2016 Survey of Health Insurance Marketplace Assister Programs and Brokers

The new system for Marketplace enrollment assistance under the Affordable Care Act (ACA) is becoming well established.  Some 5,000 Assister Programs helped consumers apply for financial assistance and select health plans for 2016 during the third Open Enrollment (OE3).  Eighty-seven percent of Programs have been in operation three years, and 7 in 10 of three year Programs report most or nearly all of their staff have also worked all three years.  Eighty-four percent of brokers certified to sell non-group Marketplace health plans this year also have worked all three Open Enrollments.  As this system of in-person help matures, important distinctions are emerging among entities which could provide opportunities to develop strategies for identifying and building on those that accomplish the most.  At the same time, substantial challenges face many Assister Programs and brokers that hinder their ability to help consumers access and successfully enroll in health coverage.

This report is based on findings from the 2016 Kaiser Family Foundation survey of Health Insurance Marketplace Assister Programs and Brokers.  The online survey was conducted from February 11 to March 4, 2016 as OE3 concluded.  As was the case in prior years, federal and state-operated Marketplaces provided contact information for directors of their Assister Programs, all of whom were invited to participate.  In addition, most Marketplaces provided contact information for brokers certified to sell their qualified non-group health plans, and for the second year, a sample of brokers was also invited to participate in the survey.

Assister Programs combined helped an estimated 5.3 million consumers during the third Open Enrollment, roughly a 10% decline from last year. This decline is significant in light of concerns over the slowing rate of annual Marketplace enrollment growth.  It may be that some already-enrolled consumers didn’t seek help again this year, particularly those in Medicaid who face a more straightforward annual redetermination process in many states, or those who elected to auto-renew their qualified health plans.  It may also be that other factors, including lack of public awareness and affordability concerns, affect the extent to which eligible uninsured individuals seek help.  Survey respondents described several key challenges, including limited resources and inherent complexities in the application and plan choice process that may also constrain the reach and productivity of Assister Programs.

Most help from Assister Programs was provided by those with very large caseloads.  About 1 in 4 Assister Programs helped more than 1,000 consumers during OE3, accounting for 80% all consumers helped by Assister Programs.  By contrast, 30% of Assister Programs helped 100 or fewer consumers during OE3, and these small caseload Programs account for just 1% of all consumers helped by Assister Programs.  Large caseload Programs include all types of Assister Programs – Navigator, Federally Qualified Health Center (FQHC) and Certified Application Counselor (CAC).  These large Programs are distinguished from smaller ones in several respects.  Large Programs were more likely to help consumers with more complex needs such as language translation (28% of large caseload Programs vs 8% of small caseload Programs), immigration-related problems (23% vs 5%), problems reporting income or household size (56% vs 35%).  In addition, large caseload Programs were more likely to help resolve Marketplace data verification problems (96% vs 81%).  Large caseload Programs were also more likely to engage in outreach activities, to help consumers resolve post-enrollment problems, and to coordinate with other Assister Programs.

Enrollment assistance shifted toward renewing consumers in 2016, though most who sought in- person help still were uninsured.  Last year 53% of Assister Programs said most or nearly all consumers they helped were new Marketplace participants.  This year, 29% said this was the case.  Increasingly Programs are serving a mix of new and renewing consumers – evidence that consumers need help to remain covered, not just to enroll for the first time.  At the same time, a majority of Programs say that most of their clients were uninsured when they sought help.  This may indicate some consumers are returning at Open Enrollment having lost their Marketplace coverage during the year.  In addition, it suggests Assister Programs remain focused on reaching the uninsured.

Some capacity shortages continue.  Overall 79% of Programs said they could serve everyone who sought help throughout OE3, but 21% had to turn some away during surge weeks in December and January.  This is unchanged from 2015.  Among large caseload Programs, 30% had to turn away at least some consumers.

Enrollment assistance remains time intensive.  For the third year, it took 90 minutes on average to help consumers enroll for the first time and like last year, it took 60 minutes on average to help renewing consumers.  Like last year, most Programs (71%) said they could help most consumers complete the plan selection process.  Also like last year, most consumers who seek help have limited understanding of the ACA and difficulty understanding insurance and comparing plan choices.  Complexity in the application itself also challenges many consumers, according to Assisters.

Assister Programs helped hundreds of thousands of consumers with Marketplace real-time data verification problems.  Programs helped nearly 230,000 consumers resolve problems related to Marketplace identity proofing.  The automated federal identity proofing system, based on credit reporting data, poses challenges for consumers without established credit history, and those who cannot pass it can face significant delays in applying for Marketplace coverage.  Several state Marketplaces have streamlined the system, including by authorizing certified Assisters to visually verify identity documents.  SBM Programs were more likely than FFM Programs (22% vs 14%) to say identity proofing problems usually could be resolved quickly during the initial visit.

Marketplaces also conduct real time verification of applicants’ immigration status and income, matching it to online data sources.  This system also poses challenges to certain consumers, for example, those who are self-employed or experience other income volatility.  In 2015, the federal Marketplace alone terminated coverage for 500,000 individuals who could not resolve data match inconsistencies (DMI) related to immigration, and reduced subsidies for 1.2 million individuals who could not resolve DMI related to income.  Nationwide, Assister Programs helped an estimated 172,000 consumers with immigration-related DMI during OE3, and 259,000 consumers with income-related DMI. This may be an indication that the volume of DMI problems is declining overall, or it may signify that many consumers faced with such problems are not getting in-person help to resolve them.  Small caseload Programs were more likely than large caseload Programs to say they would not help consumers resolve immigration DMI (19% vs. 4%) or income DMI problems (11% vs. 3%).

Medicaid file transfers can still pose challenges, especially in federal Marketplace states.  Nearly all SBM states have a single, integrated system that makes eligibility determinations for both Medicaid and Marketplace coverage. In contrast, in the 34 FFM and 4 SBM states that rely on healthcare.gov for Marketplace eligibility and enrollment functions, electronic accounts must be transferred between the federal and state systems to provide coordinated enrollment across Programs. Eight FFM states have authorized the federal system to make final Medicaid eligibility determinations, which can expedite the enrollment process. In the remaining 30 FFM states and 4 SBM states that use healthcare.gov, the federal system assesses Medicaid eligibility and the final determination is completed by the state.  Among Programs in assessment states, 34% said the Medicaid eligibility determination was usually completed in a timely manner.  Many Programs said they will try to expedite the process by helping clients file a separate application for Medicaid (44% in assessment states and 13% in determination states).  Among Programs that helped clients complete separate applications, 46% said one follow up visit was typically required, 20% said 2 or more follow up visits were typical.

Significant numbers of Assister Programs (37%) and brokers (53%) said most clients had questions about health plans that were not answered by information on the Marketplace web site. Most Assister Programs (61%) and brokers (67%) said most or nearly all consumers had difficulty understanding basic insurance concepts.  This number is down from years one and two (75%), though still substantial.  Two-thirds of Assister Programs said most QHP-eligible clients could select a plan during the initial visit; the rest said at least one follow up visit was needed.  Brokers made similar observations.

During the year, Assister Programs also helped consumers enroll through special enrollment periods (SEP) and resolve post-enrollment problems.  Assisters helped at least 830,000 consumers enroll through SEPs between Open Enrollments last year.  This is a 30% increase over SEP help we estimated following OE1 – possibly because OE1 was much longer leaving fewer months available for SEP sign ups.  Assister Programs also helped at least 349,000 consumers report mid-year changes (e.g. in income) to the Marketplace last year.  And Programs provided post-enrollment help to at least 745,000 consumers between OE2 and OE3.  Like last year, once enrolled many consumers needed help if coverage was terminated unexpectedly, claims were denied, their provider was not in the plan network or their medication was not on the plan formulary. Again as was the case last year, when Assister Programs can’t help resolve post enrollment problems on their own, usually they do not refer to Consumer Assistance Programs, but more often refer consumers back to their health plan or to the Marketplace call center.

