Moving Beyond Blind Men and Elephants: Providing Estimated Total Annual Costs Improves Health Insurance Decision-making

Wednesday, June 15, 2016: 12:00 PM
G60 (Huntsman Hall)

Author(s): Andrew Barnes; Thomas Rice; Yaniv Hanoch; Sharon K. Long

Discussant: Karen Stockley

Objective:  Selecting from among a menu of health insurance plans is challenging for many consumers in large part because real prices (i.e. total estimated annual spending) are not transparent. This study examines the effect of providing such prices on health insurance decision-making using a series of hypothetical choice experiments on a national sample.

Study Design: All participants were given two prompts presented in random order asking which of three insurance plans would best meet their needs if they expected to require health care over the next year that would lead to $1,000 in hospital bills in one scenario and $25,000 in hospital bills in another scenario.  Half of participants were randomly assigned to a condition where total estimated annual costs were provided for each plan in addition to annual premiums, deductibles, out-of-pocket maximums, and hospital co-insurance rates. Half did not receive the total cost information.  After making an insurance choice, participants were asked which plan features were most important in their decision.

Sample: 7,648 individuals ages 18-64 responding to the fall 2015 Health Reform Monitoring Survey (HRMS), an internet-based survey tracking implementation of the Affordable Care Act.

Results: Among those not receiving total estimated annual cost information, in the $1,000 hospital bill scenario 33.4% of participants chose the cost-minimizing plan whereas 42.8% chose the cost-minimizing plan in the $25,000 hospital bill scenario. In both scenarios, individuals assigned to the total estimated cost condition were more likely to choose the cost minimizing plan after adjustment for age, race/ethnicity, gender, education, family income, health status, and insurance status (10.6 and 3.0 percentage points respectively, p<0.01 each).  In the $1,000 scenario, participants provided total estimated annual costs were more likely to report expected out-of-pocket spending was most important in their plan selection (6.9 percentage points, p<0.01) and less likely to report annual premiums were most important (-8.0 percentage points, p<0.01) compared to controls. Participants who were younger, non-White, had less education, lower family incomes, those reporting difficulty affording medical bills recently, and those with less confidence in health insurance terminology benefited the most from the provision of total annual cost information (p<0.05 each).

Conclusions:  With many consumers in insurance markets choosing plans that offer them insufficient coverage given their expected health care needs and failing to switch out of poorly chosen plans, total cost calculators have the potential to improve the quality of health insurance choices, particularly among vulnerable populations.