Download Promising Approaches From Behavioral Economics to Improve

ORIGINAL ARTICLE
Promising Approaches From Behavioral
Economics to Improve Patient Lung
Cancer Screening Decisions
Andrew J. Barnes, PhD a , Lauren Groskaufmanis, MPH b, Norman B. Thomson III, MD, MBA c
Abstract
Lung cancer is a devastating disease, the deadliest form of cancer in the world and in the United States. As a consequence of CMS’s
determination to provide low-dose CT (LDCT) as a covered service for at-risk smokers, LDCT lung cancer screening is now a covered
service for many at-risk patients that first requires counseling and shared clinical decision making, including discussions of the risks and
benefits of LDCT screening. However, shared decision making fundamentally relies on the premise that with better information,
patients will arrive at rational decisions that align with their preferences and values. Evidence from the field of behavioral economics
offers many contrary viewpoints that take into account patient decision making biases and the role of the shared decision environment
that can lead to flawed choices and that are particularly relevant to lung cancer screening and treatment. This article discusses some of the
most relevant biases, and suggests incorporating such knowledge into screening and treatment guidelines and shared decision making
best practices to increase the likelihood that such efforts will produce their desired objectives to improve survival and quality of life.
Key Words: Behavioral economics, lung cancer, cancer screening, smoking
J Am Coll Radiol 2016;13:1566-1570. Copyright 2016 American College of Radiology
INTRODUCTION
Lung cancer is a devastating disease, the deadliest form of
cancer in the world and in the United States (US), expected
to account for 158,080 deaths (26.5% of all cancer deaths)
in the US in 2016, with 224,390 newly diagnosed cases this
year [1]. The 5-year survival rate remains a dismal 17.7%,
with only slight improvement over the past four decades,
despite advances in diagnosis and treatment [1]. The
annual direct costs (ie, use of resources for cancer care) of
lung cancer in 2020 are projected to be in the range of
$14.7 billion to $18.8 billion [2]. Costs due to lost time
and productivity are more difficult to model but are likely
larger than direct costs [3].
a
Department of Health Behavior and Policy, Virginia Commonwealth
University, Richmond, Virginia.
b
School of Medicine, Duke University, Durham, North Carolina.
c
Department of Radiology and Imaging, Medical College of Georgia,
Augusta University, Augusta, Georgia.
Corresponding author and reprints: Andrew J. Barnes, PhD, Department of
Health Behavior and Policy, Virginia Commonwealth University, 830 East
Main Street, 9th floor, Richmond, VA 23219; e-mail: Andrew.barnes@
vcuhealth.org.
The authors have no conflicts of interest related to the material discussed in
this article.
A sea change in lung cancer care occurred with the
release of the National Lung Screening Trial finding of a
20% reduction in lung cancer mortality among smokers
with annual low-dose CT (LDCT) screening [4]. As a result
of the United States Preventive Services Task Force’s
recommendation in favor of LDCT lung cancer screening
for smokers and the CMS’s determination to provide
LDCT as a covered service for at-risk smokers, LDCT
lung cancer screening is now a covered service for many atrisk patients. The CMS coverage determination requires a
prescreening visit with a qualified medical provider to
confirm eligibility and to engage in counseling and shared
clinical decision making to include a discussion of the risks
and benefits of LDCT screening [5]. The CMS
requirement states: “As part of the counseling and shared
decision making visit, we are requiring that, among other
things, shared decision making (including the use of one
or more decision aids) includes information on benefits,
harms, follow-up diagnostic testing, over-diagnosis, false
positive rate and total radiation exposure” [5]. This
requirement for shared decision making as a prerequisite
to a screening examination represents a new step for
CMS that reflects rising consumerism in US health care.
ª 2016 American College of Radiology
1566
1546-1440/16/$36.00
n
http://dx.doi.org/10.1016/j.jacr.2016.09.004
The decisions confronting smokers and patients with
positive LDCT screening tests are complex. The balance
of harms and benefits as patients navigate the complexities of short-term follow-up versus biopsy, risks and
benefits of various biopsy and staging procedures, and
benefits and harms of various surgical, chemotherapeutic,
and radiation treatment options is extremely complex and
difficult to model. Patient and caregiver preferences and
concerns must be included in the decision to screen, diagnose, and treat lung cancer. Provider, patient, and
caregiver biases can distort optimum outcomes.
