use a conservative estimate of 4 separate plane rides in one trip by air, then we find that the average American has a per year risk of 1 in 20 million of dying in a commercial plane crash. With 3 car trips per day this same individual has a per year risk of 1 in 13,351 of dying in an automobile collision.
So you are actually about 1,500 times more likely per year to die in a car crash than in to a plane crash!
Why is there such a discrepancy between the perceived risk of an activity and the actual risk? There are many answers to this question, but today we will examine a few key factors that make the human organism a poor actuarial.
Specifically, we will focus on five psychological components of risk perception: valence theory, the base rate fallacy, the regression fallacy, availability heuristics, and anchoring heuristics. For those unfamiliar, a heuristic approach to problem solving refers to the use of intuition or previous experience to make a best guess as to a particular solution. An example of heuristic problem solving would be a hockey player, such as myself, extending his experiential knowledge of swinging a hockey stick to the game of golf (with rather poor results I might add).
First, let’s discuss valence theory. Valence theory refers to the effect that emotions have on our perception of risk. For example, an anxious or fearful affective state will lead people to assess risk with a more pessimistic slant. The converse is also true: if a person is in a pleasant or happy mood, then they will tend towards a more optimistic assessment of a given risk. (Lerner & Keltner, 2000)
To understand valence theory, let’s look at an example. Researchers were able to create a study environment such that half of the participants were made to anticipate a feeling of regret following a gamble and half received no preconditioned expectation. The group that was made to feel anticipatory regret gambled a significantly smaller sum of money than the control group despite the identical odds of the game between the two groups. (Lerner & Keltner, 2000)
Valence theory also accounts for unrelated affective states. For example, researchers have found that the same pessimistic or optimistic risk assessment bias can be generated based on the weather, the emotional content of a movie, or the experience of a stressful exam.
Next up, the base rate fallacy. The base rate fallacy refers to our tendency to focus on sensational cases rather than the general base rate of an event (Bar-Hillel, 1980). For example, taking the population of the United States to be about 300 million, let’s estimate that about 0.1% will be contagious with the seasonal flu at any given time. Thus, approximately 300,000 individuals would theoretically be contagious at any given time. Now let’s place you on an airplane and make your neighbor sneeze. What are the odds that he has the flu?
The base rate fallacy will make you focus on the sneeze as a more important indicator than the actual base rate. I will allow that the sneeze does increase the odds that your neighbor has the flu, but with a base rate of 0.1% even a 10-fold increase would only get you to a likelihood of 1%.
Next, we turn to the regression fallacy. The regression fallacy refers to our tendency to perceive variations from the mean as being caused by something other than chance. An excellent example of the regression fallacy can be seen in the so-called, “Sports Illustrated jinx”.
An athlete is often featured on the cover of Sports Illustrated after a particularly outstanding performance. The jinx suggests that the cover feature is responsible for the athlete’s subsequent plummet back into athletic mediocrity. In reality, all that is really happening in this instance is that the athlete is regressing towards their mean performance. The performance that earned them the cover was a natural variation of their average performance. Rather than a fall from grace, the athlete is actually obeying the natural laws of statistics. (Gilovich, 2008)
Our next principle of human statistical error is availability heuristics. Availability heuristics describes our habit of estimating events that we can easily imagine as more likely than events that are more difficult to envision (Tversky & Kahneman, 1974). For an example, let’s imagine that a hypothetical man living in your state is killed by a lightning strike. News outlets statewide run story after story about his tragic demise. As a viewer, you are given ample material to envision this unfortunate man’s last moments in vivid detail.
Now, I ask you, what is more likely, dying from a lightning strike or being executed in the electric chair? Because of availability heuristics, you will instantly respond that being struck by lightning is far more likely. And yet, you are 40% more likely to be executed in the electric chair than die from a lightning strike (National Safety Council, 2014).