On average, brokers each helped 110 consumers apply for Marketplace policies during the third Open Enrollment, unchanged from last year. In addition, nearly all brokers also sold policies off the Marketplace, on average 48 during OE3, also statistically unchanged from last year.  Like last year, brokers served consumers with somewhat different characteristics than those helped by Assister Programs and provided somewhat different kinds of help.  Compared to Assister Programs, brokers were less likely to help uninsured individuals (30% of brokers said most clients were uninsured vs. 56% of Assister Programs) or consumers who lack Internet at home (60% of brokers said few/no clients lacked Internet vs. 24% of Assister Programs). Forty-eight percent of brokers helped Latino clients vs. 76% of Assister Programs.  In addition, brokers were less likely to help consumers with Medicaid applications (47% did so vs. 89% of Assister Programs.)  Brokers reported higher rates of client continuity from one year to the next; 64% of brokers said most clients they helped this year were people they had helped the year before, vs. 40% of Assister Programs.

Brokers in FFM states initiate about half of Marketplace applications using alternative enrollment sites. The FFM permits use of alternative enrollment channels that meet federal minimum standards.  On average, brokers said they started about 26% of FFM applications directly on insurance company websites and 23% of FFM applications on private web broker sites.  By comparison, SBM brokers initiated two-thirds of QHP applications on the Marketplace website.  Permitting direct enrollment through alternative channels was adopted with the intent of maximizing public awareness and enrollment opportunities and encouraging technology advances such as new plan comparison tools and apps for mobile devices.  In follow up interviews, some brokers cited technology advantages of these enrollment channels including easier data entry and the availability of “dashboard” features to help them track all clients.  Others said that not having to set up a healthcare.gov account saved time.  Still others noted this shortcut could later prove disadvantageous if consumers needed to follow up with the Marketplace but did not have an account.

Beyond logistics, several brokers mentioned that some alternative enrollment channels also offer non-QHP products, such as cancer policies, short term policies, and other “excepted benefit” products that do not have to follow ACA market rules, such as the prohibition on pre-existing condition exclusions.  Alternative enrollment channels made it simpler to obtain premium quotes and enroll consumers in these products, as well.  CMS is working on improved ways to monitor the sale of QHPs through alternative enrollment channels.  It does not track sale of other types of products through these channels.

Some insurers are ending or reducing broker commissions, especially for SEP policies.  Nearly half of brokers (49%) say at least some insurers have stopped paying commissions on all Marketplace policies; 17% say most or all of the insurers they do business with have taken this action.  More often brokers (60%) say at least some insurers have stopped paying commissions on Marketplace policies sold outside of Open Enrollment to consumers eligible for SEPs; 33% of brokers say most or all insurers have stopped paying SEP commissions for Marketplace policies.  Insurers report that SEP enrollees have higher health care claims on average than people who sign up during open enrollment, and therefore want to discourage use of SEPs. Changes to SEP commissions appear to be taking place more often in FFM states than in SBM states.  Nearly half of brokers in FFM states (46%) say most or all insurers they regularly do business with have ended commissions on SEP policies, compared to 10% of brokers in SBM states.  Twenty-nine percent of FFM brokers say no insurers have ended SEP commissions on Marketplace policies, compared to 61% of SBM brokers.

Regulators in several SBM states have prohibited these commission changes.  Other state regulators and CMS, which directly regulates insurance in five FFM states, have not taken such action.  The net impact on consumer access to coverage is not clear.  Some brokers commented they will continue to help consumers enroll in QHPs during SEPs, even if unpaid, as a public service and to earn client good will.  Others said they will consider selling other coverage, such as short-term non-renewable policies, to SEP-eligible consumers instead.

Most Assister Programs (65%) and brokers (55%) said OE3 went better than OE2.  This year, respondents were also asked to rate the ACA overall out of a possible 10 points.  On average, Assister Programs rated the ACA 6.5, while brokers on average gave a rating of 4.5.  To make the ACA work better, respondents were also asked to select the top three changes they would recommend.  Changes most frequently recommended by Assister Programs among their top three were to (1) reduce health plan cost sharing (named by 51% of Programs as one of their top three), (2) expand Medicaid eligibility in all states (32%), and (3) expand premium subsidies for Marketplace plans (30%).  Changes most frequently recommended by brokers among their top three were to (1) increase broker commissions (named by 47% as one of their top three), (2) repeal the law altogether (28%), and (3) reduce health plan cost sharing (28%).

About the Assister Programs and Brokers Described in this Report

Several types of Assister Programs provide outreach and enrollment assistance in the Marketplace.

Navigator refers to Assister Programs that contract directly with State Marketplaces or with federally facilitated Marketplace to provide free outreach and enrollment assistance to consumers.   The ACA requires all Marketplaces to establish Navigator Programs and to finance Navigators using Marketplace operating revenue.  Some states use a different name to describe these Programs, though in this report all Assister Programs funded directly by Marketplaces are referred to as Navigators.  CMS provided $67 million for Navigators to work in 34 FFM and FPM Marketplaces this year, compared to $60 million last year and $67 million in year one.¹    SBM state funding for Navigators exceeded $100 million in year one, that amount declined by about 15% in year two. ²   Moving forward, funding for Navigators has become more ad hoc in at least some SBM states.  The Connecticut Marketplace, for example, no longer provides for year-round Navigators.  Instead, during Open Enrollment, Access Health CT staff and temporary hires provide in-person enrollment assistance through temporary storefront sites and public libraries.  During the rest of the year help is available through volunteer CAC assister Programs and the state’s ombudsman office.  In Colorado, the Marketplace provides roughly 20 percent of resources for its Navigator Program and applies for philanthropic grants for the rest.

Certified Application Counselor (CAC) refers to Assister Programs that are recognized by a Marketplace but do not receive funding from a Marketplace.  This designation was created prior to the first Open Enrollment – when funding for Marketplace-paid Assisters, at least in the FFM, was still uncertain – to ensure that willing volunteer Programs would also be available to help.  CACs must be sponsored by an organization that will attest to the Marketplace that all of its individual Assisters meet minimum requirements.  CACs also must provide help to consumers free of charge.  Under federal rules, CACs are not required to engage in all activities required of Navigators, and they are not required to undergo training as extensive as that required for Navigators. All Marketplaces are required to recognize and certify CAC Programs, and states have flexibility to establish additional rules for CAC Programs.  Marketplaces are not required to provide funding to CACs; most of these Programs are primarily privately funded, supported by their own sponsoring organizations and other outside sources such as foundations.

Federally Qualified Health Center (FQHC) Programs are operated by health centers funded by the Health Resources and Services Administration (HRSA).  FQHCs treat patients regardless of ability to pay and, prior to enactment of the ACA, actively helped patients apply for Medicaid, CHIP, or other available coverage.  For the first year of ACA implementation, HRSA awarded $208 million to FQHCs to support enrollment assistance.  In the second year, HRSA made permanent enrollment assistance grants to FQHCs, which now total about $150 million per year.  All FQHC Assisters are required to complete at least the level of training required of CACs.  About 5% of FQHCs also serve as Navigators and so received Marketplace funding in addition to HRSA grants.  For purposes of this report, FQHCs that also receive Marketplace funding are referred to as Navigators.

