Racial/ethnic minorities are disproportionately at risk for adverse health and financial consequences due to lower health insurance literacy compared with white enrollees.
Objectives: To measure Connecticut’s Affordable Care Act qualified health plan enrollees’ health insurance literacy (HIL) by race, ethnicity, and language preference.
Study Design: Statewide landline and cell phone telephonic survey.
Methods: Geographically balanced cohort that oversampled black and Hispanic enrollees. Questions tested enrollees’ knowledge of basic health insurance terminology and their use. Survey data were supplemented by deidentified administrative data from the state’s health insurance exchange.
Results: Overall, subjects answered 62% of 13 questions correctly. The percentages of correct answers were 53% for black enrollees, 50% for Hispanic enrollees, 74% for white enrollees, and 45% for Spanish-speaking enrollees. The differences by race, ethnicity, and language preference were statistically significant. Overall, enrollees with a college education scored higher across all demographic groups, but disparities by race and ethnicity persisted.
Conclusions: Health insurance terminology and use rules confuse consumers, especially racial and ethnic minorities. Differences in HIL may be a previously underrecognized source of healthcare disparities because even minor errors can result in delayed care or unanticipated medical bills. Low HIL can diminish the practical value of health insurance and exacerbate perceptions of health insurance as offering insufficient value for premium price. Additional research on ways to improve HIL and investments in insurance navigation support for black and Hispanic enrollees are needed.
Am J Manag Care. 2019;25(3):e71-e75Takeaway Points
Consumers can derive disparate value from identical plans based on their health insurance literacy (HIL). Our study measured HIL among qualified health plans’ enrollees. Although gaps in HIL were evident among all enrollees, racial/ethnic minorities had significantly lower HIL than white enrollees. Low HIL coupled with complex health plan rules present a barrier to care and put minorities at constant risk for unpredictable financial liabilities.
Three corrective measures are needed to enhance the value of insurance for all:
With the implementation of health insurance exchanges established by the Affordable Care Act (ACA), more than 20 million Americans gained access to health insurance. Of those, about 10 million enrolled in private insurance plans called qualified health plans (QHPs) that met certain ACA requirements—37% for the first time.1 In doing so, the newly insured entered a complex system of jargon, rules, limits, exceptions, and exclusions. Policy makers understood from the outset that health insurance literacy (HIL) would be a key factor in the long-term success of the ACA. In 2011, an HIL roundtable coined the following working definition: HIL “measures the degree to which individuals have the knowledge, ability, and confidence to find and evaluate information about health plans, select the best plan for their own (or their family’s) financial and health circumstances, and use the plan once enrolled.”2
To be of practical value, HIL requires consumers to have basic health and financial literacy and basic numeracy skills to calculate out-of-pocket expenses, decide whether they can afford to act, and know when they have been incorrectly billed, for instance. For this reason, HIL must be viewed as a unique skill, without which consumers cannot rationally choose or use health insurance or realize the full value of their policies. Several national studies have shown that consumers, even highly educated ones, have difficulty choosing and using their health insurance2-4 and that HIL levels vary widely across population groups. Racial and ethnic minorities, young adults, and those with limited English language proficiency are especially disadvantaged.5 Poor HIL is compounded by insurance product complexity.6,7 These conditions can widen existing disparities in health and well-being among minorities.
In 2013, Connecticut launched its health insurance exchange, Access Health Connecticut (AHCT). An extensive statewide consumer outreach campaign resulted in a significant reduction in the state’s uninsured rate by 2015.8 For 2 consecutive years after the first open enrollment period, AHCT surveys revealed that compared with 34% of white enrollees, 40% of black and 44% of Hispanic enrollees in QHPs had not used their insurance; additionally, compared with 19% of white enrollees, 46% of black enrollees and 52% of Hispanic enrollees did not have a primary care provider. These statistics prompted a need to measure enrollees’ HIL, a potential contributing cause of disparate utilization. This paper presents the results of the first study of HIL among QHP enrollees by race, ethnicity, and language preference in Connecticut.
