“Pick a Plan and Roll the Dice”: A qualitative study of consumer experiences selecting a health plan in the non-group market

Highlights

• •We interviewed 36 US adults with negative experiences with non-group insurance plans.
• •Participants reported facing substantial challenges to choosing an insurance plan.
• •Barriers included understanding key terms, finding information, and comparing plans.
• •Consumers of non-group insurance plans desire customized decision support.

Abstract

Background

For consumers without access to employer-sponsored or public insurance, health plan choices in the non-group (individual) insurance market that do not meet consumer needs have the potential for negative downstream implications for health and financial well-being.

Objective

This qualitative interview study sought to understand consumers’ experiences and challenges with choosing a non-group health plan, among those who later had negative experiences with the plan they chose.

Methods

We conducted semi-structured telephone interviews with a purposive sample of 36 participants from a large regional health insurance carrier in three states who enrolled in non-group plans in 2017 (21 in Affordable Care Act (ACA) Marketplace plans and 15 enrolled off-Marketplace). Participants were included if they reported negative experiences using their plan after enrollment, such as higher-than-expected medical costs. Interviews explored challenges choosing a plan; information needed for choosing; usefulness of available tools; and preferred format for interventions to improve plan choice experiences. We analyzed interview transcripts using thematic content analysis.

Results

Study participants reported experiencing substantial challenges to choosing an insurance plan. Key barriers included understanding insurance terms, finding relevant information, and making comparisons across plans. Participants valued the ability to make comparisons across carriers when using the Marketplace websites but were less satisfied with customer service. Suggestions for improvement included greater standardization of plans and language and availability of customized one-on-one assistance.

Conclusion

Findings from this study suggest that health plan selection in the non-group market presents challenges to consumers that may be addressed through enrollment assistance and improved presentation of information. Personalized assistance to find and choose coverage may lead to plan choices that better meet consumer needs and increase confidence choosing a plan in subsequent enrollment periods.

1. Introduction

Choosing a health insurance plan can be difficult [1], [2], especially for individuals and families in the non-group (individual) insurance market. People without access to insurance coverage through their employer or who are not eligible for public insurance (such as Medicare or Medicaid) need to seek insurance coverage in the non-group insurance market. Absent involvement from employers who narrow down choice options, fewer supports exist to facilitate plan choice for consumers. Shopping for health insurance requires consumers to predict their expected health care needs and estimate out-of-pocket costs across a variety of insurance carriers and plan types. These estimations are particularly challenging for those with low health insurance literacy [3], [4], [5], [6], [7]. Chronic health conditions also make choosing a plan difficult [8], as multiple dimensions of health care needs must be considered, including access to providers and prescription drug coverage.

The Affordable Care Act (ACA) included reforms to increase insurance coverage in the non-group market. This included creation of health insurance exchanges (or Marketplaces) that are state-based or federally run; states participating in the federal Marketplace provide plan options on HealthCare.gov. Consumers can shop for and purchase non-group plans on the Marketplaces and obtain income-based premium and cost-sharing subsidies. The ACA Marketplaces often include standardized plan presentation and provide access to certified enrollment assisters (brokers and navigators). Both HealthCare.gov and state-based Marketplaces use web-based platforms that allow consumers to sort and compare qualified plans from different insurance carriers in their state based on selected attributes such as premium, deductible, or metal tier (gold, silver, bronze in descending actuarial value). Consumers can also purchase unsubsidized non-group plans off-Marketplace directly from insurance carriers, where the availability of decision support and comparative tools can vary. Since the ACA was implemented, coverage in the non-group market increased to 13.8 million enrollees in 2018, (9.9 million through Marketplaces and 3.9 million off-Marketplace) [9]; non-group enrollment increased during the COVID-19 pandemic to 19.6 million, of which 14.5 million were enrolled through Marketplaces, after the American Rescue Plan Act increased subsidy amounts and eligibility [10]. This additional enrollment includes many previously uninsured individuals who may have limited experience selecting health insurance plans.

