Going for gold: behavioral science reveals new biases in ACA exchange shopping

A new New England Journal of Medicine commentary by Peter A. Ubel, M.D., David A. Comerford, Ph.D., and Eric Johnson, Ph.D. highlights significant flaws in the way information is presented to insurance shoppers on state and federal exchange websites. The authors present original survey data to support the argument that subtle aspects of current website designs inappropriately bias decision making. The authors make their case most strongly in an analysis of the well-known gold, silver and bronze labels:

Consider the decision to lump health plans into categories with names such as bronze (for low monthly premiums and high out-of-pocket costs) and gold (for higher monthly premiums and lower out-of-pocket costs). These labels could have unintended effects on people’s attitudes toward which plans are best. After all, gold, silver, and bronze convey best, second best, and third best through association with sporting events, but the best plan for one enrollee will be different from the best plan for another.

To test whether such associations might influence people’s perceptions of insurance plans, two of us recruited a convenience sample of participants from public buses in Durham, North Carolina, and asked them which category of plans they would look at first if they were shopping for health insurance. To half the people, we described the gold plans as having higher monthly premiums and lower out-of-pocket costs — the language used by many exchanges. For the other half, we switched the gold and bronze plans, describing the gold plans as having lower monthly premiums and higher out-of-pocket costs.

(Note: For a prescient discussion of what gold, silver and bronze mean for plan financing, see Galen Benshoof’s Project Millennial post titled “On the Value of Health Insurance Plans.”) Ubel et al continued:

It shouldn’t matter whether plans are called bronze or gold. Instead, the plans’ attributes — high or low monthly premiums and high or low out-of-pocket costs — should determine people’s choices. The labels are arbitrary. Nevertheless, among participants who were below the median in mathematical ability, the majority said they preferred gold plans over bronze plans, regardless of which plan was labeled as gold.

The article, which raises additional concerns regarding the bias created through design, closes with a call for design and research communities to partner more closely in the creation and evaluation of exchange platforms. For starters, in the next open enrollment period, the authors recommend doing away with “gold, silver, bronze” labels.

The authors also suggest the addition of a new feature to the exchange websites: the ability to estimate total costs under a series of plausible scenarios. In a follow-up podcast, Ubel noted that the usefulness of the estimator could be increased if it gave both a “likely spend” for each shopper, calculated based on past utilization patterns, and some kind of a worst-case scenario figure. Additionally, he added that many shoppers are interested in knowing what providers would be “in network” if they selected one plan over another, which is not always readily available in current designs.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.