United States Government Accountability Office Report to Congressional Requesters FEDERAL HEALTH March 2018 INSURANCE EXCHANGE CMS Needs to Ensure Complete, Accurate Data on Terminations of Coverage for Nonpayment of Premiums GAO-18-269 March 2018 FEDERAL HEALTH INSURANCE EXCHANGE CMS Needs to Ensure Complete, Accurate Data on Terminations of Coverage for Nonpayment of Highlights of GAO-18-269, a report to Premiums congressional requesters Why GAO Did This Study What GAO Found CMS has noted that it is important for In 2015, 9.2 million individuals enrolled in the federal health insurance exchange enrollees to maintain continuous health in 37 states. Eligible individuals (e.g., U.S. citizens or those lawfully present in the insurance coverage to ensure the United States) are able to enroll in health coverage during the annual open stability of the FFE. Certain rules allow enrollment period. Outside of open enrollment, eligible individuals may enroll in for enrollment flexibilities—such as coverage or change their coverage selection during special enrollment periods. special enrollment periods and a 3- Individuals may enroll under a special enrollment period if, for example, they lost month grace period that is allowed their coverage from another source, such as Medicaid or an employer, or due to before coverage is terminated for relocation. Under federal regulations, enrollees may not sign up for coverage recipients of federal income-based under a special enrollment period citing loss of coverage if the coverage was lost subsidies who default on their due to nonpayment of premiums. premiums. However, some issuers have stated that these rules could be About half (53 percent) of the 2015 federally facilitated exchange (FFE) enrollees misused, resulting in non-continuous maintained continuous health insurance coverage throughout the year—that is, coverage. There are little data on the they began coverage between January 1 and March 1, 2015, and maintained it extent to which enrollees maintain through December 31, 2015. These individuals had an average of 11.6 months continuous coverage during a year of coverage. The remaining 47 percent of FFE enrollees started their coverage and, more specifically, on the extent to later or ended it during the year; they averaged 5.0 months of coverage. which coverage is terminated for Enrollees could have voluntarily ended coverage—due to gaining other nonpayment of premiums. coverage, for example—or have had it terminated by the Centers for Medicare & GAO examined (1) the extent to which Medicaid Services (CMS) or the issuers of coverage for valid reasons, including FFE enrollees maintained coverage in losing eligibility for exchange coverage or for nonpayment of premiums. 2015 and (2) the extent to which CMS has reliable data on termination of CMS does not have reliable data on issuer-generated terminations of coverage enrollees’ coverage for nonpayment of for enrollees’ nonpayment of premiums. Although CMS and issuers share data premiums. GAO analyzed CMS’s 2015 on the terminations each generates and reconcile their data on a monthly basis FFE enrollment data (the most recent to ensure data accuracy, the agency does not require issuers to consistently year of available data); interviewed report data on the reasons for terminations. Officials told us they do not track CMS officials and selected issuers; these data because they are not critical to ensure the accuracy of the federal and reviewed applicable laws and subsidy amounts—which is the main function of the monthly reconciliation guidance from CMS. process. Further, CMS lacks a transparent process to ensure the accuracy of these data, as the monthly reconciliation files transmitted between CMS and What GAO Recommends issuers do not include a place to capture data on termination reasons. Issuers GAO recommends that CMS ensures it said that they are therefore unable to ascertain whether data they provide on the has (1) complete data on terminations reasons for termination match CMS’s data, and thus they cannot make of coverage for nonpayment of corrections where necessary. premiums; and (2) a transparent process to reconcile discrepancies and The agency’s lack of reliable data on terminations for nonpayment limits its ability ensure the accuracy of these data. The to effectively oversee certain federal regulations. For example, because CMS is Department of Health and Human not systematically tracking these data, it cannot tell whether enrollees applying Services concurred with both for coverage under a special enrollment period had lost their coverage for recommendations. nonpayment of premiums—in which case they would be ineligible for the special enrollment period per federal regulations. CMS could capitalize on its existing process, already familiar to issuers, by adding a variable that captures data on termination reasons to the monthly reconciliation file. By taking this step, in addition to requiring issuers to report these data, CMS could help ensure it has reliable and transparent data on terminations of enrollee coverage for View GAO-18-269. For more information, contact John E. Dicken at (202) 512-7114 or nonpayment of premiums, and it could use these data to assess the effects of dickenj@gao.gov. CMS policies and the overall stability of the exchange. United States Government Accountability Office Contents Letter 1 Background 6 Just over Half of FFE Enrollees Maintained Continuous Coverage throughout 2015; Length of Coverage Varied by Enrollee Characteristics 11 CMS Lacks Complete and Transparent Data on Terminations of Enrollee Coverage for Nonpayment of Premiums 16 Conclusions 19 Recommendations 19 Agency Comments and Our Evaluation 19 Appendix I Demographic and Coverage Characteristics of Federally Facilitated Exchange Enrollees, 2015 21 Appendix II Average Length of Coverage for Federally Facilitated Exchange Enrollees, 2015 26 Appendix III Comments from the Department of Health and Human Services 30 Appendix IV GAO Contact and Staff Acknowledgments 33 Tables Table 1: Demographic Characteristics of Continuously Covered, and All Other, Federally Facilitated Exchange Enrollees, 2015 21 Table 2: Coverage Characteristics for Continuously Covered, and All Other, Federally Facilitated Exchange Enrollees, 2015 22 Table 3: Proportion of Federally Facilitated Exchange Enrollees Who Maintained Continuous Coverage throughout 2015, by State 24 Table 4: Average Length of Coverage for Federally Facilitated Exchange Enrollees by Demographic Characteristics, 2015 26 Table 5: Average Length of Coverage for Federally Facilitated Exchange Enrollees by Coverage Characteristics, 2015 27 Page i GAO-18-269 Federal Health Insurance Exchange Enrollment Table 6: Average Length of Coverage for Federally Facilitated Exchange Enrollees by State, 2015 28 Figures Figure 1: Data Transfer between CMS and Issuers for Initial Enrollment, Subsequent Updating, and Reconciliation Processes in the Federally Facilitated Exchange 9 Figure 2: Length of Coverage for Federally Facilitated Exchange Enrollees, 2015 12 Figure 3: Length of Coverage for Federally Facilitated Exchange Enrollees by Initial Enrollment Period, 2015 14 Abbreviations APTC Advance Premium Tax Credit CMS Centers for Medicare & Medicaid Services FFE federally facilitated exchange HHS Department of Health and Human Services OEP open enrollment period PPACA Patient Protection and Affordable Care Act SEP special enrollment period This is a work of the U.S. government and is not subject to copyright protection in the United States. The published product may be reproduced and distributed in its entirety without further permission from GAO. However, because this work may contain copyrighted images or other material, permission from the copyright holder may be necessary if you wish to reproduce this material separately. Page ii GAO-18-269 Federal Health Insurance Exchange Enrollment Letter 441 G St. N.W. Washington, DC 20548 March 9, 2018 The Honorable Orrin Hatch Chairman Committee on Finance United States Senate The Honorable Greg Walden Chairman Committee on Energy and Commerce House of Representatives The Honorable Fred Upton House of Representatives The Patient Protection and Affordable Care Act (PPACA) required, beginning in 2014, the establishment of exchanges (or marketplaces) in each state where eligible consumers can directly compare and select among a variety of qualified health plans offered by participating private issuers. 1 The Centers for Medicare & Medicaid Services (CMS)—the agency within the Department of Health and Human Services (HHS) that oversees the federally facilitated exchange (FFE) in 39 states that elected to use the FFE in 2017—has noted that, while consumers enrolled in health insurance need to maintain continuous coverage for the entire year to ensure a stable and affordable marketplace, exchange rules should also allow individuals the flexibility to make changes to their enrollment in response to life circumstances. 2 However, some issuers offering health 1 Pub. L. No. 111-148, §§ 1501, 10106, 124 Stat. 119, 244, 909 (2010), as amended by the Health Care and Education Reconciliation Act of 2010 (HCERA), Pub. L. No. 111-152, §§ 1002, 1004, 124 Stat. 1029, 1032, 1034 (2010) (codified at 26 U.S.C. § 5000A). In this report, references to PPACA include any amendments made by HCERA. To be eligible to enroll in the exchanges, an individual must be a U.S. citizen or national or otherwise be lawfully present in the U.S.; not be incarcerated (unless pending disposition of the charges); and reside in the state in which the exchange operates. 