Consumers in the non-group market, particularly those eligible for Marketplace subsidies, may be a more socioeconomically and clinically vulnerable population than other commercially insured groups [11], [12]. While a goal of the ACA was to help consumers obtain affordable coverage that meets their needs, surveys demonstrate considerable challenges for consumers choosing non-group health insurance and negative experiences using their plan, including difficulty seeing a desired provider, unexpected high costs, and financial burden [8], [13], [14], [15]. However, we lack an in-depth understanding of health plan decision making and barriers for consumers choosing non-group plans post-ACA, particularly for those with negative experiences using these plans. Understanding how someone with negative plan experiences had chosen their plan can shed light on how the plan selection process could be improved to foster better downstream outcomes. Therefore, in this study we sought to explore the experiences and challenges faced by these consumers.

2. Materials and methods

2.1. Overview

This qualitative study occurred within a larger mixed-methods project to examine consumer preferences and behaviors in the non-group insurance market. We conducted a baseline survey from May to March 2017 after 2017 open enrollment (November 2016–January 2017) among subscribers aged 18–63 years in non-group insurance plans offered on and off the ACA Marketplaces in Maine, New Hampshire, and Massachusetts by a large non-profit regional insurance carrier [8]. Maine and New Hampshire participate in the federal Marketplace through HealthCare.gov, while Massachusetts uses the state-based Massachusetts Health Connector, both of which provided sortable information on plan attributes, access to certified enrollment assisters, and the ability to search for providers within plan networks. At the time of the study, HealthCare.gov offered more than 30 non-group plans in New Hampshire and Maine, and at least 10 more were available off-Marketplace [8], while the Massachusetts Health Connector streamlined its offerings [16], [17] and provided more standardization of plans within metal tiers. Participants could be enrolled as individuals or with other family members. From this population, we selected a random sample of 7,206 subscribers stratified by state and enrollment on or off-Marketplace; 2,029 participants completed the survey for an American Association for Public Opinion Research (AAPOR) response rate type 2 of 29 % [8]. The baseline survey collected self-reported sociodemographic and clinical information and assessed health insurance literacy using a 13-item measure [18] of the participant’s confidence with insurance terms and insurance-related activities such as identifying covered providers. Among the 1,859 baseline survey respondents who gave permission to be recontacted, we conducted a follow-up survey one year later from March to July 2018 after 2018 open enrollment (November 2017–January 2018). Follow-up surveys were completed by 1,223 participants; the response rate was 18 % of the population initially sampled, and 60 % of those who completed the baseline survey) [8]. The follow-up survey included questions about participants’ experiences in their 2017 insurance plan and their current coverage source for 2018. Survey data were linked with enrollment and benefits information from the insurance carrier.

2.2. Participants

Participants for the present study were eligible if they reported one or more negative experiences using their 2017 plan in the follow-up survey. We were interested in identifying participants who had a range of negative experiences in their plan over the course of the year following enrollment that could have been affected by different factors during the selection process. Negative experiences included: reporting higher-than-expected medical care costs; delaying or forgoing health care due to cost; experiencing health care-related financial burden; wishing they had help choosing their plan; or rating their overall experience using their plan as fair or poor. We excluded those who left the non-group market in 2018. Of the remaining 1068 subscribers, 741 agreed to be approached for interviews.

Among the 741 eligible, we selected a purposive sample with diversity with respect to age, chronic conditions, state, and plan characteristics. Between August 2018 and November 2018, we invited 74 survey participants to be interviewed, with the a priori goal of completing 30 interviews. We completed interviews with 36, adding 6 additional interviews to achieve thematic saturation, which occurred when the final 5 interviews did not yield substantially new information about choosing a non-group health insurance plan [19]. Participants received a $50 gift card incentive. This study was approved by the Harvard Pilgrim Health Care Institutional Review Board and follows Standards for Reporting Qualitative Research Guidelines [20].

2.3. Data collection

We developed a semi-structured interview guide that we piloted with three members with non-group plans from the insurance carrier (see Supplement). Open-ended questions explored the following domains: challenges choosing a plan; information needed for choosing a plan; usefulness of available tools and suggestions for tools to help choose a plan; and preferred format for interventions to improve plan choice experiences. The choice of domains was informed by follow-up survey findings about negative plan experiences [8] and was intended to include elements of the choosing process that could potentially influence later experiences using the plan. Four investigators trained in qualitative methods conducted telephone interviews which were audio-recoded and transcribed verbatim. Interviews lasted 44 min on average (range: 23–79 min).

