fbpx
Edition 28 - General Edition

Taking The Next Step: Leveraging Behavioral Economics for Health Exchange Re-Enrollees

By Swathi Raman

Citation

Raman S. Taking the next step: leveraging behavioral economics for health exchange re-enrollees. Harvard Public Health Review. 2021; 28.

Taking The Next Step: Leveraging Behavioral Economics for Health Exchange Re-Enrollees

Background

With the recent close of the 2019 Open Enrollment Period, the federal health exchange marks the end of six tumultuous years – years characterized by recurrent system-wide crashes, highly politicized rhetoric, significant cutbacks in funding, and, most recently, a shorter 45-day enrollment window (Mangan, 2017). The Affordable Care Act (ACA) itself is facing judicial challenges that may threaten to topple the legislation (Goodnough & Pear, 2018). Despite these changes, the Centers for Medicare and Medicaid services (CMS) reports that over 8.4 million consumers have selected plans using the Healthcare.gov platform in the 2019 Open Enrollment period (November 1 – December 15, 2018), of which over 75%, or 6.4 million consumers, were existing consumers renewing their coverage (CMS Newsroom, 2018).

 

This population of “re-enrollees,” or consumers who have coverage from the previous year and actively select the same or a new plan, has been growing each year. In spite of the increasing significance of this population, current outreach efforts remain focused on reaching new health insurance consumers, rather than evaluating and optimizing decision-making for re-enrollees. These re-enrollees are part of an existing, accessible channel. Such a population is ripe for evaluation of measurable outcomes and interventions to improve these outcomes, specifically the estimated cost savings from the previous year’s insurance plan.

 

This article seeks to fill a key gap in applying behavioral economics to the health exchange experience of re-enrollees. The burgeoning field of behavioral economics has effectively applied its principles to address diverse health issues, including vaccination rates, cancer-screening outreach, and antibiotic prescribing patterns (Buttenheim & Asch, 2016; Hallsworth et al., 2016; Mehta et al., 2018). Applying these principles in the context of re-enrollment decisions on the federal health exchanges could improve coverage and savings for both consumers and the healthcare system as a whole.

 

This article will aim to (1) identify hassle factors, default bias, and anchoring, associated with re-enrollment decision-making and propose a system of smart defaults to address these cognitive biases. This article will then (2) examine source effects and processing pathways that can be influenced by the incorporation of physician recommendations and key stakeholder support, and the potential to shift social norms.

Identifying Hassle Factors, Default Bias

In behavioral economics, hassle factors are defined as barriers to optimal decision-making. In determining what plan would be ideal for re-enrollees, consumers need to understand their current and future healthcare needs. However, consumers may find difficulty in realistically assessing these needs, given constraints in money, time, resources, and available mental capacity.  Since good health is never a permanent state of affairs, evaluating one’s past healthcare utilization can prove to be a powerful hassle factor, resulting in difficulties optimizing decision-making.

 

Additionally, these factors may lead to default bias, as consumers find it easier to re-enroll in their previous year’s health insurance plan. Studies have indicated that consumers, especially lower income individuals, are more likely to stick to highly inefficient, “financially dominated” plans and rarely switch due to consumer inertia (Bhargava, Loewenstein, & Sydnor, 2015).

 

Several inefficiencies tend to arise when individuals attempt to evaluate available plans. A major bottleneck appears to be anchoring of individuals to a specific plan or part of a plan in lieu of  comprehensively evaluating a new plan’s characteristics. This likely reflects the complicated choice architecture of health exchange plans and the unavailability of tools with which a consumer may effectively compare key elements of competing plans.

 

Moreover, the framing of plans based on metal rating (gold, silver, or bronze) may create false anchors in the minds of consumers. Meanwhile, presentation of weekly versus monthly premium rates, ordering of plans based on price or rating, and general obfuscation that makes it difficult for consumers to compare “apples to apples” may lead to choice overload for the consumer (Ubel, Comerford, & Johnson, 2015).

Introducing Smart Defaults

Designing algorithms that provide added insight into consumers’ interactions with healthcare systems would allow consumers to overcome hassle factors and default bias and more accurately engage in health exchange decision making. Incorporating healthcare utilization information and answers from consumers about current health conditions, number of emergency department visits, demographics, or other relevant factors would help inform such an algorithm that highlights plans especially relevant for a given individual. This would generate the so-called “smart defaults.”

 

Financially dominated and suboptimal plans could be significantly less emphasized. Individuals could be given feedback based on the effectiveness of their choice selection and how this compares to other re-enrollees.

 

This data-driven design would not be completely revolutionary – retail giants, like Amazon, consistently utilize their recommendation engines to generate precise email or on-site guidance to inform future decision making (Amazon Web Services, 2019). Recommendations are even given based on previous purchases and the purchases of other customers who bought similar items. While there is difficulty in collecting this information, promoting data linkage with healthcare delivery and financing systems could increase the salience of plan characteristics to individuals.

Leveraging Peripheral Processing Pathways

While there are many efforts made to enhance enrollment on the exchanges, such as notable figures in politics and entertainment, consumers actually might respond better to feedback and advice from their physicians or clinical experts. This hypothesis is based on the behavioral economics concept of central and peripheral processing. Those with limited expertise are more likely to interpret information presented to them by the characteristics of the source (peripheral pathway) compared to the information’s intrinsic logic (central pathway).