Federal Enrollment Assistance Program (FEAP) refers to Assister Programs that contracted with CMS to provide supplemental enrollment help within FFM and FPM states in selected communities where large numbers of uninsured individuals reside.  Duties and requirements of FEAPs are similar to those of federal Navigators except that FEAPs provide “surge” assistance.  Most have rolled back staff and operations since Open Enrollment ended.  In this report, unless otherwise indicated, description of findings about Navigators will include FEAPs because the two types are so similar.  For the 2016 coverage year, CMS awarded contracts totaling about $29 million to two organizations to establish FEAPs in 10 states.³   FEAP contracts were initiated for the 2014 plan year and have been renewed subsequently.  CMS will continue to contract with FEAPs in year four, though the contract amount and work sites have not yet been determined.

Finally, in addition to Marketplace Assister Programs, the ACA authorized creation of state-based ombudsman programs, also called Consumer Assistance Programs, or CAPs.  The law requires CAPs to provide outreach and public education and provide enrollment assistance to consumers in the Marketplace.  In addition, CAPs must help all state residents resolve questions and disputes with their private health insurance coverage, including helping consumers to appeal denied claims.  The ACA requires Marketplace Assisters to refer consumers with post-enrollment problems to state CAPs.  The law provided initial funding for states to establish CAPs and 35 were established in 2010.  However no new appropriations have been enacted since and most CAPs have not received any new federal funding since 2012.⁴  Pending additional federal funding, many CAPs remain operational, albeit at reduced levels.

Broker refers to a state-licensed professional who sells private health insurance to individuals and/or businesses.  Brokers are sometimes called agents or producers. To sell non-group or small group health plans offered through a state Marketplace, brokers must register with the Marketplace annually, sign a participation agreement, and complete required training.  This year more than 80,000 brokers in federal Marketplace states and 30,000 in state-based Marketplaces were certified to help consumers apply for financial assistance financial assistance and explain coverage options.  Brokers are paid a commission by the health insurance company offering the policy that the consumer selects.  Typically insurers pay commissions when a policy is first issued and at renewal for at least several years.  Brokers also offer ongoing services to consumers once they’re covered, including help with post-enrollment questions and help buying other insurance products or financial services.

In all, more than 5,000 Marketplace Assister Programs provided outreach and enrollment help to consumers during the third Open Enrollment.  This total is based on Program data provided by all state and federal Marketplaces, and represents a 9% increase in the number of Programs operating a year ago.

Once again, most Assister Programs that help people enroll through the Marketplace are not funded by Marketplaces. Navigators and FEAPs, which are funded directly by the Marketplace, comprise about 11% of total Programs.  Assister Programs in FQHCs, primarily supported by HRSA grants, comprised another 24% and CAC Programs, which generally receive little or no public funding, comprised 65% (Figure 1).  The distribution of Assister Program types is similar to that observed last year

An estimated 30,400 Assisters together helped about 5.3 million people during the third Open Enrollment period.  The total number of full-time equivalent (FTE) Assisters remained the same this year, and the total number of consumers helped by Assister Programs fell by about 10% compared to last year.

The vast majority of consumers helped (80%) received assistance from Programs that helped more than 1,000 people.  These large caseload Programs constituted 23% of all Assister Programs this year.  Programs with mid-size caseloads (that helped between 100 and 1,000 consumers) account for 47% of all Programs and helped 19% of all consumers who received assistance.  Thirty percent of all Programs had small caseloads, helping 100 or fewer consumers during Open Enrollment; together these small caseload Programs helped just 1% of the total number of consumers who received assistance (Figure 2).

Large caseload Programs include all three types – Navigator, FQHC, and CAC Programs.  This year 37% of Navigator, 31% of FQHC, and 16% of CAC Programs said they helped more than 1,000 people during Open Enrollment.  All three types of Assister Programs are also found among medium- and small-caseload Programs.  Marketplaces only contract with Navigator Programs, however, and so generally do not provide the same level of support and monitoring for most of the large caseload Programs that are responsible for most consumer assistance.

Large caseload Programs had more staff (14.5 FTEs on average) compared to small caseload Programs (average 2.2 FTEs).  The average number of consumers helped per FTE was also much greater in large caseload Programs (251) compared to small caseload Programs (16).

In addition, large caseload Programs differed from smaller Programs in the amount and type of work they did and the types of problems they helped consumers address.  Large caseload Programs were more likely than small caseload Programs to help consumers with more complex needs such as language translation (28% vs 8%), immigration-related problems (23% vs 5%), problems reporting income or household size (56% vs 35%).  In addition, large caseload Programs were more likely to help resolve Marketplace data verification problems (96% vs 81%) and more often could help consumers successfully resolve identity proofing problems (97% vs 89%).  Large caseload Programs were also more likely to engage in outreach activities (94% vs. 54%), to help consumers resolve post-enrollment problems (88% vs 53%), and to coordinate with other Assister Programs on enrollment events (39% vs 12%).

Most Assister Programs have now operated for three years.  This year 94% of Programs are returning from last year and 87% have been in operation since the first Open Enrollment.  Staff tenure is also increasing.  Nearly seven in ten three-year-old Programs report that most or all of their staff worked during all three Open Enrollments.  More experienced Programs may be more familiar with their Marketplace systems and procedures and may have developed closer ties with communities they serve (Table 1).

Most Assister Programs generally don’t coordinate with each other. Although two-thirds say coordination with other Programs improves effectiveness, most (59%) rarely if ever coordinate with other Programs.  Among those that do regularly coordinate with others, 95% said coordination is key to effectiveness.  In general, Navigators were more likely to coordinate with other Programs on activities such as planning outreach and enrollment events and resolving complex cases.  One-third of small caseload Programs said they never coordinated with others, compared to 10% of large caseload Programs.

Assister Program budgets are mostly modest.  Twenty seven percent of all Programs reported having an annual budget for consumer assistance of $50,000 or less.  Twenty-nine percent had annual budgets between $50,000 and $500,000.  Only 5% of Programs reported annual budgets larger than $500,000.  CACs tended to have the smallest budgets (Table 2).

Navigators were more likely to receive most of their funding from the Marketplace, while FQHCs relied more heavily on grants from HRSA.  CACs were most likely to rely on re-programmed resources from their sponsoring organization or other private sector support.

Funding uncertainty continues for some Programs.  Thirty-two percent of Assister Programs are not at all certain funding will be available next year, and 35% are only somewhat certain. This finding held across all types of Assister Programs and in FFM and SBM states. Marketplaces are required by law to pay Navigators out of operating revenue, though most fund consumer assistance year by year instead of dedicating a portion of revenue for this purpose.  FQHC Programs receive ongoing funding from HRSA.  Overall about six in ten returning Programs report their budget this year is about the same as it was last year.  Twenty-eight percent say this year’s budget is less than last year and 35% say it is less than in year one.  Navigator Programs were more likely than others to report budget increases.

The need for in-person assistance remains strong.  About eight in ten Assister Programs said most-to-nearly-all consumers sought help because they lacked confidence to apply for coverage and financial assistance on their own.  As well, about eight in ten Programs said most-to-nearly-all consumers needed help evaluating plan choices.  Fewer Assister Programs this year said that most consumers sought help with technical problems related to the Marketplace website, a sign that Marketplace IT systems continue to improve.  But similar numbers report that most consumers also had problems with various aspects of the application process, including questions about how to report their income, family status, or citizenship/immigration status.  There was a drop in the share of Programs reporting most clients had limited understanding of the ACA, though this remains a leading reason consumers seek help.  In addition, this year there was an increase in Programs who said most of their clients sought help renewing coverage (Figure 3).

Enrollment assistance shifted toward renewing consumers in 2016.  This year, renewing consumers made up a substantial share of the caseload for most Assister Programs.  Twenty-nine percent of Assister Programs this year reported that most or nearly all consumers they helped were new to the Marketplace.  By comparison, last year 53% of Assister Programs said most or nearly all consumers helped were new to the Marketplace (Figure 4).  Although auto-renewal is an option, the Centers for Medicare and Medicaid Services (CMS) reported that 60% of plan renewals for 2016 were active renewals.⁵   This finding suggests many consumers believe they need in-person help to remain enrolled in Marketplace health plans and maintain their subsidies, not just to enroll for the first time.