2 Each state may operate its own exchange or elect to use the FFE. (For purposes of this report we are treating the District of Columbia as a state.) States that operate their own exchanges may also use the FFE for certain functions. In 2017, the 12 states that operated their own exchanges and did not use the FFE were: California, Colorado, Connecticut, the District of Columbia, Idaho, Maryland, Massachusetts, Minnesota, New York, Rhode Island, Vermont, and Washington. Page 1 GAO-18-269 Federal Health Insurance Exchange Enrollment insurance plans in the FFE have noted that certain of these rules could also undermine enrollees’ incentives to maintain continuous coverage. Specifically, issuers have raised concerns about enrollment that occurs outside of the open enrollment period (OEP), which generally runs near the start of the coverage year, and about enrollees ending their coverage before the end of the year. Individuals may use special enrollment periods (SEP) to sign up for coverage outside of open enrollment if they experience a triggering event, such as losing coverage from another source including Medicaid or an employer. 3 However, issuers have noted that an SEP could be misused if individuals, instead of signing up during OEP, delayed enrollment until they were sick and had imminent health care costs to be covered. Some issuers also noted that CMS could be allowing individuals whose coverage was terminated due to nonpayment of premiums to re-enroll under an SEP due to loss of coverage although, under federal regulations, this is not allowed. 4 Finally, some issuers have stated that enrollees receiving Advance Premium Tax Credits (APTC), which reduce the cost of coverage for eligible individuals, may intentionally stop paying their premiums before the end of the year to take 3 45 C.F.R. §§ 155.410,155.420 (2016). The triggering events for SEPs have changed over time; according to CMS, as of 2017, there are six broad categories: 1) loss of qualifying coverage from other sources (e.g., employer, Medicare, Medicaid); 2) change in household size (e.g., through marriage, birth or adoption of a child); 3) change in primary place of living (e.g., relocation to a new zip code); 4) change in eligibility for exchange coverage or in help paying for coverage (e.g., becoming newly eligible for FFE coverage or changes in household income); 5) enrollment or plan error on the part of the exchange; and 6) other situations (e.g., becoming incapacitated). The open enrollment period generally ran from November 1 of the prior year to January 31 of the coverage year, but the end date has changed in recent years. Specifically, for coverage year 2015, it ended on February 15 of the coverage year; in 2016 and 2017, it ended on January 31 of the coverage year; and for coverage years 2018 and beyond, it will end on December 15 of the prior year. 4 45 C.F.R. § 155.420(e) (2016). Page 2 GAO-18-269 Federal Health Insurance Exchange Enrollment advantage of the 3-month coverage grace period allowed by law while still meeting requirements for PPACA’s individual mandate. 5 There is, however, little data about the extent to which these specific problems have occurred. 6 Further, enrollees may stop paying premiums and terminate their coverage for valid reasons including unexpected reductions in household income or gaining other government or employer-sponsored coverage, and other research has found that short periods of coverage were common in the individual market even prior to PPACA. 7 Officials from both CMS and issuers indicated that individuals often do not provide notification that they are ending their coverage due to a change of circumstance. Instead, they stop paying premiums and let their exchange policies “passively” terminate. You asked us to examine the extent to which exchange enrollees maintained coverage and paid their premiums during the plan year. This report: 5 PPACA provides premium tax credits for individuals purchasing coverage through the exchanges if they meet applicable income requirements and are not eligible for coverage from another source. In this report, we refer to advance payments of these federal subsidies to issuers—rather than directly to enrollees—as APTCs. APTC recipients are entitled to receive a 3-month grace period if they default on their premium payments. For non-APTC recipients, grace periods are governed by state law and are typically 1 month. If APTC recipients do not pay their outstanding premiums during the grace period, the individuals are responsible for any medical costs incurred during the last 2 months of the grace period. PPACA required that, with some exceptions, individuals maintain qualifying health insurance during the year or pay a tax penalty. This requirement is referred to as the individual mandate. In December 2017, a law repealing the tax penalty imposed for failure to comply with the individual mandate was enacted. As a result, beginning January 1, 2019, individuals who fail to comply with the individual mandate will no longer face a tax penalty. See Pub. L. No. 115-97, § 11081 (Dec. 22, 2017). 6 Some research has reported more general enrollment trends. For example, one industry study estimated that for a group of 13 issuers, 17 percent of enrollment through June 2015 was through an SEP. The study also indicated that SEP enrollees were more likely to drop coverage—5 percent per month, compared to 3.5 percent for other enrollees. See Oliver Wyman, “Special Enrollment Periods and the Non-Group, ACA-Compliant Market,” New York, NY, Feb. 2016. According to a survey of 2,763 consumers eligible for individual coverage, approximately 21 percent reported stopping paying their premiums in 2015. Approximately 87 percent of individuals who reported stopping paying their premiums also reported repurchasing a plan in 2016, and about half of these purchased the same plan. The survey did not report on why the individuals dropped coverage. See McKinsey & Company, “2016 OEP: Reflection on Enrollment,” (Washington, DC; May 2016). 7 See, for example, Henry J. Kaiser Family Foundation, “How Many People Have Nongroup Health Insurance?” Menlo Park, Calif (2013). Page 3 GAO-18-269 Federal Health Insurance Exchange Enrollment 1. describes the extent to which enrollees who purchased health insurance coverage through the FFE in 2015 maintained their coverage during the plan year; and 2. evaluates the extent to which CMS has reliable data on the rate at which enrollees in the FFE are terminated from coverage due to nonpayment of premiums. To describe the extent to which enrollees who purchased health insurance coverage through the FFE in 2015 maintained their coverage during the plan year, we obtained and analyzed enrollment and demographic data from CMS. 8 According to CMS officials, 2015 was the most recent full year of data available at the time of our review. Specifically, CMS provided us with data from its centralized enrollment system, the Multidimensional Insurance Data Analytics System, for each enrollee who obtained health insurance coverage—that is, selected a plan and paid their first premium—through the FFE for coverage year 2015. 9 These data included, among other information, the enrollees’ coverage start and end dates; type of plan, premium, and any APTC amounts; and demographic information reported by enrollees, including gender, state of residence, and household income. 10 We used these to analyze the extent to which enrollees maintained continuous FFE coverage throughout 2015—which we defined as coverage that began from January 1, 2015, to March 1, 2015, and ended on December 31, 8 We focused our review on the 37 states that used the FFE in 2015 because enrollment data from state-based exchanges that did not use the FFE were not readily available. In 2015, the 14 states that operated their own exchanges and did not use the FFE were: California, Colorado, Connecticut, the District of Columbia, Hawaii, Idaho, Kentucky, Maryland, Massachusetts, Minnesota, New York, Rhode Island, Vermont, and Washington. 9 CMS provided data on all individuals who were ever enrolled in 2015 coverage through the FFE, including any enrollees whose coverage was ultimately terminated by the agency because citizenship status or other key information could not be verified. According to the Social Security Administration’s May 2017 Death Master File, 25,599 of the enrollees who obtained 2015 health insurance coverage through the FFE died during 2015. We excluded these enrollees from our analysis. We also excluded dental coverage for the purposes of our analysis. 10 We did not examine potential differences in health costs or the health status of individuals in the two groups because these data were not available from CMS’s centralized enrollment system. Page 4 GAO-18-269 Federal Health Insurance Exchange Enrollment 2015. 11 We compared the demographic and coverage characteristics of enrollees who maintained continuous coverage throughout 2015 with those of all other enrollees. We also examined the demographic and coverage characteristics of all FFE enrollees in 2015. Further, CMS provided us with FFE enrollment data for the time period of January 1, 2016, through June 30, 2016, which we used to analyze the extent to which the 2015 FFE enrollees re-enrolled for coverage year 2016. 