2.4. Data analysis

We analyzed data iteratively via thematic content analysis [19]. In the first, inductive phase of analysis, three investigators independently coded a subset of interviews to identify broad topics of discussion. We compared and refined these codes, organizing them into a standardized codebook. We then practiced applying the codes to transcripts independently until we reached a high level of agreement (>90 %). One investigator then applied codes to all transcripts systematically using Nvivo Version 12 (QSR International, Melbourne, Australia). A second investigator reviewed the coded transcripts to identify disagreements in coding, which the team resolved via discussion. In the second, deductive phase of analysis, we considered data code-by-code, describing emerging patterns thematically.

3. Results

Just over half of our sample was female, most were aged 40–64 years (Table A.1, Table A.2), and 47 % were enrolled in a family plan, with insurance coverage extending to a spouse and or dependents. One third of the sample reported that someone sharing their insurance plan had a chronic illness. The majority purchased their insurance on the Marketplace rather than off-Marketplace.

We identified three categories of themes: (1) Decision Making Environment, representing the external setting in which participants must choose a plan; (2) Decision Making, representing the processes of selecting a plan; and (3) Suggestions, representing participant recommendations for improving the process of choosing a plan.

3.1. Decision making environment

Theme 1: Decision support services offered different levels of knowledge, helpfulness, and objectivity (Table B.1). Participants reported support from four sources: (1) insurance brokers, who are licensed insurance agents who help consumers enroll in insurance plans; (2) carrier customer service; (3) navigators, who are trained staff from community organizations who are certified by the Marketplace to provide enrollment assistance; and (4) Marketplace customer service. By law, only insurance brokers and carrier customer service representatives can make plan recommendations to consumers.

Participants appreciated how brokers facilitated enrollment by speaking with customer service representatives and creating HealthCare.gov accounts on their behalf. While a few participants had concerns that brokers could be biased or incentivized, others relied on their expertise to make an informed choice. Generally, participants felt that navigators were knowledgeable and an unbiased source of understandable information. However, several participants found that navigators were not easily accessible in their neighborhood or provided information that was too general.

“[Navigators] were throwing ideas out there at me. They weren’t narrowing it down enough to suit my needs.”

Most participants reported interacting with a customer service representative from either the Marketplace or the carrier. Experiences with the former were largely negative, with long wait times and inconsistent or incomplete information; Marketplace customer service representatives often referred participants to website tools or the carrier for definitive answers about the details of plan benefits. In contrast, participants found the carrier’s customer service representatives to be more accessible, knowledgeable, and able to tailor information directly to the participant’s question.

Theme 2: Participants valued being able to make side-by-side comparisons on the Marketplace but were challenged by other operational aspects of Marketplace websites. Being able to compare plans from different carriers on the Marketplace website based on their attributes (such as premium or deductible level) helped participants narrow down options and choose a plan. Some participants found the Marketplace to be user-friendly, while others found the enrollment process to be time-consuming because the website was not intuitive, information was hard to locate, and webpage loading lagged.

“On the site [I was] able to side by side compare what the different rates were and the different things that were offered…[the Marketplace website] had multiple columns and that was very helpful to be able visualize what the differences were between them, compare apples to apples.”

Theme 3: Participants used the Marketplace to shop for plans, but some ultimately purchased their plans off-Marketplace. Some participants used the Marketplace as a tool to make side-by-side comparisons and narrow down choices before purchasing a plan directly from the insurance carrier off-Marketplace, where they felt that customer service was better and plan information more accurate. Participants also purchased plans directly from the carrier after using the Marketplace to shop if they were ineligible for a subsidy or had fewer plan choices available through the Marketplace.

“I started withhealthcare.gov. It allowed me to see everything. I tried to order [my plan] through that [website]. It didn’t work. I ended up calling [the insurance carrier directly]…It was a couple dollars one way or the other…[the insurance company] was more helpful..”

Theme 4: Provider finders were often useful, but other decision support tools were less helpful. Many participants used look-up tools to make sure their provider was in a plan’s network. Most found these tools to be well-designed and easy to use, although many also called their provider’s office directly for verification.

“You can stay in thehealthcare.govwebsite and put in your doctor’s name…they will say these three plans he’s covered under…”

Participants universally reported that cost estimator tools, which ask consumers to predict health care needs and calculate associated out-of-pocket costs in different plan options, were unhelpful because they were confusing to interpret, too hypothetical, and not personalized enough. The perceived utility of website tools varied depended on a participant’s level of comfort with technology and need for support.