The telephonic survey target was 500 enrollees stratified by county and race/ethnicity. The sample was extracted from a pool of more than 66,000 insured AHCT QHP enrollees inclusive of the 2013, 2014, and 2015 enrollment periods. Based on results of previous national studies that showed lower HIL among racial/ethnic minorities, the survey oversampled black and Hispanic enrollees. Oversampling was defined as a preponderance of people of color compared with the frequency in the general population using the 2010 Connecticut Census as a reference. Therefore, the stratified sample target was 50% white, 25% black, and 25% Hispanic (vs 70%, 10%, and 14%, respectively, in Census) from Connecticut’s 8 counties.
The survey was conducted by an independent third-party vendor by random selection from the sample pool until the stratified targets were met. Telephone outreach included landline and cell phone numbers. Interviews were conducted in English or Spanish based on enrollee preference. Participants were offered a $5 gift card for their participation. The survey was fielded from July 13, 2016, to July 29, 2016.
The 25-minute survey included 13 questions about knowledge and use of health insurance terminology. Responses were augmented with additional administrative data by AHCT. A fully deidentified data set was used for statistical analysis.
Analysis consisted of descriptive statistics, the χ2 test for categorical or ordinal variables, independent samples t tests for 2-group comparisons, univariate analysis of variance for more than 2 groups, and linear regressions for continuous outcomes, with and without categorical copredictors. Analyses were done in Stata 15 (StataCorp, LLC; College Station, Texas), and statistical significance for all tests was established at P <.05.
The study was exempted by the University of Connecticut Institutional Review Board because only deidentified data were used in the analysis.
A total of 516 participants completed the survey, and 506 answered questions about race and ethnicity. Unless otherwise stated, all results are based on the 506 complete responders, 49% of whom were white, 27% Hispanic, and 25% black, very close to the target. Thirty-nine percent were male and 96% were between the ages of 25 and 64 years. Nineteen percent answered the survey in Spanish. All but 1 of the 93 participants who chose Spanish as their preferred language identified themselves as Hispanic. A total of 85% had a household income of less than $55,000. Totals of 53% and 41% had enrolled in QHPs by telephone and online, respectively. The distribution of income by race/ethnicity showed a statistically significantly (P <.05) higher income among white enrollees.
Respondents from 8 Connecticut counties were collapsed into 4 groups because the number of responders in some counties was too low for robust statistical analysis. The sample included residents from Hartford County (27%), Fairfield County (34%), New Haven County (21%), and the other 5 Connecticut counties (18%). A total of 63% of participants had either some college or a bachelor’s degree and 37% either had completed high school or had some high school education. Detailed baseline characteristics of the surveyed population and differences by race/ethnicity are shown in Table 1.
The survey probed enrollees’ recognition of the meaning of the words premium, deductible, co-pay, out-of-pocket limit, provider network, and appeal in the context of health insurance. It also included questions probing enrollees’ understanding of how to use health insurance terminology and 1 case scenario requiring calculation using co-pay and deductible to figure out enrollee out-of-pocket cost. Black respondents had lower average HIL correct scores by 20.48 percentage points (95% CI, —24.49 to –16.47; P <.001) than white respondents (53.29% vs 73.76%, respectively), whereas Hispanic respondents had lower average HIL correct scores by 23.48 percentage points (95% CI, —27.38 to –19.59; P <.001) than white respondents (50.28% vs 73.76%, respectively).
The size of the race/ethnicity differences shrank slightly when controlling for education and income, but differences remained significant (black vs white: —14.67% [95% CI, –18.66% to –10.68%]; Hispanic vs white: –16.83% [95% CI, –20.85% to –12.81%]). The difference in the average percent correct HIL responses of Hispanic and black respondents was small and nonsignificant. Controlling for covariates did not change this conclusion. Overall, enrollees with a college education scored higher across all demographic groups, but disparities by race and ethnicity persisted (Figure).