Even so, most who sought help during the third Open Enrollment were uninsured. This year, a majority of Assister Programs (55%) reported that most to nearly all of the consumers they helped were uninsured at the time they sought assistance.  This is significantly lower than 83% of Programs last year and 89% of Programs in year one who said most people they helped were uninsured at the time they sought assistance (Figure 5).  Even with the shift toward helping Marketplace enrollees renew coverage, a primary focus of Assister Programs continues to be on enrolling the uninsured.

Demand for consumer assistance sometimes exceeded capacity.  Twenty-one percent of Programs said they could not help all who sought it during the third Open Enrollment period overall. Similar numbers this year and last year had to turn away at least some consumers during the final weeks of Open Enrollment when a surge in demand always happens (Figure 6).

This year capacity to meet demand was especially stretched among large caseload Programs (30% had to turn at least some consumers away) compared to small caseload Programs (16%).  Programs that reported a budget decrease in 2016 were more likely to say demand for help far exceeded capacity compared to Programs whose budget stayed the same or increased from the prior year (12% vs. 4%).

Eligibility and enrollment assistance remains time-intensive.  As was the case in years one and two, Assister Programs reported that, on average, it took 90 minutes to help consumers who were applying to the Marketplace for the first time.  In addition, like last year, Programs said it took one hour on average to help consumers who were returning to the Marketplace to renew or change coverage (Figure 7).

Why does the average time required for in-person help remain the same, even as Assisters and consumers gained experience and Marketplace websites have grown more reliable?  The answer may lie in the inherent complexity of applying for coverage and financial assistance under the ACA.  This year’s survey provides new data about specific aspects of the eligibility and enrollment process that can be challenging – real time verification by the Marketplace of applicants’ identity and application information, coordination between the Marketplace and state Medicaid programs, and the process of comparing and selecting Marketplace QHPs.

Real Time Verification by the Marketplace

On-line applications were expected to be easier and faster to complete because Marketplace websites would be able to verify consumer’s information in real time, matching it to other online databases.  However, sometimes IT system problems, or the fact that some consumers don’t have matching information in the databases Marketplaces check, result in significant delays and may even pose a barrier to enrollment for some people.    

Many consumers sought help with Marketplace identity-proofing requirements.  To protect against enrollment fraud, Marketplaces first verify in real time the identity of applicants before they can submit an application for coverage and financial assistance.  FFM states use an automated remote identity proofing process (RIDP) that compares applicant information against credit files and other online data sources.  Many SBM states also use the federal RIDP system.  One study found that certain groups of individuals are especially likely to have difficulty completing RIDP, including young adults and recent immigrants with limited credit history.⁶   The same study observed some SBM states have adopted alternatives or modifications to the RIDP system to streamline the process of proving one’s identity and expedite resolution of problems when they arise.

During the third Open Enrollment period, Assister Programs helped at least 230,000 consumers with identity-proofing problems. Depending on Program size, on average between 3% and 10% of consumers helped by Assister Programs during Open Enrollment encountered identity-proofing problems.

Overall, about 90% of Programs that helped consumers with identity proofing problems said they usually could help resolve them.  Those in SBM states were more likely than in FFM states (22% vs. 14%) to say they usually could resolve problems quickly during the initial visit.  This is probably because several SBM states use an alternative to RIPD.  For example in Colorado and California, certified Assisters can visually verify an applicant’s identification document, upload a copy to the Marketplace, and then proceed immediately with the application.

Across all states, most Programs said that when identity proofing problems arose they usually added significantly to visit time or necessitated a follow up visit (Table 3).  Programs in SBM states were less likely than in FFM states (3% vs. 10%) to say identity proofing problems usually could not be resolved.  Across all Marketplaces, large caseload Programs were less likely than small Programs (3% vs 11%) to say these problems usually could not be resolved.

Overall, about one in four Assister Programs said they would like more training in resolution of online identity proofing problems.  Large caseload Programs were twice as likely as small caseload Programs (35% vs 17%) to say they would like more training on this topic.

Many consumers sought help for Marketplace data match inconsistencies (DMI). Marketplaces also require real time verification of consumers’ citizenship or immigration status and income.  Applicant information is matched against other online data, for example, held by the Social Security Administration, Department of Homeland Security, and Internal Revenue Service. When the Marketplace can’t verify application information online, consumers receive a notice of data match inconsistency (DMI) and are provided a temporary eligibility determination based on information they submitted.  Consumers can enroll in coverage right away, but must provide additional documentation to the Marketplace within 90 days, otherwise their coverage or subsidies may be terminated.  During 2015, the FFM terminated coverage for 500,000 individuals with citizenship or immigration DMI, and terminated or reduced premium tax credits and cost sharing subsidies for 1.2 million consumers with income DMI.⁷ 

Most Assister Programs said they helped consumers resolve DMI problems relating to immigration or citizenships during the third Open Enrollment.  We estimate these Programs helped at least 172,000 consumers with immigration-related DMIs.  In addition, most Programs reported helping consumers with DMI problems related to income.  We estimate at least 259,000 consumers sought in-person help with income-related DMIs from these Programs.

In comparison to the number of FFM enrollees who could not resolve DMI problems last year and who lost coverage or subsidies as a result, these estimates suggest that either the number of consumers affected by DMI fell substantially this year, or most consumers with DMI are not getting help from Assister Programs.

Most Assister Programs will try to help consumers resolve DMI problems when they arise, though smaller Programs are more likely than large caseload Programs to refer consumers elsewhere for help or advise them to resolve inconsistencies on their own.  In general, large caseload Programs were also more likely to know the resolution of their client’s DMIs compared to smaller Programs (Table 4).

Marketplace notices about DMI may also present a challenge.  For example, in case of an income-related DMI, FFM notices list examples of documents consumers might submit to verify different sources of projected income but do not specify which would be most appropriate for the individual applicant.⁸   With respect to DMI notices related to immigration or citizenship, 39% of Assister Programs said most of the time it was not clear what documentation the Marketplace wanted to the consumer to provide; 29% of Assister Programs responded this way with respect to income-related DMI notices.

One-in-five Assister Programs overall said they would like more training on the resolution of DMI problems.  Large caseload Programs were more likely to want such training (about one in three) compared to small Programs (about one in seven).

Medicaid-Marketplace Coordination

Depending on the state, consumers also needed extra help enrolling in Medicaid.  The ACA outlines a “no wrong door” approach to applying for coverage and requires a “single streamlined” application for financial assistance that can be used to determine eligibility for both QHP subsidies and Medicaid or the Children’s Health Insurance Program (CHIP).  All 13 State-based Marketplaces that do not use healthcare.gov have integrated their eligibility systems with Medicaid, eliminating the need to transfer data between systems to make eligibility determinations for coverage.  Eight FFM states allow healthcare.gov to determine Medicaid eligibility, though files are then transferred to state Medicaid agencies to complete enrollment.

In the remaining 30 states, healthcare.gov assesses Medicaid eligibility, then transfers the consumer’s file to the state Medicaid program for a final eligibility determination and to complete enrollment. Among Programs in states with integrated eligibility systems or in FFM determination states, 74% said the Medicaid enrollment process was usually completed in a timely manner.  By contrast, 34% of Programs in assessment states said that the Medicaid eligibility determination was completed in a timely manner.  To expedite the process, 44% of Programs in assessment states and 13% in determination states said they would help clients who were assessed eligible complete a separate Medicaid application.

