Behavioral Economics
We focus on 3 concepts from behavioral economics: judgment, decision-making, and choice architecture. The Table includes concepts in each area, with examples of how each can be applied to increase cancer screening in low-income and minority populations. Many of these ideas can be easily incorporated into strategies suggested by The Guide to Community Preventive Services (Community Guide) (an evidence-based, accessible resource for health promotion and disease prevention, with specific coverage of breast, cervical, and colorectal cancer, compiled by an expert task force using systematic reviews as the basis for its recommendations), including evidence-based client- and provider-directed interventions to promote cancer screening, client reminders (eg, letters, post cards), small media (eg, brochures, newsletters), reducing structural barriers, provider assessment and feedback, one-on-one education, group education, and reducing client out-of-pocket costs.
Judgment
Judgment refers to the subjective assessment of the probability of outcomes or events. Decades of psychological and behavioral economics research has shown that judgments, particularly those in which people have limited experience or information, can be greatly influenced by seemingly irrelevant factors. Three aspects of judgment are relevant to cancer screening: availability, representativeness, and unrealistic optimism.
Availability refers to a tendency to judge the likelihood of future events, such as a cancer diagnosis, based on how easy it is to imagine them or call up similar events in memory. For example, women may be more likely to focus on their risk of breast cancer after the diagnosis of a family member or close friend because this experience is highly "available" in memory.
Representativeness refers to judgments about the probability or frequency of an event based on its resemblance to one's past experiences or assumptions. If clinicians and health educators can provide memorable messages about cancer screening and show that people similar to the target audience undergo screening, then both availability and representativeness can be used to increase the perception of risk and highlight the benefits of screening.
Unrealistic optimism occurs when individuals have unreasonably low estimates of their own susceptibility to harm or overly high estimates of their chances of success or benefit. For example, people tend to underestimate their chances of receiving a diagnosis of cancer. Personally relevant messages that provide accurate information about cancer risk and the true prevalence of screening should increase uptake of screening tests, in part, by counteracting a tendency for unrealistic optimism.
Decision-making
Decision-making examines the choices people make when faced with multiple options. The manner in which information is presented can influence decisions. Framing effects occur when the way choices are presented influences the options individuals prefer. In the classic example from early work in behavioral economics, people were presented with a scenario in which a disease is expected to kill 600 people and then given the choice of taking a course of action that would save 200 people or a course of action that had a one-third probability of saving 600 people and a two-thirds probability that no one would be saved. A majority chose the certain option of saving 200 people even though the probabilities are identical. Message framing promoting cancer screening also appears to affect the decision to be screened, particularly when messages are tailored with personal information. For example, women from underserved populations responded more favorably to letters from their primary care physicians that addressed their specific mammography beliefs, stage of change, perceived barriers, and risk factors than to a generic letter encouraging breast cancer screening. Several studies also suggest that framing the choice of colorectal screening tests should focus on the relative mortality risk reduction, sensitivity and specificity, preparation, and pain associated with each type of test.
Fairly consistent context effects also have been observed wherein people are frequently drawn to the one option in a set that dominates the others — or, if no single option dominates, the middle option between two extremes. For instance, colonoscopy often dominates the options for colorectal cancer screening among recommending physicians because of its effectiveness in detecting and preventing cancer, but compared with FOBT it may be seen as an extreme option among patients concerned about inconvenient preparation and invasiveness. This is an instance in which the context of the choices may differ depending on the perspective of an expert compared with that of a patient, which argues for shared decision-making and informed patient choice.
Health decisions also can be influenced by loss aversion, or a stronger preference for avoiding losses than acquiring gains. A consistent aversion to loss is a central finding in behavioral economics. Framing breast cancer screening messages in terms of loss aversion (eg, if a woman is not screened appropriately, she may risk loss of her health, her breasts, or her life) may have a minimal advantage over gain-framed messages. Loss-framed messages for diagnostic follow-up of abnormal cervical cancer screening are associated with heightened distress, particularly among women with attentional styles highly attuned to medical threats, but no difference in adherence compared with neutral or positive frames.
Messages about screening may also be framed to appeal to one's sense of fairness. People who perceive the health care system as unfair (eg, have high levels of perceived discrimination or medical mistrust) are generally less trusting and less adherent to recommendations for preventive health behaviors such as cancer screening, though some exceptions to this finding have been noted. Addressing issues of trust and other barriers through improved communication, cultural competence, engagement in medical decision-making, and by providing better information regarding screening options may improve screening in underserved populations. Research suggests that people will often prefer to take nothing than to take an offer they believe is unfair. In the case of colorectal cancer screening, for example, there is a perception among some patients that a colonoscopy is a higher-quality procedure than an FOBT, which suggests that low-income and minority adults, especially those who mistrust medical providers or have experienced discrimination in past health care encounters, could reject screening altogether if the only test they were offered was one they perceived to be inferior to other options. A similar concern about fairness might be attached to mammography and prostate-specific antigen screening in light of recent USPSTF recommendations and the burden of mortality among African Americans. Understanding individuals' perceptions about the fairness of requiring (or recommending against) a particular test and their preferences for screening could help insurers communicate with specific subgroups about the appropriateness of various types of cancer screening.