12 We assessed the reliability of the data provided by CMS in several ways, including discussing the reliability of the data with CMS officials, reviewing relevant CMS data manuals and other documentation, performing manual and electronic tests of the data to identify any outliers or anomalies, and comparing the data with data from published sources. We also interviewed representatives from a small selection of issuers to discuss the reliability of relevant issuer-reported data maintained in CMS’s centralized enrollment system. 13 We determined that the data were sufficiently reliable for the purposes of our reporting objectives. To evaluate the extent to which CMS has reliable data on the rate at which FFE enrollees are terminated from coverage due to nonpayment of premiums, we examined data from CMS’s enrollment system on the reported reasons for termination of enrollee coverage in 2015. For a small selection of enrollees identified by CMS as having their coverage terminated for nonpayment, we examined the termination reasons listed in issuer data. 14 We also interviewed officials from CMS, industry 11 We picked a start date of no later than March 1, 2015, because open enrollment ended on February 15, 2015, so individuals enrolling during the open enrollment time frame would generally have had coverage start dates on or before March 1. In addition, individuals who began coverage by March 1, 2015, and remained enrolled through the end of the year would have had at least 10 months of continuous coverage, which fulfilled PPACA’s individual mandate for health coverage. Individuals could have held multiple exchange policies purchased through the FFE during this time frame; we considered them to have maintained coverage as long as they did not have any uncovered days in between policies. To examine the average number of months enrollees held coverage, we assumed that months are 30.4 days long, as this was the average month length in 2015. 12 According to CMS officials, data for the remainder of 2016 were not available at the time of our request because it takes at least 6 months for enrollment data to stabilize. 13 For example, we interviewed issuer representatives about the reliability of coverage end dates, which are revised and transmitted to CMS by issuers when issuers terminate policies. We interviewed representatives from Aetna, Anthem, Blue Cross Blue Shield of North Carolina, Health Care Service Corporation, and Independence Blue Cross. These issuers together held 26 percent of total exchange enrollment in 2015. 14 Specifically, we obtained data on a selection of between 25 and 30 enrollees each from Aetna, Anthem, and Health Care Service Corporation. Page 5 GAO-18-269 Federal Health Insurance Exchange Enrollment stakeholders, and selected issuers to learn about how the data on termination reasons are collected. 15 In addition, we reviewed relevant PPACA provisions and federal regulations and guidance, including technical guidance from CMS governing the transfer of information between CMS and issuers participating in the FFE, and determined whether the agency’s relevant policies and procedures are consistent with standards for internal control in the federal government. 16 We conducted this performance audit from July 2016 to March 2018 in accordance with generally accepted government auditing standards. Those standards require that we plan and perform the audit to obtain sufficient, appropriate evidence to provide a reasonable basis for our findings and conclusions based on our audit objectives. We believe that the evidence obtained provides a reasonable basis for our findings and conclusions based on our audit objectives. The exchanges (including the FFE and those operated by individual Background states) provide a seamless, single point of access for eligible individuals to enroll in qualified health plans. For the FFE, CMS established a website—Healthcare.gov—as the public portal through which individuals may apply for coverage and select and enroll in health plans, which are offered at different levels of coverage, or “metal tiers”—bronze, silver, gold, and platinum—that reflect the percentage of covered medical expenses estimated to be paid by the insurer. 17 The data that individuals provide in their application is stored in the FFE’s centralized enrollment system, which is maintained by CMS. 15 The industry stakeholders we interviewed were America’s Health Insurance Plans and the Blue Cross Blue Shield Association. 16 GAO, Standards for Internal Control in the Federal Government, GAO-14-704G (Washington, D.C.: September 2014). Internal control is a process effected by an entity’s oversight body, management, and other personnel that provides reasonable assurance that the objectives of an entity will be achieved. 17 For example, bronze plans tend to have the lowest premiums, but subject consumers to the highest out-of-pocket costs (such as deductibles) when they receive health care services, while platinum plans tend to have the highest premiums and the lowest out-of- pocket costs. Individuals may also apply in person, over the phone, or by mailing in a paper application. Page 6 GAO-18-269 Federal Health Insurance Exchange Enrollment Although CMS oversees the centralized enrollment system, both CMS and issuers have shared responsibility for enrollment and coverage functions once individuals apply for coverage: • CMS is responsible for determining an individual’s eligibility for coverage and income-based federal subsidies, enrolling the individual, and processing subsequent coverage changes or terminations. For example, individuals may change their existing coverage by signing up under an SEP due to the birth of a child or relocation, or they may voluntarily terminate their coverage, or CMS may terminate coverage if the agency is unable to verify key information such as citizenship status. CMS is also responsible for making payments for APTCs and determining whether an enrollee is eligible for any cost-sharing reductions that lower enrollees’ out-of- pocket costs for expenses, such as deductibles and copayments. • Issuers are responsible for, among other things, collecting premiums from enrollees, arranging for coverage through provider networks, and paying claims. Issuers are also responsible for processing, and notifying CMS of, terminations related to nonpayment of premiums or fraud. 18 As a result of this shared responsibility, CMS and issuers notify each other of coverage updates by transferring data back and forth through electronic files known as “transaction files.” It is critical that both issuers and CMS have consistent, accurate, and current information on enrollees, because monthly APTC payments are based on enrollment data in CMS’s centralized system. Federal regulations require CMS to reconcile enrollment information with issuers on at least a monthly basis. 19 Accordingly, CMS and issuers reconcile certain key data elements on a monthly basis through an automated enrollment reconciliation process, in which issuer and CMS data are compared and discrepancies are resolved. Through this process, APTC amounts and their effective dates are compared and reconciled. CMS’s data system is considered to be correct when considering discrepancies on overall enrollment counts or with key data elements, such as coverage start and end dates between issuer and CMS data. Therefore, CMS will not change the APTC 18 These are generally the only types of terminations of coverage that issuers are allowed to make. Issuers told us that terminations for fraud are rare. 19 See 45 C.F.R. § 155.400(d) (2016). Page 7 GAO-18-269 Federal Health Insurance Exchange Enrollment payments based on issuers’ data that may differ from CMS’s data unless there are significant discrepancies. 20 There are several specific steps involved in transferring data between CMS and issuers for initial enrollment, subsequent updates, and reconciliation (see fig. 1 for a high-level overview of the data transfer process): • Initial enrollment: CMS forwards an outbound electronic transaction file to the issuer with information on the applicant, the plan selection, the premium, and the APTC amount. Once the issuer receives the initial premium payment, the issuer sends an inbound electronic transaction file back to CMS to confirm the enrollment. Issuers may not refuse to issue coverage to an individual CMS has deemed eligible once that individual has made the initial premium payment. Transaction files are transmitted electronically on a daily basis. • Subsequent changes/terminations: Subsequent changes to the individual’s coverage may be initiated by enrollees, CMS, or issuers. For example, enrollees may request changes to their coverage through the portal if they experience a change in circumstance (such as needing to enroll under an SEP due to the birth of a child, or to terminate their coverage if they move to a different state); CMS may terminate coverage if the agency cannot verify key eligibility information (such as citizenship status); or issuers may terminate coverage if enrollees fail to pay their premiums. If CMS initiates changes in coverage, it notifies issuers through subsequent outbound transaction files, and similarly, if issuers initiate changes they notify CMS through subsequent inbound transaction files. • Monthly reconciliation: CMS sends issuers a snapshot of key elements of the enrollment data in its centralized enrollment system in an outbound reconciliation file. Issuers compare the data from the file to their enrollment systems and identify missing enrollments or other discrepancies. Issuers make updates as necessary and send CMS an inbound reconciliation file with information about current enrollees, cancellations, and terminations in their systems. 