“I don’t get all those cost estimator things. I don’t know. Just doesn’t work for me.”

3.2. Decision making process

Theme 5: The task of synthesizing information to inform a plan choice was challenging for many participants (Table B.2). The large amount of information needed to be found, understood, and applied to their own health care needs to choose a plan created a feeling of information overload for many. Difficulty interpreting health insurance jargon and terminology was especially common. Other participants reported spending many hours over several days searching for information from multiple sources, compiling information into a useable format or personalized spreadsheet, and adjusting their budgets before ultimately making a choice. The amount of time and difficulty finding and interpreting plan information led many respondents to feel frustrated with the amount of effort needed to choose an insurance plan.

“You have to do your research, and it’s not trivial. It’s a lot of work to make an informed choice.”

Theme 6: Uncertainty about future health care needs and the risk of picking the “wrong” plan were concerning to many participants. Many participants expressed concern that choosing the wrong plan would lead to unexpected costs, the need to switch doctors, or adverse health outcomes. The inability to predict health care needs and unexpected events such as emergency department visits was a source of worry.

“There’s a certain level of anxiety about picking the wrong plan and then paying so much out of pocket and you didn’t realize. Or a certain prescription has to be changed because of it.”

Alternatively, being in good health allowed some to feel that their choice of insurance plan carried a lower risk since they did not anticipate needing much health care. Some participants worried about choosing a plan that would later be discontinued.

Theme 7: Participants’ confidence in their ability to choose a plan varied and increased with experience. Most participants believed in their ability to overcome obstacles to finding and synthesizing plan information to make a good choice. Confidence was particularly high among those with a background in health care or human resources that gave them an understanding of health insurance terminology. Most participants reported that although choosing a non-group plan was complex and intimidating initially, once they had successfully chosen a plan, the decision became easier in subsequent open enrollment periods. Prior support from a knowledgeable person such as a navigator gave some participants the knowledge and confidence to complete the process on their own the following year.

“In the next two enrollment periods I did it on my own. I was familiar enough with it that I didn’t need to go see [navigator].”

A small subset was discouraged by the process and lacked the self-efficacy to make an informed decision, leading them to forgo researching plans, default to remaining in their existing plan, or choose randomly among low-premium plans. Health insurance literacy was generally lower for these participants, and they struggled to identify what would enable them to make a more informed decision.

“It’s just easier to pick a package and then roll the dice and see what ends up getting covered.”

3.3. Suggestions for improvement

Theme 8: participants wanted information to be simplified and standardized to allow for direct comparison across plans and carriers (Table B.3). Participants suggested improvements to the information provided for choosing a plan, including reducing the use of jargon to describe health plan information, providing a glossary of definitions with examples, and making plan names standardized and explanatory.

“[I want] better detailed information about what things actually are. Because we were looking at lower premium plans that have coinsurance. And you go, ‘That’s great’, but then you say, ‘What’s coinsurance?’”

Theme 9: participants wanted more user-friendly, customized decision support. Participants recommended that Marketplaces and carriers provide opportunities for information to be customized, and that they create a more user-friendly web experience.

“The ideal thing [would be] to have somebody that could say, ‘Let’s look at your situation. Based on the last couple of years of your medical bills and your general health, it looks like the best recommendation would be to go with this plan.’ That would be helpful.”

4. Discussion

This qualitative study of participants who had negative experiences using their non-group health insurance plans finds that consumers have considerable challenges finding, understanding, and synthesizing plan information when selecting health plans. These challenges occurred both on and off the ACA Marketplace. However, study participants found several features that made them feel better about their choices: personalized assistance from brokers and navigators, presentation of information that allowed for direct comparisons across plans and carriers, customer service with knowledge of benefits details, and provider finders. While some participants were overwhelmed and lacked self-efficacy in selecting a plan, others felt that prior experience helped them feel more confident in their plan choices.

Participants sought greater use of plain language and avoidance of insurance jargon on Marketplace and carrier websites. Use of common names and terminology across plans and carriers could allow health insurance plan consumers to more easily compare “apples-to-apples.” Some study participants were not aware that solutions such as filters to compare plans on attributes and glossaries of insurance terms already exist, or they may have forgotten having seen them by the time of the interview; regardless, design and marketing could be improved to be more effective in conveying terminology for lasting retention.

