Practical Implications of Behavioral Economics for Dermatology Practice

Patient behavior is a critical issue that affects whether patients develop skin disease (for example, the ultraviolet [UV] exposure behavior that causes skin cancer) and whether patients with skin disease improve or recover with treatment (as their adherence to treatment determines their outcomes). Addressing the behavioral choices that patients make is an important component of dermatologic practice. The evolving area of behavioral economics that involves the psychology of the choices that patients make, and which was the basis for recent Nobel Prizes in Economics, has important practical implications for dermatologic practice.

Standard, traditional economic thinking modeled humans as completely rational utility maximizers. Experimental behavioral psychologists who have examined the choices that real people make have found that people's behavior diverges from this model in a long list of critically important ways (Table 1). Understanding these 'predictably irrational' characteristics of human thinking has implications for nearly all aspects of dermatology practice.

Table 1. Common Cognitive Biases That Lead People to Behave Differently from a Rational Utility-maximizing Model.

Consider the issue of loss aversion. People are bothered much more by losses than they value equivalent gains. Presenting information as a loss has the potential to be more meaningful than presenting the same information as a gain. Patients would probably be much less excited about a drug that would help them live a year longer than one that would prevent them from dying a year earlier. So, rather than explaining to patients that using sunscreen has a benefit in enabling them to look better for longer, we might choose to frame the use of sunscreen as helping to prevent the loss of their healthful, youthful-looking complexion.

Salience, the quality of standing out relative to other items, also has a major impact on people's thinking. People have a far greater fear of a terrorist attack on an airplane than they do of dying in a car accident, even though the latter is a far more common form of death. The terrorist attack stands out much more in the mind; the more graphic the image is, the greater the impact. We can put this to use in encouraging better sun protection by changing how we discuss skin cancer. Telling patients that sunscreen can prevent skin cancer does not have the same salience as telling patients that "sunscreen can help prevent you from developing a golf-ball-sized, ulcerating, odious skin cancer of your nose that requires removal of your nose, leaving a gaping hole in the center of your face, but don't worry, you would be able to wear a nice latex nasal prosthesis to cover the hole". The description of the specific possibility of a golf-ball-sized nasal tumor can have greater impact than the description of skin cancer in general, even though skin cancer in general is far more likely and includes the special case of the golf-ball-sized tumor.

Anchoring refers to the tendency to 'anchor' on a reference, even a random one, when making decisions. When people were asked to first think about the last two digits of their social security number and then state what they thought the value of an item was - people with higher numbers in the social security number stated higher values for the item. This can be used to affect the level of risk that patients perceive a risk to be. For example, telling patients that "the risk of this medicine is greater than the risk of a lightning strike", will appear much safer to patients than telling them that "the risk of this medication is less than a coin flip". Marketers use this same bias in getting people to purchase products, marking the price up high initially, then giving the consumer a lower price on discount. The initial high price provides an anchor that makes the discounted item appear to be of great value. Pricing cosmetic dermatology services could benefit from similar strategies.

Context has a huge effect on people's thinking. The frame of mind that people are in changes the way in which they view things and their resulting behaviors. Putting people in a religious frame of mind - even if they aren't religious - affects their likelihood of cheating. A study examined this phenomenon by administering a test to people on which they had the ability to cheat. The investigators found that by asking people to try to name some of the Ten Commandments before taking the test, cheating was nearly completely eliminated. This phenomenon might be useful for determining whether patients are taking their medication regularly; by discussing with them their religious affiliation before asking them about medication use, more accurate information on medication use can be anticipated.

People have a profound tendency to prefer and defend the ideas that they make themselves, along with a tendency to resist the ideas of others, particularly when those ideas are externally imposed. When we give medications to patients, there is a considerable likelihood that patients will be poorly adherent to the treatment regimen. It could be helpful to give strategies to patients that they can use to remind them to use the treatment, but if those strategies are pushed onto people, adherence may be paradoxically reduced. To the extent that we can, it may be helpful to elicit ideas on how patients will remember to use the treatment. If it is their idea (even if we help them come up with it), it is potentially going to be more effective.

Although standard economic theory focuses heavily on quantitative measures of value, real people are affected by the meaning of what they do. This was studied by paying research subjects to put together Lego models. The authors proposed that a sense of meaning was a hidden motivational driver of productivity in the labor-producing workforce. To study this, they recruited college students who were asked to perform the simple repetitive task of building pre-defined Lego models according to a per-unit wage schedule that declined on each successive model completion. Under one condition, the 'Meaningful' condition, subjects made multiple copies of the same model and placed each one on a desk such that the subject could observe steady accumulation of Lego models, whereas under the 'Sisyphus' condition, once subjects completed the model, the pieces were taken apart and given to them to make again. The total lack of meaning in the second condition led the students to put fewer models together and to require higher payment for the ones that they did put together. Fortunately for dermatologists, the feedback we get from patients for making their lives better gives our work tremendous meaning. And we don't just do dermatology; we are dermatologists. It is important that we share meaning with all our office staff; knowing that they, too, are contributing to making patients' lives better, which will build their morale and encourage them in their work efforts.

Counterintuitive conclusions arise with the observation that humans violate the assumptions of rational choice theory. These 'irrational' choices have important implications for dermatology practice.