21 CMS then performs 20 CMS stated that starting with May 2016 payments, issuers would be paid the policy- based payments as calculated based on CMS data, except in cases of extreme variation (greater than 25 percent) from the manual payment amount calculated. In those cases, CMS will work with issuers to resolve the discrepancy. 21 Issuers may also dispute the results of the automated comparison and submit the dispute to CMS for resolution. Page 8 GAO-18-269 Federal Health Insurance Exchange Enrollment an automated comparison of the data in the inbound reconciliation files with its centralized enrollment system and identifies any further discrepancies that may need to be resolved either by CMS or issuers. If necessary, CMS makes further updates to its data. Figure 1: Data Transfer between CMS and Issuers for Initial Enrollment, Subsequent Updating, and Reconciliation Processes in the Federally Facilitated Exchange Notes: PPACA provides certain subsidies to eligible individuals, including Advance Premium Tax Credits (APTC), that reduce premium costs. In an April 2017 final rule, CMS implemented several actions that, in part, responded to issuer concerns about special enrollment periods and Page 9 GAO-18-269 Federal Health Insurance Exchange Enrollment stability of enrollment. 22 Specifically, CMS stated that the agency would require documentation from all individuals applying to enroll in coverage under an SEP to verify their eligibility for the SEP prior to enrollment. 23 CMS also stated that, starting in June 2017, it would allow issuers, subject to state law, to apply a new premium payment to an individual’s past due payments before applying that premium towards a new enrollment. 24 CMS stated that issuers would be allowed to refuse to provide coverage to an enrollee applying under an SEP due to loss of existing coverage if the issuer had previously terminated the enrollee’s coverage for nonpayment of premiums, unless the enrollee paid the past due premiums. CMS further stated that this provision was intended to encourage individuals to maintain continuous coverage rather than start and stop coverage (and thereby avoid incurring past due premiums). 25 22 See 82 Fed. Reg. 18,346 (April 18, 2017) (codified at 45 C.F.R. pts. 147, 155, and 156). 23 The final rule expanded on previous agency efforts to curb potential abuses of SEPs, including a special enrollment confirmation process that CMS had begun in June 2016, under which CMS required documentation of enrollees’ eligibility for the most common SEP categories. Enrollees who were selected for review and could not prove their eligibility would have their coverage terminated. Since implementing the process, CMS reported seeing a 20 percent drop in SEP enrollments in 2016 over 2015. See CMS: “Pre- Enrollment Verification for Special Enrollment Periods,” (Dec. 20, 2016), accessed August 28, 2017, https//www.cms.gov/CCIIO/resources/fact-sheets-and-faqs/index.html. 24 The final rule applies to past debt owed for coverage from an issuer in the same controlled group, as defined by the Internal Revenue Code, within the past 12 months. 25 In the April 2017 final rule, CMS also stated that the agency would explore other policies that could encourage enrollees to maintain continuous coverage, such as requiring maintenance of continuous coverage without a 63-day break if individuals wish to avoid the pre-existing condition exclusion and allowing waiting periods to be imposed under certain circumstances. Page 10 GAO-18-269 Federal Health Insurance Exchange Enrollment Approximately 4.9 million enrollees (53 percent of the 9.2 million FFE Just over Half of FFE enrollees in 2015) maintained continuous coverage throughout the year— Enrollees Maintained that is, their coverage began between January 1 and March 1, 2015, and lasted through December 31, 2015. These individuals therefore had 10 or Continuous Coverage more months of continuous coverage, with an average length of coverage throughout 2015; of 11.6 months. Most of these enrollees (83 percent) re-enrolled in coverage by June 2016. Length of Coverage Varied by Enrollee The remaining 4.3 million enrollees (47 percent of the FFE enrollees in 2015) did not maintain continuous FFE coverage throughout the year, as Characteristics defined above. 26 The average length of coverage for these enrollees was about 5.0 months and, for most (72 percent), coverage ended prior to the end of the year. (See fig. 2 for information on enrollee length of coverage.) Of the 4.3 million enrollees, 38 percent re-enrolled in exchange coverage for 2016, although enrollees that held coverage through the end of the year—regardless of their length of coverage—were far more likely to have re-enrolled than enrollees whose coverage ended prior to the year’s end. 27 26 This category includes individuals whose coverage CMS terminated during the year because the agency could not verify their citizenship status or other key information. CMS reported that, during 2015, the agency terminated coverage for about 500,000 enrollees whose citizenship status or other key information could not be verified. See CMS, December 31, 2015 Effectuated Enrollment Snapshot (Baltimore, Md.: Mar. 11, 2016). Of the 4.3 million enrollees, about 264,000 had gaps in coverage—that is, they dropped and resumed coverage through the FFE at least once during the year, with at least one day without coverage in between. Their average length of total coverage across all separate periods of coverage was 8.3 months. The total number of uncovered days in between these enrollees’ periods of exchange coverage ranged from 1 to 333 days, with an average of 66 days. 27 Specifically, of the 4.3 million enrollees, 81 percent of those with coverage ending December 31, 2015, re-enrolled in coverage by June 2016, as compared to 21 percent for those whose coverage ended earlier. This higher rate for enrollees with coverage ending December 31, 2015, is likely in part because some enrollees who remained enrolled in December 2015 were automatically re-enrolled in coverage for 2016. (If eligible consumers do not make an active re-enrollment selection, CMS automatically re-enrolls them in their existing plan if it remains available. If that plan is no longer available, CMS generally automatically re-enrolls consumers in a similar plan if it identifies one, or the consumers would have to actively re-enroll if CMS does not identify a similar plan.) Page 11 GAO-18-269 Federal Health Insurance Exchange Enrollment Figure 2: Length of Coverage for Federally Facilitated Exchange Enrollees, 2015 Notes: We defined enrollees with 10 or more months of continuous coverage in 2015 as those who began coverage by March 1, 2015, and maintained this coverage through December 31, 2015. We defined enrollees with gaps in coverage as those who had multiple periods of FFE coverage in 2015 with one or more uncovered days in between. For all other enrollees, to determine total months covered, we assumed months to be 30.4 days long and rounded to the nearest month. We excluded dental coverage and the 25,599 individuals who died in 2015, per data from the Social Security Administration, from this analysis. In general, we did not find notable differences in attributes of enrollees’ coverage (for example, by benefit level of selected plan or monthly premium after APTC) or enrollee demographics when comparing the two groups of enrollees—those who maintained continuous coverage throughout 2015, and those who did not. (For data on coverage and demographics of FFE enrollees who did maintain continuous coverage throughout 2015 and those who did not, see app. I.) However, in examining the demographic and coverage characteristics of all FFE enrollees, we found that enrollees with certain characteristics Page 12 GAO-18-269 Federal Health Insurance Exchange Enrollment tended to remain covered for a longer period of time in 2015 compared to other enrollees. 28 For example: • Enrollment period. Enrollees who enrolled during the open enrollment period had a higher average length of coverage than enrollees who enrolled through an SEP—9.1 months compared to 5.2 months (see fig. 3). 29 However, more individuals who enrolled through an SEP remained enrolled through December 31, 2015, compared to individuals who enrolled during open enrollment—72 percent compared to 64 percent. 30 28 We did not test whether there was statistical correlation between these characteristics and average length of coverage. Since our focus was the length of time for which enrollees remained enrolled in exchange coverage, we excluded the approximately 264,000 enrollees (3 percent) that dropped and resumed exchange coverage during the year. In addition, about 25 percent of enrollees had multiple records in 2015. (New records were generated when enrollees reported changes in circumstances, such as changes in income, or when enrollees made changes to their coverage, such as adding a family member after the birth of a child.) For these, we used information from their first record to examine their characteristics, although some characteristics, such as eligibility for APTC and benefit level of selected plan, could have changed between records. 