In the years following implementation of the ACA, Marketplaces and carriers have added decision support tools such as provider and drug finders, cost estimators, and quality ratings to help consumers select plans [21]. Many study participants prioritized continuity with existing providers and utilized provider finders but found their utility was limited by incorrect or outdated information [21], [22]. Other decision support tools, such as drug finders and quality indicators, were infrequently used or not perceived as effective. Despite the fact that use of personalized decision support tools has been shown to be associated with cost-saving plan choices and increased satisfaction, those who might benefit most from these tools are less likely to use them [23], [24], [25]. Use of tools could be enhanced if they were easier to find and simpler to use, and included definitions of insurance terms paired with example situations and less jargon [25]. Simpler tools may be more effective; for example, showing only total cost estimates rather than other financial details led consumers to choose lower cost plans without a decrease in plan quality [26], [27], [28].

Even with enhanced tools, meeting the needs of consumers who are overwhelmed, lack confidence, and feel disengaged from the selection process may be a challenge. We found that study participants’ lack of confidence in their ability to make an optimal plan choice overlapped with lacking health insurance literacy [5]. Low health insurance literacy is not uncommon [3], [29], [30], but some study participants with low health insurance literacy were still actively engaged in choosing a plan or sought support services for answers to their questions. Gaining mastery experiences over time in successive open enrollment periods or working with a navigator or broker appeared to help to increase confidence in making an informed choice.

In addition to improving satisfaction and retention, optimizing health plan choice in the non-group market is important given the ramifications for downstream health care access, quality, and affordability. Misunderstanding insurance benefits or incorrectly estimating out-of-pocket costs or eligibility for subsidies when choosing a plan may lead to burdensome downstream costs, unmet need, greater Emergency Department use, and worse overall health [10], [11], [12], [13], [14], [31], [32], [33]. The uncertainty expressed by participants about future health care needs and associated fear of unexpected downstream health care costs could drive consumers to over-insure and choose plans with higher premiums, even if healthy [34]. Understanding and addressing barriers to choosing a plan might help mitigate negative downstream experiences using coverage. This may be even more important when people must find coverage in the non-group market after job loss, such as during economic downturns as in the COVID-19 pandemic [35]. However, it is notable that in our sample of participants who reported negative experiences after enrollment, not all expressed difficulty choosing their plan. This suggests that other elements of non-group market plan design and costs may also contribute to adverse experiences in non-group plans, despite informed choices [36].

4.1. Policy implications

Our findings suggest areas where ACA Marketplace and non-group market policies could improve the health plan selection process. Study participants valued real-time, one-on-one aid from assisters who could answer questions, clarify insurance terms, and make recommendations tailored to their health care needs and preferences. Although the ACA requires that Marketplaces provide access to certified enrollment assisters such as navigators and brokers, federal funding cutbacks to navigator programs during this time may have constrained access; funding for navigators has more recently been increased [37]. Navigators are also prohibited from making plan recommendations. Consumers would benefit from expanded capacity of trained, in-person assisters who can make personalized plan recommendations, especially for those who are new to the non-group market (24 % of Marketplace enrollees in 2019) or have low health insurance literacy [8], [38], [39]. Targeting outreach to these groups could help them gain mastery experiences that carry into active, informed choices in future years and less need for support in subsequent open enrollment periods.

Although consumers enrolled in health insurance plans both on and off-Marketplace experienced challenges choosing a plan in our study, those who purchased a plan on the Marketplace found it a helpful tool for making standardized comparisons between plans from different carriers, even if they subsequently enrolled off Marketplace. However, other features of the Marketplace warrant improvement, such as website organization and customer service knowledge of plan benefits. Improving navigator capacity in subsequent years has kept more enrollment through the Marketplace [37], but expansions in the number of carriers and plans [16] may make it harder for Marketplace customer service to provide the level of plan information that could be obtained off-Marketplace from the carrier. Other political and policy changes in the years following our study, such as withdrawal of federal funding for cost-sharing subsidies, have likely changed the plan choice options and decision making in the Marketplaces [40], [41]. As the ACA evolves over time, policy makers should consider how to preserve and enhance the features of the Marketplace that help consumers choose non-group coverage.