Screening messages can also be framed in terms of social and cultural norms. Such messaging may describe the majority of a peer group engaging in behaviors such as screening. One study in men found that descriptive norms (ie, "what others like me do") were predictive of cancer screening attitudes and intention, particularly when subjective norms (ie, "what significant others expect me to do") were low. Those who received information that screening was common reported greater intention to be screened, rated their probability of being screened as higher, and were more likely to leave their names and addresses to get more information about screening. This finding has particular relevance for underserved populations, who may be unlikely to encounter frequent messages from family and friends regarding cancer screening.
Affect, or emotional response, can have a significant influence on decisions as well. Fear of cancer diagnosis, pain, and embarrassment have been recurring themes in the literature on barriers to cancer screening, particularly among the underserved. It may be effective to frame health messages to elicit positive affect about screening or counter negative affect regarding screening expectations. For example, describing the relief of knowing more about one's health status or the sense of pride that comes with taking care of one's health may be more effective than addressing the fear of a cancer diagnosis.
Finally, people tend to discount future rewards, or focus on immediate gratification rather than longer-term benefits. Therefore, it may be effective to reframe the costs of screening as minimal and identify immediate benefits, such as peace of mind or the ability to prevent cancer. Another strategy from behavioral economics that may offset future discounting is providing incentives to be screened. This allows screening to have an immediate and tangible benefit that may offset the inconvenience or discomfort of screening.
Choice Architecture
Synthesizing these principles, scholars in behavioral economics have proposed a set of strategies to counter some of the flaws in human judgment and decision-making in the face of uncertainty and risk. They suggest that optimal systems for high-stakes decisions will have 1) proper alignment of incentives, 2) an appreciation of how individuals understand the consequences of their decisions, 3) sensible default options, 4) appropriate feedback, 5) allowance for expected errors, and 6) a clear presentation of information for making complex choices.
Choice architecture, defined as "organizing the context in which people make decisions", provides a promising means for encouraging screening. As competing choices grow in number, structuring complex choices can increase the quality of decision-making. As noted above, emphasizing aspects of colorectal cancer screening such as mortality risk reduction, specificity, and pain may help to organize options for patients. Default options exploit the tendency to accept the status quo when an option is presented as standard or prescribed. Defaults are powerful tools in promoting behaviors while respecting patient self-determination. Studies of interventions (ie, vaccination, HIV testing, organ donation) that use default options suggest that they may be a viable strategy to increase cancer screening. Substantial increases in participation across several behavioral domains are often observed by simply making the desired behavior the default option and requiring people to opt out, though the use of an opt-in strategy may be preferable for more controversial procedures such as HPV vaccination, where parents are more likely to consent when they can opt in. Providers could select 1 test and screening interval as the default and automatically schedule appointments for procedures while allowing patients to opt out.
Choice architecture also emphasizes feedback on performance. Information technologies that allow medical practices to track, remind, and provide feedback to both providers and patients are especially promising strategies for increasing cancer screening. Ideally, positive feedback should be provided when screening is completed, but prompts should also be given when screening opportunities are missed or when screening is overdue. Financial incentives to patients can also be effective in promoting screening. Incentives are among the most effective methods for increasing colorectal cancer screening in primary care. On the basis of promising findings in colorectal cancer screening and recommendations from the Community Guide regarding removal of structural barriers, incentives may be more appealing to low-income people if they could partially or fully cover costs incurred before screening (eg, transportation, time off of work, childcare) as opposed to being given in cash or noncash forms after screening. This may counteract the tendency to discount future rewards by adding proximal, immediate benefits of screening. Patients should also be assisted in understanding the consequences of not being screened and the consequences of various screening intervals and screening tests through informed decision-making that relies on both traditional presentations of risk and the principles of behavioral economics outlined above. Finally, a well-designed system should include an allowance for errors, making it easy for individuals to correct the "error" of not being screened with multiple reminders or on-demand scheduling and testing.
Future Directions for "Nudging" Underserved Populations Toward Cancer Screening
Research. Although we have noted some evidence supporting the principles of behavioral economics in cancer screening interventions, many research questions remain. There is some evidence that tailoring information to be consistent with patient characteristics and beliefs improves responsiveness to cancer screening recommendations, but fine-tuned experimental manipulations of behavioral economic factors such as availability, representativeness, affect, and social and cultural norms are needed as well. To test whether framing effects significantly alter attitudes and behaviors, randomized experiments should examine emotional reactions (eg, affect), information processing (eg, understanding), judgments and beliefs (eg, about cancer, about screening tests, about barriers), intentions (eg, to be screened), and actual screening behavior in response to various presentations of information. Research is also needed to test the speculation regarding fairness and the potential rejection of screening methods perceived as inferior.
Systems-level research should evaluate the effectiveness of organizing clinical practice to include elements of choice architecture, such as default options, feedback, incentives, and allowance for errors. In the area of incentives in particular, future research might also explore whether more invasive forms of screening require greater incentives among low-income adults, as might be expected based on research in other populations.
Policy and Practice. In light of the promising evidence of effectiveness, changes in policy and practice may not need to wait on additional evidence to be implemented. For example, leading cancer organizations and federal agencies should consider advocating a "default option" approach to scheduling cancer screening appointments (with an opportunity to opt out) that is integrated into electronic medical records, particularly for providers and systems who serve underserved populations. Incentives delivered in ways that both facilitate (eg, transportation and child care) and reward (eg, gift cards) screening behavior should also be considered. At the very least, systems and providers should incorporate the recommendations made in this article and by others to include appropriate framing in their communications with patients regarding cancer screening decisions.