Default options have extremely powerful effects on human choice behavior. In their book Nudge, Richard Thaler and Cass Sunstein describe a variety of ways in which socially desirable outcomes have been encouraged simply by making a particular choice the default. Consider retirement planning; too few employees sign up to make contributions to their companies' retirement plans. One way to address this problem is to offer employees more retirement plan choices and options on how to invest their retirement contributions; however, this approach actually reduces the rate of employee participation in retirement schemes. Companies that automatically enroll their new employees in the 401(k) retirement plan obtain much higher enrollment rates than those that require employees to 'opt in' to participate. People tend to go with the default options; the more complicated the decision, the more likely they are to just stick with the default.

Similar effects are seen in organ donation programs. Countries that use an 'opt-out' system have much higher organ donation rates than those that use an 'opt-in' system. In a similar way, by using the default option phenomenon to present what they believe is the best treatment choice as a default, physicians are likely to influence patients to choose that option. For instance, a dermatologist can say to a psoriasis patient, 'For patients like you, the usual choice is Enbrel.' If the physician goes on to describe a long list of alternative, more complicated options, patients might be even more likely to accept the default option. Presenting a long list of options tends to overwhelm the patient; too many choices increases the probability of choosing the default option. When Medicare Part D was started in Maine, 46 options were offered, and many patients were so overwhelmed by the decision that they did not sign up. Maine developed a system to automatically switch patients into the option that would provide the best coverage, leading to much better outcomes.

Negativity bias is the tendency to place much more weight on observed negative events than on statistically more common positive events. One negative story about a brand or product can outweigh much larger numbers of positive incidents. For example, a single negative story about a Tesla Model S battery catching on fire drove Tesla's stock price down by 10% in 2 days, whereas the stock price rose by only 2% when Tesla received a 5-star crash test rating based on much more data. Even when most reviews are positive, stakeholders become extremely concerned about one or two very negative reviews on websites such as DrScore.com (physician offices), Amazon (books), or Yelp (restaurants). Although patients generally rate their doctors very highly, with an average rating of 9.3 out of 10, it takes large numbers of positive ratings to undo the psychological effect of one very negative rating.

Negativity bias is a problem for patients' perceptions of medications. If a patient has heard one horror story about a medication from a friend or relative, the patient may be very unwilling to try the medication despite statistics showing the likelihood of benefit from the treatment. If the physician is aware that the patient has a strong bias against a particular medication, the physician can compensate by recommending a different medication. Even if the alternative medication is very similar, the physician may be able to create a positive impression of that medication by emphasizing its differences from the disfavored medication.

The Easterlin Paradox is the finding that, beyond a certain modest income level, additional income is not associated with increased happiness. In addition, people tend to misjudge the level of spending that will actually increase their happiness. In Happy Money, Elizabeth Dunn and Michael Norton argue that people could get more happiness out of their spending by investing in connections with others rather than spending money on oneself, and by buying more experiences rather than things. Patients who join support groups such as the National Psoriasis Foundation (NPF) not only become more knowledgeable than other patients, but also have a highly beneficial support network of fellow patients. By working together to raise money and demand political support for psoriasis-related causes, NPF members develop strong bonds and enhance their happiness. Doing good deeds together for the entire psoriasis community raises happiness much more than self-interested activities done alone.

To exploit the happiness benefits of positive outcomes, perhaps medical treatment should be presented to patients as an experience. This is already a common practice in cosmetic dermatology, and should be emulated in clinical dermatology. Presenting the expected improvement in psoriasis symptoms by stressing the activities and experiences that could result from skin clearing, rather than just changes in the lesions, is likely to give patients a better sense of the potential benefits of treatment.

On a similar note, the hedonic treadmill concept suggests that a person's happiness easily adjusts to new circumstances, whether circumstances improve or decline. People compare themselves to a reference group of similar peers, and may not feel any more successful or enduringly happy when they move up, since they then adopt more successful people as their reference group. Thus, when patients have an initial improvement in their disease, they may feel great about it. However, after some time they may lose satisfaction with that degree of improvement and may begin to feel their treatment is 'no longer working'.

Mental accounting is another cognitive bias in which people mentally place money in separate accounts. People are more likely to travel across town to get a pen for $5 instead of $15 than to get a suit for $490 instead of $500. Objectively, whether it is worth traveling across town to save $10 is not related to the total price of the item, but psychologically, the benefit of saving $10 seems much greater when it is a greater fraction of the total price. Amazingly, Thaler found that even when the buyer was purchasing both items simultaneously and paying the exact same total amount regardless of which was the discounted item, there was still a major difference in willingness to travel. Separate mental accounts may explain why some patients do not mind paying $10 for an over-the-counter medication, but may balk at paying $10 for a better prescription treatment because they think it should be fully covered by insurance. If patients are missing their phototherapy appointments, it may be helpful to determine whether they have a mentally separate 'gas money' account that gets depleted when gasoline prices rise. If so, a home light unit is probably better.

The cognitive biases that underlie 'irrational' choices are ubiquitous. With a sound knowledge of these biases, we can help address the issues of human behavior that play a large role in patients' satisfaction and treatment outcomes.