Follow up interviews with Assister Program directors were conducted to learn why they created separate Medicaid applications.  Some observed that direct applications were often processed faster than transferred files.  Others cited difficulty in obtaining confirmation from healthcare.gov that transfers were successfully completed.  Still others noted that applying directly to Medicaid in some states can also expedite application for other benefits, such as the Supplemental Nutrition Assistance Program. Some directors said they pre-screen consumers, then submit the application to the Marketplace or the state Medicaid portal depending on the coverage the individual will most likely be eligible for. Among Assister Programs that typically help consumers complete a separate Medicaid application, two-thirds said at least one additional follow up visit was needed to complete the separate application (Table 5).

Comparing and Selecting Health Plans

The process of comparing and selecting health plans can also be complex.  The sheer number of plan choices can be one reason.  People living in FFM states, on average, had a choice of 50 Marketplace plans this year.⁹  Research on plan choice finds that having more than 10 options makes it harder for consumers to compare and evaluate.¹⁰  In addition, relatively small variations in QHPs can sometimes be meaningful to consumers.  For example, while most QHPs (not modified by cost sharing reductions) have annual deductibles well in excess of $1,000 per person, many plans impose separate deductibles for at least some services, and many exempt key services, such as primary care visits, from the deductible.^11 ,12  Another survey found that most consumers said it was somewhat or very easy to compare Marketplace plans generally; 74% found it easy to compare premiums, 69% found it easy to compare plan cost sharing features, and 60% found it easy to compare plan provider networks.¹³ 

Marketplace health plan information sometimes leaves consumer questions unanswered. This year, 37% of Programs said consumers often or almost always had questions about health plans that were not answered by information on the Marketplace website (Figure 8). This is an increase from last year (31%) and attributable mostly to Programs in FFM states.  Thirty-seven percent of FFM Programs this year, compared to 27% last year, said client’s health plan questions often were unanswered by information on healthcare.gov. Most Marketplaces have developed new plan comparison tools since the first Open Enrollment, for example, to sort options based on participating providers or to estimate consumer out-of-pocket expenses.  For 2017, new standardized plan choices may be offered in the FFM to simplify plan comparison by consumers.  Improvements to summary of benefits and coverage (SBC), a plain language summary of health plan provisions, were also approved and will be implemented in 2018.  These changes may make it easier for consumers to understand and compare plan choices in the future.

One in four Assister Programs say they would like additional training on qualified health plan features and how to distinguish differences between plan options.

Insurance literacy limitations among consumers persist.  This year most Assister Programs (62%) said most or almost all of their clients needed help understanding basic insurance terms and concepts such as “deductible” and “in-network service.”  This is an improvement from 74% of Programs who answered this way in the first two years, though still evidence of widespread limitations (Figure 9).  Three-in-ten Assister Programs would like additional training in health insurance literacy.

Assister Programs knew the plan choice for most/nearly all of QHP-eligible clients.  This year again, when asked how often they knew the plan choice of their QHP-eligible clients, 71% of Assister Programs said this was the case for most or almost all such clients – the same number who reported this in year two and an increase over year one, when website breakdowns required Assisters to spend most appointment time helping consumers with the application. Large caseload Programs were more likely than smaller Programs to observe the plan choice for most or nearly all clients who were eligible for QHPs (82% vs. 62%).  Two-thirds of Programs said most or nearly all of their QHP-eligible clients were able to complete the plan selection during the initial visit.  The other 34% said clients typically required multiple visits.

Returning Assister Programs helped at least 830,000 consumers with special enrollment periods in the past year.  Again this year we asked returning Assister Programs about help they provided consumers outside of Open Enrollment periods.  Most Programs were available throughout the year to help consumers who became eligible for special enrollment periods (SEP) or who needed to report other mid-year income or family changes to the Marketplace in order to update their application for subsidies.

Large caseload Programs helped more people with SEPs and reporting other mid-year changes compared to small caseload Programs (Table 6).  Nationwide, we estimate Assister Programs helped at least 830,000 consumer apply for SEPs in 2015, which is a 30% increase over the amount of SEP assistance reported for 2014.  This change may be due to the fact that the first Open Enrollment extended through April of 2014, leaving fewer remaining months that year for SEP to arise.

Assister Programs also helped consumers report mid-year changes in their subsidy eligibility, though fewer people came in for this type of help.  Large caseload Programs, again, provided more of this type of help (Table 46).  Nationwide, we estimate Assister Programs helped at least 349,000 consumers report mid-year changes to the Marketplace in 2015.

Assister Programs provided post-enrollment help to at least 745,000 consumers between the second and third Open Enrollment periods.  This year, nearly all returning Assister Programs also offered to help consumers with post-enrollment problems, though they are not required to do so.  Large caseload Programs provided most of this assistance (Table 7).

Consumers sought help with premium payment and invoicing problems, claims denials, and when their health providers were not in-network.  Consumers also returned for help because they did not understand how to use their health coverage (Figure 10).  Like last year, most Assister Programs (70%) said they could help consumers successfully resolve post-enrollment problems most of the time; 25% said they succeeded just some of the time and 5% said not very often.

The ACA requires Navigators to refer consumers with post-enrollment problems to state Consumer Assistance Programs, or CAPs.  However, federal funding for CAPs has not continued, and while many remain operational, Marketplace Assisters mostly refer consumers with post-enrollment problems elsewhere.  When asked where they refer consumers with post-enrollment problems they cannot resolve, only 19% of Assister Programs mentioned CAPs.  Instead, like last year, Assisters mostly referred consumers to the Marketplace Call Center (77%) or back to their health plan (59%) (Figure 11).

For the second year, the survey included health insurance brokers who were certified by the Marketplace to help consumers apply for non-group coverage.  Most, though not all state Marketplaces provided contact information for at least some of their certified brokers.  As a result, survey findings may not reflect experiences generalizable to the nation as a whole.

Enrollment Assistance by Brokers

Virtually all (92%) of brokers who sold non-group coverage in the Marketplace this year had done so last year and 84% were registered with the Marketplace during the first Open Enrollment period.

Most brokers who sold Marketplace coverage (82%) also sold policies outside of the Marketplace.   On average, brokers reported helping 158 consumers, both in and outside of the Marketplace, with eligibility and enrollment during the third Open Enrollment period.  On average, brokers helped more than twice as many clients apply for coverage through the Marketplace (110) compared to outside of the Marketplace (48) (Figure 12).

Some brokers were busier than others.  Most (56%) said they helped up to 50 Marketplace consumers during this Open Enrollment period, while 26% of brokers said they helped more than 100.  In Medicaid expansion states, brokers helped an average of 94 people enroll in Marketplace plans during Open Enrollment; in non-expansion states, the average was 138.  These findings are similar to what brokers reported last year.

Though we cannot make estimates of the number of consumers helped by brokers nationally due to methodological limitations, brokers clearly play a significant role in helping consumers to enroll in Marketplace coverage.  For example, the California Marketplace reported last year that 43% of new enrollees in 2015 were broker-assisted.¹⁴ 

This year brokers helped more renewing consumers than new enrollees.  This year 52% of brokers said most consumers they helped during Open Enrollment were clients who were returning to the Marketplace to renew or change their QHP and 21% said most clients they helped were new to the Marketplace.  By comparison, last year 49% of brokers said most consumers they helped were new to the Marketplace and 26% said most clients were renewing (Figure 13).

In addition, this year, brokers say fewer consumers were uninsured at the time they sought help.  Thirty percent of brokers this year said most or nearly all consumers they helped were uninsured, compared to 50% last year (Figure 14).