4.2. Limitations

Several limitations should be noted. First, our sample was drawn from a single insurance carrier in one region of the country during a single year, limiting transferability; other insurance carriers may offer resources, assistance, and benefit designs that differ in their usefulness and costs. Subsequent ACA and Marketplace policy changes in the years following our study altered the number of participating plans and carriers and the availability of subsidies and navigators, which likely influenced consumer choice and experiences [37], [40], [41]. However, our study includes both state-based and federally facilitated ACA Marketplaces in three states. Second, our sample was limited to those who reported negative experiences using non-group plans, as we wanted to know about plan choice experiences for these consumers. By intentionally selecting those with negative experiences, we may have been less likely to uncover positive attributes associated with choosing plans on and off Marketplace. The experiences of study respondents, for whom negative experiences were more prevalent than among non-respondents, may not reflect the non-group population with less negative experiences. Recall of experiences may have been impacted by the timing of interviews more than a year after enrolling in the plan. Finally, although our qualitative approach is well-suited to gaining a greater understanding of participants’ experiences purchasing health insurance in the non-group market, our findings cannot quantify the prevalence of those views.

5. Conclusion

Health plan selection in the non-group market can be challenging for consumers on and off the ACA Marketplaces. Health plans and the ACA Marketplaces have an opportunity to improve plan choice experiences through greater standardization of plans and language, improved customer service, availability of customized one-on-one assistance, and outreach to consumers who lack experience and confidence to actively choose a plan. Supporting consumers during the plan selection process may lead to plan choices that better meet their needs, reduce frustration and worry, improve downstream plan experiences, and increase confidence choosing a plan in subsequent enrollment periods.

Funding

This project was supported by grant number R01HS024700 from the Agency for Healthcare Research and Quality. The content is solely the responsibility of the authors and does not necessarily represent the official views of the Agency for Healthcare Research and Quality.

CRediT authorship contribution statement

Elena Faugno: Writing – original draft, Validation, Project administration, Methodology, Formal analysis, Data curation. Melissa B. Gilkey: Writing – review & editing, Supervision, Conceptualization. Lauren A. Cripps: Writing – review & editing, Validation, Project administration, Data curation. Anna Sinaiko: Writing – review & editing, Supervision, Funding acquisition, Conceptualization. Alon Peltz: Writing – review & editing. Jon Kingsdale: Writing – review & editing, Supervision, Conceptualization. Alison A. Galbraith: Writing – review & editing, Validation, Supervision, Methodology, Funding acquisition, Conceptualization.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A.

See Fig. A.1, Table A.1, Table A.2 and Table B.1, Table B.2, Table B.3.

Table A.1.

Characteristics of study participants (n = 36).

	%
Sex
Female	58
Male	42
Age (years)
18–39	25
40–64	75
Family income*
<250 % FPL	25
251–400 % FPL	22
>400 % FPL	39
Family member with chronic condition	33
2017 plan type
Individual	53
Family	47
State of residence
Massachusetts	36
New Hampshire	33
Maine	31
2018 plan source
Marketplace	58
Off-Marketplace	42
Switched insurance plan in 2018	67
Switched insurance carrier in 2018	28
Somewhat or very confident on all health insurance literacy measures	22
Negative plan experiences
In the prior year, total out-of-pocket costs for medical care for participant and their family were higher from what they had expected when they chose their plan	54
In the prior year, participant or family member on plan delayed or didn’t get a doctor visit, test, prescription, or any other medical care because of the cost	61
Health care-related financial burden: In the prior year, participant had trouble paying or was unable to afford any of their or their family’s medical bills OR Participant not confident that they could afford the care they would need if they or a family member got sick or had an accident	75
When choosing the current plan, participant wished they had more help in figuring out which plan would be best for them	42
Participant rated their experience getting health insurance as fair or poor	50

^*Data is missing from n = 4 participants.

Table A.2.

Characteristics of respondents vs non respondents (n = 71).