 

Simply providing consumers with information does not improve re-enrollment decisions (Barnes, Hanoch, & Rice, 2015); taking advantage of source characteristics could encourage uptake of this information. Exemplifying the divide between central and peripheral processing, the same advice given by physicians appears to have greater potential to improve insurance utilization decisions (Barnes, Hanoch, & Rice, 2016). The same was observed in a National Health Service (NHS) study attempting to improve inappropriate antibiotic prescription; sending high prescribing providers a letter from the NHS medical director with readily available educational information resulted in improved practice (Hallsworth et al., 2016). Giving insurance re-enrollment consumers feedback from respected sources may have a similarly powerful effect.

Incorporating Key Stakeholders to Shift Social Norms

At a minimum, from a choice architecture point of view, simply tagging suitable plans on the exchanges as endorsed by hypothetical physicians may improve re-enrollment decisions. This could work in hand with the proposed “smart default” algorithm, enhancing the efficacy of such an intervention. Coupling the personalization of the choices available and mitigation of choice overload with “physician recommendations” for the ideal plans may further help steer consumers in the right direction, among the numerous options available.

 

The ethics and specifics of such an exchange recommendation system would have to be explored in detail. However, it does appear that coupling messages regarding re-enrollment with the actual or virtual voices of leaders in clinical practice may be a powerful and low-cost method of improving decision-making.

 

A valid concern in taking advantage of this peripheral processing pathway is that physicians may not have the expertise or time to do this work. However, incorporating social workers, health system navigators, or allied health professionals with the “endorsement” of physicians in clinical practices may be extremely effective. A system of “referrals” and “consults” for re-enrollment and insurance education similar to that used in sending patients to see physicians in other medical specialties may be effective.

 

This could both empower non-clinical members of the care team and significantly improve insurance decision making for re-enrollees. Eventually, a method such as this may create a strong social norm for improved health insurance literacy and education to complement clinical practice. This is already the case with the growing influence of incorporating social work and case management to complement the delivery of healthcare.

Conclusion

Effective policy does not act in an ideal world; its implementation must account for the real world. Passing monumental health legislation, allocating funding for subsidies, and setting up exchanges is insufficient if these initiatives are not effectively utilized by target populations. Implementation and effective uptake are critical.

 

From a policy perspective, the exchanges cannot be optimized without acknowledging flaws in human rationality. This is especially true when it comes to re-enrollment decisions. Overcoming biases preventing effective decision making and leveraging biases to promote uptake of information would go a long way towards optimizing the effects of the exchanges. Incorporating simple behavioral economics principles could maximize positive impact in a cost-effective manner, for a population that truly needs it.

References

  1. Amazon Web Services. (2019). Real-time product recommendations. Retrieved from https://aws.amazon.com/mp/scenarios/bi/recommendation/
  2. Barnes, A. J., Hanoch, Y., & Rice, T. (2015). Determinants of coverage decisions in health insurance marketplaces: Consumers’ decision-making abilities and the amount of information in their choice environment. Health Services Research, 50(1), 58-80. doi:10.1111/1475-6773.12181 [doi]
  3. Barnes, A. J., Hanoch, Y., & Rice, T. (2016). Can plan recommendations improve the coverage decisions of vulnerable populations in health insurance marketplaces? PloS One, 11(3), e0151095. doi:10.1371/journal.pone.0151095 [doi]
  4. Bhargava, S., Loewenstein, G., & Sydnor, J. (2015). Do individuals make sensible health insurance decisions? evidence from a menu with dominated options. National Bureau of Economic Research Working Paper, Retrieved from https://www.nber.org/papers/w21160
  5. Buttenheim, A. M., & Asch, D. A. (2016). Leveraging behavioral insights to promote vaccine acceptance: One year after Disneyland. JAMA Pediatrics, 170(7), 635-636. doi:10.1001/jamapediatrics.2016.0192 [doi]
  6. CMS Newsroom. (2018). Weekly enrollment snapshot: Week 7 fact sheet. Retrieved from https://www.cms.gov/newsroom/fact-sheets/weekly-enrollment-snapshot-week-7
  7. Goodnough, A., & Pear, R. (2018, Dec 14,). Texas judge strikes down Obama’s affordable care act as unconstitutional. Retrieved from https://www.nytimes.com/2018/12/14/health/obamacare-unconstitutional-texas-judge.html
  8. Hallsworth, M., Chadborn, T., Sallis, A., Sanders, M., Berry, D., Greaves, F., Davies, S. C. (2016). Provision of social norm feedback to high prescribers of antibiotics in general practice: A pragmatic national randomised controlled trial. Lancet (London, England), 387(10029), 1743-1752. doi:10.1016/S0140-6736(16)00215-4 [doi]
  9. Mangan, D. (2017, Dec 21,). Obamacare enrollment blows away expectations at nearly 9 million, despite short sign-up window. Retrieved from https://www.cnbc.com/2017/12/21/government-reveals-final-obamacare-enrollment-numbers-for-2017.html
  10. Mehta, S. J., Khan, T., Guerra, C., Reitz, C., McAuliffe, T., Volpp, K. G., . . . Doubeni, C. A. (2018). A randomized controlled trial of opt-in versus opt-out colorectal cancer screening outreach. The American Journal of Gastroenterology, doi:10.1038/s41395-018-0151-3 [doi]
  11. Ubel, P., Comerford, D., & Johnson, E. (2015). Healthcare.gov 3.0 — behavioral economics and insurance exchanges. NEJM Perspective, (372), 695-698. Retrieved from https://www.nejm.org/doi/full/10.1056/NEJMp1414771