Enrollment assistance was also time intensive for brokers.  Like Assister Programs, brokers reported it took, on average, 1 to 2 hours to help a new Marketplace consumer enroll in coverage, and just over an hour to help a returning consumer.  On average, brokers encountered 8 clients with identity proofing problems during Open Enrollment. Similar to Assister Programs, brokers in SBM states were twice as likely to report these problems could be resolved quickly during the initial visit (28%) compared to brokers in FFM states (14%).  Also, brokers reported 12 clients, on average, encountered DMI problems related to immigration and 21 clients, on average, encountered DMI problems related to income.  Even more often than Assister Programs, brokers said that Marketplace DMI notices were unclear; 54% said immigration DMI notices were unclear most or nearly all of the time, 60% said this about DMI income notices.

Similar to Assister Programs, 27% of brokers said that for OE3 overall, they were unable to help all who asked for it and had to turn at least some consumers away.  Brokers were much more likely to say that demand exceeded their capacity in early December, just prior to the deadline for selecting or renewing coverage for January 1. Thirty-one percent of brokers found it hard to serve all consumers during this surge period, compared to 17% who said demand for help exceeded their capacity during the final two weeks of Open Enrollment.

Between Open Enrollments, brokers helped consumers with SEPs and post-enrollment problems.  On average, brokers helped about 27 SEP-individuals enroll in coverage, or less than one per week, about the same number they reported for the prior year.  Nearly all (94%) brokers will help clients with post-enrollment problems that may arise.  Between the second and third Open Enrollment periods, brokers report they helped 47 clients, on average, with post-enrollment problems, similar to the number they reported last year.

Comparing Activities of Brokers and Assister Programs

Like last year, brokers generally engaged in similar consumer assistance activities as Assister Programs, but with emphasis on different services.  For example, the vast majority of both brokers and Assister Programs said they help consumers compare and select QHPs, apply for premium tax credits, and resolve post-enrollment problems.  But as was also the case last year, compared to Assister Programs, brokers were less likely to engage in outreach and public education activities (40% vs 76%) and less likely to help consumers apply for exemptions from the individual mandate (24% vs 50%).  Compared to Assister Programs brokers were more likely to help small businesses select coverage (29% vs 4%).

Compared to Assister Programs, when clients received a notice of data match inconsistency from the Marketplace, brokers were somewhat less likely to help the consumer; 76% said they will help consumers resolve immigration-related DMI, compared to 89% of Assister Programs.  Brokers were also less likely, compared to Assister Programs, to help individuals eligible for Medicaid and CHIP (47% vs 89%).  Brokers who said they helped consumers with Medicaid applications were more likely to be from SBM states, where Marketplace eligibility systems are better integrated with Medicaid.

Also similar to Assister Programs, most brokers said they would like to receive additional training on a range of topics, including tax related issues, Marketplace appeals and renewal procedures, Medicare, and Medicaid.

Comparing Clients of Brokers and Assister Programs

Similar to Assister Programs, brokers overwhelmingly said consumers they helped had limited understanding of the ACA and limited health insurance literacy.  In other respects, though, broker clients differed somewhat from consumers served by Assister Programs.  For example,

• 85% of brokers said few or none of their clients needed language translation help, compared to 54% of Assister Programs
• 60% of brokers said few or none of their clients lacked internet at home, compared to 24% of Assister Programs
• 48% of brokers said they helped Latino clients, compared to 76% of Assister Programs
• 30% of brokers said most or nearly all clients they served were uninsured when they sought help, compared to 56% of Assister Programs
• 8% of brokers said most or nearly all clients had income low enough to qualify for Medicaid, compared to 42% of Assister Programs.

Brokers also were more likely than Assister Programs (64% vs 40%) to say most of the consumers they helped this year were people whom they had also helped during the previous Open Enrollment period.

Use of Alternative Enrollment Channels

This year we asked brokers about their use of Marketplace websites vs. alternative enrollment channels when helping clients apply for Marketplace health plans.  Federal regulations permit direct enrollment by individuals through insurance company websites into the Marketplace health plans they offer, and also through private web broker sites which are required to display all qualified health plans offered in the Marketplace.  Permitting direct enrollment into QHPs through these alternative channels was adopted with intent of maximizing public awareness and enrollment opportunities and to encourage technological innovations such as plan compare tools and apps for mobile devices.  State Marketplaces can decide whether to allow enrollment into QHPs through insurance company or web broker sites; the FFM allows use of these alternative enrollment channels.

Alternative enrollment channels must meet federal standards for accuracy and completeness of health plan information.  Alternative enrollment channels may also market other products that are not QHPs (for example, supplemental policies, accident-only policies, dread-disease policies); those that do must clearly distinguish non-QHP products from QHPs and indicate that federal premium and cost sharing subsidies only apply to QHPs.  In addition, alternative enrollment channels in the federal Marketplace must follow other standards.  All insurance companies that sell plans on the FFM have the capability of enrolling consumers directly on their own website.  According to CMS staff, fewer than two dozen web broker sites have been certified and are actively used in FFM states currently.¹⁵   To be certified, sponsors of these sites must complete training similar to that required of brokers and attest to CMS that their site meets all other standards.

In FFM states, consumers (or their brokers) who enroll through alternative channels enter information about themselves, their dependents, income.  Then they are re-directed to healthcare.gov which determines eligibility to participate in the Marketplace and for subsidies.  Finally, consumers are re-directed back to the alternative enrollment site where they can view the resulting net cost of health plan options, select a plan, and enroll.

On average, FFM brokers initiated 51% of QHP applications on healthcare.gov, 26% on insurance company sites, and 23% on web broker sites.  SBM brokers initiated two-thirds of QHIP applications on the state Marketplace site, 19% on insurance company sites, and 15% on web broker sites (Figure 15).

Some survey participants volunteered additional information about the pros and cons of using alternative enrollment channels.  Some noted technical and functionality advantages of alternative enrollment channels they used.  For example, some web broker sites provide a dashboard to track client accounts.  Ease of data entry was also mentioned as an advantage in some alternative enrollment channels; for example, some alternative channels provide a single screen to enter data on all family members, compared to healthcare.gov which requires data entry on separate screens for each household member.  Some alternative channels permit consumers to apply even if they do not have an email address. By contrast, healthcare.gov requires consumers to provide an email address to open an account, and this is burdensome for some consumers who do not have internet at home.  Brokers also said they sometimes started applications on other channels when the FFM site was functioning slowly, or vice versa.

Several mentioned one disadvantage of using alternative enrollment channels, namely that when an application is started off healthcare.gov, the consumer does not have an account created on the FFM site. Without an account, consumers can’t access the healthcare.gov Message Center, for example, where important updates about requests for additional documentation and other information are posted and easily accessible. Another broker observed that, without a healthcare.gov account, consumers who need to report mid-year changes can only submit them by phone to the call center.

Finally, some brokers noted that some alternative channels also sell non-QHP products and can provide quotes for these products along with the QHP.  A few said that they sell a significant volume of accident-only policies, cancer policies, short term policies, and other excepted benefit products to consumers who feel they need added protection from high cost sharing under QHPs.   Excepted benefit policies are not required to follow ACA market rules, such as the prohibition on pre-existing condition exclusions.  Currently, CMS does not require alternative channels to report data on non-QHP products sold to QHP enrollees.  Staff say they are working on improved ways to monitor the sale of QHP products through alternative enrollment channels.¹⁶ 

Changing Broker Commissions

Some insurers are ending or reducing broker commissions, especially for SEP policies.  Nearly half of brokers (49%) said at least some insurers have stopped paying commissions on all Marketplace policies; 17% said most or all of the insurers they do business with have taken this action.  Twenty-nine percent of brokers reported most or all insurers they do business with have reduced commissions on all Marketplace policies, while 14% said most or all insurers have reduced commissions on certain Marketplace policies, such as gold policies.  Late in 2015, United HealthCare announced first that it would cut agent commissions from as much as 10% to 2%, then, effective in 2016, suspend commissions entirely for the sale of Marketplace policies.¹⁷  In response several other insurers announced they, too, would end or reduce broker commissions for at least some Marketplace policies or enrollments.¹⁸ 

More often, brokers reported insurers were terminating or reducing commissions for policies sold to people eligible for SEP.  Insurers report that SEP enrollees have higher health claims on average than people who sign up during Open Enrollment, and therefore want to discourage use of SEPs.¹⁹   Sixty percent of brokers said at least some insurers have stopped paying commissions on Marketplace policies sold outside of Open Enrollment.  One-third reported most or all insurers have stopped paying SEP commissions for Marketplace policies (Table 8).