Among the 172 enrollees with a high school education or less who answered the “premium” question, 26% (n = 45) identified the premium as “the best type of health insurance you can buy” (incorrect answer) and 8% (n = 14) chose “a bonus you get at the end of the year if you stay covered” (incorrect answer). Among those who took the survey in Spanish, 56% had the correct answer compared with 80% who took it in English (Table 2). The impact of low HIL was amplified when patients needed to combine insurance features such as deductible and co-pay. The case scenario of estimating out-of-pocket costs illustrates this point: “Suppose that under your health insurance policy, hospital expenses are subject to a $1000 deductible and a $250 per day co-pay. You get sick and are hospitalized for 4 days, and the bill comes to $6000. How much of that hospital bill will you have to pay yourself?” The correct answer is $2000. Overall, only 31% of respondents had the correct answer. Of those who underestimated their out-of-pocket cost, 11% thought that they would owe $0 and another 31% thought that the amount would be $1000. Another 17% of respondents estimated their out-of-pocket liability to be 2 or 3 times higher than the correct amount.
Our results confirm previous reports that insurance features that involve calculations, such as combining co-pay and deductible to determine out-of-pocket exposure, pose greater difficulty than concepts like “appeal” or “premium.”
Low HIL is a substantial barrier to selecting and using commercial health insurance. From the consumer perspective, having health insurance that one cannot fully understand or use with confidence can justifiably erode the perceived value of being insured. The problem is especially burdensome for people of color.
Our survey shows that consumers frequently confuse vocabulary terms, such as premium or deductible, which they know and use in everyday life without great difficulty, but that in the context of health insurance have a different meaning. For example, when asked to choose the best definition of the word premium (in the context of health insurance), 25% selected the wrong answer. Considering that the word premium has 6 potential uses as a noun, 3 as an adjective, and 2 in idiomatic expressions, choosing the wrong answer is understandable. In everyday life, the word premium most commonly denotes something of high value, and the thesaurus synonym is bonus. Not surprisingly, 20% (n = 101) respondents chose either “a bonus you get at the end of the year for staying insured” or “the best type of health insurance you can buy.” Only 5% (n = 23) responded that they did not know.
It is plausible that members anticipating an elective hospital admission who erroneously overestimate their out-of-pocket costs—as 20% of black, 17% of Hispanic, and 19% of white enrollees did—could decide to forgo or postpone needed care for fear that they would not be able to afford it. On the other hand, 50% of black and 66% of Hispanic enrollees underestimated their out-of-pocket costs. Among those whose language preference is Spanish, only 9% could calculate their out-of-pocket expense accurately. Similar case scenarios could be constructed with every wrong answer. In nearly every case, the problem was more severe for people of color or with limited English proficiency. Additional qualitative research and empirical data are needed to shed light onto these hypothetical case scenarios. There are multiple reports of the difficulty of selecting a plan, even among experts such as health economists and other highly educated consumers. Peer-reviewed articles, testimonials, and the grey literature frequently highlight the complexity of our current commercial insurance system as a barrier to the selection and use of health plans. Together, the deficits in HIL documented in this survey compose one of several root causes of the discrepancy between consumers’ expected benefits and their actual experience.
A limitation of our study is the lack of information about enrollees’ previous insurance status and year of enrollment. It is possible that new enrollees or those with the shortest exposure to insurance have greater difficulty with insurance terminology than those with previous experience. Nonetheless, our results suggest that a concerted effort to test educational approaches that improve HIL among all Connecticut residents, but most especially among minorities and those with less than a high school education, is warranted. Should HIL education prove to be effective, a broad-based educational campaign with contributions from private and public organizations could reduce the overall and disparate HIL gap over time. In the meantime, increased and better point-of-care health insurance navigation support by trained “insurance coaches” could partially overcome HIL barriers. In the best of cases, we believe that HIL education and insurance navigation support are only palliative measures rather than root-cause solutions to efficient selection and use of overly complex health insurance products. Regretfully, current policy curtailing funding for marketing, communication, and support for ACA enrollment will only exacerbate disparities.