Patients considering LDCT lung cancer screening
must learn the intricacies of prevalence, incidence, trueand false-positive findings, and true- and false-negative
findings, and appreciate that for 1,000 LDCT-screened
patients, 100 to 200 will have positive scans but only
10 to 20 will have lung cancer. Only a third of newly
diagnosed lung cancers are expected to be curable. The
question of over-diagnosis and unexpected and potentially clinically significant findings adds to the complexity
and the challenge of fully informing patients and caregivers of the pros and cons of screening and follow-up
testing. Decision support tools have been developed by
the National Comprehensive Cancer Network [6] and
the University of Michigan [7] to assist in shared
clinical decision making. These tools assist in framing
the basic concepts and manage the complex decisions
implicit in choosing to screen and in choosing
alternatives when navigating a positive screen finding.
However, such decision aids are fundamentally reliant
on the premise that with better information, patients
will arrive at rational decisions that align with their
preferences and values. Evidence from the field of
behavioral economics offers many contrary viewpoints
that take into account decision making biases on the
part of both patients and providers and the role of the
shared decision environment that can lead to flawed
choices. For guidance in better understanding the role of
biases and decision environments in shared decision
making, and opportunities to intervene and improve lung
cancer screening and treatment decisions, we turn to
behavioral economics.
BRIEF OVERVIEW OF BEHAVIORAL
ECONOMICS
Behavioral economics is a branch of economics that
challenges the fundamental assumption that humans
behave as fully informed and rational actors. Rather,
behavioral economics, as a discipline, combines the
fundamentals of economic theory with insights from
psychology about the common biases that influence decision making. Behavioral economics understands decision making as a process with predictable biases [8].
Given the predictable nature of decision making biases,
health policy can be crafted to anticipate and counteract
biases to produce socially desirable outcomes.
Health care has long been identified as a sector where
the actors—patients, providers, and payers—fail to
behave in a rational manner. As described in Irrationality
in Health Care: What Behavioral Economics Reveals about
What We Do and Why, many features of this market make
it uniquely resistant to description via traditional economic models [9]. For example, owing to insurance,
patients, the consumers of health care goods and
services, pay for only a fraction of the services they
consume with little salience to the cost of consumption
[9]. Additionally, the price of services is unclear, and
both patients and payers generally lack the ability to
compare goods and services on the basis of value and
price [9]. There is an inherent information bias in the
consumption and selection of health services; patients
must rely on providers to select the procedures and
medications that are necessary [10]. Theories based on
a rational decision maker fail to explain why people eat
poorly when they do not want to be obese, or fail to
take prescribed medication that they have gone to a
physician to obtain [10]. Given the unique aspects of
this market, individuals are particularly prone to
irrationality and bias in decision making.
Using behavioral economics to anticipate these “irrational” choices allows for the formulation of more
realistic and effective policy [11]. Once policy is no
longer predicated on a rational actor, possibilities
emerge for novel approaches to delivering care,
incentivizing health behavior, promoting evidence-based
medicine, and communicating crucial health information [11,12]. Behavioral economics holds promise as a
tool for designing the framework for health care
choices—called “choice architecture.” By taking
advantage of expected decision biases, individuals can
be guided or “nudged” toward wiser choices without
restricting choice freedom [8].
PATIENT BIASES AFFECTING MEDICAL
DECISIONS
Patients often make shared health decisions in collaboration with a health care provider. Frequently, multiple
treatment options exist for a given condition and the
Journal of the American College of Radiology
Barnes, Groskaufmanis, Thomson n Behavioral Economics and Lung Cancer Screening Decisions
1567
provider and patient must reach a consensus about which
option to pursue [13]. This is an inherently biased
process, because the provider has unique expertise
about the situation and is expected to use that expertise
to guide the patient toward optimal treatment [13].
Behavioral economics has valuable explanatory power
at the level of the individual actor. Among the individual
actors, patients arguably have the greatest impact on
health outcomes. Interactions with the health system and
health care providers contribute to only a fraction of the
overall health of an individual; lifestyle choices, which are
largely under the control of the patient actor, are a major
component of overall health. In 2010, for example, it was
estimated that 42.7% of the cancers that occurred in
Britain could be attributed to exposures owing to lifestyle
factors, such as obesity, smoking, and exercise [14].
Powerfully, behavioral economics can be used to design
choices that guide individuals toward healthier decisions
without denying them the ability to choose less healthy
alternatives [15,16].