	Respondents %	Non-Respondents %
	(n = 36)	(n = 35)
Sex
Female	58	51
Male	42	49
Age (years)
18–39	25	46
40–64	75	54
Family income*
<250 % FPL	25	34
251–400 % FPL	22	26
>400 % FPL	39	40
Family member with chronic condition	33	31
2017 plan type
Individual	53	54
Family	47	46
State of residence
Massachusetts	36	9
New Hampshire	33	43
Maine	31	48
2018 plan source
Marketplace	58	80
Off-Marketplace	42	20
Switched insurance plan in 2018	67	43
Somewhat or very confident on all health insurance literacy measures	22	11
Negative plan experiences
In the prior year, total out-of-pocket costs for medical care for participant and their family were higher from what they had expected when they chose their plan	54	49
In the prior year, participant or family member on plan delayed or didn’t get a doctor visit, test, prescription, or any other medical care because of the cost	61	51
Health care-related financial burden: In the prior year, participant had trouble paying or was unable to afford any of their or their family’s medical bills OR Participant not confident that they could afford the care they would need if they or a family member got sick or had an accident	75	63
When choosing the current plan, participant wished they had more help in figuring out which plan would be best for them	42	60
Participant rated their experience getting health insurance as fair or poor	50	37

^*Data is missing from n = 4 participants in the respondent category.

Table B.1.

Participant’s Perception of the Decision-Making Environment.

Theme	Illustrative Quotation
Theme 1: Decision support services offered different levels of knowledge, helpfulness, and objectivity	“I just found [the navigator] very reassuring and very knowledgeable, and to me, it made a whole lot of difference. I just understood it better seeing her explain it, eyeball to eyeball. And I could ask questions, as many as I needed to have her frame it in a way that I understood.” “[Navigators] were throwing ideas out there at me. They weren't narrowing it down enough to suit my needs.” “If you're calling [the insurance company], these people are trained for that insurance. Whereas, if you're calling the [Marketplace], they don’t really have as much knowledge as the specific insurance companies.”
Theme 2: Participants valued side-by-side comparisons on the Marketplace but were challenged by other operational aspects	“On the site [I was] able to side by side compare what the different rates were and the different things that were offered…[the Marketplace website] had multiple columns and that was very helpful to be able to kind of visualize what the differences were between them, compare apples to apples.” “I think the structure itself is pretty easy…you just put in your income and then it gives you these choices, it pretty much does all the work for you, you just have to go down through the list and choose.” “Some things about the site aren’t 100 percent intuitive. But they do have like sort of these table charts where it basically lines up the different categories of plans and you can just look at—it just tells you the different costs and prices and deductibles.” “You'll click on a certain button to go to the next page and then it brings you in a circle, and then all the sudden you're back to the first page again. You're like “Well, how did that happen?” I think sometimes the way they're set-up is confusing considering there's like a million people using them. They should just be as cut-and-dry as possible.” “I noticed that when I call the [government website help line] every representative has a different answer for the same thing”
Theme 3: Participants used the Marketplace to shop for plans, but some ultimately purchased their plans off-Marketplace	“I started withhealthcare.gov. It allowed me to see everything. I tried to order [my plan] through that [website]. It didn't work. I ended up calling [the insurance carrier directly]…It was a couple dollars one way or the other…[the insurance company] was more helpful…I would advise somebody to first of all look at the exchange so that they can see everything. And then buy directly from the provider, because I…think that one tends to get better service from people to whom you write a check.” “The reason I decided to not [purchase] onhealthcare.govwas that my daughters are adopted and it's very difficult if your kids are adopted to get health care because when you go directly to [insurance company], they believe that they're my children and if I went to the Exchange, then I sort of needed to prove it and that's not easy to do because you can't meet with people” “I did go toHealthcare.govto see what sort of plans, what sort of pricing they had because we weren't going to get a subsidy anyway, I definitely, even if it was just a few dollars cheaper, I didn't want to deal with them, I just felt like it was easier to-- and so I just compared the prices in general with [insurance company site] to see if there was any plan with a higher deductible maybe that would be better cost and apparently there isn't.”
Theme 4: Provider finders were often useful, but other decision support tools were less helpful	“Now…you can stay in thehealthcare.govwebsite and put in your doctor’s name…they will say these three plans he’s covered under or this hospital, these two plans are contracted with that hospital. So those tools have gotten way better…it doesn't take hardly any time at all to get that info.” “The only difficulty that I would say that we have encountered in the past when changing [insurance plans] is that sometimes they'll indicate that some doctors are taking patients. And then, when we've called their offices, they've said that they're not taking patients.” “[Cost estimator] just seemed -- a little too hypothetical for me. I might've played with it for five or ten minutes but it felt like this wasn't really helping me that much.” “I don't get all those cost estimator things. I don't know. Just doesn't work for me.”