Finally we arrive at anchoring heuristics. Anchoring heuristics refers to the tendency to place too much weight on the first piece of information we receive. The classic example of this heuristic is the salesman who provides a very high sales price as the first quote. Thus, the buyer will perceive any decrease in price as a bargain because of the disproportionate weight they have assigned to the first price. (Tversky & Kahneman, 1974)
So to recap, we have discussed valence theory, the base rate fallacy, the regression fallacy, availability heuristics, and anchoring heuristics.
One final concept that I wish to examine before we utilize our new found knowledge to study the public response to Ebola involves the temporality of risk.
Human beings evolved to assess risk quickly and immediately: a friend who became violently ill after eating a particular red berry prompted us to instantly categorize these fruits as poisonous. Yet, the subtler choice between a wooly mammoth rib eye and a fern salad did not matter as much to our ancestors who possessed a life expectancy of around 30 years (Finch, 2012).
However, with the advent of technology and science, modern day humans are regularly living into their 70s. Yet we no longer shop by trekking long distances to battle a wooly mammoth. Instead we drive to our local supermarket, where the unhealthiest foods are the least expensive. As a result of our increased lifespan and the overabundance of cheap calories (among many other things), cancer and heart disease are now the leading causes of mortality.
About 1 in 7 Americans will die from heart disease and another 1 in 7 from cancer. 1 in 8,357 people will die from air and space transport. Small private planes can account for the majority of the risk from air and space transport, but for simplicity’s sake we will include them in the following metric. Taken together, cancer and heart disease are 2,388 times more likely to kill you than air travel. And yet, there are very few people with a phobia of double bacon cheeseburgers. (National Safety Council, 2014)
As we can see, effects that are measured in years are given much less weight than effects measured in hours.
Now let’s use our newly acquired understanding of human statistical error to examine the General Public’s (myself included) overestimation of Ebola risk.
First and foremost, I think we can all agree that the media coverage of the Ebola virus in America has been infused with a healthy dose of anxiety and fear. I am not suggesting that this is inherently wrong. As I acknowledged earlier, Ebola is, without question, a terrifying disease. However, this concoction of negative emotions, running 24 hours a day on every news channel has a profound emotional contagion effect. We are unable to escape the negative emotional valence and thus, perform our risk assessments with a pessimistic bias.
Additionally, the media has portrayed the Ebola cases in such vivid detail that we are unable to avoid the base rate fallacy and the influence of availability heuristics. Despite the base rate currently resting at 3 active (ill on American soil before quarantine) cases in a population of 300 million, we focus on the highly “available” cases rather than the base rate of 1 in 100 million.
The regression fallacy is responsible for the public perception that the healthcare system is incapable of containing the spread of Ebola. Rather than representing outliers, the two cases of healthcare workers infected with Ebola are being treated as the new normal. Instead of the more likely scenario in which our healthcare system studies its mistakes and learns to safely contain infected patients, we focus on the variant case errors as the new standard. We ignore the fact that many (admittedly, not all) medical teams around the globe have been fully capable of treating infected patients without becoming infected themselves.
Anchoring heuristics explains why news coverage in the days that followed the first healthcare worker’s diagnosis with Ebola had such a substantial impact on public perception and future news coverage. The news stories were terrifying and ran on a continuous loop for days. The focus on this exceptional case anchored the public’s perception of Ebola as being highly transmissible and contagious. Despite weeks of work trying to repair this misconception, officials have been unable to get the proverbial genie back in the lamp.
And finally, media coverage has reinforced the immediacy of the Ebola problem. Not only does the disease kill quickly, it also is spreading quickly. The temporal nature of Ebola rings every evolutionary alarm bell rattling around in our collective consciousness. And yet a far more insidious villain lurks forgotten in the nether regions of our mind. By this time next year about 20,000 Americans will have died from the seasonal flu, however, influenza has become so commonplace and feels so temporally distant that it is largely ignored despite this statistic (CDC, 2013).
So I shouldn’t be afraid of Ebola?