By understanding the patient as a nonrational actor
with predictable biases, it becomes easier to understand
why individuals do not sign up to become organ donors
even though they would like to [17], or why they fail to
exercise when they would like to lose weight [18]. Policies
that anticipate patient biases can be more effective in their
efforts to promote healthier behavior. Some of the most
well-documented biases of patient actors are presented
below.
Key biases of patient actors include the following:
n
n
Loss aversion: People generally weigh losses more
heavily than equivalent gains. Thus, patients will be
likely to weigh the small cost of a preventive service
(eg, getting the LDCT screen) more heavily than the
benefits of avoiding a costly illness in the future [19].
Although patients face zero cost sharing for many
preventive services under the Affordable Care Act,
including a number of cancer screenings, the indirect
costs of transportation and missed work time, among
others, create perceived cost barriers to screening.
Evidence from behavioral economics suggests that
LDCT screening barriers could be mitigated if
incentives were designed to reduce these barriers (eg,
cover transportation costs or provide gift cards or
lotteries to increase screening rates) [20].
Present bias: Individuals overly weigh the outcomes
that occur in the present and overly discount the
potential outcomes of the future. Thus, patients may
be more likely to weigh the immediate challenges of
1568
n
n
n
n
undergoing a screening test—transportation, time off
work, discomfort—more heavily than the long-term
reduction in risk provided by the screening [21].
Framing effect: Framing is the expression of logically
equivalent information in different ways. For example,
negative lung cancer risk framing states that a patient
has an 80% chance of dying, whereas positive framing
states that the same patient has a 20% chance of survival. Patients respond more favorably to the same
information, framed in a positive manner [22].
Empathy gaps: Empathy gaps occur when individuals
make choices in a certain emotional state that they
would not make in a different emotional state. For
example, cigarette smokers who are not deprived of a
cigarette underestimate the amount that they will be
willing to pay for a cigarette after being deprived [23].
Such empathy gaps may bias patients’ lung cancer
screening and treatment preferences in a shared
decision making context. Shared decision making
could, therefore, occur with a timing gap. The timing
gap, as an evaluation tool, allows the patient to review
the information and carefully consider the benefits and
harms before acting. Reflection in an environment
other than the doctor’s office, which may be an
emotionally charged environment, could better align
patient choices with desired outcomes.
Mental accounting: Individuals value the same absolute
amount differently, depending on the reference point.
For example, a $10 discount on a $20 purchase will be
viewed as less valuable than a $10 discount on a $10
purchase [24]. In the context of presenting lung cancer
risks after LDCT screens, how the risks are framed
affects patient risk perceptions and decision making.
Framing can be a useful evaluation tool; it can
emphasize either the benefits of screening or the
harms of overtreatment [22]. Appropriate framing
guidelines, available to radiologists, could serve as a
valuable evaluation tool for patients considering
LDCT.
Decision fatigue: Choosing among large numbers of
options is overwhelming. Paradoxically, individuals
faced with more options are less likely to select options
that satisfy their needs [25]. Thus, in shared decision
making contexts, patients may quickly become cognitively overloaded when multiple treatment pathways
are proposed, increasing the likelihood that they
employ heuristics rather than systemic and analytic
decision strategies. When faced with complex decisions, or a large number of options, individuals use
mental shortcuts or “heuristics” to reach decisions.
Journal of the American College of Radiology
Volume 13 n Number 12PB n December 2016
n
This is beneficial because it allows for maximization of
efficiency, but overreliance can result in unintended
outcomes [26]. Further, when faced with many
options in a shared decision environment, patients
may be more likely to experience decision regret [27].
Default bias: If a default decision is made for an individual, even if an individual is free to go against the
default without consequences, the individual will be
more likely to select this option [28]. Thus, even in
shared decision making, patients’ ultimate choice will
be sensitive to the screening or treatment option
presented by the provider as the default. Shared
decision making for LDCT could consider taking
advantage of default bias by providers presenting a
default option that maximizes benefits and minimizes
risks. As argued by Purnell et al. [20], radiologists
could advocate to nudge patients into LDCT
screening appointments by leveraging EHR systems
to schedule them automatically, and underserved
populations that typically bear the brunt of
disparities in access to cancer care and cancer
outcomes may especially benefit from such
behaviorally informed interventions. Creating such
defaults can be used to nudge patients toward a more
socially and personally beneficial outcome. However,
any default screening or treatment option (including
no treatment) should be frequently reevaluated as
further research is produced and a more definitive
scientific consensus is reached regarding the benefits
and harms of the default versus alternative options.