29 For this analysis, we considered any enrollee who enrolled by February 22, 2015, to have enrolled during the open enrollment period (although the open enrollment period ended on February 15, 2015, CMS extended it for certain individuals who attempted but could not complete enrollment by this date). However, some of these enrollees also qualified for SEPs when they enrolled. When we consider these enrollees to have enrolled via SEP, the average length of coverage for enrollees who enrolled via SEP becomes 7.2 months, while the average length of coverage for those that enrolled during the open enrollment period remains at 9.1 months. 30 This difference could have resulted from the fact that OEP enrollees enrolled earlier in the year and thus had more time to experience any changes in their circumstances, such as loss of eligibility for exchange coverage, or gaining eligibility for other coverage, prior to the end of the year. Page 13 GAO-18-269 Federal Health Insurance Exchange Enrollment Figure 3: Length of Coverage for Federally Facilitated Exchange Enrollees by Initial Enrollment Period, 2015 Notes: There were 9.2 million federally facilitated exchange enrollees in coverage year 2015, excluding dental coverage and the 25,599 individuals who died in 2015, per data from the Social Security Administration. We considered any enrollee who enrolled in 2015 exchange coverage by February 22, 2015, to have enrolled during the open enrollment period, and any enrollee who enrolled later in the year and had a special enrollment reason code on their record to have enrolled via a special enrollment period. Enrollment period could not be classified for about 1 percent of enrollees due to data anomalies. To determine total months covered, we assumed months to be 30.4 days long and rounded to the nearest month. Since our focus was the length of time for which enrollees remained enrolled in exchange coverage, we excluded the approximately 264,000 enrollees (3 percent) that dropped and resumed exchange coverage during the year. Page 14 GAO-18-269 Federal Health Insurance Exchange Enrollment • Age. Enrollees aged 55 or older had the highest average length of coverage, while those aged 25 to 34 had the lowest—9.2 months compared to 7.8 months. • Reported household income. APTC-eligible enrollees who reported having a household income between 301 and 400 percent of the federal poverty level had the highest average length of coverage, while those who reported having a household income less than, or equal to, 100 percent of the federal poverty level had the lowest—8.9 months compared to 8.0 months. 31 • Eligibility for APTC. Enrollees who were eligible for APTC had a higher average length of coverage than enrollees who were not eligible for APTC—8.6 months compared to 7.8 months • Benefit level of selected plan. 32 Enrollees who selected higher- benefit, gold plans had the highest average length of coverage, while enrollees who selected lower-benefit catastrophic, plans had the lowest—8.8 months compared to 6.7 months. Enrollees who selected silver plans—the most common plan selection—had an average length of coverage of 8.6 months. • State of residence. Enrollees residing in Maine had the highest average length of coverage, while enrollees residing in Mississippi had the lowest—9.4 months compared to 8.0 months. See appendix II for additional data on the average length of coverage for enrollees by various characteristics. 31 Certain individuals earning up to 400 percent of the federal poverty level are eligible to receive APTC that can reduce premium costs for plans purchased on an exchange. The APTC amount is calculated on a sliding scale whereby individuals with lower income can receive a larger amount. We did not obtain data on reported household income for APTC-ineligible enrollees as according to CMS officials, the FFE does not collect income information for individuals who do not request financial assistance. 32 Issuers offering exchange plans must cover certain categories of benefits at standardized levels of coverage, which are categorized by “metal tiers” of bronze, silver, gold, or platinum, depending on the portion of health care costs expected to be paid by the health plan. In addition to these metal tiers, catastrophic plans are available to certain individuals if they are under age 30 or the lowest-cost bronze plan costs more than 8 percent of their household income. Page 15 GAO-18-269 Federal Health Insurance Exchange Enrollment CMS’s data on terminations of enrollee coverage due to nonpayment of CMS Lacks Complete premiums are not complete and accurate. CMS officials told us that they and Transparent Data collect some information from issuers on their terminations of enrollee coverage for nonpayment of premiums. When issuers terminate policies, on Terminations of the inbound transaction files they send to CMS must include, among Enrollee Coverage for other elements, a revised coverage end date taking the termination into account. CMS uploads these data into its centralized FFE enrollment Nonpayment of system. 33 However, while the issuers may also include codes that Premiums designate the reasons for the terminations, there is no requirement for them to consistently do so. Data on termination codes may therefore not be consistently reported by issuers. CMS officials told us that data on reasons for termination are not tracked because they are not critical to ensure the accuracy of APTC payments—which is the main function of the reconciliation process. Officials stated that key essential variables that CMS does track are whether coverage is effectuated (that is, whether the first premium payment has been made), whether the enrollee is eligible for APTC payments, and whether coverage was terminated. In addition, when issuers do report termination reason codes, these data are not always accurate. Specifically, CMS told us that, historically, issuers may have incorrectly used the nonpayment termination code for other types of terminations, and two issuers we interviewed acknowledged having done so. We compared data on terminations for nonpayment from CMS’s centralized enrollment system with data we obtained from three issuers for a small selection of enrollees. 34 We found that for one large issuer operating in multiple states, the CMS data indicated that coverage for 18 of the 26 enrollees that we examined had been terminated for nonpayment of premiums, while the issuer data indicated that coverage had been terminated for other reasons, in most cases because it had expired at the end of the year. The issuer indicated 33 This revised coverage end date must take into account grace periods, as appropriate. For APTC recipients, CMS regulations specify that the effective date for termination of coverage is the last day of the first month of the 3-month grace period. 45 C.F.R. § 155.430(d)(4) (2016). In other words, if an APTC recipient pays through March and then stops paying premiums, the enrollee would receive coverage for the 3-month grace period through June 30th. However, if no premiums are paid after that grace period, the enrollee would be retroactively terminated with a coverage end date of April 30th and be responsible for any medical costs incurred after that date. 34 We obtained data on a selection of between 25 and 30 enrollees each from Aetna, Anthem, and Health Care Service Corporation. Page 16 GAO-18-269 Federal Health Insurance Exchange Enrollment that it likely reported these year-end terminations to CMS incorrectly as terminations for nonpayment of premiums. 35 CMS has recently taken actions that may improve the reliability of data on terminations for nonpayment, but these actions do not ensure the data are consistently reported and recorded by CMS. Specifically, in July 2017, CMS indicated that it would add new codes to the transaction files for issuers to use to help prevent inaccurate reporting of the nonpayment termination code. CMS told us that it expects issuers to begin using the new codes in 2018. CMS’s data on terminations for nonpayment therefore may be more reliable beginning in 2018. However, CMS has not required issuers to report the termination reasons in the transaction files because, according to CMS officials, these data are not essential to tracking the accuracy of APTC payments. The agency also does not have plans in the near future to use the data in tracking trends in enrollment and termination of enrollee coverage in the FFE to assess the overall stability of the exchange. 36 Further, CMS does not have a transparent, systematic process for issuers to ensure that data on terminations they initiate due to nonpayment are complete and accurate in the CMS system. Issuers we interviewed told us that they are unable to ascertain whether CMS is correctly updating the FFE enrollment system with the termination reason codes issuers provide when policies are terminated. While issuers can determine from the monthly reconciliation files whether CMS has updated certain issuer data for enrollees whose coverage was terminated (for example, the revised coverage end date), the files do not capture data on reasons for termination. Therefore, issuers are unable to determine if the CMS FFE data on termination reason codes matches theirs and make corrections where necessary. 