Changes in SEP commissions appear to be taking place more often in FFM states than in SBM states.  Nearly half of brokers in FFM states (46%) reported most or all insurers they regularly do business with have ended commissions on SEP policies, compared to 10% of brokers in SBM states.  Twenty-nine percent of FFM brokers reported no insurers have ended SEP commissions on Marketplace policies, compared to 61% of SBM brokers. So far, authorities in several SBM states have prohibited such broker commission reductions.  The Connecticut Insurance Department ruled against United’s action for 2016 on grounds that broker commissions had been incorporated into health plan rate filings the state had already approved.²⁰   Colorado’s regulator ruled that elimination of broker commissions on certain policies, including SEP policies, would constitute discrimination and an unfair marketing practice.²¹  The Kentucky Department of Insurance issued an advisory opinion that failure to pay agent commissions in accordance with filed rates would be a violation of the Insurance Code.²²   The California Marketplace is considering new requirements for participating insurers to pay the same commission rates for all their policies year round.²³ 

In other states, including those directly regulated by CMS, which have not blocked commission modifications, the effect on access to coverage remains to be seen.  Some agents who volunteered information after the survey said they would continue to help SEP-eligible consumers enroll in major medical health plans, even if they aren’t paid a commission, because it’s the right thing to do and because they hope consumers will ask for help renewing coverage at the next Open Enrollment, when commissions would apply.  Others said they would consider selling short-term non-renewable policies to SEP-eligible consumers instead.

Assister Programs and brokers were asked, in general, how the third Open Enrollment compared to the second.  Both acknowledged improvements: 65% of Assister Programs and 55% of brokers said OE3 went much better or somewhat better than OE2.

In addition, this year the survey asked both Assister Programs and brokers to rate the ACA overall on a scale of 1-10, with 10 signifying the law is working perfectly and 1 that it is not working at all.  Respondents were also offered a menu of possible ways to change the law and asked to select the top three changes they would recommend.

Assister Programs gave the ACA a rating of 6.5 out of 10, on average.  The top three recommended changes by Assister Programs were:

• Reduce health plan deductibles and cost sharing (51% included this among top three changes; 23% named this change as their first priority);
• Expand Medicaid eligibility in non-expansion states (32% named among top three changes; 16% designated as first priority); and
• Expand subsidies (30% named among top three; 12% as first priority) (Figure 16).

Brokers gave the ACA a rating of 4.5 out of 10, on average.  Their top priority changes were:

• Increase broker commissions (47% included among the top three changes; 20% named this as the top priority);
• Repeal the law altogether (28% listed among top three changes; 20% as first priority): and
• Reduce health plan deductibles and cost sharing (28% included among their top three changes; 6% named as first priority (Figure 17).

The new ACA system for in-person enrollment assistance through Marketplaces is becoming well established.  The vast majority of Programs have operated for three years and most of their staff have worked all three years, as well.  With tenure comes increasing expertise with Marketplace rules and procedures and familiarity with communities served.  There are now opportunities to build on the strengths of the most seasoned Programs and Assisters – perhaps offering more in-depth training and continuing education to develop specialized skills.

Fewer consumers were helped by Assister Programs this year.  Perhaps not coincidentally, the annual rate of Marketplace enrollment growth slowed this year, as well.  Investing in consumer assistance could help to increase enrollment, although those investments have to compete against other needs in federal, state, and marketplace budgets.  Evidence suggests consumers’ need for in-person help won’t go away any time soon:  an increasing share of consumers seeking help this year were renewing vs. applying for the first time; most still have limited understanding of health insurance and the ACA; and many still lack confidence to apply on their own.  There is also substantial churn in Marketplace enrollment – for example, as people gain or lose jobs with health benefits – creating an influx of new consumers seeking coverage and in-person help between Open Enrollment periods.

Uncertainty is also a challenge for Assister Programs, with one in three not certain that funding will be available next year.   The FFM has reduced funding uncertainty by adopting multi-year agreements with Assister Programs, though the amount of funding is decided year-by-year.

The survey reveals that the bulk of consumer help through Assister Programs is provided by a minority of large Programs–80% of all consumers helped in OE3 were served by just one-quarter of all Assister Programs.  These large-caseload Programs include Navigators, which contract directly with Marketplaces, and FQHC and CAC Programs, which are certified by Marketplaces but not necessarily as familiar to Marketplace officials.  Large caseload Programs may provide the greatest opportunity for improving consumer assistance in the future; however, these programs face resource constraints and were the most likely to say that demand for help exceeded their capacity to provide it, especially during surge times.

Assisters continue to report that it takes 90 minutes on average to help new Marketplace participants, and 60 minutes on average for returning consumers.  That the process remains time intensive, even after IT systems have improved, indicates how complicated the application process can be for consumers.  Consumers face particular challenges when “real time” data verification and file transfers don’t work, and significant delays and enrollment barriers can result.  It appears that many, if not most individuals who experience data verification difficulties are not being helped by Assister Programs.

Brokers, who have emerged as an important avenue for marketplace enrollment, are, not surprisingly, concerned about the loss of revenue as insurance companies reduce or end commissions, actions taken most often this year for SEP enrollments.  Millions of consumers are estimated to become eligible for SEPs during the year, but only a fraction take up the opportunity to enroll.²⁴   SEP enrollments can help offset normal churn of individuals who return to group health plans or public coverage during the year.  Loss of broker commissions, combined with adoption of new SEP eligibility verification requirements by the FFM, could dampen Marketplace enrollment.

The survey shows that brokers in FFM states rely heavily on alternative enrollment channels, especially to the extent these offer enhanced functionality.  However, little is known about the experiences of consumers who apply through them or how often consumers buy other products through these sites, such as short-term policies or plans that target specific diseases.

Finally, Assisters and brokers on the front lines have valuable insights into how health reforms are working for consumers.  Lower cost sharing in Marketplace health plans was identified as a priority by both Assister Programs and brokers.  Reducing cost sharing presents tradeoffs – increasing premiums or government subsidies for low-income consumers – but could also be a factor helping to sustain enrollment growth.

The Kaiser Family Foundation 2015 Survey of Health Insurance Marketplace Assister Programs and Brokers was designed and analyzed by KFF researchers and administered by Davis Research. This nationwide survey was conducted through an online questionnaire from February 11, 2016 through March 4, 2016.

To recruit Assister Program survey participants, we asked officials CMS and from States operating SBM or FPM Marketplaces to provide contact information for the directors of their certified Assister Programs. In addition, we requested contact information for the directors of enrollment assistance activities in each of the FQHCs from HRSA.  All Assister Programs received an email with a link to the survey inviting the director to participate. In the event the person receiving the survey was not the appropriate person to complete it, they were asked to provide the contact name and email for the appropriate person within their organization.