LOOKING AHEAD
A crucial consideration when evaluating the role of
behavioral economics is the ultimate power of choice
architecture in crafting behavior. Proponents of libertarian paternalism argue that small interventions that
incorporate the principles of behavioral economics, also
known as “nudges,” can promote socially advantageous
behavior through choice architecture, without reductions
in freedom [8]. By taking advantage of typical biases of
decision making that patients exhibit, architects can
guide these decision makers towards choices that are in
their (or society’s) best interest, without relying on
mandates [23]. Behaviorally informed interventions may
achieve the strongest gains in LDCT screening among
underserved populations by alleviating barriers to
understanding and evaluating screening and treatment
options and accessing services. Such interventions
will also be most effective among patients with a high
degree of trust that their provider is indeed considering
the patients’ best interests in any nudges given during
screening decisions.
Nevertheless, the limitations to behaviorally informed
interventions must be considered. Many changes to the
choice architecture, including defaults or other types of
nudges, have been criticized for being overtly paternalistic. They might seem—rightly or wrongly—to manipulate patients into making lung cancer screening and
treatment decisions that go against their wishes. This
raises the ethics of intentionally using behavioral economics to “nudge” patients toward one decision over
another and must be balanced with the principles of
shared decision making put forward by medical professional societies and patient advocates. The behaviorally
informed interventions presented in this paper may strike
the right balance with ethicists when cancer screening
decisions are steered toward improving patient health
when unintentional patient decision making biases
impede their ability to adequately obtain and process the
information needed to make these decisions. Further,
despite the best of intentions, providers might use
behavioral economic tools such as defaults incorrectly,
such as by nudging people toward screening and treatment decisions that would make patients worse off.
Additionally, it is sometimes inefficient to solely take
advantage of biases to change behavior. If a health issue is
pressing or drastic enough, a combination of mandates
and strong incentives must be used. For example,
although the principles of behavioral economics should
be used to craft smoking cessation interventions, high
cigarette taxes or raising the minimum legal age to purchase will undoubtedly have a greater impact on smoking
rates than carefully crafted cessation programs [29].
Looking forward, the increased utilization of electronic health records provides a new and exciting venue
for interventions that help physicians and patients make
appropriate choices in general, and regarding LDCT in
particular. Electronic health records provide a novel
mechanism for crafting choice architecture. Similarly,
mobile health (mHealth) platforms provide opportunities
to improve health decisions by leveraging actor-specific
insights from behavioral economics. Knowledge of the
inherent biases of decision making will allow for better
design of systems that support less biased, more socially
beneficial decisions. Conversely, inattention to potential
biases could result in the creation of systems that reinforce these biases and facilitate poor choices.
Overall, behavioral economics provides a useful tool
for understanding the behavior of actors in the health
Journal of the American College of Radiology
Barnes, Groskaufmanis, Thomson n Behavioral Economics and Lung Cancer Screening Decisions
1569
system at a time of rapid change and can be a valuable
tool set to incorporate into LDCT shared decisionmaking paradigms. Incorporating knowledge about patient biases into screening and treatment guidelines and
shared decision making best practices increases the likelihood that such efforts will produce their desired objectives to improve survival and quality of life and place
patients and families at the center of their health care
decisions.
TAKE-HOME POINTS
-
-
-
-
Recent LDCT coverage policies require counseling
and shared clinical decision making, including discussions of the risks and benefits of LDCT
screening.
Shared decision making fundamentally relies on the
premise that with better information, patients will
arrive at rational decisions that align with their
preferences and values.
Evidence from the field of behavioral economics
suggests that patient decision making biases can lead
to flawed choices.
Nudges, among other behavioral economic interventions, can leverage patient biases to increase
LDCT screening.
ADDITIONAL RESOURCES
Additional resources can be found online at: http://dx.
doi.org/10.1016/j.jacr.2016.09.004.
REFERENCES
1. National Cancer Institute. Surveillance, Epidemiology, and End Results program. Available at: http://seer.cancer.gov/statfacts/html/lungb.
html. Accessed May 14, 2016.
2. Mariotto AB, Yabroff KR, Shao Y, Feuer EJ, Brown ML. Projections
of the cost of cancer care in the United States: 2010-2020. J Natl
Cancer Inst 2011;103:117-28.
3. Yabroff KR, Lund J, Kepka D, Mariotto A. Economic burden of
cancer in the United States: estimates, projections, and future research.