37 Some issuers told us they had requested that CMS add a variable to capture data on termination reasons in the monthly reconciliation files sent to issuers. CMS officials stated that the agency is in the initial stages of exploring whether this would be feasible for CMS 35 For the remaining two issuers, the data from CMS generally matched the issuers’ data. 36 In November 2017, officials told us that CMS is exploring whether to require issuers to report termination reasons in the transaction files, but these discussions are in the very early stages and no decisions have been made yet. 37 While issuers can send updated enrollee information to CMS through another transaction file, there is still no mechanism for issuers and CMS to systematically verify the accuracy of data on reasons for terminations. Page 17 GAO-18-269 Federal Health Insurance Exchange Enrollment and issuers, but that it will require significant resources and time to develop. Although CMS’s recent changes may improve its data, they do not ensure the agency will have complete and transparent data on terminations for nonpayment of premiums. According to federal internal control standards, federal agencies should obtain and use relevant, reliable data to achieve their objectives. 38 Without complete and accurate data, CMS may be allowing enrollees who lost exchange coverage for nonpayment of premiums to re-enroll under SEPs although, under federal regulations, these individuals are ineligible to do so. 39 Issuers reported that this had occurred. 40 CMS officials told us that the agency is exploring options to have its system automatically prevent certain enrollees with prior terminations for nonpayment from enrolling in coverage under an SEP for loss of existing coverage, but noted that this functionality would depend on receiving reliable data on terminations for nonpayment from issuers. 41 Further, without reliable data, CMS may not be able to assess the effects of its April 2017 policy allowing issuers to apply enrollees’ new premium payments toward unpaid premiums over the past 12 months. This is because the agency lacks the complete and accurate data that would be 38 See GAO, Standards for Internal Control in the Federal Government, GAO-14-704G (Washington, D.C.: September 2014). Internal control is a process effected by an entity’s oversight body, management, and other personnel that provides reasonable assurance that the objectives of an entity will be achieved. 39 Under CMS’s regulations, individuals may not enroll under an SEP for loss of minimum essential coverage if they lost their coverage due to nonpayment of premiums. However, it is possible that individuals who lost coverage due to nonpayment of premiums and later re-enrolled in exchange coverage via this type of SEP had lost other minimum essential coverage, such as Medicaid, in between. 40 Issuers were not able to provide us data on the extent to which this had occurred. According to CMS, issuers lack complete data available on circumstances that could lawfully allow such a consumer to re-enroll under a SEP, such as loss of Medicaid and employer sponsored insurance. 41 CMS noted that in the absence of such a system, the agency is now allowing issuers (or issuers in the same control group) to refuse to effectuate new coverage in cases of enrollees who have a record of termination due to nonpayment of premiums unless the individual pays the past due premiums. See 82 Fed. Reg. 18346, 18362-63 (April 18, 2017). However, this change may still not ensure proper use of SEPs. Specifically, issuers lack complete data on CMS’s eligibility decisions. In addition, one issuer told us that their systems may not be able to identify enrollees whose coverage they had terminated for nonpayment of premiums because, once the enrollee’s coverage was terminated and the balance due written off, the enrollee’s records were no longer retained. Page 18 GAO-18-269 Federal Health Insurance Exchange Enrollment necessary to ensure that issuers are correctly identifying enrollees terminated for nonpayment. In its role as administrator of the FFE, it is important for CMS to assess Conclusions the overall stability of the exchange by, among other things, tracking trends in enrollment and termination of enrollee coverage and addressing issuers’ concerns, where appropriate, to ensure their continued participation in the exchange. Issuers have raised concerns that the SEP regulations potentially allow individuals to enroll in coverage despite having their coverage terminated for nonpayment of premiums. However, CMS does not have the data needed to determine the extent of these problems. While CMS has made some efforts to improve the accuracy of the agency’s data on terminations for nonpayment, it has not indicated whether the agency will require issuers to consistently and accurately report these data. Moreover, CMS has no way to ensure the reliability and transparency of the data, because the existing process—the exchange of monthly reconciliation files between CMS and issuers—does not have a place to capture these data. CMS could capitalize on this existing process, already familiar to issuers, by adding a variable that captures data on termination reasons to the monthly reconciliation file and tracking its accuracy. By taking this step, in addition to requiring issuers to report these data, CMS could help ensure it has reliable and transparent data on terminations of enrollee coverage for nonpayment of premiums, and it could use these data to assess the effects of CMS policies and the overall stability of the exchange. We are making the following two recommendations to CMS: Recommendations • The Administrator of CMS should ensure that CMS has complete data on terminations of enrollee coverage for nonpayment of premiums by requiring issuers to report these data. (Recommendation 1) • The Administrator of CMS should provide a transparent process for issuers and CMS to systematically reconcile discrepancies in their data on terminations of enrollee coverage for nonpayment of premiums. (Recommendation 2) We provided a draft of this report to HHS. HHS provided written Agency Comments comments, which are reprinted in appendix III. HHS concurred with our and Our Evaluation first recommendation to require issuers to report data on terminations of enrollee coverage for nonpayment of premiums. HHS noted that it Page 19 GAO-18-269 Federal Health Insurance Exchange Enrollment currently collects information on termination reasons on enrollment transactions with issuers, and that it would review the requirements for collection of these data to identify possible improvements. HHS also concurred with our second recommendation to ensure a transparent process for issuers and CMS to systematically reconcile discrepancies in their data on terminations of enrollee coverage for nonpayment of premiums. HHS stated that it would consider how to incorporate reconciliation of these data into its existing monthly data reconciliation process with issuers, balancing issuer and agency burdens against the benefits of doing so. As agreed with your office, unless you publically announce the contents of the report earlier, we plan no further distribution until 30 days from the report date. At that time, we will send copies of this report to the Secretary of Health and Human Services and other interested parties. In addition, the report is available at no charge on the GAO website at http://www.gao.gov. If you or your staff have any questions about this report, please contact me at (202) 512-7114 or dickenj@gao.gov. Contact points for our Offices of Congressional Relations and Public Affairs may be found on the last page of this report. GAO staff who made major contributions to this report are listed in appendix IV. John E. Dicken Director, Health Care Page 20 GAO-18-269 Federal Health Insurance Exchange Enrollment Appendix I: Demographic and Coverage Appendix I: Demographic and Coverage Characteristics of Federally Facilitated Exchange Enrollees, 2015 Characteristics of Federally Facilitated Exchange Enrollees, 2015 Table 1 provides information on demographic characteristics for federally facilitated exchange (FFE) enrollees that maintained continuous coverage throughout 2015—defined as beginning coverage by March 1, 2015, and maintaining it without any gaps through December 31, 2015—and for all other 2015 FFE enrollees. Table 2 provides information on the characteristics of these enrollees’ coverage. Table 3 provides the extent to which enrollees maintained continuous coverage throughout 2015 by their state of residence. Table 1: Demographic Characteristics of Continuously Covered, and All Other, Federally Facilitated Exchange Enrollees, 2015 Characteristic Enrollees who maintained All other continuous coverage enrollees (%) throughout 2015 (%) Age as of December 0-17 years 8 11 31, 2015 18-24 8 10 25-34 15◆ 21◆ 35-44 15 16 45-54 22 19 55+ 32◆ 23◆ Gender Female 55 55 Male 45 45 Number of family 1 67◆ 74◆ members on policya 2+ 33◆ 26◆ Reported household <=100% 1 1 income as a 101-150 38 39 percentage of the federal poverty levela,b 151-200 24 22 201-250 12 11 251-300 7 6 