Yes and no. I am as guilty of these statistical errors as anyone else. I can feel the anxiety creeping up my spine every time I hear the three-syllable word that has become synonymous with fear. But I have made the conscious decision to turn off the TV, close the app, and not open the newspaper. I came to the conclusion that the news is more dangerous than the disease.
The right amount of anxiety spurs us into action, while too much corrodes our resolve. I think we should be afraid of Ebola in that it is estimated to infect as many as 10,000 new people each week by December in West Africa. And with a mortality rate upwards of 60%, West Africa will experience nearly 6,000 deaths per week as the New Year arrives. This is a staggering number of lives lost, and these statistics are the ones that should scare us.
But allow us to transform our fear into an empathic resolve. Let us feel the fear of our fellow man in Africa and use this experience to fuel our desire to help.
The world is currently uniting to fight this tremendous disease, and it is my hope that by clearing the fog of perceptual biases each of us might join the fight with a clear mind and a steady resolve.
American Psychiatric Association. (2013). DSM 5. American Psychiatric Association.
Bar-Hillel, M. (1980). The base-rate fallacy in probability judgments. Acta Psychologica, 44(3), 211-233.
Barnett, A. (2009). Aviation safety and security. Wiley Encyclopedia of Operations Research and Management Science.
Blalock, G., Kadiyali, V., & Simon, D. H. (2009). Driving fatalities after 9/11: a hidden cost of terrorism. Applied Economics, 41(14), 1717-1729.
Department of Transportation. (2014). Transportation Safety by the Numbers | Bureau of Transportation Statistics. Retrieved from http://www.rita.dot.gov/bts/sites/rita.dot.gov.bts/files/publications/by_the_numbers/transportation.
Centers for Disease Control. (2010). CDC – Epidemiology & Geographic Distribution – Eastern Equine Encephalitis. Retrieved from http://www.cdc.gov/easternequineencephalitis/tech/epi.html
Centers for Disease Control. (2013) Estimating Seasonal Influenza-Associated Deaths in the United States: CDC Study Confirms Variability of Flu. Retrieved from http://www.cdc.gov/flu/about/disease/us_flu-related_deaths.htm.
Finch, C. E. (2012). Evolution of the human lifespan, past, present, and future: phases in the evolution of human life expectancy in relation to the inflammatory load. Proceedings of the American philosophical society, 156(1), 9.
Floyd, M. F., Gibson, H., Pennington-Gray, L., & Thapa, B. (2004). The effect of risk perceptions on intentions to travel in the aftermath of September 11, 2001. Journal of Travel & Tourism Marketing, 15(2-3), 19-38.
Gilovich, T. (2008). How we know what isn’t so. Simon and Schuster.
Insurance Institute for Highway Safety. (2012). Fatality Facts. Retrieved from http://www.iihs.org/iihs/topics/t/general-statistics/fatalityfacts/state-by-state-overview
Lerner, J. S., & Keltner, D. (2000). Beyond valence: Toward a model of emotion-specific influences on judgement and choice. Cognition & Emotion, 14(4), 473-493.
National Household Travel Survey. (2009). Retrieved from http://nhts.ornl.gov/index.shtml.
National Safety Council. (2014). Injury Facts.
Rothbaum, B. O., Hodges, L., Smith, S., Lee, J. H., & Price, L. (2000). A controlled study of virtual reality exposure therapy for the fear of flying. Journal of consulting and Clinical Psychology, 68(6), 1020.
Savage, I. (2013). Comparing the fatality risks in United States transportation across modes and over time. Research in Transportation Economics, 43(1), 9-22.
Taylor, J. E., Alpass, F., Stephens, C., & Towers, A. (2011). Driving anxiety and fear in young older adults in New Zealand. Age and ageing, 40(1), 62-66.
Tversky, A., & Kahneman, D. (1974). Judgment under uncertainty: Heuristics and biases. science, 185(4157), 1124-1131.
WHO. (2014). WHO | Ebola response roadmap. Retrieved from http://www.who.int/csr/resources/publications/ebola/response-roadmap/en/.