To analyze results, we assigned Assister Programs to one of four types based on their primary source of funding.  The first type, Navigators, were those identified by Marketplace officials contracted with and received grant funding directly from the Marketplace.     The second type, FEAP, were those identified by CMS as contractors that operate in certain FFM states and that otherwise act as Navigators.  We tracked FEAP responses separately in the survey, but for most data analysis presented in this report we combined responses of FEAPs and Navigators.   The third type, FQHCs, were those that received grant funding from HRSA to provide enrollment assistance. We identified FQHCs using the contact list provided by HRSA.  A small percentage of FQHC Programs receive both HRSA grants and Marketplace Navigator grant funding; these were categorized as Navigators for our analysis.  All other Assister Programs certified to provide assistance in Marketplaces were designated as CACs.

A total of 5,094 Programs were invited to participate in the study, and 688 Programs responded and were included (for a response rate of 13.5%).  Because response rates varied by Program type, data were weighted to reflect the distribution in the initial sample by Program type and Marketplace type; for our analysis, FFM and FPM Marketplaces were grouped together.  (FFM + FPM, and SBM).  Weighted and unweighted proportions of the final sample by Program type are shown in the table below.

Nationwide Estimates

Using responses provided by Assister Programs in the study, we were able to estimate the number of Assister Program staff and the number of consumers they helped with eligibility and enrollment in Medicaid/CHIP and Qualified Health Plans during the second Open Enrollment period nationwide, by extrapolating response data to the national level.  Survey participants were asked to provide the number of full-time equivalent Assisters in their Program and the number of consumers helped.  Respondents who did not provide a numeric value for the number of consumers helped were asked to estimate a number using a range of options.  In making our calculation, we used the midpoint value for responses that provided a range of numbers of consumers helped.   Non-responses were imputed based on the type of Assister Program. A limitation of our national-estimates methodology is that outliers in our response data (i.e. assister programs that helped over 10,000 people during open enrollment, or who had more than 100 staff), when extrapolated to the national level may have an outsize influence on our estimates of total helped and total assister staff nationwide.

We also surveyed the work of Assister Programs outside of Open Enrollment as they helped people apply for Special Enrollment Periods, report mid-year changes to the Marketplace, and resolve post-enrollment problems.  Using response data provided by returning Assister Programs, we were able to estimate the number of people nationally who received help from Assister Programs between the first and second Open Enrollment periods with each of these types of issues.

Brokers

To recruit brokers in the Federally-Facilitated Marketplace (FFM) states, we obtained contact information from a file of brokers in the FFM states, made publicly available through healthcare.gov.²⁵   To obtain broker contact information from the SBM and FPM states, we asked Marketplaces to provide contact information, and when that was not provided, compiled contact information that was publicly available on Marketplace websites.   As we estimate that there are tens of thousands of brokers selling non-group Marketplace policies nationwide, we drew a sample of 9,432 brokers based on their distribution by Marketplace type (FFM, FPM, or SBM).  Our general sampling rule was to randomly select 10% of all contacts in each state; we oversampled in ten states where we had fewer than 500 contacts to begin with.  Because we did not have a complete sample of Marketplace brokers in all states, we were not able to compute national estimates of the numbers of consumers helped by brokers.

Out of the 9,432 brokers who were invited to participate in the study, 418 responded and were included (for a response rate of 4%).

Toplines and Margin of Sampling Error

Survey toplines with overall frequencies of both Assister Programs and Brokers for all survey questions are available at https://www.kff.org/health-reform/report/2016-survey-of-health-insurance-marketplace-assister-programs-and-brokers.

The sample size and margin of sampling error (MOSE) for the total sample and key subgroups of Assister Programs are shown in the table below. All statistical tests of significance account for the effect of weighting.

 

1. Center for Consumer Information and Insurance Oversight, “Navigator Grant Recipients for States with Federally-facilitated or State Partnership Marketplace,” available at http://www.cms.gov/CCIIO/Programs-and-Initiatives/Health-Insurance-Marketplaces/Downloads/Navigator-Grantee-Summaries-UPDATED-05-05-15.pdf. ↩︎
2. In year 3, the 14 SBM states were California, Colorado, Connecticut, District of Columbia, Hawaii, Idaho, Kentucky, Maryland, Massachusetts, Minnesota, New York, Rhode Island, Vermont and Washington.  The 3 consumer assistance FPM states were Delaware, New Hampshire and West Virginia.  Arkansas and Illinois were approved for status as a consumer assistance FPM in year one, but have since ceased providing state support for consumer assistance.  The FPM states were included with FFM states for this analysis. ↩︎
3. During the third Open Enrollment period, FEAPs operated in Arizona, Florida, Georgia, Indiana, Louisiana, North Carolina, New Jersey, Ohio, Pennsylvania, and Texas. ↩︎
4.   Twelve CAP programs received limited supplemental grants for FY 2015:  California, Connecticut, District of Columbia, Maine, Maryland, Massachusetts, Michigan, Mississippi, Missouri, New York, North Carolina, and Vermont. ↩︎
5. https://www.cms.gov/Newsroom/MediaReleaseDatabase/Fact-sheets/2015-Fact-sheets-items/2015-12-30.html ↩︎
6. http://www.cbpp.org/research/remote-identity-proofing-impacts-on-access-to-health-insurance ↩︎
7. https://www.cms.gov/Newsroom/MediaReleaseDatabase/Fact-sheets/2016-Fact-sheets-items/2016-03-11.html ↩︎
8. http://kffhealthnews.org/news/paperwork-inconsistencies-causing-thousands-to-lose-obamacare-subsidies/ ↩︎
9. https://aspe.hhs.gov/basic-report/health-plan-choice-and-premiums-2016-health-insurance-marketplace ↩︎
10. See for example A Frakt, “Too Many Choices,” Academy Health, 2013, available at http://blog.academyhealth.org/too-many-choices/ ↩︎
11. Patient Cost Sharing in Marketplace Plans, 2016, Kaiser Family Foundation, available at https://modern.kff.org/health-costs/issue-brief/patient-cost-sharing-in-marketplace-plans-2016/ ↩︎
12. “Five Facts About Deductibles” CMS blog post November 17, 2015, available at https://blog.cms.gov/2015/11/ ↩︎
13. 2016 Survey of Non-Group Health Insurance Enrollees, Wave 3, Kaiser Family Foundation, available at https://modern.kff.org/health-reform/poll-finding/survey-of-non-group-health-insurance-enrollees-wave-3/ ↩︎
14. http://www.coveredca.com/news/ ↩︎
15. Personal communication, April 29, 2016 ↩︎
16. Personal communication, April 29, 2016. ↩︎
17. http://insurancenewsnet.com/oarticle/unitedhealthcare-to-stop-paying-insurance-agents-for-selling-aca-health-plans ↩︎
18. http://www.usatoday.com/story/news/nation/2016/03/31/insurers-cut-commissions-restrict-when-and-what-plans-people-buy/82210946/ ↩︎
19. http://kffhealthnews.org/news/licking-wounds-insurers-accelerate-moves-to-limit-health-law-enrollment/ ↩︎
20. http://ctmirror.org/2016/02/12/state-says-unitedhealthcare-cant-ax-broker-commissions/ ↩︎
21. http://csahu.org/images/B-4_87_Prohibition_on_Differing_Commission_Structures_for_the_Sale_of_Health_Benefit_Plans.pdf ↩︎
22. http://insurance.ky.gov/Documents/AdvOp16_01AgentCommissionPayments010616.pdf ↩︎
23. http://www.benefitspro.com/2016/04/08/covered-california-posts-agent-comp-draft ↩︎
24. See for example, M Buettgens et al, “More than 10 Million Uninsured Could Obtain Marketplace Coverage through Special Enrollment Periods,” November 2015, available at http://www.urban.org/sites/default/files/alfresco/publication-pdfs/2000522-More-than-10-Million-Uninsured-Could-Obtain-Marketplace-Coverage-through-Special-Enrollment-Periods.pdf ↩︎
25. https://localhelp.healthcare.gov/ ↩︎