301-350 4 3 351-400 2 2 Unknown 12 16 Legend: ◆ = difference of 5 or more percentage points between enrollees who maintained continuous coverage throughout 2015 and all other enrollees. Source: GAO analysis of data from the Centers for Medicare & Medicaid Services. | GAO-18-269 Notes: There were 9.2 million federally facilitated exchange enrollees in coverage year 2015, excluding dental coverage and the 25,599 individuals who died in 2015, per data from the Social Security Administration. Of the 9.2 million enrollees, 4.9 million maintained continuous coverage throughout the year, which we defined as coverage that began by March 1, 2015, and ended on December 31, 2015. The remaining 4.3 million enrollees did not maintain continuous coverage throughout the year, and included about 264,000 enrollees with gaps in coverage—that is, enrollees who dropped and then resumed coverage through the federally facilitated exchange in 2015 with a Page 21 GAO-18-269 Federal Health Insurance Exchange Enrollment Appendix I: Demographic and Coverage Characteristics of Federally Facilitated Exchange Enrollees, 2015 gap of at least one day in between. Where enrollees had multiple records, we used information from their first record to examine their characteristics. a For our analyses of number of family members and reported household income, where multiple family members were included on a policy, we used data from the principal policy holder’s record only and excluded the other family members from analysis. b These data reflect CMS’s calculation of enrollees’ adjusted federal poverty level using the 2014 poverty guidelines published by the Department of Health and Human Services. In 2014, the federal poverty level for a family of 4 in the 48 contiguous states and the District of Columbia was $23,850. The “unknown” category consists of enrollees who were not eligible for advance premium tax credits. According to CMS officials, the federally facilitated exchange does not collect income information for individuals who do not request financial assistance. Table 2: Coverage Characteristics for Continuously Covered, and All Other, Federally Facilitated Exchange Enrollees, 2015 Characteristic Enrollees who maintained All other continuous coverage enrollees (%) throughout 2015 (%) Enrollment perioda,b Open enrollment 97◆ 66◆ Special enrollment 2◆ 32◆ Reason for special Loss of minimum 50◆ 56◆ enrollmenta essential coverage Attesting to denial of 35◆ 23◆ Medicaid eligibility Change in eligibility 6 4 for premium tax credits or cost- sharing reductions Moved to new 3 3 service area Adoption or baby 2 2 born in household Otherc 5◆ 12◆ Number of plans 1 85 86 held during the year 2+ 15 14 Number of issuers 1 96 96 from which plans 2+ 4 4 were held Benefit level of Catastrophic <1 1 selected plana Bronze 19 21 Silver 70 69 Gold 7 7 Platinum 3 3 Eligibility for Eligible 91 88 advance premium Ineligible 9 12 tax creditsa Page 22 GAO-18-269 Federal Health Insurance Exchange Enrollment Appendix I: Demographic and Coverage Characteristics of Federally Facilitated Exchange Enrollees, 2015 Characteristic Enrollees who maintained All other continuous coverage enrollees (%) throughout 2015 (%) Eligibility for cost- Eligible 75 74 sharing reductionsa Ineligible 25 26 Monthly premium <$50 27 29 after advance 50-75 10 10 premium tax creditsa,d 76-100 8 8 101-150 13 13 151-200 10 10 >200 33 29 Re-enrollment in Re-enrolled 83◆ 38◆ exchange coverage e Did not re-enroll 17◆ 62◆ for 2016 Legend: ◆ = difference of 5 or more percentage points between enrollees who maintained continuous coverage throughout 2015 and all other enrollees. Source: GAO analysis of data from the Centers for Medicare & Medicaid Services. | GAO-18-269 Notes: There were 9.2 million federally facilitated exchange enrollees in coverage year 2015, excluding dental coverage and the 25,599 individuals who died in 2015, per data from the Social Security Administration. Of the 9.2 million enrollees, 4.9 million maintained continuous coverage throughout the year, which we defined as coverage that began by March 1, 2015, and ended on December 31, 2015. The remaining 4.3 million enrollees did not maintain continuous coverage throughout the year, and included about 264,000 enrollees with gaps in coverage—that is, enrollees who dropped and then resumed coverage through the federally facilitated exchange in 2015 with a gap of at least one day in between. a For these analyses, where enrollees had multiple records, we used information from their first record to examine their characteristics. b We considered any enrollee who enrolled in 2015 exchange coverage by February 22, 2015, to have enrolled during the open enrollment period, and any enrollee who enrolled later in the year and had a special enrollment reason specified on their record to have enrolled via a special enrollment period. Percentages in this category do not sum to 100 because according to CMS officials, the enrollment period for 1 percent of 2015 exchange enrollees could not be classified due to data anomalies. c The “other” category includes special enrollment granted due to circumstances such as marriage, owing a tax penalty for not having health coverage in 2014, or other exceptional circumstances. d For our analysis of premiums, where multiple family members were included on a policy, we used data from the principal policy holder’s record only, and excluded the other family members from analysis. We calculated premium amounts at the policy level; where multiple family members were on a policy, per-enrollee costs would be lower. e According to CMS officials, data for the latter half of 2016 were not available from CMS at the time of our review. We therefore examined whether individuals had re-enrolled in exchange coverage for 2016 as of June 30, 2016. Page 23 GAO-18-269 Federal Health Insurance Exchange Enrollment Appendix I: Demographic and Coverage Characteristics of Federally Facilitated Exchange Enrollees, 2015 Table 3: Proportion of Federally Facilitated Exchange Enrollees Who Maintained Continuous Coverage throughout 2015, by State State Enrollees who maintained continuous coverage throughout 2015 (%) Maine 65 Montana 63 South Dakota 61 Alaska 59 North Dakota 59 Nebraska 58 New Hampshire 58 Virginia 58 West Virginia 58 Oregon 57 Wisconsin 57 Arkansas 56 Delaware 56 Indiana 56 Michigan 56 Missouri 56 Kansas 56 Alabama 55 New Mexico 55 North Carolina 55 Pennsylvania 55 Utah 55 Illinois 54 Louisiana 54 Ohio 54 South Carolina 54 Arizona 53 Wyoming 53 New Jersey 52 Oklahoma 52 Tennessee 52 Florida 51 Georgia 51 Iowa 51 Page 24 GAO-18-269 Federal Health Insurance Exchange Enrollment Appendix I: Demographic and Coverage Characteristics of Federally Facilitated Exchange Enrollees, 2015 State Enrollees who maintained continuous coverage throughout 2015 (%) Mississippi 49 Nevada 49 Texas 48 Source: GAO analysis of data from the Centers for Medicare & Medicaid Services. | GAO-18-269 Notes: There were 9.2 million federally facilitated exchange enrollees in coverage year 2015, excluding dental coverage and the 25,599 individuals who died in 2015, per data from the Social Security Administration. Of the 9.2 million enrollees, 4.9 million maintained continuous coverage throughout the year, which we defined as coverage that began by March 1, 2015, and ended on December 31, 2015. The remaining 4.3 million enrollees did not maintain continuous coverage throughout the year, and included about 264,000 enrollees with gaps in coverage—that is, enrollees who dropped and then resumed coverage through the federally facilitated exchange in 2015 with a gap of at least one day in between. Where enrollees had multiple records, we used information from their first record to determine their state of residence. Page 25 GAO-18-269 Federal Health Insurance Exchange Enrollment Appendix II: Average Length of Coverage for Appendix II: Average Length of Coverage for Federally Facilitated Exchange Enrollees, 2015 Federally Facilitated Exchange Enrollees, 2015 Table 4 provides information on average length of coverage for all 9.2 million federally facilitated exchange enrollees in 2015 by various demographic characteristics. Table 5 provides information on average length of coverage for these enrollees by characteristics of the enrollees’ coverage. Table 6 provides information on average length of coverage for enrollees by their state of residence. Table 4: Average Length of Coverage for Federally Facilitated Exchange Enrollees by Demographic Characteristics, 2015 Characteristic Average length of coverage (in months) Age as of December 31, 2015 0-17 years 7.9 18-24 8.1 25-34 7.8 35-44 8.3 45-54 8.7 55+ 9.2 Gender Female 8.5 Male 8.4 Number of family members on 1 8.2 policya 2+ 8.9 Reported household income as a <=100% 8.0 percentage of the federal poverty 101-150 8.3 levela, b 151-200 8.6 201-250 8.7 251-300 8.8 301-350 8.9 351-400 8.9 Unknown 7.8 Source: GAO analysis of data from the Centers for Medicare & Medicaid Services. | GAO-18-269 Notes: There were 9.2 million federally facilitated exchange enrollees in coverage year 2015, excluding dental coverage and the 25,599 individuals who died in 2015, per data from the Social Security Administration. Since our focus was the length of time for which enrollees remained enrolled in exchange coverage, we excluded the approximately 264,000 enrollees (3 percent) that dropped and