Cancer Epidemiol Biomarkers Prev 2011;20:2006-14.
4. National Lung Screening Trial Research Team, Aberle DR,
Adams AM, et al. Reduced lung-cancer mortality with low-dose
computed tomographic screening. N Engl J Med 2011;365:395-409.
5. Jensen TS, Chin J, Ashby L, Hermansen J, Hutter JD. Decision memo
for screening for lung cancer with low dose computed tomography
(LDCT). Centers for Medicare and Medicaid Services, Baltimore,
Maryland; 2015. CAG-00439N.
1570
6. National Comprehensive Cancer Network. Lung cancer screening.
NCCN guidelines for patients version 1.2016. Available at: https://www.
nccn.org/patients/guidelines/lung_screening/index.html#. Accessed May
16, 2016.
7. University of Michigan. Lung cancer CT screening. Should I get
screened? Available at: http://www.shouldiscreen.com/. Accessed May
16, 2016.
8. Thaler RH, Sunstein CR. Nudge: Improving decisions about health,
wealth, and happiness. New Haven: Yale University Press; 2008.
9. Hough DE. Irrationality in health care: What behavioral economics
reveals about what we do and why. Stanford, CA: Stanford University
Press; 2013.
10. Rice T. The behavioral economics of health and health care. Annu Rev
Public Health 2013;34:431-47.
11. Galizzi MM. What is really behavioral in behavioral health policy? And
does it work? Appl Econ Perspect Policy 2014;36(1):25-60.
12. Loewenstein G, Asch DA, Friedman JY, Melichar LA, Volpp KG. Can
behavioural economics make us healthier? BMJ 2012;344:e3482.
13. Bloom G, Standing H, Lloyd R. Markets, information asymmetry and
health care: Towards new social contracts. Soc Sci Med 2008;66(10):
2076-87.
14. Parkin DM, Boyd L, Walker LC. 16. The fraction of cancer attributable to lifestyle and environmental factors in the UK in 2010. Br J
Cancer 2011;105 Suppl 2:S77-81.
15. Wisdom J, Downs JS, Loewenstein G. Promoting healthy choices:
Information versus convenience. Am Econ J Appl Econ 2010;2(2):
164-78.
16. Thaler RH, Sunstein CR. Libertarian paternalism. Am Econ Rev
2003;93(2):175-9.
17. van Dalen HP, Henkens K. Comparing the effects of defaults in organ
donation systems. Soc Sci Med 2014;106:137-42.
18. Zimmerman FJ. Using behavioral economics to promote physical activity. Prev Med 2009;49(4):289-91.
19. Trivedi AN, Rakowski W, Ayanian JZ. Effect of cost sharing on
screening mammography in Medicare health plans. N Engl J Med
2008;358(4):375-83.
20. Purnell JQ, Thompson T, Kreuter MW, McBride TD. Behavioral
economics: “Nudging” underserved populations to be screened for
cancer. Prev Chronic Dis 2015;12:E06.
21. Chapman GB, Elstein AS. Valuing the future: temporal discounting of
health and money. Med Decis Making 1995;15(4):373-86.
22. Tversky A, Kahneman D. The framing of decisions and the psychology
of choice. Science 1981;211:453-8.
23. Loewenstein G. Hot-cold empathy gaps and medical decision making.
Health Psychol 2005;24:S49-56.
24. Shampanier K, Mazar N, Ariely D. Zero as a special price: The true
value of free products. Market Sci 2007;26(6):742-57.
25. Sethi-Iyengar S, Huberman G. How Much Choice Is Too Much?
Contributions to 401(k) Retirement Plans. In: Pension design and
structure: new lessons from behavioral finance. Oxford University
Press; Oxford, England. 2004:83-95.
26. Besedes T, Deck C, Sarangi S, Shor M. Age effects and heuristics in
decision making. Rev Econ Stat 2012;94(2):580-95.
27. Iyengar SS, Lepper MR. When choice is demotivating: Can one desire
too much of a good thing? J Pers Soc Psychol 2000;79(6):995-1006.
28. Johnson EJ, Goldstein D. Medicine. Do defaults save lives? Science
2003;302:1338-9.
29. Ross H, Blecher E, Yan L, Hyland A. Do cigarette prices motivate
smokers to quit? New evidence from the ITC survey. Addiction
2011;106(3):609-19.
Journal of the American College of Radiology
Volume 13 n Number 12PB n December 2016