Given the difficulty selecting and using health insurance among even highly educated consumers, a more fundamental solution is needed: a drastic simplification of plan designs, fewer but more meaningful choices, and efficient defaults. An example of simpler insurance is the Basic Health Plan deployed successfully in New York and Minnesota9 or the Connecticut Medicaid plan. These plans either exclude cost sharing or have substantially simplified cost-sharing rules.
Low HIL disadvantages racial/ethnic minorities disproportionately. A demonstrably effective HIL educational campaign, coupled with point-of-care health insurance navigation support and efforts to simplify health insurance designs, can enhance the value of health insurance for all.Author Affiliations: University of Connecticut Health Disparities Institute (VGV, BB, EC, DOS, JF), Hartford, CT.
Source of Funding: Grant from the Connecticut Health Foundation (CHF).
Author Disclosures: Dr Villagra reports board membership for CHF, from which he did not accept grant compensation for this study; grants received, none of which included compensation for this role; and meeting/conference attendance as part of CHF board activities. The remaining authors report no relationship or financial interest with any entity that would pose a conflict of interest with the subject matter of this article.
Authorship Information: Concept and design (VGV, BB, EC, JF); acquisition of data (VGV); analysis and interpretation of data (VGV, BB, EC, JF); drafting of the manuscript (VGV, DOS); critical revision of the manuscript for important intellectual content (VGV, DOS); statistical analysis (BB, EC); provision of patients or study materials (VGV); obtaining funding (VGV, JF); administrative, technical, or logistic support (DOS, JF); and supervision (VGV).
Address Correspondence to: Victor G. Villagra, MD, University of Connecticut Health Disparities Institute, 241 Main St, 5th Floor, Hartford, CT 06106. Email: firstname.lastname@example.org.REFERENCES
1. Carman KG, Eibner C, Paddock SM. Trends in health insurance enrollment, 2013-15. Health Aff (Millwood). 2015;34(6):1044-1048. doi: 10.1377/hlthaff.2015.0266.
2. Quincy L. Making health insurance cost-sharing clear to consumers: challenges in implementing health reform’s insurance disclosure requirements. Issue Brief (Commonw Fund). 2011;2:1-17.
3. Loewenstein G, Friedman JY, McGill B, et al. Consumers’ misunderstanding of health insurance. J Health Econ. 2013;32(5):850-862. doi: 10.1016/j.jhealeco.2013.04.004.
4. Paez KA, Mallery CJ. A little knowledge is a risky thing: wide gap in what people think they know about health insurance and what they actually know. American Institutes for Research website. air.org/sites/default/files/Health%20Insurance%20Literacy%20brief_Oct%202014_amended.pdf. Published October 2014. Accessed January 30, 2018.
5. Long SK, Goin D. Large racial and ethnic differences in health insurance literacy signal need for targeted education and outreach. Urban Institute website. hrms.urban.org/briefs/literacy-by-race.html. Published February 6, 2014. Accessed January 30, 2018.
6. Villagra VG. Enhancing the value of health insurance by making it simpler. Reform to Transform website. reformtotransform.files.wordpress.com/2015/11/r2t-policy-briefs-2-0-simplifying-health-insurance-final2.pdf. Published October 2015. Accessed January 30, 2018.
7. Bhargava S, Loewenstein G. Choosing a health insurance plan: complexity and consequences. JAMA. 2015;314(23):2505-2506. doi: 10.1001/jama.2015.15176.
8. Barnett JC, Vornovitsky MS. Health Insurance Coverage in the United States: 2015. Washington, DC: United States Census Bureau; 2016. census.gov/content/dam/Census/library/publications/2016/demo/p60-257.pdf. Accessed February 2, 2018.
9. Tolbert J, Antonisse A, Dorn S. Improving the affordability of coverage through the Basic Health Program in New York and Minnesota. Kaiser Family Foundation website. files.kff.org/attachment/Issue-Brief-Improving-the-Affordability-of-Coverage-through-the-Basic-Health-Program-in-Minnesota-and-New-York. Published December 8, 2016. Accessed February 2, 2018.