then resumed exchange coverage during the year from our analysis. Where enrollees had multiple records, we used information from enrollees’ first record in 2015 to determine their characteristics. Our analysis of average length of coverage assumes a month to be 30.4 days long. a For our analyses of number of family members and reported household income, where multiple family members were included on a policy, we used data from the principal policy holder’s record only, and excluded the other family members from analysis. b These data reflect CMS’s calculation of enrollees’ adjusted federal poverty level using the 2014 poverty guidelines published by the Department of Health and Human Services. In 2014, the federal poverty level for a family of 4 in the 48 contiguous states and the District of Columbia was $23,850. Page 26 GAO-18-269 Federal Health Insurance Exchange Enrollment Appendix II: Average Length of Coverage for Federally Facilitated Exchange Enrollees, 2015 The “unknown” category consists of enrollees who were not eligible for advance premium tax credits. According to CMS officials, the federally facilitated exchange does not collect income information for individuals who do not request financial assistance. Table 5: Average Length of Coverage for Federally Facilitated Exchange Enrollees by Coverage Characteristics, 2015 Characteristic Average length of coverage (in months) Enrollment perioda,b Open enrollment 9.1 Special enrollment 5.2 Reason for special Loss of minimum essential coverage 6.9 enrollmenta Attesting to denial of Medicaid eligibility 8.0 Change in eligibility for premium tax 7.7 credits or cost-sharing reductions Moved to new service area 6.5 Adoption or baby born in household 7.2 Otherc 6.4 Number of plans 1 8.3 held during the year 2+ 9.8 Number of issuers 1 8.4 from which plans 2+ 10.4 were held Benefit level of Catastrophic 6.7 selected plana Bronze 8.2 Silver 8.6 Gold 8.8 Platinum 8.5 Eligibility for advance Eligible 8.6 premium tax creditsa Ineligible 7.8 Eligibility for cost- Eligible 8.5 sharing reductionsa Ineligible 8.4 Monthly premium <$50 8.1 after advance 50-75 8.3 premium tax creditsa,d 76-100 8.4 101-150 8.4 151-200 8.4 >200 8.8 Re-enrollment in Re-enrolled 9.9 exchange coverage Did not re-enroll 6.2 for 2016e Source: GAO analysis of data from the Centers for Medicare & Medicaid Services. | GAO-18-269 Page 27 GAO-18-269 Federal Health Insurance Exchange Enrollment Appendix II: Average Length of Coverage for Federally Facilitated Exchange Enrollees, 2015 Notes: There were 9.2 million federally facilitated exchange enrollees in coverage year 2015, excluding dental coverage and the 25,599 individuals who died in 2015, per data from the Social Security Administration. Since our focus was the length of time for which enrollees remained enrolled in exchange coverage, we excluded the approximately 264,000 enrollees (3 percent) that dropped and then resumed exchange coverage during the year from our analysis. Our analysis of average length of coverage assumes a month to be 30.4 days long. a For these analyses, where enrollees had multiple records, we used information from their first record to examine their characteristics. b We considered any enrollee who enrolled in 2015 exchange coverage by February 22, 2015, to have enrolled during the open enrollment period, and any enrollee who enrolled later in the year and had special enrollment reason code on their record to have enrolled via a special enrollment period. c The “other” category includes special enrollment granted due to circumstances such as marriage, owing a tax penalty for not having health coverage in 2014, or other exceptional circumstances. d For our analysis of premiums, where multiple family members were included on a policy, we used data from the principal policy holder’s record only, and excluded the other family members from analysis. We calculated premium amounts at the policy level; where multiple family members were on a policy, per-enrollee costs would be lower. e According to CMS officials, data for the latter half of 2016 were not available when we requested data from CMS. We therefore examined whether individuals had re-enrolled in exchange coverage for 2016 as of June 30, 2016. Table 6: Average Length of Coverage for Federally Facilitated Exchange Enrollees by State, 2015 State Average length of coverage (in months) Maine 9.4 Montana 9.3 South Dakota 9.2 Alaska 9.1 North Dakota 9.1 New Hampshire 9.0 Oregon 9.0 Virginia 8.9 Wisconsin 8.9 Indiana 8.8 Nebraska 8.8 New Mexico 8.8 West Virginia 8.8 Alabama 8.7 Arkansas 8.7 Delaware 8.7 Kansas 8.7 Michigan 8.7 Missouri 8.7 Utah 8.7 Page 28 GAO-18-269 Federal Health Insurance Exchange Enrollment Appendix II: Average Length of Coverage for Federally Facilitated Exchange Enrollees, 2015 State Average length of coverage (in months) Illinois 8.6 North Carolina 8.6 Pennsylvania 8.6 South Carolina 8.6 Arizona 8.5 Ohio 8.5 Tennessee 8.5 Wyoming 8.5 Louisiana 8.4 New Jersey 8.4 Oklahoma 8.4 Florida 8.3 Georgia 8.3 Iowa 8.3 Nevada 8.2 Texas 8.1 Mississippi 8.0 Source: GAO analysis of data from the Centers for Medicare & Medicaid Services. | GAO-18-269 Notes: There were 9.2 million federally facilitated exchange enrollees in coverage year 2015, excluding dental coverage and the 25,599 individuals who died in 2015, per data from the Social Security Administration. Since our focus was the length of time for which enrollees remained enrolled in exchange coverage, we excluded the approximately 264,000 enrollees (3 percent) that dropped and then resumed exchange coverage during the year from our analysis. Where enrollees had multiple records, we used information from enrollees’ first record in 2015 to determine their state of residence. Our analysis of average length of coverage assumes a month to be 30.4 days long. Page 29 GAO-18-269 Federal Health Insurance Exchange Enrollment Appendix III: Comments from the Appendix III: Comments from the Department of Health and Human Services Department of Health and Human Services Page 30 GAO-18-269 Federal Health Insurance Exchange Enrollment Appendix III: Comments from the Department of Health and Human Services Page 31 GAO-18-269 Federal Health Insurance Exchange Enrollment Appendix III: Comments from the Department of Health and Human Services Page 32 GAO-18-269 Federal Health Insurance Exchange Enrollment Appendix IV: GAO Contact and Staff Appendix IV: GAO Contact and Staff Acknowledgments Acknowledgments John Dicken, (202) 512-7114 or dickenj@gao.gov GAO Contact In addition to the contact named above, William Hadley (Assistant Staff Director), Iola D’Souza (Analyst in Charge), Richard Lipinski, Peter Mann- Acknowledgments King, and Priyanka Sethi Bansal made key contributions to this report. Also contributing were Muriel Brown, Laurie Pachter, and Emily Wilson. (101013) Page 33 GAO-18-269 Federal Health Insurance Exchange Enrollment The Government Accountability Office, the audit, evaluation, and investigative GAO’s Mission arm of Congress, exists to support Congress in meeting its constitutional responsibilities and to help improve the performance and accountability of the federal government for the American people. GAO examines the use of public funds; evaluates federal programs and policies; and provides analyses, recommendations, and other assistance to help Congress make informed oversight, policy, and funding decisions. GAO’s commitment to good government is reflected in its core values of accountability, integrity, and reliability. The fastest and easiest way to obtain copies of GAO documents at no cost is Obtaining Copies of through GAO’s website (https://www.gao.gov). Each weekday afternoon, GAO GAO Reports and posts on its website newly released reports, testimony, and correspondence. To have GAO e-mail you a list of newly posted products, go to https://www.gao.gov Testimony and select “E-mail Updates.” Order by Phone The price of each GAO publication reflects GAO’s actual cost of production and distribution and depends on the number of pages in the publication and whether the publication is printed in color or black and white. Pricing and ordering information is posted on GAO’s website, https://www.gao.gov/ordering.htm. Place orders by calling (202) 512-6000, toll free (866) 801-7077, or TDD (202) 512-2537. Orders may be paid for using American Express, Discover Card, MasterCard, Visa, check, or money order. Call for additional information. Connect with GAO on Facebook, Flickr, Twitter, and YouTube. Connect with GAO Subscribe to our RSS Feeds or E-mail Updates. Listen to our Podcasts. Visit GAO on the web at https://www.gao.gov. Contact: To Report Fraud, Website: https://www.gao.gov/fraudnet/fraudnet.htm Waste, and Abuse in Automated answering system: (800) 424-5454 or (202) 512-7470 Federal Programs Orice Williams Brown, Managing Director, WilliamsO@gao.gov, (202) 512-4400, Congressional U.S. Government Accountability Office, 441 G Street NW, Room 7125, Relations Washington, DC 20548 Chuck Young, Managing Director, youngc1@gao.gov, (202) 512-4800 Public Affairs U.S. Government Accountability Office, 441 G Street NW, Room 7149 Washington, DC 20548 James-Christian Blockwood, Managing Director, spel@gao.gov, (202) 512-4707 Strategic Planning and U.S. Government Accountability Office, 441 G Street NW, Room 7814, External Liaison Washington, DC 20548 Please Print on Recycled Paper.