Table B.2.

Participant’s Decision-Making Process.

Theme	Illustrative Quotation
Theme 5: synthesizing information to inform a plan choice was challenging	“I just find the insurance world so daunting, because I don’t understand… this is a copay and a premium and you have to use up so much of your deductible, and this is in the deductible, that’s not in the deductible. I find it very confusing.” “You have to do your research, and it’s not trivial. It’s a lot of work to make an informed choice.” “I don’t have a sense that I thoroughly investigated all of the options, it was just too much for me” “Well I think it's pretty hard [to learn about the available health plans], you look at even on the government website, they give you all these options and it's like Russian roulette, nobody can figure that stuff out.” “People are busy. People are living their lives. They don’t want to be an expert on insurance.”
Theme 6: uncertainty about future health care needs and the risk of picking the “wrong” plan was concerning	“There’s a certain level of anxiety about picking the wrong plan and then paying so much out of pocket and you didn’t realize. Or a certain prescription has to be changed because of it.” “Trying to figure out how healthy you're going to be the next year and whether you think you're going to need to be in an emergency room at the hospital or you need to see an outpatient specialist or what you think you need so that you can juggle the numbers around and come up with the best thing…just, we need national healthcare.” “One plan has a high deductible but less copays, another one has higher copays but less deductible, you're really gambling on what's going to happen to you over the next year but I guess that's what insurance is, a gamble.” “As a family we're very healthy and we don't go to the doctor very often. …I've never hit the deductible, which means that I'm saving [on premium costs of the difference between the high and low deductible plan], so going with the higher deductible is worthwhile” “It'll be two years in November that I was diagnosed with Hodgkin's Lymphoma. I would rather pay a little bit more money each month for my insurance to make sure that I have the best coverage because when you have that diagnosis, you never know what's going to happen a year from now…and you just don't want to be left with any big financial surprises, you just want to make sure that you have the best coverage that you can afford.”
Theme 7: confidence in the ability to choose a plan varied and increased with experience	“Once you’ve done it a couple of times, you feel a little more confident, and it goes quicker the next time.” “In the next two enrollment periods I did it on my own. I was familiar enough with it that I didn’t need to go see [navigator].” “I feel like I’ve learned a lot about insurance that I never knew because of this Affordable Care Act and shopping for insurance.” “It’s just easier to pick a package and then roll the dice and see what ends up getting covered.” “I don’t really know [what advice to give] because I wouldn’t say that I’m confident that I’m doing the right thing”

Table B.3.

Participant Recommendations for Improving the Process of Choosing a Plan.

Theme	Illustrative Quotation
Theme 8: provide simplified, standardized information	“[I want] better detailed information about what things actually are. Because we were looking at lower premium plans that have coinsurance. And you go, ‘That's great’, but then you say, ‘What's coinsurance?’” “I think if there was kind of like a key for people… explaining kind of the basic terms, so people don’t have to do that [look it up themselves]…at least they have something sitting there saying ‘a deductible is this, and a coinsurance is this.’” “One of the biggest problems with [insurance company’s websites] is that if [they’re] going to offer 25 plans, it's not easy to get back to them, and they have stupid names. They should be called 1, 2, 3 and then you go back and check 1, or you can check 6; as opposed to Blue Plus 3285, which makes it hard to get back to something [to compare].” “[T]o do the comparison of all three [different plans] is difficult, because there is not a standard language used by every single plan, and to try to navigate that and figure out what exactly that means you either have to call and speak to somebody and maybe get transferred four times and get the answer finally, but it's so time-consuming…so there needs to be a language standard plan so that the consumer knows exactly what they're getting for exactly how much.”
Theme 9: provide user-friendly, customized decision support	“The ideal thing [would be] to have somebody that could say, ‘Let’s look at your situation. Based on the last couple of years of your medical bills and your general health, it looks like the best recommendation would be to go with this plan.’ That would be helpful.” “[Have an example of how to compare plans by] showing a printout of the page. For example, ‘This is what it's going to look like when you go on the website, don't be afraid, click on this, click on that.’ That kind of thing.”

FPL = Federal Poverty Level.

FPL = Federal Poverty Level.

Fig. A1.

Timeline of Surveys and Interview.

Appendix B. Supplementary material

The following are the Supplementary material to this article: