United States Government Accountability Office Report to the Ranking Member, Committee on Finance, U.S. Senate MEDICAID September 2018 Access to Health Care for Low-Income Adults in States with and without Expanded Eligibility GAO-18-607 September 2018 MEDICAID Access to Health Care for Low-Income Adults in States with and without Expanded Eligibility Highlights of GAO-18-607, a report to the Ranking Member, Committee on Finance, U.S. Senate Why GAO Did This Study What GAO Found Under PPACA, states could choose to According to the 2016 National Health Interview Survey (NHIS), an estimated 5.6 expand Medicaid coverage to certain million uninsured, low-income adults—those ages 19 through 64—had incomes uninsured, low-income adults. As of at or below the income threshold for expanded Medicaid eligibility as allowed December 2017, 31 states and the under the Patient Protection and Affordable Care Act (PPACA). Estimates from District of Columbia chose to expand this nationally representative survey showed that about 1.9 million of the 5.6 Medicaid to cover these adults, and 19 million uninsured, low-income adults lived in states that chose to expand states did not. Medicaid under PPACA, while the remaining 3.7 million lived in non-expansion GAO was asked to provide information states—those that did not choose to expand Medicaid. In 2016, over half of about the demographic characteristics uninsured, low-income adults were male, over half were employed, and over half of and access to health care services had incomes less than 100 percent of the federal poverty level in both expansion for low-income adults—those with and non-expansion states. household incomes less than or equal to 138 percent of the federal poverty The 2016 NHIS estimates showed that low-income adults in expansion states level—in expansion and non- were less likely to report having any unmet medical needs compared with those expansion states. This report describes in non-expansion states, and low-income adults who were insured were less 2016 national survey estimates of (1) likely to report having unmet medical needs compared with those who were the number and demographic uninsured. Among the low-income adults who were uninsured, those in characteristics for low-income adults expansion states were less likely to report having any unmet medical needs who were uninsured in expansion and compared with those in non-expansion states. non-expansion states, (2) unmet medical needs for low-income adults in Low-Income Adults Who Reported Having Any Unmet Medical Need in Expansion and Non- Expansion States and by Insurance Status, 2016 expansion and non-expansion states and by insurance status, (3) barriers to health care for low-income adults in expansion and non-expansion states and by insurance status, and (4) having a usual place of care and receiving selected health care services for low-income adults in expansion and non-expansion states and by insurance status. GAO obtained 2016 NHIS estimates from the National Center for Health Statistics (NCHS), the federal agency within the Department of Health and Human Services that maintains these survey data. Notes: The difference between all low-income adults in expansion and non-expansion states was statistically significant at p < 0.05 (a common indictor that denotes statistical significance). The NHIS is a household interview survey difference between low-income adults who were uninsured in expansion and non-expansion states designed to be a nationally was statistically significant at p < 0.05. representative sample of the civilian, Differences between low-income adults who were uninsured in expansion or non-expansion states non-institutionalized population and low-income adults who were insured—Medicaid or private health insurance—in expansion or residing in the United States. non-expansion states were statistically significant at p < 0.05. Estimates were calculated for The 2016 NHIS estimates also showed that low-income adults in expansion demographic characteristics for states were less likely to report financial barriers to needed medical care and uninsured, low-income adults. In other types of health care, such as specialty care, compared with those in non- addition, estimates were calculated for expansion states, and low-income adults who were insured were less likely to unmet medical needs, barriers to report financial barriers to needed medical care compared with those who were health care, and having a usual place uninsured. United States Government Accountability Office of care and receiving selected health Among low-income adults who were uninsured, those in expansion states were services for low-income adults in less likely to report financial barriers to needed medical care compared with expansion and non-expansion states those in non-expansion states. and by insurance status The Low-Income Adults Who Reported Financial Barriers to Needed Medical Care in Expansion estimates were based on responses and Non-Expansion States and by Insurance Status, 2016 to selected survey questions. GAO selected these survey questions from the Family and Adult Access to Health Care and Utilization and another section of the 2016 NHIS. GAO took steps to assess the reliability of the 2016 NHIS estimates, including interviewing NCHS officials and examining the data for logical errors. GAO determined that the data were sufficiently reliable for the purposes of its analyses. The Department of Health and Human Services provided technical comments on a draft of this report, Notes: The difference between all low-income adults in expansion and non-expansion states was which GAO incorporated as statistically significant at p < 0.05. The difference between low-income adults who were uninsured in appropriate. expansion and non-expansion states was statistically significant at p < 0.05. Differences between low-income adults who were uninsured in expansion or non-expansion states and low-income adults who were insured—Medicaid or private health insurance—in expansion or non-expansion states were statistically significant at p < 0.05. Finally, the 2016 NHIS estimates showed that low-income adults in expansion states were more likely to report having a usual place of care to go when sick or needing advice about their health and receiving selected health care services compared with those in non-expansion states. The estimates also showed that low-income adults who were insured were generally more likely to report having a usual place of care and receiving selected health care services compared with those who were uninsured. Among the uninsured, relatively similar percentages of low-income adults in expansion and non-expansion states reported having a usual place of care. Similarly, estimates showed that relatively similar percentages of low-income adults who were uninsured in expansion and non- expansion states reported receiving selected health care services, such as receiving a flu vaccine or a blood pressure check. Low-income Adults Who Reported Having a Usual Place of Care in Expansion and Non- Expansion States and by Insurance Status, 2016 Notes: The difference between all low-income adults in expansion and non-expansion states was statistically significant at p < 0.05. Differences between low-income adults who were uninsured in expansion or non-expansion states and low-income adults who were insured—Medicaid or private health insurance—in expansion or non-expansion states were statistically significant at p < 0.05. View GAO-18-607. For more information, contact Carolyn L. Yocom at (202) 512-7114 or yocomc@gao.gov. Contents Letter 1 Background 6 Survey Estimates Showed 5.6 Million Uninsured, Low-Income Adults Had Qualifying Incomes for Expanded Medicaid Coverage 10 Survey Estimates Showed Low-Income Adults in Expansion States and Those Who Were Insured Were Less Likely to Report Any Unmet Medical Needs 13 Survey Estimates Showed Low-Income Adults in Expansion States and Those Who Were Insured Were Less Likely to Report Financial Barriers to Health Care 15 Survey Estimates Showed Low-Income Adults in Expansion States and Those Who Were Insured Were Generally More Likely to Report Having a Usual Place of Care and Receiving Selected Health Care Services 18 Agency Comments and Our Evaluation 22 Appendix I Objectives, Scope, and Methodology 24 Appendix II Status of Medicaid Eligibility Expansion by States, as of 2017 33 Appendix III Estimates of Demographic Characteristics and Health Status in Expansion and Non-Expansion States 36 Appendix IV Estimates of Any Unmet Medical Needs in Expansion and Non- Expansion States and by Insurance Status 40 Appendix V Estimates of Barriers to Health Care in Expansion and Non-Expansion States and by Insurance Status 44 Page i GAO-18-607 Medicaid Expansion and Access to Care Appendix VI Estimates on Place of Care and Services in Expansion and Non- Expansion States and by Insurance Status 53 Appendix VII GAO Contact and Staff Acknowledgments 63 Tables Table 1: Demographic Characteristics of Uninsured, Low-Income Adults in Expansion and Non-Expansion States, 2016 10 Table 2: States Classified as Expansion States and Non- Expansion States for this Study 26 Table 3: Sample Sizes and Population Estimates of Low-Income Adults by State Expansion Status 28 Table 4: Comparison Groups of Low-Income Adults for Tests of Statistically Significant Differences in Selected Measures of Access to Care 30 Table 5: Status of Medicaid Eligibility Expansion, by State, as of 2017 33 Table 6: Demographic Characteristics of Uninsured, Low-Income Adults in Expansion and Non-Expansion States, 2016 37 Table 7: Health Status of Uninsured, Low-Income Adults in Expansion and Non-Expansion Status, 2016 39 Table 8: Percentage of Low-Income Adults in Expansion and Non- Expansion States Who Reported Any Unmet Medical Needs, 2016 41 Table 9: Percentage of Low-Income Adults in Expansion and Non- Expansion States Who Reported Any Unmet Medical Needs, by Insurance Status, 2016 43 Table 10: Percentage of Low-Income Adults in Expansion and Non-Expansion States Who Reported Financial Barriers to Different Types of Needed Health Care Services, 2016 46 Table 11: Percentage of Low-Income Adults in Expansion and Non-Expansion States Who Reported Financial Barriers to Different Types of Needed Health Care Services, by Insurance Status, 2016 47 Table 12: Percentage of Low-Income Adults in Expansion and Non-Expansion States Who Reported Taking Actions to Save Money on Prescriptions, 2016 49 Page ii GAO-18-607 Medicaid Expansion and Access to Care Table 13: Percentage of Low-Income Adults in Expansion and Non-Expansion States Who Reported Taking Actions to Save Money on Medications, by Insurance Status, 2016 49 Table 14: Percentage of Low-Income Adults in Expansion and Non-Expansion States Who Reported Non-Financial Barriers to Health Care, 2016 51 Table 15: Percentage of Low-Income Adults in Expansion and Non-Expansion States Who Reported Non-Financial Barriers to Health Care, by Insurance Status, 2016 51 Table 16: Percentage of Low-Income Adults in Expansion and Non-Expansion States Who Reported Having a Usual Place of Care, 2016 55 Table 17: Percentage of Low-Income Adults in Expansion and Non-Expansion States Who Reported Having a Usual Place of Care, by Insurance Status, 2016 56 Table 18: Percentage of Low-Income Adults in Expansion and Non-Expansion States Who Had a Usual Place of Care by the Type of Place They Usually Went for Care, 2016 57 Table 19: Percentage of Low-Income Adults in Expansion and Non-Expansion States with a Usual Place of Care by the Type of Place They Usually Went for Care, by Insurance Status, 2016 57 Table 20: Percentage of Low-Income Adults in Expansion and Non-Expansion States Who Reported Receiving Selected Health Care Services, 2016 58 Table 21: Percentage of Low-Income Adults in Expansion and Non-Expansion States Who Reported Receiving Selected Health Care Services, by Insurance Status, 2016 59 Table 22: Percentage of Low-Income Adults in Expansion and Non-Expansion States Who Reported Visiting or Speaking to a Health Care Professional about their Health, 2016 60 Table 23: Percentage of Low-Income Adults in Expansion and Non-Expansion States Who Reported Seeing or Speaking to a Health Care Provider About their Health, by Insurance Status, 2016 61 Figures Figure1: National Medicaid Enrollment by Eligible Population, December 2017 7 Page iii GAO-18-607 Medicaid Expansion and Access to Care Figure 2: Status of Medicaid Eligibility Expansion by State, as of 2017 9 Figure 3: Health Status of Uninsured, Low-Income Adults in Expansion and Non-Expansion States, 2016 12 Figure 4: Percentage of Low-Income Adults in Expansion and Non-Expansion States and by Insurance Status Who Reported Any Unmet Medical Need, 2016 14 Figure 5: Percentage of Low-Income Adults in Expansion and Non-Expansion States and by Insurance Status Who Reported Financial Barriers to Different Types of Needed Health Care Services, 2016 16 Figure 6: Percentage of Low-Income Adults in Expansion and Non-Expansion States and by Insurance Status Who Reported Having a Usual Place of Care, 2016 19 Figure 7: Percentage of Low-Income Adults in Expansion and Non-Expansion States and by Insurance Status Who Reported Using Selected Health Care Services, 2016 21 Abbreviations CMS Centers for Medicare & Medicaid Services FPL federal poverty level HHS Department of Health and Human Services NCHS National Center for Health Statistics NHIS National Health Interview Survey PPACA Patient Protection and Affordable Care Act 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 iv GAO-18-607 Medicaid Expansion and Access to Care Letter 441 G St. N.W. Washington, DC 20548 September 13, 2018 The Honorable Ron Wyden Ranking Member Committee on Finance United States Senate Dear Mr. Wyden: Historically, eligibility for Medicaid—a federal-state health care financing program—has been limited to certain categories of low-income individuals, including children, parents, pregnant women, and individuals who have disabilities or who are aged 65 and older. Beginning in 2014, the Patient Protection and Affordable Care Act (PPACA) gave states the option of expanding Medicaid eligibility beyond these categories to include certain adults with incomes that do not exceed 138 percent of the federal poverty level (FPL). 1 As of December 2017, there were 31 “expansion states”—those states and the District of Columbia that chose to expand Medicaid eligibility—and 19 “non-expansion states”—those that had not expanded Medicaid eligibility to this additional adult population. Several years have passed since PPACA gave states the option of expanding Medicaid eligibility. You asked us to provide the most recently available information on the demographic characteristics of uninsured, low-income adults, and the extent to which low-income adults are accessing health care services in two groups of states: expansion states 1 Under PPACA, enacted on March 23, 2010, states may opt to expand their Medicaid programs to cover non-elderly, non-pregnant adults who are not eligible for Medicare, and whose income does not exceed 133 percent of the FPL beginning January 1, 2014. Pub. L. No. 111-148, 124 Stat. 119 (2010), as amended by the Health Care and Education Reconciliation Act of 2010, Pub. L. No. 111-152, 124 Stat. 1029 (2010). PPACA also provides for a 5 percent disregard when calculating income for determining Medicaid eligibility, which effectively increases income eligibility from 133 percent of FPL to 138 percent of FPL. The FPL is based on household income and family size, using the U.S. Census Bureau’s poverty thresholds. In 2016, 138 percent of FPL for an individual was $17,231 and for a family of four was $33,897. PPACA also permitted an early expansion option, whereby states could expand eligibility for this population, or a subset of this population, starting on April 1, 2010. Page 1 GAO-18-607 Medicaid Expansion and Access to Care and non-expansion states. 2 This report describes what 2016 national survey estimates showed regarding: 1. the number and demographic characteristics for low-income adults who were uninsured in expansion and non-expansion states, 2. unmet medical needs for low-income adults in expansion and non- expansion states and by insurance status, 3. barriers to health care for low-income adults in expansion and non- expansion states and by insurance status, and 4. having a usual place of care and receiving selected health care services for low-income adults in expansion and non-expansion states and by insurance status. To address our research objectives, we used data from the 2016 National Health Interview Survey (NHIS), from the Centers for Disease Control and Prevention, National Center for Health Statistics (NCHS). 3 These data were the most recent data available when we conducted our analyses. To describe the number and demographic characteristics for low-income adults—individuals ages 19 to 64, with incomes that did not exceed 138 percent FPL—who were uninsured in expansion and non-expansion states, we requested that NCHS calculate estimates based on responses to survey questions on demographic characteristics. 4 Responses to survey questions were calculated as an estimated percentage of the total population for three groups of low-income adults: those in expansion 2 In this report, we use the term “low-income adults” to refer to individuals, ages 19 to 64, with household incomes that do not exceed 138 percent of FPL. 3 The NHIS is a cross-sectional, household interview survey designed to be a nationally representative sample that covers the civilian, noninstitutionalized population residing in the United States. Several segments of the population are not included in the NHIS. Examples of persons excluded from the survey are patients in long-term care facilities, persons on active duty with the Armed Forces (though their dependents are included), persons incarcerated in the prison system, and U.S. nationals living in foreign countries. 4 We asked NCHS researchers to produce estimates from the NHIS data because state identifiers, which are not in the public use file, were required to conduct these analyses. For the purposes of these NHIS analyses, we excluded low-income adults who were noncitizens, were covered by Medicare, only received health care services through military health care or through the Indian Health Service, or had Supplemental Social Security Income. We also excluded adult females from the Sample Adult file who reported being pregnant. Page 2 GAO-18-607 Medicaid Expansion and Access to Care states, in non-expansion states, and for all states. 5 We also requested that NCHS test for statistically significant differences between expansion states and non-expansion states. Using these estimates, we summarized the uninsured, low-income adult population by expansion states, non- expansion states, and all states combined for demographic characteristics including race and ethnicity, gender, and employment status. To describe unmet medical needs, barriers to health care, and having a usual place of care and receiving selected health care services for low- income adults in expansion and non-expansion states and by insurance status, we requested that NCHS also calculate estimates using the 2016 NHIS. Estimates were based on responses to selected survey questions and composite measures—NCHS-developed measures based on 5 We classified the 30 states and the District of Columbia that expanded their Medicaid program prior to July 1, 2016, as expansion states. The states classified as expansion states were Alaska, Arizona, Arkansas, California, Colorado, Connecticut, Delaware, Hawaii, Illinois, Indiana, Iowa, Kentucky, Maryland, Massachusetts, Michigan, Minnesota, Montana, Nevada, New Hampshire, New Jersey, New Mexico, New York, North Dakota, Ohio, Oregon, Pennsylvania, Rhode Island, Vermont, Washington, and West Virginia. The remaining 20 states were classified as non-expansion states, including Louisiana, which expanded Medicaid on July 1, 2016. The other non-expansion states were Alabama, Florida, Georgia, Idaho, Kansas, Maine, Mississippi, Missouri, Nebraska, North Carolina, Oklahoma, South Carolina, South Dakota, Tennessee, Texas, Utah, Virginia, Wisconsin, and Wyoming. For 3 states, we included NHIS respondents for part of calendar year 2016. Two of the 3 states—Alaska and Montana—we excluded respondents who were surveyed during the first 6 months after the state expanded Medicaid to allow for the effect of expansion to take place. For Alaska, which expanded Medicaid on September 1, 2015, we excluded responses from those surveyed from January through February 2016. For Montana, which expanded Medicaid on January 1, 2016, we excluded responses from those surveyed from January through June 2016. Because Louisiana was classified as a non-expansion state, we included NHIS respondents interviewed from January through June 2016, when Louisiana was a non-expansion state. See R.A. Cohen and E.P. Zammitti, Coverage, Access, and Utilization by Medicaid Expansion Status: Estimates from the National Health Interview Survey, United States, 2016, (Hyattsville, Md.: National Center for Health Statistics, December 2017), accessed December 12, 2017, https://www.cdc.gov/nchs/health_policy/coverage_and_access.htm; and R.A. Cohen and E.P. Zammitti, Access and Utilization by Medicaid Expansion Status for Low-Income Adults Aged 19-64: Estimates from the National Health Interview Survey, United States, 2016, (Hyattsville, Md.: National Center for Health Statistics, January 2018), accessed January 17, 2018, https://www.cdc.gov/nchs/health_policy/coverage_and_access.htm. Page 3 GAO-18-607 Medicaid Expansion and Access to Care responses to NHIS questions on related topics. 6 We selected survey questions and their related composite measures from the Access to Health Care and Utilization and Health Behaviors sections of the 2016 NHIS. 7 These survey questions and composite measures allowed us to summarize access to health care for low-income adults in terms of unmet medical needs, barriers to health care, and having a usual place of care and receiving selected health care services. Responses to survey questions and composite measures were calculated as an estimated percentage of the total population for eight groups of low-income adults: (1) those in expansion states, (2) those in non-expansion states, (3) those who had Medicaid in expansion states, (4) those who had Medicaid in non-expansion states, (5) those who had private health insurance in expansion states, (6) those who had private health insurance in non- expansion states, (7) those who were uninsured in expansion states, and (8) those who were uninsured in non-expansion states. 8 We also 6 The analysis included two composite measures: (1) any unmet medical needs, which was based on six underlying survey questions that asked respondents about whether during the past 12 months they needed medical care but did not get it because they could not afford it; delayed seeking medical care because of worry about the cost; or did not get prescription medicines, mental health care or counseling, eyeglasses or dental care; and (2) any non-financial barriers to health care, which is based on five underlying survey questions that asked respondents whether they delayed care in the past 12 months for any of the following reasons; could not get through on the telephone; could not get an appointment soon enough; waited too long to see the doctor after arriving at the doctor’s office; the clinic/doctor’s office was not open when respondent could get there; and did not have transportation. Estimates of the composite measure on non-financial barriers to health care and the underlying survey question are in appendix V. 7 NHIS data are organized into several data files. Files used to develop estimates for our study include the Person and Sample Adult files. 8 Health insurance classification was based on a hierarchy of mutually exclusive categories in the following order: private health insurance, Medicaid, other coverage, and uninsured. Low-income adults with more than one coverage type were assigned the first appropriate category in the hierarchy, and those with other coverage were excluded from these analyses. Low-income adults were classified as having private health insurance if they reported that they were covered by any comprehensive private health insurance plan, including health maintenance and preferred provider organization plans. Private health insurance excludes plans that pay for only one type of service, such as accidents or dental care. Low-income adults were classified as having Medicaid if they reported that they were covered by Medicaid or by state-sponsored health plans with no premiums or if it is not known if a premium is charged. Low-income adults were classified as uninsured if they did not have any private health insurance or were not enrolled in Medicare, Medicaid, the Children’s Health Insurance Program, a state-sponsored or other government-sponsored health plan, or a military plan. An adult was also defined as uninsured if they had only a private plan that paid for one type of service, such as accidents or dental care. Page 4 GAO-18-607 Medicaid Expansion and Access to Care requested that NCHS test for statistically significant differences between the groups of low-income adults, specifically: • all low-income adults in expansion states compared with all low- income adults in non-expansion states; • uninsured, low-income adults in expansion states compared with each of the four groups of insured, low-income adults—low-income adults who had Medicaid in expansion states, low-income adults who had Medicaid in non-expansion states, low-income adults who had private health insurance in expansion states, and low-income adults who had private health insurance in non-expansion states; • uninsured, low-income adults in non-expansion states compared with each of the four groups of insured, low-income adults; and • uninsured, low-income adults in expansion states compared with uninsured, low-income adults in non-expansion states. 9 We took steps to assess the reliability of the 2016 NHIS estimates, including interviewing NCHS officials and checking frequency distributions for missing estimates, outliers, and obvious errors. Based on this work, we determined that the estimates were sufficiently reliable for the purposes of our reporting objectives. Appendix I provides additional information on our scope and methodology. We conducted this performance audit from May 2017 through September 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. 9 We also requested that NCHS test for statistically significant differences between low- income adults who had Medicaid in expansion states and those who had Medicaid in non- expansion states, and between low-income adults who had private health insurance in expansion states and those who had private health insurance in non-expansion states. We did not request that NCHS test for significant differences between low-income adults who had Medicaid in expansion or non-expansion states and low-income adults who had private health insurance in expansion or non-expansion states. Estimates of all survey questions and composite measures we reviewed and results of all tests for statistically significant differences are in appendixes IV through VI. Page 5 GAO-18-607 Medicaid Expansion and Access to Care In fiscal year 2016, Medicaid covered an estimated 72.2 million low- Background income and medically needy individuals in the United States, and Medicaid estimated expenditures totaled over $575.9 billion. 10 The federal government matches most state expenditures for Medicaid services on the basis of a statutory formula. 11 States receive higher federal matching rates for certain services or populations, including an enhanced matching rate for Medicaid expenditures for individuals who became eligible for Medicaid under PPACA. 12 Of the $575.9 billion in estimated expenditures for 2016, the federal share totaled over $363.4 billion and the states’ share totaled $212.5 billion. 13 The Centers for Medicare & Medicaid Services (CMS)—a federal agency within the Department of Health and Human Services (HHS)—and states jointly administer and fund the Medicaid program. States have flexibility within broad federal requirements to design and implement their Medicaid programs. States must submit a state Medicaid plan to CMS for review and approval. A state’s approved Medicaid plan outlines the services provided and the groups of individuals covered. While states must cover certain mandatory populations and benefits, they have the option of covering other categories of individuals and benefits. 14 PPACA permitted states to expand coverage to a new population—non- elderly, non-pregnant adults who are not eligible for Medicare and whose income does not exceed 138 percent of the FPL. This expansion 10 In fiscal year 2016, the estimated Medicaid total enrollment included enrollees in all states, the District of Columbia, and five United States territories. See Department of Health & Human Services, Office of the Actuary, Centers for Medicare & Medicaid Services, 2016 Actuarial Report on the Financial Outlook for Medicaid. 11 Under this formula, known as the Federal Medical Assistance Percentage, the federal government pays a share of Medicaid expenditures based on each state’s per capita income relative to the national average. Federal law specifies that the Federal Medical Assistance Percentage will be no lower than 50 percent and no higher than 83 percent. 12 States that chose to expand their Medicaid programs under PPACA receive a federal match of 100 percent beginning in 2014 for expenditures for newly eligible low-income adults, gradually diminishing to 90 percent by 2020. Certain states that expanded Medicaid coverage for low-income adults prior to the enactment of PPACA in 2010 may also receive an enhanced federal match for Medicaid expenditures for this expansion population. 13 See Office of the Actuary, 2016 Actuarial Report. 14 Mandatory benefits include inpatient and outpatient hospital services, and laboratory and x-ray services, among others. Optional benefits include prescription drugs, physical therapy, and dental services, among others. Page 6 GAO-18-607 Medicaid Expansion and Access to Care population comprised 20 percent of total Medicaid enrollment in 2017. (See fig. 1.) Figure1: National Medicaid Enrollment by Eligible Population, December 2017 Note: Medicaid enrollment numbers are from CMS-64 enrollment data and may differ from other data sources. Average monthly enrollment was tabulated from the CMS-64 enrollment report for fiscal year 2017, which was accessed on April 4, 2018. a Expansion enrollees refers to individuals ages 19 to 64, without children, who are not disabled or pregnant, and have household incomes that did not exceed 138 percent of the federal poverty level. In 2017, 138 percent of the federal poverty level was $22,411 for a two-person household in the contiguous United States. b Other adults refers to adults who are not aged and not disabled. As of December 2017, 31 states and the District of Columbia had expanded Medicaid eligibility to the new coverage population allowed under PPACA and 19 states had not. 15 Figure 2, an interactive map, 15 Louisiana expanded Medicaid eligibility on July 1, 2016. For the purposes of these NHIS analyses, Louisiana was classified as a non-expansion state and we included low-income adults interviewed from January through June 2016 when Louisiana was a non-expansion state. Page 7 GAO-18-607 Medicaid Expansion and Access to Care illustrates states’ Medicaid expansion status. See appendix II for additional information on figure 2. Page 8 GAO-18-607 Medicaid Expansion and Access to Care Figure 2: Status of Medicaid Eligibility Expansion by State, as of 2017 Interactivity instructions: Roll over on each state to see population and Medicaid enrollment data for 2017. See Appendix II for the non-interactive, printer-friendly version. WA NH MT ME ND OR MN MA VT ID WI SD NY WY MI RI IA PA NE NV OH UT IL IN CO WV CT CA KS MO VA KY NC TN AZ OK NM AR NJ SC MS AL GA LAa TX DE MD FL DC AK Hawaii Expansion states (32) Non-expansion states (19) Sources: GAO summary of U.S. Census Bureau, Centers for Medicare & Medicaid Services, and Kaiser Family Foundation data (data); Map Resources (map). | GAO-18-607 Notes: Census population estimates are as of July 1, 2017. Medicaid enrollment numbers come from CMS-64 enrollment data and may differ from other data sources. Average monthly enrollment was tabulated from the CMS-64 enrollment report for calendar year 2017 that was accessed on April 4, 2018. a Louisiana expanded Medicaid eligibility on July 1, 2016. For the purpose of the NHIS analysis, Louisiana was classified as a non-expansion state and we only included low-income adults interviewed from January through June 2016 when Louisiana was a non-expansion state. Page 9 GAO-18-607 Medicaid Expansion and Access to Care According to the NHIS estimates, 5.6 million low-income adults were Survey Estimates uninsured in 2016. Of these 5.6 million, an estimated 1.9 million Showed 5.6 Million uninsured, low-income adults resided in expansion states, compared with an estimated 3.7 million in non-expansion states. Estimates of uninsured, Uninsured, Low- low-income adults comprised less than 1 percent of the total population Income Adults Had for all expansion states and 3 percent of the total population for all non- expansion states. 16 Qualifying Incomes for Expanded NHIS estimates also showed that over half of uninsured, low-income adults were male, over half were employed, and over half had incomes Medicaid Coverage less than 100 percent FPL. For some demographic characteristics, there were some statistically significant differences between uninsured, low- income adults in expansion states compared with these adults in non- expansion states. For example, expansion states had significantly larger percentages of uninsured, low-income males than non-expansion states. (See table 1.) See table 6 in appendix III for additional demographic characteristics of uninsured, low-income adults. Table 1: Demographic Characteristics of Uninsured, Low-Income Adults in Expansion and Non-Expansion States, 2016 Expansion states Non-expansion states All states Demographic characteristics Percent (standard error) Percent (standard error) Percent (standard error) N = 1,915,000 N = 3,721,000 N = 5,639,000 Poverty statusa Less than 100% FPL 59 (3.28)b 69 (2.05)b 65 (1.81) b 100% to less than or equal to 138% FPL 42 (3.28) 32 (2.05)b 35 (1.81) Sex Male 63 (2.33)b 51 (1.49)b 55 (1.31) Female 37(2.33)b 49 (1.49)b 45 (1.31) Race and ethnicity Hispanic 22 (2.64) 20 (3.26) 20 (2.30) Non-Hispanic, white only 53 (3.40) 45 (2.70) 48 (2.14) b b Non-Hispanic, black only 17 (2.47) 30 (2.32) 26 (1.73) Non-Hispanic, Asian only NA 1 (0.57) 2 (0.60) 16 The 2016 total population for expansion and non-expansion states were derived from the U.S. Census Bureau, Annual Estimates of the Resident Population: April 1, 2010, to July 1, 2017. See https://factfinder.census.gov/faces/tableservices/jsf/pages/productview.xhtml?pid=PEP_20 17_PEPANNRES&src=pt, accessed on June 11, 2018. Page 10 GAO-18-607 Medicaid Expansion and Access to Care Expansion states Non-expansion states All states Demographic characteristics Percent (standard error) Percent (standard error) Percent (standard error) N = 1,915,000 N = 3,721,000 N = 5,639,000 Non-Hispanic, other 3 (1.02) 4 (0.98) 4 (0.73) Employment status Employed 64 (2.43) 56 (1.60) 58 (1.37) Unemployed 14 (1.85) 17 (1.35) 16 (1.10) b b Not in workforce 22 (2.17) 28 (1.60) 26 (1.30) Legend: NA = Not available because of estimates did not meet the NCHS standards for accuracy and precision. Source: GAO summary of the 2016 NHIS estimates produced by NCHS. │GAO-18-607 Notes: Estimates reflect questions included in the Family Core component of the National Health Interview Survey (NHIS). Estimates are based on household interviews of a sample of the civilian, noninstitutionalized U.S. population. Estimates were not available when they did not meet the National Center for Health Statistics’ (NCHS) standards for accuracy and precision. The percentages may not sum to 100 percent due to rounding. For the purposes of these NHIS analyses, low-income adults are individuals ages 19 to 64, with family incomes that did not exceed 138 percent of the federal poverty level (FPL). These analyses excluded low-income adults who were noncitizens, were covered by Medicare, only received health care services through military health care or through the Indian Health Service, or had Supplemental Social Security Income. We also excluded adult females from the Sample Adult file who reported being pregnant. An individual was defined as uninsured if they did not have any private health insurance, Medicare, Medicaid, Children’s Health Insurance Program, state-sponsored or other government-sponsored health plan, or military plan. An individual was also defined as uninsured if they had only a private plan that paid for one type of service, such as accidents or dental care. For the purposes of the NHIS analyses, the 30 states and the District of Columbia that expanded their Medicaid programs prior to July 1, 2016, were classified as expansion states. The remaining 20 states were classified as non-expansion states. a The FPL is based on family income and family size, using the U.S. Census Bureau’s poverty thresholds. The 2016 NHIS imputed income files were used to help create the poverty variable, and this variable is based on reported and imputed family information. See https://www.cdc.gov/nchs/nhis/nhis_2016_data_release.htm. b Difference between expansion and non-expansion states was statistically significant at p < 0.05. Estimates from the 2016 NHIS showed some statistically significant differences in the health status of uninsured, low-income adults in expansion and non-expansion states. In particular, expansion states had a larger percentage of these adults who reported that their health was “good” and a smaller percentage who reported their health as “fair or poor” than those in non-expansion states. However, the percentages of uninsured, low-income adults with responses of “excellent or very good” in both expansion and non-expansion states were large—47 percent or larger, and the differences between the two groups of states were not statistically significant. (See fig. 3.) See table 7 in appendix III for additional information about the health status for uninsured, low-income adults. Page 11 GAO-18-607 Medicaid Expansion and Access to Care Figure 3: Health Status of Uninsured, Low-Income Adults in Expansion and Non- Expansion States, 2016 Notes: Estimates reflect those included in the Family and Sample Adult Core components of the National Health Interview Survey (NHIS). Estimates are based on household interviews of a sample of the civilian, noninstitutionalized U.S. population. For the purposes of these NHIS analyses, low-income adults are individuals ages 19 to 64, with family incomes that did not exceed 138 percent of the federal poverty level. These analyses excluded low-income adults who were noncitizens, were covered by Medicare, only received health care services through military health care or through the Indian Health Service, or had Supplemental Social Security Income. We also excluded adult females from the Sample Adult file who reported being pregnant. An individual was defined as uninsured if they did not have any private health insurance, Medicare, Medicaid, Children’s Health Insurance Program, state-sponsored or other government-sponsored health plan, or military plan. An individual was also defined as uninsured if they had only a private plan that paid for one type of service, such as accidents or dental care. For the purposes of these NHIS analyses, the 30 states and the District of Columbia that expanded their Medicaid programs prior to July 1, 2016, were classified as expansion states. The remaining 20 states were classified as non-expansion states. For Good and for Fair or Poor, the difference between expansion and non-expansion states was statistically significant at p < 0.05. Page 12 GAO-18-607 Medicaid Expansion and Access to Care The 2016 NHIS estimates showed that smaller percentages of low- Survey Estimates income adults in expansion states reported having any unmet medical Showed Low-Income needs compared with those in non-expansion states; and smaller percentages of those who were insured reported having any unmet Adults in Expansion medical needs compared with those who were uninsured, regardless of States and Those where they lived, for example: Who Were Insured • Low-income adults in expansion and non-expansion states. Were Less Likely to Twenty-six percent of low-income adults in expansion states reported having any unmet medical needs, compared with 40 percent of those Report Any Unmet in non-expansion states. 17 Medical Needs • Low-income adults who were insured and uninsured. Thirty-four percent or less of the low-income adults who had Medicaid or private health insurance in expansion or non-expansion states reported Access to Health Care: Measuring Any having any unmet medical needs, compared with 50 percent or more Unmet Medical Needs The National Center for Health Statistics, the of those who were uninsured in expansion or non-expansion states. 18 federal agency that conducts the National Further, among the uninsured, 50 percent of low-income adults living Health Interview Survey (NHIS), developed a in expansion states reported any unmet medical needs, compared composite measure on any unmet medical needs, which was based on six survey with 63 percent of those in non-expansion states. 19 (See fig. 4.) See questions on respondents’ ability to afford tables 8 and 9 in appendix IV for estimates of the composite measure different types of needed health care services. These questions asked whether in the past 12 we reviewed on any unmet medical needs. months respondents could not afford medical care at any time; delayed seeking medical care due to worries about costs; or could not afford needed prescription drugs, mental health or counseling, dental care, or eyeglasses. Source: GAO summary of a selected NHIS composite measure. | GAO-18-607 17 This difference was statistically significant at p < 0.05. 18 These differences were statistically significant at p < 0.05. 19 This difference was statistically significant at p < 0.05. Page 13 GAO-18-607 Medicaid Expansion and Access to Care Figure 4: Percentage of Low-Income Adults in Expansion and Non-Expansion States and by Insurance Status Who Reported Any Unmet Medical Need, 2016 Notes: Estimates are for a composite measure, which was based on responses to questions from the Access to Health Care and Utilization sections of the Family and Sample Adult Core components of the 2016 National Health Interview Survey (NHIS). For the purposes of these NHIS analyses, low- income adults are individuals ages 19 to 64, with family incomes that did not exceed 138 percent of the federal poverty level. These analyses excluded low-income adults who were noncitizens, were covered by Medicare, only received health care services through military health care or through the Indian Health Service, or had Supplemental Social Security Income. We also excluded adult females from the Sample Adult file who reported being pregnant. Low-income adults had any unmet medical needs if they reported that in the past 12 months they did not get needed medical care because they could not afford it; delayed seeking medical care due to worries about costs; or needed but did not get because they could not afford prescription medicines, mental health care or counseling, dental care, or eyeglasses. For the purposes of these NHIS analyses, the 30 states and the District of Columbia that expanded their Medicaid programs prior to July 1, 2016, were classified as expansion states. The remaining 20 states were classified as non-expansion states. Health insurance classification was based on a hierarchy of mutually exclusive categories in the following order: private health insurance, Medicaid, other coverage, and uninsured. Adults with more than one coverage type were assigned the first appropriate category in the hierarchy and those with other coverage were excluded from these analyses. The difference between all low-income adults in expansion and non-expansion states was statistically significant at p < 0.05. The difference between low-income adults who were uninsured in expansion and non-expansion states was statistically significant at p < 0.05. Differences between low-income adults who were uninsured in expansion or non-expansion states and low-income adults who were insured—Medicaid or private health insurance—in expansion or non-expansion states were statistically significant at p < 0.05. Page 14 GAO-18-607 Medicaid Expansion and Access to Care The 2016 NHIS estimates showed that smaller percentages of low- Survey Estimates income adults in expansion states reported financial barriers to needed Showed Low-Income health care compared with those in non-expansion states; and smaller percentages of those who were insured reported financial barriers to Adults in Expansion needed health care compared with those who were uninsured, regardless States and Those of where they lived, for example: Who Were Insured • Low-income adults in expansion and non-expansion states. Nine Were Less Likely to percent of low-income adults in expansion states reported that they could not afford needed medical care, compared with 20 percent of Report Financial low-income adults in non-expansion states. 20 Barriers to Health • Low-income adults who were insured and uninsured. Twelve Care percent or less of low-income adults who had Medicaid or private health insurance in expansion or non-expansion states reported financial barriers to needed medical care, compared with 27 percent Access to Health Care: Measuring Financial Barriers to Health Care or more of those who were uninsured in expansion or non-expansion The 2016 National Health Interview Survey states. 21 In addition, among low- income adults who were uninsured, (NHIS) asked respondents whether they did a smaller percentage of those who lived in expansion states reported not obtain different types of needed health financial barriers to two of the six needed health care services care services in the past 12 months because they could not afford the service. Services compared with those who lived in non-expansion states. 22 (See fig. 5.) included: See tables 10 through 13 in appendix V for estimates of all survey • medical care, questions we reviewed on financial barriers to health care. • specialty care, • prescription medications, • mental health care or counseling, • dental care, and • eyeglasses. Source: GAO summary of selected NHIS questions. | GAO-18-607 20 This difference was statistically significant at p < 0.05. 21 These differences were statistically significant at p < 0.05. 22 For financial barriers to medical care and prescription drugs, differences were statistically significant at p < 0.05. For mental health care, dental care, specialty care, and eyeglasses, differences were not statistically significant at p < 0.05. Page 15 GAO-18-607 Medicaid Expansion and Access to Care Figure 5: Percentage of Low-Income Adults in Expansion and Non-Expansion States and by Insurance Status Who Reported Financial Barriers to Different Types of Needed Health Care Services, 2016 Page 16 GAO-18-607 Medicaid Expansion and Access to Care Notes: Estimates are for selected questions, which were from the Access to Health Care and Utilization sections of the Family and Sample Adult Core components of the 2016 National Health Interview Survey (NHIS). For the purposes of these NHIS analyses, low-income adults are individuals ages 19 to 64, with family incomes that did not exceed 138 percent of the federal poverty level. These analyses excluded low-income adults who were noncitizens, were covered by Medicare, only received health care services through military health care or through the Indian Health Service, or had Supplemental Social Security Income. We also excluded adult females from the Sample Adult file who reported being pregnant. For the purposes of these NHIS analyses, the 30 states and the District of Columbia that expanded their Medicaid programs prior to July 1, 2016, were classified as expansion states. The remaining 20 states were classified as non-expansion states. Health insurance classification was based on a hierarchy of mutually exclusive categories in the following order: private health insurance, Medicaid, other coverage, and uninsured. Adults with more than one coverage type were assigned the first appropriate category in the hierarchy and those with other coverage were excluded from these analyses. For all six service types, the difference between all low-income adults in expansion and non- expansion states was statistically significant at p < 0.05. For two of the six service types—medical care and prescription medications—the difference between low-income adults who were uninsured in expansion and non-expansion states was statistically significant at p < 0.05. For all six service types, with four exceptions, differences between low-income adults who were uninsured in expansion or non-expansion states and low-income adults who were insured—Medicaid or private health insurance—in expansion or non-expansion states were statistically significant at p < 0.05. For mental health care, differences between low-income adults who were uninsured in expansion states and low-income adults who had Medicaid in expansion or non-expansion states were not statistically significant at p < 0.05. For eyeglasses, differences between low-income adults who were uninsured in expansion or non-expansion states and those who had Medicaid in non- expansion states were not statistically significant at p < 0.05. The 2016 NHIS also collected information on non-financial barriers to health care. Specifically, the survey asked whether respondents had delayed health care due to non-financial reasons, such as they lacked transportation, were unable to get through on the phone, were unable to get a timely appointment, experienced long wait time at the doctor’s office, or were not able to get to a clinic or doctor’s office when it was open. 23 The 2016 NHIS showed that the same or similar percentages of low-income adults in expansion and non-expansion states reported delaying care due to a lack of transportation or other non-financial reasons. 24 Further, generally similar or larger percentages of low-income adults with insurance reported delaying care due to non-financial reasons, compared with those who were uninsured. 25 See tables 14 and 15 in 23 NCHS developed a composite measure on any non-financial barriers to health care, which was based on these five survey questions. 24 Any differences in the percentage of low-income adults in expansion and non-expansion states that reported delaying care due to a lack of transportation, long wait times at the doctor’s office, or other non-financial reason were not statistically significant at p < 0.05. 25 Estimates were not available for all insured and uninsured low-income adult populations for all survey questions on non-financial barriers we reviewed, because estimates did not meet NCHS’s standards for accuracy and precision. See table 15 in appendix V. Page 17 GAO-18-607 Medicaid Expansion and Access to Care appendix V for estimates of low-income adults in expansion and non- expansion states and by insurance status on non-financial barriers to health care. The 2016 NHIS estimates showed that a larger percentage of low-income Survey Estimates adults in expansion states reported having a usual place of care Showed Low-Income compared with those in non-expansion states; and larger percentages of those who were insured reported having a usual place of care compared Adults in Expansion with those who were uninsured, regardless of where they lived, for States and Those example: 26 Who Were Insured • Low-income adults in expansion and non-expansion states. Were Generally More Eighty-two percent of the low-income adults in expansion states reported having a usual place of care when they were sick or needed Likely to Report advice about their health, compared with 68 percent of those in non- Having a Usual Place expansion states. 27 of Care and • Low-income adults who were insured and uninsured. Seventy- eight percent or more of those who had Medicaid or private health Receiving Selected insurance in expansion or non-expansion states reported having a Health Care Services usual place of care, compared with 46 percent or less of those who were uninsured in expansion or non-expansion states. 28 Among the uninsured, similar percentages of low-income adults in expansion and non-expansion states reported having a usual place of care. 29 (See Access to Health Care: Having a Usual fig. 6.) See tables 16 through 19 in appendix VI for estimates of all Place of Care The 2016 National Health Interview Survey survey questions we reviewed on having a usual place of care. (NHIS) asked respondents about whether they had a place they usually go when sick or need advice about their health. Source: GAO summary of selected NHIS questions. | GAO-18-607 26 Low-income adults who reported the emergency department as their usual place of care were considered to not have a usual place of care. 27 This difference was statistically significant at p < 0.05. 28 These differences were statistically significant at p < 0.05. 29 This difference was not statistically significant at p < 0.05. Page 18 GAO-18-607 Medicaid Expansion and Access to Care Figure 6: Percentage of Low-Income Adults in Expansion and Non-Expansion States and by Insurance Status Who Reported Having a Usual Place of Care, 2016 Notes: Estimates are for a selected survey question, which was from the Health Care Access and Utilization section of the Sample Adult Core component of the 2016 National Health Interview Survey (NHIS). For the purposes of these NHIS analyses, low-income adults are individuals ages 19 to 64, with family incomes that did not exceed 138 percent of the federal poverty level. These analyses excluded low-income adults who were noncitizens, were covered by Medicare, only received health care services through military health care or through the Indian Health Service, or had Supplemental Social Security Income. We also excluded adult females from the Sample Adult file who reported being pregnant. Low-income adults who reported the emergency department as their usual place of care were considered to not have a usual place of care. For the purposes of these NHIS analyses, the 30 states and the District of Columbia that expanded their Medicaid programs prior to July 1, 2016, were classified as expansion states. The remaining 20 states were classified as non-expansion states. Health insurance classification was based on a hierarchy of mutually exclusive categories in the following order: private health insurance, Medicaid, other coverage, and uninsured. Adults with more than one coverage type were assigned the first appropriate category in the hierarchy and those with other coverage were excluded from these analyses. The difference between all low-income adults in expansion and non-expansion states is statistically significant at p < 0.05. Differences between low-income adults who were uninsured in expansion or non-expansion states and low-income adults who were insured—Medicaid or private health insurance—in expansion or non-expansion states were statistically significant at p < 0.05. Page 19 GAO-18-607 Medicaid Expansion and Access to Care Survey Estimates Showed The 2016 estimates showed that larger percentages of low-income adults Low-Income Adults in in expansion states reported receiving selected health care services, such as a flu vaccine, compared with those in non-expansion states; and larger Expansion States and percentages of those with insurance reported receiving selected health Those Who Were Insured care services compared with those who were uninsured, regardless of Were Generally More where they lived, for example: Likely to Report Receiving • Low-income adults in expansion and non-expansion states. Selected Services Thirty-one percent of low-income adults in expansion states reported receiving flu vaccinations, compared with 24 percent of those in non- Access to Health Care: Measuring Receipt expansion states. 30 of Selected Health Care Services The 2016 National Health Interview Survey • Low-income adults who were insured and uninsured. Forty-three (NHIS) asked respondents whether they had received certain health care services in the percent or more of low-income adults who had Medicaid or private past 12 months, including: health insurance in expansion or non-expansion states reported • having their blood cholesterol checked by receiving blood cholesterol checks, compared with 28 percent or less a doctor, nurse, or other health of low-income adults who were uninsured in expansion or non- professional; expansion states. 31 Among the uninsured, generally similar • having their blood pressure checked by a doctor, nurse, or other health professional; percentages of low-income adults in expansion and non-expansion • having a fasting test for high blood sugar states reported blood cholesterol checks, flu vaccines, and other or diabetes; selected services. 32 (See fig. 7.) See tables 20 and 21 in appendix VI • getting a flu vaccine by both shot and for estimates of all survey questions we reviewed on selected health nasal spray; and care services. • visiting a hospital emergency department. Source: GAO summary of selected NHIS questions. | GAO-18-607 30 This difference was statistically significant at p < 0.05. Differences between percentages of low-income adults in expansion states and those in non-expansion states who reported receiving other selected services, such as having their blood sugar checked, were not statistically significant at p < 0.05. 31 These differences were statistically significant at p < 0.05. 32 Any differences between the percentages of low-income adults who were uninsured in expansion and non-expansion states who reported receiving selected health care services were not statistically significant at p < 0.05. Page 20 GAO-18-607 Medicaid Expansion and Access to Care Figure 7: Percentage of Low-Income Adults in Expansion and Non-Expansion States and by Insurance Status Who Reported Using Selected Health Care Services, 2016 Notes: Estimates are based on selected survey questions, which were from the Access to Health Care and Utilization sections of the Sample Adult Core component of the 2016 National Health Page 21 GAO-18-607 Medicaid Expansion and Access to Care Interview Survey (NHIS). For the purposes of these NHIS analyses, low-income adults are individuals ages 19 to 64, with family incomes that did not exceed 138 percent of the federal poverty level. These analyses excluded low-income adults who were noncitizens, were covered by Medicare, only received health care services through military health care or through the Indian Health Service, or had Supplemental Social Security Income. We also excluded adult females from the Sample Adult file who reported being pregnant. For the purposes of these NHIS analyses, the 30 states and the District of Columbia that expanded their Medicaid programs prior to July 1, 2016, were classified as expansion states. The remaining 20 states were classified as non-expansion states. Health insurance classification was based on a hierarchy of mutually exclusive categories in the following order: private health insurance, Medicaid, other coverage, and uninsured. Adults with more than one coverage type were assigned the first appropriate category in the hierarchy and those with other coverage were excluded from these analyses. For blood cholesterol check, blood pressure check, and flu vaccine, the differences between all low- income adults in expansion states and non-expansion states was statistically significant at p < 0.05. For all five services with four exceptions, differences between low-income adults who were uninsured in expansion or non-expansion states and low-income adults who were insured—Medicaid or private health insurance—in expansion or non-expansion states were statistically significant at p < 0.05. For receiving a blood sugar check, the difference between low-income adults who were uninsured in non- expansion states and low-income adults who had private health insurance in non-expansion states was not statistically significant at p < 0.05. For visiting the hospital emergency department, the difference between low-income adults who were uninsured in non-expansion states and those who had Medicaid in expansion states was not statistically significant at p < 0.05, and the differences between the low-income adults who were uninsured in expansion states and those who had private health insurance in expansion or non-expansion states were not statistically significant at p < 0.05. a Includes vaccines by both shot and nasal spray. The 2016 NHIS also asked respondents whether they visited or had spoken to a health care professional about their health, including: • a general doctor, such as a general practitioner, family doctor, and internist; • a nurse practitioner, physician’s assistant, or midwife; and • a doctor who specializes in a particular disease, with the exception of obstetricians, gynecologists, psychiatrists, and ophthalmologists. See tables 22 and 23 in appendix VI for estimates of low-income adults in expansion and non-expansion states and by insurance status on contacting health care professionals. We provided a draft of this report to HHS for comment. HHS provided Agency Comments technical comments, which we incorporated as appropriate. and Our Evaluation As agreed with your offices, unless you publicly announce the contents of this 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, the appropriate congressional Page 22 GAO-18-607 Medicaid Expansion and Access to Care committee, and other interested parties. In addition, this report is available at no charge on the GAO website at http://www.gao.gov. If you are your staff members have any questions about this report, please contact me at (202) 512-7114 or yocomc@gao.gov. Contact points for our Offices of Congressional Relations and Public Affairs may be found on the last page of this report. Major contributors to this report are listed in appendix VII. Sincerely yours, Carolyn L. Yocom Director, Health Care Page 23 GAO-18-607 Medicaid Expansion and Access to Care Appendix I: Objectives, Scope, and Appendix I: Objectives, Scope, and Methodology Methodology To describe national survey estimates of (1) the number and demographic characteristics of uninsured, low-income adults in expansion and non-expansion states; (2) unmet medical needs for low-income adults in expansion and non-expansion states and by insurance status; (3) barriers to health care for low-income adults in expansion and non- expansion states and by insurance status; and (4) having a usual place of care and receiving selected health care services for low-income adults in expansion and non-expansion states and by insurance status, we used data from the 2016 National Health Interview Survey (NHIS). 1 The 2016 NHIS were the most recent data available when we conducted our analyses. This appendix describes the data source, study population, analyses conducted, study limitations, and data reliability assessment. Data Source The NHIS collects demographic, health status, health insurance, health care access, and health care service use data for the civilian, noninstitutionalized U.S. population. It is an annual, nationally representative, cross-sectional household interview survey. 2 NHIS interviews are conducted continuously throughout the year for the National Center for Health Statistics (NCHS), which is a federal agency within the Department of Health and Human Services that compiles statistical information to help guide health policy decisions. Interviews are conducted in respondents’ homes, and interviewers may conduct follow- up interviews over the telephone to complete an interview. Information about some NHIS respondents, such as information about their health status, may be obtained through an interview with another family member on behalf of the respondent. 1 The estimates of demographic characteristics and access to health care from the 2016 NHIS for this study were produced by Robin A. Cohen, and Emily P. Zammitti of National Center for Health Statistics’ Division of Health Interview Statistics based on the analysis plans we provided. See R.A. Cohen and E.P. Zammitti, Coverage, Access, and Utilization by Medicaid Expansion Status: Estimates from the National Health Interview Survey, United States, 2016, (Hyattsville, MD.: National Center for Health Statistics, December 2017), accessed December 12, 2017, https://www.cdc.gov/nchs/health_policy/coverage_and_access.htm; and R.A. Cohen and E.P. Zammitti, Access and Utilization by Medicaid Expansion Status for Low-Income Adults Aged 19-64: Estimates from the National Health Interview Survey, United States, 2016, (Hyattsville, MD.: National Center for Health Statistics, January 2018), accessed January 17, 2018, https://www.cdc.gov/nchs/health_policy/coverage_and_access.htm. 2 Several segments of the U.S. population are not included in the NHIS. Examples of persons excluded from the survey are patients in long-term care facilities, persons on active duty with the Armed Forces (although their dependents are included), persons incarcerated in the prison system, and U.S. nationals living in foreign countries. Page 24 GAO-18-607 Medicaid Expansion and Access to Care Appendix I: Objectives, Scope, and Methodology NHIS data are organized into several data files. Estimates used for our study are based on data with the 2016 Family and Sample Adult Core components of the 2016 NHIS. Sociodemographic, insurance, and select health care access and utilization variables were defined using data collected in the Family Core component of the survey, which includes data on every household member for the families participating in NHIS. Other measures of health care access and utilization examined in this study are based on data collected in the Sample Adult Core component. In this component, the respondent (i.e., the sample adult) is randomly selected from among all adults aged ≥18 years in the family. A proxy respondent might respond for the sample adult if, because of health reasons, the sample adult is physically or mentally unable to respond themselves. The 2016 imputed income files were used to define poverty thresholds, which is based on reported and imputed family income. The NHIS publicly released data files for 2016 include data for 40,220 households containing 97,169 persons, and the total household response rate was 67.9 percent. Study Population For this study we asked NCHS to provide estimates of low-income, non- elderly adults, which we defined as individuals ages 19 to 64, with family incomes that did not exceed 138 percent of the federal poverty level (FPL). 3 We also requested that estimates be provided separately for respondents based on whether they resided in an expansion or non- expansion state, and whether they were covered by private health insurance, Medicaid, or had no insurance. We gave NCHS specifications for the definition of low-income, non-elderly adults; the states that should be classified as expansion or non-expansion states in calendar year 2016; and the respondents who should be classified as having private health insurance, Medicaid, or no insurance. 3 We chose this definition because under the Patient Protection and Affordable Care Act (PPACA), states may opt to expand their Medicaid programs to cover non-elderly, non- pregnant adults who are not eligible for Medicare, and whose income does not exceed 133 percent of the FPL beginning January 1, 2014. Pub. L. No. 111-148, 124 Stat. 119 (2010), as amended by the Health Care and Education Reconciliation Act of 2010, Pub. L. No. 111-152, 124 Stat. 1029 (2010). PPACA also provides for a 5 percent disregard when calculating income for determining Medicaid eligibility, which effectively increases income eligibility from 133 percent of FPL to 138 percent of FPL. The FPL is based on family income and family size, using the U.S. Census Bureau’s poverty thresholds. In 2016, 138 percent of FPL for an individual was $17,231 and for a family of four was $33,897. PPACA also permitted an early expansion option, whereby states could expand eligibility for this population, or a subset of this population, starting on April 1, 2010. Page 25 GAO-18-607 Medicaid Expansion and Access to Care Appendix I: Objectives, Scope, and Methodology We asked NCHS to exclude respondents who were noncitizens, were covered by Medicare, only received health care services through military health care or through the Indian Health Service, or had Supplemental Social Security Income. We also excluded adult females from the Sample Adult file who responded they were pregnant at the time of the interview. In addition, we asked NCHS to exclude individuals for which information was missing—not recorded or not provided during the interview—on health insurance coverage (Medicaid, private health insurance, Indian Health Service, military health care, or no health insurance), receipt of Supplemental Social Security Income, and U.S. citizenship. We classified individuals in our study population as residing in an expansion or non-expansion state based on their state of residence when they were interviewed for the 2016 NHIS. We classified the 30 states and the District of Columbia that expanded their Medicaid eligibility before July 1, 2016, as expansion states. The remaining 20 states were classified as non-expansion states. Louisiana expanded Medicaid coverage on July 1, 2016; therefore, we classified it as a non-expansion state. We decided not to classify Louisiana as an expansion state because we allowed a 6- month period for the effects of expansion to appear. Therefore, for Louisiana we only included NHIS respondents interviewed from January through June 2016 when Louisiana was a non-expansion state. Similarly, for two expansion states—Alaska and Montana—we only included individuals who were interviewed March through December 2016 and July through December 2016, respectively, after the state expanded Medicaid to allow for a 6-month time period for the effect of expansion to take place. (See table 2.) Table 2: States Classified as Expansion States and Non-Expansion States for this Study Expansion states (date of expansion) Non-expansion states a Alaska (September 1, 2015) Alabama Arizona (January 1, 2014) Florida Arkansas (January 1, 2014) Georgia California (January 1, 2014) Idaho Colorado (January 1, 2014) Kansas Connecticut (January 1, 2014) Louisianab Delaware (January 1, 2014) Maine District of Columbia (January 1, 2014) Mississippi Hawaii (January 1, 2014) Missouri Illinois (January 1, 2014) Nebraska Page 26 GAO-18-607 Medicaid Expansion and Access to Care Appendix I: Objectives, Scope, and Methodology Expansion states (date of expansion) Non-expansion states Indiana (February 1, 2015) North Carolina Iowa (January 1, 2014) Oklahoma Kentucky (January 1, 2014) South Carolina Maryland (January 1, 2014) South Dakota Massachusetts (January 1, 2014) Tennessee Michigan (April 1, 2014) Texas Minnesota (January 1, 2014) Utah c Montana (January 1, 2016) Virginia Nevada (January 1, 2014) Wisconsin New Hampshire (August 15, 2014) Wyoming New Jersey (January 1, 2014) New Mexico (January 1, 2014) New York (January 1, 2014) North Dakota (January 1, 2014) Ohio (January 1, 2014) Oregon (January 1, 2014) Pennsylvania (January 1, 2015) Rhode Island (January 1, 2014) Vermont (January 1, 2014) Washington (January 1, 2014) West Virginia (January 1, 2014) Source: GAO summary of Kaiser Family Foundation’s data | GAO-18-607 Note: For this study we classified the 30 states and the District of Columbia that expanded their Medicaid programs before July 1, 2016, as expansion states. We classified the remaining 20 states as non-expansion states. a We excluded individuals interviewed in Alaska from January through February 2016 to allow for a 6- month time period for the effects of the expansion to take place. b We only included individuals interviewed in Louisiana from January through June 2016 (the 6-month period before Louisiana expanded Medicaid on July 1, 2016), and classified Louisiana as a non- expansion state. c We excluded individuals interviewed in Montana from January through June 2016 to allow for a 6- month time period for the effects of the expansion to take place. Table 3 below illustrates the sample size and population estimates of low- income sample adults by expansion state, non-expansion state, and national total. Page 27 GAO-18-607 Medicaid Expansion and Access to Care Appendix I: Objectives, Scope, and Methodology Table 3: Sample Sizes and Population Estimates of Low-Income Adults by State Expansion Status Expansion states Non-expansion states All states Sample size Population Sample size Population Sample size Population estimate estimate estimate (thousands) (thousands) (thousands) Low-Income adults 2,027 14,564 1,476 9,459 3,503 24,023 Source: GAO summary of 2016 NHIS estimates produced by National Center for Health Statistics. │ GAO-18-607 Notes: Estimates are from the Family Core and Sample Adult components of the 2016 National Health Interview Survey (NHIS). For the purposes of these NHIS analyses, low-income adults are individuals ages 19 through 64, with family incomes that did not exceed 138 percent of the federal poverty level. These analyses excluded low-income adults who were noncitizens, were covered by Medicare, only received health care services through military health care or through the Indian Health Service, or had Supplemental Social Security Income. We also excluded adult females from the Sample Adult file who reported being pregnant. For the purposes of these NHIS analyses, the 30 states and the District of Columbia that expanded their Medicaid programs before July 1, 2016, were classified as expansion states. The remaining 20 states were classified as non-expansion states. We classified NHIS respondents as having private health insurance, Medicaid, or no insurance based on the health insurance classification approach used by NCHS for NHIS. NCHS assigned NHIS respondents’ health insurance classification based on a hierarchy of mutually exclusive categories in the following order: private health insurance, Medicaid, other coverage, and uninsured. 4 Low-income adults with more than one coverage type were assigned to the first appropriate category in the hierarchy. Respondents were classified as having private health insurance if they reported that they were covered by any comprehensive private health insurance plan (including health maintenance and preferred provider organizations). Private coverage excluded plans that pay for one type of service, such as accidents or dental care. Respondents were classified as having Medicaid if they reported they were covered by Medicaid or by a state-sponsored health plan with no premiums or it was not known whether a premium was charged. Respondents were classified as being uninsured if they did not report having any private health insurance, Medicare, Medicaid, Children’s Health Insurance Program, state-sponsored or other government-sponsored health plan, or military health plan. Respondents were also classified as being uninsured if they 4 Our study excludes NHIS respondents classified in NCHS’s health insurance hierarchy as having “other coverage.” NCHS defines other coverage as insurance coverage with the Children’s Health Insurance Program, government sponsored-health plans, Medicare, or other state sponsored health plans with a premium. Page 28 GAO-18-607 Medicaid Expansion and Access to Care Appendix I: Objectives, Scope, and Methodology only had insurance coverage with a private plan that paid for one type of service, such as accidents or dental care. Analyses Conducted We gave NCHS officials specifications to calculate estimates from the 2016 NHIS for demographic characteristics, access to care, as well as composite measures of access to health care based on selected survey questions. Composite measures are NCHS-developed measures based on responses to NHIS questions covering related topics. The analysis included two composite measures: 1. any unmet medical needs, which is based on responses to six underlying survey questions that asked respondents about whether during the past 12 months they needed medical care but did not get it because they could not afford it; delayed seeking medical care because of worry about the cost; or did not get prescription medicines, mental health care or counseling, eyeglasses, or dental care due to cost; and 2. any non-financial barriers to health care, which is based on five underlying questions that asked respondents whether they delayed care in the past 12 months for any of the following reasons: could not get through on the telephone; could not get an appointment soon enough; waited too long to see the doctor after arriving at the doctor’s office; the clinic/doctor’s office was not open when respondent could get there; and did not have transportation. NCHS officials calculated our requested estimates of groups within our study population based on whether respondents resided in an expansion or non-expansion state and whether they had private health insurance, Medicaid, or were uninsured at the time of the interview. For each comparison—such as comparisons of access to health care for respondents in expansion versus non-expansion states—we asked NCHS to test for statistically significant differences. We identified a statistically significant difference when the p-value from a t-test of the difference in the estimated proportions between two study subgroups had a value of less than 0.05. To describe the number and demographic characteristics of uninsured, low-income adults, we compared estimates of selected demographic characteristics (race and ethnicity, gender, poverty status, and employment status) and reported health status for this group in expansion Page 29 GAO-18-607 Medicaid Expansion and Access to Care Appendix I: Objectives, Scope, and Methodology and non-expansion states. 5 These and other estimates of demographic characteristics and reported health status from the 2016 NHIS for uninsured, low-income adults by expansion states, non-expansion states, and all states are provided in tables 6 and 7 in appendix III. To describe unmet medical needs, barriers to health care, and having a usual place of care and receiving selected services for all low-income adults in expansion and non-expansion states and by insurance status, we asked NCHS to calculate estimates based on responses to selected NHIS questions and NCHS composite measures. We selected these survey questions and composite measures from the Family and Adult Access to Health Care and Utilization and Adult Health Behaviors sections of the 2016 NHIS. To summarize estimates of low-income adults in expansion and non-expansion states and by insurance status, responses to selected survey questions and composite measures were calculated as an estimated percentage of the relevant group’s total population for eight groups of low-income adults: (1) those in expansion states, (2) those in non-expansion states, (3) those who had Medicaid in expansion states, (4) those who had Medicaid in non-expansion states, (5) those who had private health insurance in expansion states, (6) those who had private health insurance in non-expansion states, (7) those who were uninsured in expansion states, and (8) those who were uninsured in non-expansion states. We asked NCHS to test for statistically significant differences for the estimates of access to care between selected groups of low-income adults. (See table 4.) The results of the tests for statistically significant differences for these comparison groups are in appendixes IV through VI. Table 4: Comparison Groups of Low-Income Adults for Tests of Statistically Significant Differences in Selected Measures of Access to Care Comparison Groups of Low-Income Adults All in expansion states versus All in non-expansion states Uninsured in expansion states versus Uninsured in non-expansion states Medicaid in expansion states versus Medicaid in non-expansion states Private health insurance in expansion states versus Private health insurance in non-expansion states 5 We tested for statistically significant differences between uninsured, low-income adults in expansion and non-expansion states. Page 30 GAO-18-607 Medicaid Expansion and Access to Care Appendix I: Objectives, Scope, and Methodology Uninsured in expansion states versus Medicaid in expansion states Uninsured in expansion states versus Medicaid in non-expansion states Uninsured in expansion states versus Private health insurance in expansion states Uninsured in expansion states versus Private health insurance in non-expansion states Uninsured in non-expansion states versus Medicaid in expansion states Uninsured in non-expansion states versus Medicaid in non-expansion states Uninsured in non-expansion states versus Private health insurance in expansion states Uninsured in non-expansion states versus Private health insurance in non-expansion states Source: GAO. | GAO-18-607 Notes: Tests for statistically significant differences of estimates of access to care for selected subgroups of low-income adults were produced by the National Center for Health Statistics using data from the 2016 National Health Interview Survey (NHIS). For the purposes of these NHIS analyses, low-income adults were individuals ages 19 through 64, with family incomes that did not exceed 138 percent of the federal poverty level. These analyses excluded low-income adults who were noncitizens, were covered by Medicare, only received health care services through military health care or through the Indian Health Service, or had Supplemental Social Security Income. We also excluded adult females from the Sample Adult file who reported being pregnant. For the purposes of these NHIS analyses, the 30 states and the District of Columbia that expanded their Medicaid programs before July 1, 2016, were classified as expansion states. The remaining 20 states were classified as non-expansion states. Study Limitations and Our study has some limitations. First, our study did not examine whether Data Reliability statistically significant differences in estimates of access to health care between respondents in expansion and non-expansion states were Assessment associated with the choice to expand Medicaid. 6 Second, NHIS data are based on respondent-reported data, which may be subject to potential biases and recall of participants’ use of health services and may be less accurate than administrative data or clinical data. Third, we could not 6 To examine whether differences in access to health care between respondents in expansion and non-expansion states were associated with Medicaid expansion, we would need to simultaneously examine multiple factors—such as demographics, health status, community characteristics, and other factors—that may differ between respondents in expansion and non-expansion states. Examining whether estimated differences in access to health care between respondents in expansion states and non-expansion states were associated with expansion would require other analysis methods—such as regression analysis. Regression analysis is a statistical technique in which the functional relationship between a dependent variable (such as whether a respondent had a usual place of care) and one or more independent variables (such as whether the respondent resided in an expansion state or the respondent’s level of education) can be estimated. We did not conduct a regression analysis for this study. Page 31 GAO-18-607 Medicaid Expansion and Access to Care Appendix I: Objectives, Scope, and Methodology report estimates of access to health care that did not meet NCHS’s standards of reliability or precision. 7 We assessed the reliability of NHIS data by reviewing NHIS data documentation; interviewing knowledgeable NCHS officials and academic researchers; and examining the data for logical errors, missing values, and values outside of expected ranges. We determined that the data were sufficiently reliable for the purposes of these analyses. 7 For example, estimates for four survey questions about non-financial barriers to care for uninsured, low-income adults in expansion states did not meet NCHS standards. HHS officials suggest that for more information about NCHS’s standards of reliability or precision, see Parker JD, Talih M, Malec DJ, Beresovsky V, Carroll M, Gonzalez JF, et al. National Center for Health Statistics Data Presentation Standards for Proportions. National Center for Health Statistics. Vital Health Stat 2(175). 2017. Available from: https://www.cdc.gov/nchs/data/series/sr_02/sr02_175.pdf. Page 32 GAO-18-607 Medicaid Expansion and Access to Care Appendix II: Status of Medicaid Eligibility Appendix II: Status of Medicaid Eligibility Expansion by States, as of 2017 Expansion by States, as of 2017 Under the Patient Protection and Affordable Care Act (PPACA), states may opt to expand their Medicaid programs’ eligibility to cover certain low-income adults beginning January 2014. 1 As of December 2017, 31 states and the District of Columbia had expanded their Medicaid programs as permitted under PPACA and 19 states had not. 2 Table 5 lists the states that expanded Medicaid eligibility and those that did not. It also includes state population and other Medicaid data, which is presented in the roll-over information in interactive figure 2. Table 5: Status of Medicaid Eligibility Expansion, by State, as of 2017 State Status of Medicaid Date of Medicaid State population (as of Medicaid enrollment expansion expansion July 2017) (calendar year 2017) Alabama Non-expansion None 4,874,747 1,018,770 Alaska Expansion September 1, 2015 739,795 187,461 Arizona Expansion January 1, 2014 7,016,270 1,936,263 Arkansas Expansion January 1, 2014 3,004,279 940,776 California Expansion January 1, 2014 39,536,653 11,517,800 Colorado Expansion January 1, 2014 5,607,154 1,347,895 Connecticut Expansion January 1, 2014 3,588,184 898,660 Delaware Expansion January 1, 2014 961,939 208,695 District of Columbia Expansion January 1, 2014 693,972 259,057 Florida Non-expansion None 20,984,400 4,030,767 Georgia Non-expansion None 10,429,379 1,865,535 Hawaii Expansion January 1, 2014 1,427,538 329,675 Idaho Non-expansion None 1,716,943 319,739 Illinois Expansion January 1, 2014 12,802,023 2,865,435 Indiana Expansion February 1, 2015 6,666,818 1,346,668 Iowa Expansion January 1, 2014 3,145,711 594,586 1 Specifically, states may opt to expand their Medicaid programs to cover non-elderly, non- pregnant adults who are not eligible for Medicare and whose income does not exceed 133 percent of the federal poverty level (FPL). PPACA also provides for a 5 percent disregard when calculating income for determining Medicaid eligibility, which effectively increases income eligibility from 133 percent of FPL to 138 percent of FPL. PPACA also permitted an early expansion option, whereby states could expand eligibility for this population, or a subset of this population, starting on April 1, 2010. 2 Louisiana expanded Medicaid eligibility on July 1, 2016. For the purposes of these NHIS analyses, Louisiana was classified as a non-expansion state and we only included low- income adults interviewed from January through June 2016 when Louisiana was a non- expansion state. Page 33 GAO-18-607 Medicaid Expansion and Access to Care Appendix II: Status of Medicaid Eligibility Expansion by States, as of 2017 State Status of Medicaid Date of Medicaid State population (as of Medicaid enrollment expansion expansion July 2017) (calendar year 2017) Kansas Non-expansion None 2,913,123 373,437 Kentucky Expansion January 1, 2014 4,454,189 1,340,826 Louisianaa Expansion July 1, 2016 4,684,333 1,637,972 Maine Non-expansion None 1,335,907 258,710 Maryland Expansion January 1, 2014 6,052,177 1,172,931 Massachusetts Expansion January 1, 2014 6,859,819 1,787,828 Michigan Expansion April 1, 2014 9,962,311 2,136,440 Minnesota Expansion January 1, 2014 5,576,606 1,116,804 Mississippi Non-expansion None 2,984,100 709,656 Missouri Non-expansion None 6,113,532 978,801 Montana Expansion January 1, 2016 1,050,493 238,793 Nebraska Non-expansion None 1,920,076 227,531 Nevada Expansion January 1, 2014 2,998,039 603,094 New Hampshire Expansion August 15, 2014 1,342,795 194,606 New Jersey Expansion January 1, 2014 9,005,644 1,694,461 New Mexico Expansion January 1, 2014 2,088,070 885,472 New York Expansion January 1, 2014 19,849,399 1,523,749 North Carolina Non-expansion None 10,273,419 1,678,020 North Dakota Expansion January 1, 2014 755,393 69,169 Ohio Expansion January 1, 2014 11,658,609 3,002,378 Oklahoma Non-expansion None 3,930,864 671,524 Oregon Expansion January 1, 2014 4,142,776 987,823 Pennsylvania Expansion January 1, 2015 12,805,537 2,795,591 Rhode Island Expansion January 1, 2014 1,059,639 308,038 South Carolina Non-expansion None 5,024,369 1,248,324 South Dakota Non-expansion None 869,666 101,921 Tennessee Non-expansion None 6,715,984 1,615,817 Texas Non-expansion None 28,304,596 4,331,612 Utah Non-expansion None 3,101,833 302,445 Vermont Expansion January 1, 2014 623,657 182,041 Virginia Non-expansion None 8,470,020 1,029,118 Washington Expansion January 1, 2014 7,405,743 1,822,796 West Virginia Expansion January 1, 2014 1,815,857 552,292 Wisconsin Non-expansion None 5,795,483 1,188,306 Wyoming Non-expansion None 579,315 61,303 Source: GAO summary of U.S. Census Bureau, Centers for Medicare & Medicaid Services, and Kaiser Family Foundation’s data | GAO-18-607 Page 34 GAO-18-607 Medicaid Expansion and Access to Care Appendix II: Status of Medicaid Eligibility Expansion by States, as of 2017 Note: Census population estimates are as of July 1, 2017. Medicaid enrollment numbers come from CMS-64 enrollment data and may differ from other data sources. Average monthly enrollment was tabulated from the CMS-64 enrollment report for calendar year 2017 that was accessed April 4, 2018. a Louisiana expanded Medicaid eligibility on July 1, 2016. For the purposes of these NHIS analyses, Louisiana was classified as a non-expansion state and we only included low-income adults interviewed from January through June 2016 when Louisiana was a non-expansion state. Page 35 GAO-18-607 Medicaid Expansion and Access to Care Appendix III: Estimates of Demographic Appendix III: Estimates of Demographic Characteristics and Health Status in Expansion and Non-Expansion States Characteristics and Health Status in Expansion and Non-Expansion States This appendix provides additional 2016 National Health Interview Survey (NHIS) estimates we obtained from the National Center for Health Statistics (NCHS). 1 Table 6 presents estimates of selected demographic characteristics for low-income adults who were uninsured at the time of the survey interview. 2 The table provides estimates for these adults based on whether they resided in states that expanded Medicaid eligibility as permitted under the Patient Protection and Affordable Care Act (PPACA) (referred to as expansion states) or states that did not (referred to as non- expansion states). 3 We report statistically significant differences when comparing the responses of uninsured, low-income adults in expansion and non-expansion states. 1 NCHS provides additional estimates on its website of reported demographic characteristics and selected health conditions for individuals ages 19 to 64 who were uninsured at the time of the interview and had family incomes less than or equal to 138 percent of the federal poverty level (FPL). See R.A. Cohen and E.P. Zammitti, Coverage, Access, and Utilization by Medicaid Expansion Status: Estimates from the National Health Interview Survey, United States, 2016, (Hyattsville, Md.: National Center for Health Statistics, December 2017), accessed December 12, 2017, https://www.cdc.gov/nchs/health_policy/coverage_and_access.htm. 2 In this report, low income adults are individuals ages 19 to 64, with family incomes that did not exceed 138 percent of FPL. FPL is based on family income and size, using the U.S. Census Bureau’s poverty thresholds. In 2016, 138 percent of FPL for an individual was $17,231 and for a family of four was $33,897. These analyses excluded those who were covered by Medicare, had only military health care, had only Indian Health Service, were noncitizens, or had Supplemental Social Security Income. We also excluded adult females from the Sample Adult file who reported being pregnant. An individual is defined as uninsured if they did not have any private health insurance, Medicare, Medicaid, Children’s Health Insurance Program, state-sponsored or other government-sponsored health plan, or military plan. An individual is also defined as uninsured if they had only a private plan that paid for one type of service, such as accidents or dental care. 3 Historically, Medicaid eligibility has been limited to certain categories of low-income individuals such as children, parents, pregnant women, persons with disabilities, and individuals ages 65 and older. Under PPACA, states may opt to expand their Medicaid programs to cover non-elderly, non-pregnant adults who are not eligible for Medicare and whose income does not exceed 133 percent of FPL. PPACA also provides for a 5 percent disregard when calculating income for determining Medicaid eligibility, which effectively increases income eligibility from 133 percent of FPL to 138 percent of FPL. Page 36 GAO-18-607 Medicaid Expansion and Access to Care Appendix III: Estimates of Demographic Characteristics and Health Status in Expansion and Non-Expansion States Table 6: Demographic Characteristics of Uninsured, Low-Income Adults in Expansion and Non-Expansion States, 2016 Expansion states Non-expansion states All states Demographic characteristics Percent (standard error) Percent (standard error) Percent (standard error) N = 1,915,000 N = 3,721,000 N = 5,639,000 Age group 19-34 years 53.7 (2.96) 47.7 (1.82) 49.8 (1.60) 35-49 years 23.3 (2.29) 28.2 (1.60) 26.5 (1.33) 50-64 years 23.0 (2.31) 24.1 (1.67) 23.7 (1.36) a Poverty status Less than 100% FPL 58.5 (3.28)b 68.5 (2.05)b 65.1 (1.81) 100% to less than or equal to 41.5 (3.28)b 31.5 (2.05)b 34.9 (1.81) 138% FPL Sex Male 62.7 (2.33)b 50.6 (1.49)b 54.7 (1.31) b b Female 37.3 (2.33) 49.4 (1.49) 45.3 (1.31) Race and ethnicity Hispanic 21.9 (2.64) 19.7 (3.26) 20.4 (2.30) Non-Hispanic, white only 53.3 (3.40) 45.0 (2.70) 47.9 (2.14) b b Non-Hispanic, black only 17.4 (2.47) 29.8 (2.32) 25.6 (1.73) Non-Hispanic, Asian only NA 1.3 (0.57) 2.2 (0.60) Non-Hispanic, other 3.2 (1.02) 4.3 (0.98) 3.9 (0.73) Marital status Married 26.3 (3.00) 28.3 (1.76) 27.6 (1.54) Widowed 1.3 (0.48) 2.4 (0.56) 2.1 (0.40) Divorced or separated 12.3 (1.79) 16.1 (1.21) 14.8 (1.02) Living with a partner 16.9 (2.04) 14.4 (1.63) 15.2 (1.28) Never married 43.2 (3.10) 38.8 (2.08) 40.3 (1.74) Number of family members 1 23.4 (2.30) 18.7 (1.63) 20.3 (1.34) 2-4 56.2 (2.88) 60.9 (2.08) 59.3 (1.68) 5 or more 20.4 (2.77) 20.4 (1.73) 20.4 (1.49) Employment status Employed 63.8 (2.43) 55.7 (1.60) 58.4 (1.37) Unemployed 14.3 (1.85) 16.8 (1.35) 15.9 (1.10) b b Not in workforce 22.0 (2.17) 27.5 (1.60) 25.7 (1.30) Education status Less than high school 29.5 (2.84) 29.1 (1.73) 29.3 (1.50) c High school diploma or GED 34.6 (2.65) 40.7 (1.87) 38.7 (1.52) Page 37 GAO-18-607 Medicaid Expansion and Access to Care Appendix III: Estimates of Demographic Characteristics and Health Status in Expansion and Non-Expansion States Expansion states Non-expansion states All states Demographic characteristics Percent (standard error) Percent (standard error) Percent (standard error) N = 1,915,000 N = 3,721,000 N = 5,639,000 Some college 28.1 (2.49) 25.8 (1.69) 26.5 (1.39) Bachelor’s degree or more 7.8 (1.71) 4.4 (0.78) 5.5 (0.79) Legend: NA = Not available because of estimates did not meet the National Center for Health Statistics’ (NCHS) standards for accuracy and precision. Source: GAO summary of 2016 NHIS estimates produced by NCHS. │GAO-18-607 Notes: Estimates reflect questions included in the Family Core component of the National Health Interview Survey (NHIS). Estimates are based on household interviews of a sample of the civilian, noninstitutionalized U.S. population. Estimates were not available when they did not meet NCHS’s standards for accuracy and precision. For the purposes of these NHIS analyses, low-income adults are individuals ages 19 through 64, with family incomes that did not exceed 138 percent of the federal poverty level. These analyses excluded low-income adults who were noncitizens, were covered by Medicare, only received health care services through military health care or through the Indian Health Service, or had Supplemental Social Security Income. We also excluded adult females from the Sample Adult file who reported being pregnant. An individual was defined as uninsured if they did not have any private health insurance, Medicare, Medicaid, Children’s Health Insurance Program, state-sponsored or other government-sponsored health plan, or military plan. An individual was also defined as uninsured if they had only a private plan that paid for one type of service, such as accidents or dental care. For the purposes of these NHIS analyses, the 30 states and the District of Columbia that expanded their Medicaid programs prior to July 1, 2016, were classified as expansion states. The remaining 20 states were classified as non-expansion states. See R.A. Cohen and E.P. Zammitti, Coverage, Access, and Utilization by Medicaid Expansion Status: Estimates from the National Health Interview Survey, United States, 2016, (Hyattsville, Md.: National Center for Health Statistics, December 2017), accessed December 12, 2017, https://www.cdc.gov/nchs/health_policy/coverage_and_access.htm. a The FPL is based on family income and family size, using the U.S. Census Bureau’s poverty thresholds. The 2016 NHIS imputed income files were used to help create the poverty variable, and this variable is based on reported and imputed family information. See https://www.cdc.gov/nchs/nhis/nhis_2016_data_release.htm. b Difference between expansion and non-expansion states was statistically significant at p < 0.05. c General Educational Development (GED) is General Educational Development high school equivalency diploma. Table 7 shows estimates of the reported health status of uninsured, low- income adults based on whether they resided in an expansion or non- expansion state. The table provides the number and percent of these adults who reported that at the time of the interview their health status was excellent or very good; good; or fair or poor. The table also shows the extent to which these adults reported whether their health status was different at the time of the interview compared to the previous year. We report statistically significant differences when comparing the responses of uninsured, low-income adults in expansion and non-expansion states. Page 38 GAO-18-607 Medicaid Expansion and Access to Care Appendix III: Estimates of Demographic Characteristics and Health Status in Expansion and Non-Expansion States Table 7: Health Status of Uninsured, Low-Income Adults in Expansion and Non-Expansion Status, 2016 Expansion states Non-expansion states All states Health status Percent (standard error) Percent (standard error) Percent (standard error) N = 1,834,000 N = 3,382,000 N = 5,219,000 Health status Excellent or very good 47.1 (4.33) 49.9 (2.79) 49.0 (2.37) a a Good 40.0 (4.52) 28.0 (2.46) 32.2 (2.30) Fair or poor 12.9 (3.04)a 22.0 (2.29)a 18.8 (1.85) Health status compared to previous year Better 15.7 (3.05) 17.0 (2.04) 16.5 (1.70) Same 75.2 (4.03) 66.5 (2.49) 69.6 (2.15) Worse NA 16.5 (2.07) 13.9 (1.75) Legend: NA = Not available because of estimates did not meet the National Center for Health Statistics’ (NCHS) standards for accuracy and precision. Source: GAO summary of 2016 NHIS estimates produced by NCHS. │GAO-18-607 Notes: Estimates reflect questions included in the Family Core and Sample Adult Core components of the National Health Interview Survey (NHIS). Estimates are based on household interviews of a sample of the civilian, noninstitutionalized U.S. population. Estimates were not available when they did not meet NCHS’s standards for accuracy and precision. For the purposes of these NHIS analyses, low-income adults are individuals ages 19 through 64, with family incomes that did not exceed 138 percent of the federal poverty level. These analyses excluded low-income adults who were noncitizens, were covered by Medicare, only received health care services through military health care or through the Indian Health Service, or had Supplemental Social Security Income. We also excluded adult females from the Sample Adult file who reported being pregnant. An individual was defined as uninsured if they did not have any private health insurance, Medicare, Medicaid, Children’s Health Insurance Program, state-sponsored or other government-sponsored health plan, or military plan. An individual was also defined as uninsured if they had only a private plan that paid for one type of service, such as accidents or dental care. For the purposes of these NHIS analyses, the 30 states and the District of Columbia that expanded their Medicaid programs prior to July 1, 2016, were classified as expansion states. The remaining 20 states were classified as non-expansion states. See R.A. Cohen and E.P. Zammitti, Coverage, Access, and Utilization by Medicaid Expansion Status: Estimates from the National Health Interview Survey, United States, 2016, (Hyattsville, Md.: National Center for Health Statistics, December 2017), accessed December 12, 2017, https://www.cdc.gov/nchs/health_policy/coverage_and_access.htm. a For Good and for Fair or Poor, the difference between expansion and non-expansion states was statistically significant at p < 0.05. Page 39 GAO-18-607 Medicaid Expansion and Access to Care Appendix IV: Estimates of Any Unmet Appendix IV: Estimates of Any Unmet Medical Needs in Expansion and Non-Expansion States and by Insurance Status Medical Needs in Expansion and Non- Expansion States and by Insurance Status This appendix provides estimates of any unmet medical needs for low- income adults—individuals ages 19 to 64, with family incomes that did not exceed 138 percent of the federal poverty level (FPL)—from the 2016 National Health Interview Survey (NHIS), which were produced by the National Center for Health Statistics (NCHS). 1 Estimates are based on a composite measure of any unmet medical needs. Table 8 shows estimates of all low-income adults in expansion and non-expansion states. 2 We also report statistically significant differences between low- income adults in expansion and non-expansion states. 1 Estimates are based on selected survey questions, which were from the Access to Health Care and Utilization and Health Behaviors sections of the Family and Sample Adult Core components of the 2016 National Health Interview Survey (NHIS). For the purposes of these NHIS analyses, low-income adults are individuals ages 19 through 64, with family incomes that did not exceed 138 percent of the federal poverty level. These analyses excluded low-income adults who were noncitizens, were covered by Medicare, only received health care services through military health care or through the Indian Health Service, or had Supplemental Social Security Income. We also excluded adult females from the Sample Adult file who reported being pregnant. NCHS provides additional estimates of access to care and utilization for low-income adults on its website. See R.A. Cohen and E.P. Zammitti, Access and Utilization by Medicaid Expansion Status for Low- Income Adults Aged 19-64: Estimates from the National Health Interview Survey, United States, 2016, (Hyattsville, MD.: National Center for Health Statistics, January 2018), accessed January 17, 2018, https://www.cdc.gov/nchs/health_policy/coverage_and_access.htm; and R.A. Cohen and E.P. Zammitti, Coverage, Access, and Utilization by Medicaid Expansion Status: Estimates from the National Health Interview Survey, United States, 2016, (Hyattsville, Md.: National Center for Health Statistics, December 2017), accessed December 12, 2017, https://www.cdc.gov/nchs/health_policy/coverage_and_access.htm. 2 We classified the 30 states and the District of Columbia that expanded their Medicaid program prior to July 1, 2016 as expansions states, and the remaining 20 states were classified as non-expansion states. States classified as expansion states were Alaska, Arizona, Arkansas, California, Colorado, Connecticut, Delaware, Hawaii, Illinois, Indiana, Iowa, Kentucky, Maryland, Massachusetts, Michigan, Minnesota, Montana, Nevada, New Hampshire, New Jersey, New Mexico, New York, North Dakota, Ohio, Oregon, Pennsylvania, Rhode Island, Vermont, Washington, and West Virginia. The 20 non- expansion states are Alabama, Florida, Georgia, Idaho, Kansas, Louisiana, Maine, Mississippi, Missouri, Nebraska, North Carolina, Oklahoma, South Carolina, South Dakota, Tennessee, Texas, Utah, Virginia, Wisconsin, and Wyoming. For 3 states, we only included NHIS respondents for part of calendar year 2016. Specifically, we excluded respondents who were surveyed during the first 6 months after a state expanded Medicaid to allow for the effect of expansion to take place. Two of the 3 states—Alaska and Montana—were classified as expansion states and the third state, Louisiana, was classified as non-expansion. For Alaska, which expanded Medicaid on September 1, 2015, we excluded responses from those surveyed in January 2016 and February 2016. For Montana, which expanded Medicaid on January 1, 2016, we excluded responses from those surveyed from January through June 2016. Because we classified Louisiana as a non-expansion state, we excluded all responses from those surveyed after the state expanded Medicaid on July 1, 2016. Page 40 GAO-18-607 Medicaid Expansion and Access to Care Appendix IV: Estimates of Any Unmet Medical Needs in Expansion and Non-Expansion States and by Insurance Status Table 8: Percentage of Low-Income Adults in Expansion and Non-Expansion States Who Reported Any Unmet Medical Needs, 2016 Expansion states Non-expansion states (standard error) (standard error) N = 14,913,000 N = 9,627,000 Any unmet medical needs 26.4 (1.40)a 40.1 (1.91)a Source: GAO summary of 2016 NHIS estimates produced by the National Center for Health Statistics. │GAO-18-607 Notes: Estimates are for a composite measure, which was based on responses to questions from the Access to Health Care and Utilization sections of the Family and Sample Adult Core components of the 2016 National Health Interview Survey (NHIS). For the purposes of these NHIS analyses, low- income adults are individuals ages 19 through 64, with family incomes that did not exceed 138 percent of the federal poverty level. These analyses excluded low-income adults who were noncitizens, were covered by Medicare, only received health care services through military health care or through the Indian Health Service, or had Supplemental Social Security Income. We also excluded adult females from the Sample Adult file who reported being pregnant. Low-income adults had any unmet medical needs if they reported that in the past 12 months they did not get needed medical care because they could not afford it; delayed seeking medical care due to worries about costs; or needed but did not get because they could not afford prescription medicines, mental health care or counseling, dental care, or eyeglasses. For the purposes of these NHIS analyses, the 30 states and the District of Columbia that expanded their Medicaid programs prior to July 1, 2016, were classified as expansion states. The remaining 20 states were classified as non-expansion states. See R.A. Cohen and E.P. Zammitti, Access and Utilization by Medicaid Expansion Status for Low-Income Adults Aged 19-64: Estimates from the National Health Interview Survey, United States, 2016, (Hyattsville, MD.: National Center for Health Statistics, January 2018), accessed January 17, 2018, https://www.cdc.gov/nchs/health_policy/coverage_and_access.htm. a Difference between low-income adults in expansion and non-expansion states was statistically significant at p < 0.05. Table 9 shows estimates of six groups of low-income adults: (1) low- income adults who were uninsured in expansion states; (2) low-income adults who were uninsured in non-expansion states; (3) low-income adults who had Medicaid in expansion states; (4) low-income adults who had Medicaid in non-expansion states; (5) low-income adults who had private health insurance in expansion states; and (6) low-income adults Page 41 GAO-18-607 Medicaid Expansion and Access to Care Appendix IV: Estimates of Any Unmet Medical Needs in Expansion and Non-Expansion States and by Insurance Status who had private health insurance in non-expansion states. 3 We also report any statistically significant differences when comparing the six groups of low-income adults, specifically: • low-income adults who were uninsured in expansion states compared with each of the four groups of low-income adults who were insured— low-income adults who had Medicaid in expansion states, low-income adults who had Medicaid in non-expansion states, low-income adults who had private health insurance in expansion states, and low-income adults who had private insurance in non-expansion states; • low-income adults who were uninsured in non-expansion states compared with each of the four groups of low-income adults who were insured; • low-income adults who were uninsured in expansion states compared with low-income adults who were uninsured in non-expansion states; • low-income adults who had Medicaid in expansion states compared with low-income adults who had Medicaid in non-expansion states; and • low-income adults who had private health insurance in expansion states compared with low-income adults who had private health insurance in non-expansion states. 4 3 Health insurance classification was based on a hierarchy of mutually exclusive categories in the following order: private health insurance, Medicaid, other coverage, and uninsured. Low-income adults with more than one coverage type were assigned the first appropriate category in the hierarchy, and those with other coverage were excluded from these analyses. Low-income adults were classified as having private health insurance if they reported that they were covered by any comprehensive private health insurance plan, including health maintenance and preferred provider organization plans. Private health insurance excludes plans that pay for only one type of service, such as accidents or dental care. Low-income adults were classified as having Medicaid if they reported that they were covered by Medicaid or by state-sponsored health plans with no premiums or if it is not known if a premium is charged. Low-income adults were classified as uninsured if they did not have any private health insurance or were not enrolled in Medicare, Medicaid, the Children’s Health Insurance Program, a state-sponsored or other government-sponsored health plan, or a military plan. An adult was also defined as uninsured if they had only a private plan that paid for one type of service, such as accidents or dental care. 4 We did not request that NCHS test for significant differences between low-income adults who had Medicaid in expansion or non-expansion states and low-income adults who had private health insurance in expansion or non-expansion states. Page 42 GAO-18-607 Medicaid Expansion and Access to Care Appendix IV: Estimates of Any Unmet Medical Needs in Expansion and Non-Expansion States and by Insurance Status Table 9: Percentage of Low-Income Adults in Expansion and Non-Expansion States Who Reported Any Unmet Medical Needs, by Insurance Status, 2016 Uninsured Medicaid Private health insurance Expansion Non-expansion Expansion states Non-expansion Expansion states Non-expansion states (standard states (standard (standard error) states (standard (standard error) states (standard error) error) error) error) N = 1,834,000 N = 3,382,000 N = 7,671,000 N = 2,160,000 N = 5,408,000 N = 4,085,000 a, b a, c b, c b, c Any 49.6 (4.53) 62.7 (3.00) 26.5 (1.85) 34.4 (3.75) 18.2 (1.85)b, c, d 24.5 (2.11)b, c, d unmet medical needs Source: GAO summary of the 2016 NHIS estimates produced by the National Center for Health Statistics. │GAO-18-607 Notes: Estimates are for a composite measure, which was based on responses to questions from the Access to Health Care and Utilization and sections of the Family and Sample Adult Core components of the 2016 National Health Interview Survey (NHIS). For the purposes of these NHIS analyses, low- income adults are individuals ages 19 through 64, with family incomes that did not exceed 138 percent of the federal poverty level. These analyses excluded low-income adults who were noncitizens, were covered by Medicare, only received health care services through military health care or through the Indian Health Service, or had Supplemental Social Security Income. We also excluded adult females from the Sample Adult file who reported being pregnant. Adults had any unmet medical needs if they reported that in the past 12 months they did not get needed medical care because they could not afford it; delayed seeking medical care due to worries about costs; or needed but did not get because they could not afford prescription medicines, mental health care or counseling, dental care, or eyeglasses. For the purposes of these NHIS analyses, the 30 states and the District of Columbia that expanded their Medicaid programs prior to July 1, 2016, were classified as expansion states. The remaining 20 states were classified as non-expansion states. Health insurance classification was based on a hierarchy of mutually exclusive categories in the following order: private health insurance, Medicaid, other coverage, and uninsured. Low-income adults with more than one coverage type were assigned the first appropriate category in the hierarchy and those with other coverage were excluded from these analyses. See R.A. Cohen and E.P. Zammitti, Coverage, Access, and Utilization by Medicaid Expansion Status: Estimates from the National Health Interview Survey, United States, 2016, (Hyattsville, Md.: National Center for Health Statistics, December 2017), accessed December 12, 2017, https://www.cdc.gov/nchs/health_policy/coverage_and_access.htm. a Difference between low-income adults who were uninsured in expansion and non-expansion states was statistically significant at p < 0.05. b Difference between low-income adults who were uninsured in expansion states and low-income adults in expansion or non-expansion states who were insured—Medicaid or private health insurance—was statistically significant at p < 0.05. c Difference between low-income adults who were uninsured in non-expansion states and low-income adults in expansion or non-expansion states who were insured—Medicaid or private health insurance—was statistically significant at p < 0.05. d Difference between low-income adults who had private health insurance in expansion and non- expansion states was statistically significant at p < 0.05. Page 43 GAO-18-607 Medicaid Expansion and Access to Care Appendix V: Estimates of Barriers to Health Appendix V: Estimates of Barriers to Health Care in Expansion and Non-Expansion States and by Insurance Status Care in Expansion and Non-Expansion States and by Insurance Status This appendix provides estimates of barriers to health care for low- income adults—individuals ages 19 to 64, with family incomes that did not exceed 138 percent of the federal poverty level (FPL)—from the 2016 National Health Interview Survey (NHIS), which we obtained from the National Center for Health Statistics (NCHS). 1 Estimates of financial barriers to needed medical, specialty, and other types of health care and prescription drugs are based on selected survey questions. Estimates of non-financial barriers to health care are based on responses to selected survey questions and a composite measure. 2 Estimates are reported for: 1 Estimates are based on selected survey questions and a composite measure, which was based on responses to some of these questions. These survey questions were from the Access to Health Care and Utilization section of the Sample Adult Core component of the 2016 National Health Interview Survey (NHIS). For the purposes of these NHIS analyses, low-income adults are individuals ages 19 through 64, with family incomes that did not exceed 138 percent of FPL. These analyses excluded low-income adults who were noncitizens, were covered by Medicare, only received health care services through military health care or through the Indian Health Service, or had Supplemental Social Security Income. We also excluded adult females from the Sample Adult file who reported being pregnant. NCHS provides additional estimates of access to care and utilization for low-income adults on its website. See R.A. Cohen and E.P. Zammitti, Access and Utilization by Medicaid Expansion Status for Low-Income Adults Aged 19-64: Estimates from the National Health Interview Survey, United States, 2016, (Hyattsville, MD.: National Center for Health Statistics, January 2018), accessed January 17, 2018, https://www.cdc.gov/nchs/health_policy/coverage_and_access.htm; and R.A. Cohen and E.P. Zammitti, Coverage, Access, and Utilization by Medicaid Expansion Status: Estimates from the National Health Interview Survey, United States, 2016, (Hyattsville, Md.: National Center for Health Statistics, December 2017), accessed December 12, 2017, https://www.cdc.gov/nchs/health_policy/coverage_and_access.htm. 2 NCHS developed a composite measure on any non-financial barrier to care based on responses to five survey questions that asked respondents whether they delayed care due to a lack of transportation, long wait times, or other non-financial reasons. Page 44 GAO-18-607 Medicaid Expansion and Access to Care Appendix V: Estimates of Barriers to Health Care in Expansion and Non-Expansion States and by Insurance Status • All low-income adults in expansion and non-expansion states. 3 We also report statistically significant differences between low-income adults in expansion and non-expansion states. • Six groups of low-income adults: (1) low-income adults who were uninsured in expansion states; (2) low-income adults who were uninsured in non-expansion states; (3) low-income adults who had Medicaid in expansion states; (4) low-income adults who had Medicaid in non-expansion states; (5) low-income adults who had private health insurance in expansion states; and (6) low-income adults who had private health insurance in non-expansion states. 4 We also report any statistically significant differences when comparing the six groups of low-income adults, specifically: 3 We classified the 30 states and the District of Columbia that expanded their Medicaid program prior to July 1, 2016 as expansion states, and the remaining 20 states were classified as non-expansion states. States classified as expansion states were Alaska, Arizona, Arkansas, California, Colorado, Connecticut, Delaware, Hawaii, Illinois, Indiana, Iowa, Kentucky, Maryland, Massachusetts, Michigan, Minnesota, Montana, Nevada, New Hampshire, New Jersey, New Mexico, New York, North Dakota, Ohio, Oregon, Pennsylvania, Rhode Island, Vermont, Washington, and West Virginia. The 20 non- expansion states are Alabama, Florida, Georgia, Idaho, Kansas, Louisiana, Maine, Mississippi, Missouri, Nebraska, North Carolina, Oklahoma, South Carolina, South Dakota, Tennessee, Texas, Utah, Virginia, Wisconsin, and Wyoming. For 3 states, we only included NHIS respondents for part of calendar year 2016. Specifically, we excluded respondents who were surveyed during the first 6 months after a state expanded Medicaid to allow for the effect of expansion to take place. Two of the 3 states—Alaska and Montana—were classified as expansion states and the third state, Louisiana, was classified as non-expansion. For Alaska, which expanded Medicaid on September 1, 2015, we excluded responses from those surveyed in January 2016 and February 2016. For Montana, which expanded Medicaid on January 1, 2016, we excluded responses from those surveyed from January through June 2016. Because we classified Louisiana as a non-expansion state, we excluded all responses from those surveyed after the state expanded Medicaid on July 1, 2016. 4 Health insurance classification was based on a hierarchy of mutually exclusive categories in the following order: private health insurance, Medicaid, other coverage, and uninsured. Low-income adults with more than one coverage type were assigned the first appropriate category in the hierarchy, and those with other coverage were excluded from these analyses. Low-income adults were classified as having private health insurance if they reported that they were covered by any comprehensive private health insurance plan, including health maintenance and preferred provider organization plans. Private health insurance excludes plans that pay for only one type of service, such as accidents or dental care. Low-income adults were classified as having Medicaid if they reported that they were covered by Medicaid or by state-sponsored health plans with no premiums or if it is not known if a premium is charged. Low-income adults were classified as uninsured if they did not have any private health insurance or were not enrolled in Medicare, Medicaid, the Children’s Health Insurance Program, a state-sponsored or other government-sponsored health plan, or a military plan. An adult was also defined as uninsured if they had only a private plan that paid for one type of service, such as accidents or dental care. Page 45 GAO-18-607 Medicaid Expansion and Access to Care Appendix V: Estimates of Barriers to Health Care in Expansion and Non-Expansion States and by Insurance Status • low-income adults who were uninsured in expansion states compared with each of the four groups of low-income adults who were insured—low-income adults who had Medicaid in expansion states, low-income adults who had Medicaid in non-expansion states, low-income adults who had private health insurance in expansion states, and low-income adults who had private insurance in non-expansion states; • low-income adults who were uninsured in non-expansion states compared with each of the four groups of low-income adults who were insured; • low-income adults who were uninsured in expansion states compared with low-income adults who were uninsured in non- expansion states; • low-income adults who had Medicaid in expansion states compared with low-income adults who had Medicaid in non- expansion states; and • low-income adults who had private health insurance in expansion states compared with low-income adults who had private health insurance in non-expansion states. 5 Financial barriers to medical, specialty, and other types of health care. Tables 10 and 11 present estimates and differences in estimates of responses to survey question that asked whether respondents did not obtain different types of needed health care services in the past 12 months because they could not afford it. Table 10: Percentage of Low-Income Adults in Expansion and Non-Expansion States Who Reported Financial Barriers to Different Types of Needed Health Care Services, 2016 Type of services low-income adults could not Expansion states Non-expansion states afford in the past 12 months (standard error) (standard error) N = 14,913,000 N = 9,627,000 Did not get needed medical care due to costs 9.4 (0.87)a 19.9 (1.45)a a Delayed seeking medical care due to costs 10.6 (0.99) 21.4 (1.61)a Needed but could not afford prescription medication 9.7 (0.90)a 18.1 (1.42)a Needed but could not afford mental health care 3.5 (0.50)a 5.8 (1.04)a 5 We did not request that NCHS test for significant differences between low-income adults who had Medicaid in expansion or non-expansion states and low-income adults who had private health insurance in expansion or non-expansion states. Page 46 GAO-18-607 Medicaid Expansion and Access to Care Appendix V: Estimates of Barriers to Health Care in Expansion and Non-Expansion States and by Insurance Status Type of services low-income adults could not Expansion states Non-expansion states afford in the past 12 months (standard error) (standard error) N = 14,913,000 N = 9,627,000 Needed but could not afford dental care 15.1 (1.12)a 22.3 (1.56)a a Needed but could not afford eyeglasses 9.6 (0.90) 14.3 (1.36)a Needed but could not afford follow-up care 4.6 (0.66)a 8.5 (1.00)a Needed but could not afford specialist care 6.1 (0.82)a 11.2 (1.27)a Source: GAO summary of the 2016 NHIS estimates produced by the National Center for Health Statistics. │GAO-18-607 Notes: Estimates are based on selected survey questions, which were from the Access to Health Care and Utilization sections of the Family and Sample Adult Core components of the 2016 National Health Interview Survey (NHIS). For the purposes of these NHIS analyses, low-income adults are individuals ages 19 through 64, with family incomes that did not exceed 138 percent of the federal poverty level. These analyses excluded low-income adults who were noncitizens, were covered by Medicare, only received health care services through military health care or through the Indian Health Service, or had Supplemental Social Security Income. We also excluded adult females from the Sample Adult file who reported being pregnant. For the purposes of these NHIS analyses, the 30 states and the District of Columbia that expanded their Medicaid programs prior to July 1, 2016, were classified as expansion states. The remaining 20 states were classified as non-expansion states. See R.A. Cohen and E.P. Zammitti, Access and Utilization by Medicaid Expansion Status for Low-Income Adults Aged 19-64: Estimates from the National Health Interview Survey, United States, 2016, (Hyattsville, MD.: National Center for Health Statistics, January 2018), accessed January 17, 2018, https://www.cdc.gov/nchs/health_policy/coverage_and_access.htm. a Difference between low-income adults in expansion and non-expansion states was statistically significant at p < 0.05. Table 11: Percentage of Low-Income Adults in Expansion and Non-Expansion States Who Reported Financial Barriers to Different Types of Needed Health Care Services, by Insurance Status, 2016 Type of service Uninsured Medicaid Private health insurance low-income adults Expansion Non-expansion Expansion Non-expansion Expansion Non-expansion could not afford in states (standard states states (standard states states (standard states (standard the past 12 error) (standard error) error) (standard error) error) error) months N = 1,834,000 N = 3,382,000 N = 7,671,000 N = 2,160,000 N = 5,408,000 N = 4,085,000 Did not get medical 27.2 (3.63)a, b 37.3 (2.64)a, c 7.7 (1.21)b, c 12.0 (2.30)b, c 5.7 (1.01)b, c, d 9.7 (1.33)b, c, d care due to costs Delayed seeking 30.6 (4.28)b 40.9 (3.09)c 7.2 (1.20)b, c 11.7 (2.28)b, c 8.7 (1.21)b, c 10.5 (1.38)b, c medical care due to costs Needed but could 23.0 (3.85)a, b 33.2 (3.11)a, c 8.8 (1.14)b, c 13.2 (2.07)b, c 6.3 (1.18)b, c 8.3 (1.34)b, c not afford prescription medication Needed but could 7.9 (2.31)b 12.8 (2.74)c 4.0 (0.75)c 3.0 (0.98)c 1.1 (0.33)b, c 1.6 (0.44)b, c not afford mental health care Page 47 GAO-18-607 Medicaid Expansion and Access to Care Appendix V: Estimates of Barriers to Health Care in Expansion and Non-Expansion States and by Insurance Status Type of service Uninsured Medicaid Private health insurance low-income adults Expansion Non-expansion Expansion Non-expansion Expansion Non-expansion could not afford in states (standard states states (standard states states (standard states (standard the past 12 error) (standard error) error) (standard error) error) error) months N = 1,834,000 N = 3,382,000 N = 7,671,000 N = 2,160,000 N = 5,408,000 N = 4,085,000 Needed but could 30.8 (4.27)b 38.6 (3.55)c 14.5 (1.41)b, c 19.8 (3.03)b, c 10.4 (1.46)b, c 10.4 (1.49)b, c not afford dental care Needed but could 20.2 (3.93)b 21.1 (2.94)c 10.2 (1.18)b, c 15.6 (2.85) 5.1 (0.98)b, c 8.2 (1.52)b, c not afford eyeglasses Needed but could 15.3 (3.27)b 17.8 (2.66)c 3.1 (0.77)b, c 2.4 (1.04)b, c 3.1 (0.76)b, c 4.0 (0.98)b, c not afford follow-up care Needed but could 16.8 (3.65)b 21.6 (3.03)c 5.7 (1.19)b, c 6.5 (1.87)b, c 3.1 (0.71)b, c 5.2 (1.25)b, c not afford specialist care Source: GAO summary of the 2016 NHIS estimates produced by the National Center for Health Statistics. │GAO-18-607 Notes: Estimates are based on selected survey questions, which were from the Access to Health Care and Utilization sections of the Family and Sample Adult Core components of the 2016 National Health Interview Survey (NHIS). For the purposes of these NHIS analyses, low-income adults are individuals ages 19 through 64, with family incomes that did not exceed 138 percent of the federal poverty level. These analyses excluded low-income adults who were noncitizens, were covered by Medicare, only received health care services through military health care or through the Indian Health Service, or had Supplemental Social Security Income. We also excluded adult females from the Sample Adult file who reported being pregnant. For the purposes of these NHIS analyses, the 30 states and the District of Columbia that expanded their Medicaid programs prior to July 1, 2016, were classified as expansion states. The remaining 20 states were classified as non-expansion states. Health insurance classification was based on a hierarchy of mutually exclusive categories in the following order: private health insurance, Medicaid, other coverage, and uninsured. Low-income adults with more than one coverage type were assigned the first appropriate category in the hierarchy and those with other coverage were excluded from these analyses. See R.A. Cohen and E.P. Zammitti, Coverage, Access, and Utilization by Medicaid Expansion Status: Estimates from the National Health Interview Survey, United States, 2016, (Hyattsville, Md.: National Center for Health Statistics, December 2017), accessed December 12, 2017, https://www.cdc.gov/nchs/health_policy/coverage_and_access.htm. a Difference between low-income adults who were uninsured in expansion and non-expansion states was statistically significant at p < 0.05. b Difference between low-income adults who were uninsured in expansion states and low-income adults in expansion or non-expansion states who were insured—Medicaid or private health insurance—was statistically significant at p < 0.05. c Difference between low-income adults who were uninsured in non-expansion states and low-income adults in expansion or non-expansion states who were insured—Medicaid or private health insurance—was statistically significant at p < 0.05. d Difference between low-income adults who had private health insurance in expansion and non- expansion states was statistically significant at p < 0.05. Financial barriers to prescription drugs. Tables 12 and 13 present estimates and differences in estimates of survey question that asked Page 48 GAO-18-607 Medicaid Expansion and Access to Care Appendix V: Estimates of Barriers to Health Care in Expansion and Non-Expansion States and by Insurance Status respondents who had been prescribed medications whether they had taken actions during the past 12 months to save money on medications. Table 12: Percentage of Low-Income Adults in Expansion and Non-Expansion States Who Reported Taking Actions to Save Money on Prescriptions, 2016 Action taken to save money on Expansion states Non-expansion states prescriptions in the past 12 (standard error) (standard error) months N = 14,913,000 N = 9,627,000 Skipped medication doses 5.1 (0.67)a 8.3 (1.08)a Took less medication 5.4 (0.70)a 8.3 (1.07)a a Delayed filling a prescription 6.7 (0.85) 11.9 (1.27)a Asked for a lower cost medication 10.4 (0.94)a 14.2 (1.20)a Source: GAO summary of the 2016 NHIS estimates produced by the National Center for Health Statistics. │GAO-18-607 Notes: Estimates are based on selected survey questions, which were from the Access to Health Care and Utilization section of the Sample Adult Core component of the 2016 National Health Interview Survey (NHIS). For the purposes of these NHIS analyses, low-income adults are individuals ages 19 through 64, with family incomes that did not exceed 138 percent of the federal poverty level. These analyses excluded low-income adults who were noncitizens, were covered by Medicare, only received health care services through military health care or through the Indian Health Service, or had Supplemental Social Security Income. We also excluded adult females from the Sample Adult file who reported being pregnant. Low-income adults who were not prescribed medication in the past 12 months were considered to have not taken any steps to save money on prescriptions. For the purposes of these NHIS analyses, the 30 states and the District of Columbia that expanded their Medicaid programs prior to July 1, 2016, were classified as expansion states. The remaining 20 states were classified as non-expansion states. See R.A. Cohen and E.P. Zammitti, Access and Utilization by Medicaid Expansion Status for Low-Income Adults Aged 19-64: Estimates from the National Health Interview Survey, United States, 2016, (Hyattsville, MD.: National Center for Health Statistics, January 2018), accessed January 17, 2018, https://www.cdc.gov/nchs/health_policy/coverage_and_access.htm. a Difference between low-income adults in expansion and non-expansion states was statistically significant at p < 0.05. Table 13: Percentage of Low-Income Adults in Expansion and Non-Expansion States Who Reported Taking Actions to Save Money on Medications, by Insurance Status, 2016 Action taken to save Uninsured Medicaid Private health insurance money on prescriptions Expansion Non-expansion Expansion Non-expansion Expansion Non-expansion in the past 12 months states states states states states states (standard (standard (standard (standard (standard (standard error) error) error) error error) error) N = 1,834,000 N = 3,382,000 N = 7,671,000 N = 2,160,000 N = 5,408,000 N = 4,085,000 Skipped medication doses 11.8 (3.31)a 13.8 (2.42)b 4.6 (0.84)a, b NA 3.5 (0.86)a, b 4.6 (1.03)a, b Took less medication 11.9 (3.31)a 13.5 (2.48)b 4.7 (0.86)a, b 7.5 (2.22) 4.1 (0.96)a, b 4.4 (0.88)a, b a b a, b b a, b, c Delayed filling a 14.5 (3.61) 17.3 (2.52) 6.3 (1.17) 9.3 (2.61) 4.6 (0.95) 8.8 (1.37) b, c prescription Page 49 GAO-18-607 Medicaid Expansion and Access to Care Appendix V: Estimates of Barriers to Health Care in Expansion and Non-Expansion States and by Insurance Status Action taken to save Uninsured Medicaid Private health insurance money on prescriptions Expansion Non-expansion Expansion Non-expansion Expansion Non-expansion in the past 12 months states states states states states states (standard (standard (standard (standard (standard (standard error) error) error) error error) error) N = 1,834,000 N = 3,382,000 N = 7,671,000 N = 2,160,000 N = 5,408,000 N = 4,085,000 Asked for a lower cost 15.7 (3.67) 19.6 (2.45)b 9.9 (1.24)b 9.8 (1.68)b 9.4 (1.44)b 12.1 (1.69)b medication Legend: NA = Not available because of estimates did not meet the NCHS’s standards for accuracy and precision. Source: GAO summary of the 2016 NHIS estimates produced by NCHS. │GAO-18-607 Notes: Estimates are based on selected survey questions, which were from the Access to Health Care and Utilization section of the Sample Adult Core component of the 2016 National Health Interview Survey (NHIS). Estimates were not available when they did not meet the National Center for Health Statistics’ (NCHS) standards for accuracy and precision. For the purposes of these NHIS analyses, low-income adults are individuals ages 19 through 64, with family incomes that did not exceed 138 percent of the federal poverty level. These analyses excluded low-income adults who were noncitizens, were covered by Medicare, only received health care services through military health care or through the Indian Health Service, or had Supplemental Social Security Income. We also excluded adult females from the Sample Adult file who reported being pregnant. Low-income adults who were not prescribed medication in the past 12 months were considered to have not taken any steps to save money on prescriptions. For the purposes of these NHIS analyses, the 30 states and the District of Columbia that expanded their Medicaid programs prior to July 1, 2016, were classified as expansion states. The remaining 20 states were classified as non-expansion states. Health insurance classification was based on a hierarchy of mutually exclusive categories in the following order: private health insurance, Medicaid, other coverage, and uninsured. Low-income adults with more than one coverage type were assigned the first appropriate category in the hierarchy and those with other coverage were excluded from these analyses. See R.A. Cohen and E.P. Zammitti, Coverage, Access, and Utilization by Medicaid Expansion Status: Estimates from the National Health Interview Survey, United States, 2016, (Hyattsville, Md.: National Center for Health Statistics, December 2017), accessed December 12, 2017, https://www.cdc.gov/nchs/health_policy/coverage_and_access.htm. a Difference between low-income adults who were uninsured in expansion states and low-income adults in expansion or non-expansion states who were insured—Medicaid or private health insurance—was statistically significant at p < 0.05. b Difference between low-income adults who were uninsured in non-expansion states and low-income adults in expansion or non-expansion states who were insured—Medicaid or private health insurance—was statistically significant at p < 0.05. c Difference between low-income adults who had private health insurance in expansion and non- expansion states was statistically significant at p < 0.05. Non-financial barriers to health care. Tables 14 and 15 present estimates and differences in estimates of the NCHS composite measure on any non-financial barriers to health care, which was based on responses to five survey questions on whether respondents delayed care in the past 12 months due to long wait times, a lack of transportation, and other non-financial reasons. Additionally, these tables present estimates and differences in estimates of responses to the composite measure’s five underlying survey questions. Page 50 GAO-18-607 Medicaid Expansion and Access to Care Appendix V: Estimates of Barriers to Health Care in Expansion and Non-Expansion States and by Insurance Status Table 14: Percentage of Low-Income Adults in Expansion and Non-Expansion States Who Reported Non-Financial Barriers to Health Care, 2016 Non-financial barrier that Expansion states Non-expansion states delayed health care in the past (standard error) (standard error) 12 months N = 14,913,000 N = 9,627,000 Anya 15.3 (1.24) 14.8 (1.40) Unable to get through on 3.6 (0.48) 2.5 (0.50) telephone Unable to get an appointment 8.0 (0.88) 6.2 (0.99) soon enough Wait time at doctor’s office was 6.4 (0.78) 6.4 (1.07) too long Clinic or doctor’s office was not 3.3 (0.48) 2.4 (0.50) open when able to go No transportation 4.3 (0.66) 5.1 (0.84) Source: GAO summary of the 2016 NHIS estimates produced by the National Center for Health Statistics. │GAO-18-607 Notes: Estimates are based on selected survey questions and their related composite measure. These survey questions were from the Access to Health Care and Utilization section of Sample Adult Core component of the 2016 National Health Interview Survey (NHIS). For the purposes of these NHIS analyses, low-income adults are individuals ages 19 through 64, with family incomes that did not exceed 138 percent of the federal poverty level. These analyses excluded low-income adults who were noncitizens, were covered by Medicare, only received health care services through military health care or through the Indian Health Service, or had Supplemental Social Security Income. We also excluded adult females from the Sample Adult file who reported being pregnant. For the purposes of these NHIS analyses, the 30 states and the District of Columbia that expanded their Medicaid programs prior to July 1, 2016, were classified as expansion states. The remaining 20 states were classified as non-expansion states. See R.A. Cohen and E.P. Zammitti, Access and Utilization by Medicaid Expansion Status for Low-Income Adults Aged 19-64: Estimates from the National Health Interview Survey, United States, 2016, (Hyattsville, MD.: National Center for Health Statistics, January 2018), accessed January 17, 2018, https://www.cdc.gov/nchs/health_policy/coverage_and_access.htm. a Low-income adults had any non-financial barrier to care if they reported delaying care in the past 12 months because they could not get through on the telephone, could not get an appointment soon enough, the wait time at the doctor’s office was too long, the doctor’s office or clinic was not open when they could get there, or did not have transportation. Table 15: Percentage of Low-Income Adults in Expansion and Non-Expansion States Who Reported Non-Financial Barriers to Health Care, by Insurance Status, 2016 Non-financial barrier that Uninsured Medicaid Private health insurance delayed healthcare in the Expansion Non- Expansion Non- Expansion Non- past 12 months states expansion states expansion states expansion (standard states (standard states (standard states error) (standard error) (standard error) (standard error) error) error) N = 1,834,000 N = 3,382,000 N = 7,671,000 N = 2,160,000 N = 5,408,000 N = 4,085,000 Anya 8.6 (2.28)b 14.0 (2.46) 17.6 (1.83)b 19.5 (3.04)b 14.2 (1.61)b 12.9 (2.16) Page 51 GAO-18-607 Medicaid Expansion and Access to Care Appendix V: Estimates of Barriers to Health Care in Expansion and Non-Expansion States and by Insurance Status Non-financial barrier that Uninsured Medicaid Private health insurance delayed healthcare in the Expansion Non- Expansion Non- Expansion Non- past 12 months states expansion states expansion states expansion (standard states (standard states (standard states error) (standard error) (standard error) (standard error) error) error) N = 1,834,000 N = 3,382,000 N = 7,671,000 N = 2,160,000 N = 5,408,000 N = 4,085,000 Unable to get through on NA 1.7 (0.68)c 4.1 (0.73)c NA 3.1 (0.75) 2.2 (0.71) phone Unable to get an appointment NA 3.0 (0.80)c 9.2 (1.37)c 8.4 (2.02)c 7.5 (1.15)c 7.6 (1.74)c soon enough Wait time at doctor’s office was NA 7.7 (2.13) 7.9 (1.27) 6.1 (1.42) 4.1 (0.93) NA too long Clinic or doctor’s office was not NA 1.9 (0.63)c 2.5 (0.57) 3.0 (1.08) 4.8 (0.92)c 2.4 (0.85) open when able to go No transportation 1.0 (0.54)b, d 6.0 (1.58)c, d 6.2 (1.10)b 9.9 (2.39)b 2.8 (0.85) 1.8 (0.71)c Legend: NA = Not available because of estimates did not meet the NCHS’s standards for accuracy and precision. Source: GAO summary of the 2016 NHIS estimates produced by NCHS. │GAO-18-607 Notes: Estimates are based on selected survey questions and their related composite measure. These survey questions were from the Access to Health Care and Utilization section of the Sample Adult Core component of the 2016 National Health Interview Survey (NHIS). Estimates were not available when they did not meet the National Center for Health Statistics’ (NCHS) standards for accuracy and precision. For the purposes of these NHIS analyses, low-income adults are individuals ages 19 through 64, with family incomes that did not exceed 138 percent of the federal poverty level. These analyses excluded low-income adults who were noncitizens, were covered by Medicare, only received health care services through military health care or through the Indian Health Service, or had Supplemental Social Security Income. We also excluded adult females from the Sample Adult file who reported being pregnant. For the purposes of these NHIS analyses, the 30 states and the District of Columbia that expanded their Medicaid programs prior to July 1, 2016, were classified as expansion states. The remaining 20 states were classified as non-expansion states. Health insurance classification was based on a hierarchy of mutually exclusive categories in the following order: private health insurance, Medicaid, other coverage, and uninsured. Low-income adults with more than one coverage type were assigned the first appropriate category in the hierarchy and those with other coverage were excluded from these analyses. See R.A. Cohen and E.P. Zammitti, Coverage, Access, and Utilization by Medicaid Expansion Status: Estimates from the National Health Interview Survey, United States, 2016, (Hyattsville, Md.: National Center for Health Statistics, December 2017), accessed December 12, 2017, https://www.cdc.gov/nchs/health_policy/coverage_and_access.htm. a Low-income adults were classified as facing any non-financial barriers to care if they reported delaying care in the past 12 months because they were unable to get through on the telephone, were unable to get an appointment soon enough, the wait time at the doctor’s office was too long, the doctor’s office or clinic was not open when they could get there, or did not have transportation. b Difference between low-income adults who were uninsured in expansion states and low-income adults in expansion or non-expansion states who were insured—Medicaid or private health insurance—was statistically significant at p < 0.05. c Difference between low-income adults who were uninsured in non-expansion states and low-income adults in expansion or non-expansion states who were insured—Medicaid or private health insurance—was statistically significant at p < 0.05. d Difference between low-income adults who were uninsured in expansion and non-expansion states was statistically significant at p < 0.05. Page 52 GAO-18-607 Medicaid Expansion and Access to Care Appendix VI: Estimates on Place of Care Appendix VI: Estimates on Place of Care and Services in Expansion and Non-Expansion States and by Insurance Status and Services in Expansion and Non- Expansion States and by Insurance Status This appendix provides estimates on having a usual place of care and receiving selected health care services for adults—individuals ages 19 to 64, with family incomes that did not exceed 138 percent of the federal poverty level (FPL)—from the 2016 National Health Interview Survey (NHIS), which we obtained from the National Center for Health Statistics (NCHS). 1 Estimates are based on responses to selected survey questions on having a usual place of care, receiving selected health care services, and contacting health care professionals. Estimates are reported for: 1 Estimates are based on selected survey questions, which were from the Access to Health Care and Utilization and Health Behavior sections of the Family and Sample Adult Core components of the 2016 National Health Interview Survey (NHIS). For the purposes of these NHIS analyses, low-income adults are individuals ages 19 through 64, with family incomes that did not exceed 138 percent of the federal poverty level. These analyses excluded low-income adults who were noncitizens, were covered by Medicare, only received health care services through military health care or through the Indian Health Service, or had Supplemental Social Security Income. We also excluded adult females from the Sample Adult file who reported being pregnant. NCHS provides additional estimates of access to care for low-income adults on its website. NCHS provides additional estimates of access to care and utilization for low-income adults on its website. See R.A. Cohen and E.P. Zammitti, Access and Utilization by Medicaid Expansion Status for Low-Income Adults Aged 19-64: Estimates from the National Health Interview Survey, United States, 2016, (Hyattsville, MD.: National Center for Health Statistics, January 2018), accessed January 17, 2018, https://www.cdc.gov/nchs/health_policy/coverage_and_access.htm; and R.A. Cohen and E.P. Zammitti, Coverage, Access, and Utilization by Medicaid Expansion Status: Estimates from the National Health Interview Survey, United States, 2016, (Hyattsville, Md.: National Center for Health Statistics, December 2017), accessed December 12, 2017, https://www.cdc.gov/nchs/health_policy/coverage_and_access.htm. Page 53 GAO-18-607 Medicaid Expansion and Access to Care Appendix VI: Estimates on Place of Care and Services in Expansion and Non-Expansion States and by Insurance Status • All low-income adults in expansion and non-expansion states. 2 We also report statistically significant differences between low-income adults in expansion and non-expansion states. • Six groups of low-income adults: (1) low-income adults who were uninsured in expansion states; (2) low-income adults who were uninsured in non-expansion states; (3) low-income adults who had Medicaid in expansion states; (4) low-income adults who had Medicaid in non-expansion states; (5) low-income adults who had private health insurance in expansion states; and (6) low-income adults who had private health insurance in non-expansion states. 3 We also report any statistically significant differences when comparing the six groups of low-income adults, specifically: 2 We classified the 30 states and the District of Columbia that expanded their Medicaid program prior to July 1, 2016 as expansion states, and the remaining 20 states were classified as non-expansion states. States classified as expansion states were Alaska, Arizona, Arkansas, California, Colorado, Connecticut, Delaware, Hawaii, Illinois, Indiana, Iowa, Kentucky, Maryland, Massachusetts, Michigan, Minnesota, Montana, Nevada, New Hampshire, New Jersey, New Mexico, New York, North Dakota, Ohio, Oregon, Pennsylvania, Rhode Island, Vermont, Washington, and West Virginia. The 20 non- expansion states are Alabama, Florida, Georgia, Idaho, Kansas, Louisiana, Maine, Mississippi, Missouri, Nebraska, North Carolina, Oklahoma, South Carolina, South Dakota, Tennessee, Texas, Utah, Virginia, Wisconsin, and Wyoming. For 3 states, we only included NHIS respondents for part of calendar year 2016. Specifically, we excluded respondents who were surveyed during the first 6 months after a state expanded Medicaid to allow for the effect of expansion to take place. Two of the 3 states—Alaska and Montana—were classified as expansion states and the third state, Louisiana, was classified as non-expansion. For Alaska, which expanded Medicaid on September 1, 2015, we excluded responses from those surveyed in January 2016 and February 2016. For Montana, which expanded Medicaid on January 1, 2016, we excluded responses from those surveyed from January through June 2016. Because we classified Louisiana as a non-expansion state, we excluded all responses from those surveyed after the state expanded Medicaid on July 1, 2016. 3 Health insurance classification was based on a hierarchy of mutually exclusive categories in the following order: private health insurance, Medicaid, other coverage, and uninsured. Low-income adults with more than one coverage type were assigned the first appropriate category in the hierarchy, and those with other coverage were excluded from this analysis. Low-income adults were classified as having private health insurance if they reported that they were covered by any comprehensive private health insurance plan, including health maintenance and preferred provider organization plans. Private health insurance excludes plans that pay for only one type of service, such as accidents or dental care. Low-income adults were classified as having Medicaid if they reported that they were covered by Medicaid or by state-sponsored health plans with no premiums or if it is not known if a premium is charged. Low-income adults were classified as uninsured if they did not have any private health insurance or were not enrolled in Medicare, Medicaid, the Children’s Health Insurance Program, a state-sponsored or other government-sponsored health plan, or a military plan. An adult was also defined as uninsured if they had only a private plan that paid for one type of service, such as accidents or dental care. Page 54 GAO-18-607 Medicaid Expansion and Access to Care Appendix VI: Estimates on Place of Care and Services in Expansion and Non-Expansion States and by Insurance Status • low-income adults who were uninsured in expansion states compared with each of the four groups of low-income adults who were insured—low-income adults who had Medicaid in expansion states, low-income adults who had Medicaid in non-expansion states, low-income adults who had private health insurance in expansion states, and low-income adults who had private insurance in non-expansion states; • low-income adults who were uninsured in non-expansion states compared with each of the four groups of low-income adults who were insured; • low-income adults who were uninsured in expansion states compared with low-income adults who were uninsured in non- expansion states; • low-income adults who had Medicaid in expansion states compared with low-income adults who had Medicaid in non- expansion states; and • low-income adults who had private health insurance in expansion states compared with low-income adults who had private health insurance in non-expansion states. 4 Having a usual place of care. Tables 16 through 19 present estimates and differences in estimates of survey questions that asked respondents about the place of care they usually go to when sick or need advice about their health and the type of place that respondents most often went. Table 16: Percentage of Low-Income Adults in Expansion and Non-Expansion States Who Reported Having a Usual Place of Care, 2016 Expansion states Non-expansion states (standard error) (standard error) N = 14,913,000 N = 9,627,000 Had a usual place of care 81.6 (1.24)a 68.1 (1.68)a Source: GAO summary of the 2016 NHIS estimates produced by the National Center for Health Statistics. │GAO-18-607 Notes: Estimates are based on a selected survey question, which was from the Health Care Access and Utilization section of the Sample Adult Core component of the 2016 National Health Interview Survey (NHIS). For the purposes of these NHIS analyses, low-income adults are individuals ages 19 through 64, with family incomes that did not exceed 138 percent of the federal poverty level. These analyses excluded low-income adults who were noncitizens, were covered by Medicare, only received health care services through military health care or through the Indian Health Service, or had 4 We did not request that NCHS test for significant differences between low-income adults who had Medicaid in expansion or non-expansion states and low-income adults who had private health insurance in expansion or non-expansion states. Page 55 GAO-18-607 Medicaid Expansion and Access to Care Appendix VI: Estimates on Place of Care and Services in Expansion and Non-Expansion States and by Insurance Status Supplemental Social Security Income. We also excluded adult females from the Sample Adult file who reported being pregnant. Low-income adults who reported that in the past 12 months the emergency department was their usual place of care were considered to not have a usual place of care. For the purposes of these NHIS analyses, the 30 states and the District of Columbia that expanded their Medicaid programs prior to July 1, 2016, were classified as expansion states. The remaining 20 states were classified as non-expansion states. See R.A. Cohen and E.P. Zammitti, Access and Utilization by Medicaid Expansion Status for Low-Income Adults Aged 19-64: Estimates from the National Health Interview Survey, United States, 2016, (Hyattsville, MD.: National Center for Health Statistics, January 2018), accessed January 17, 2018, https://www.cdc.gov/nchs/health_policy/coverage_and_access.htm. a Difference between low-income adults in expansion and non-expansion states was statistically significant at p < 0.05. Table 17: Percentage of Low-Income Adults in Expansion and Non-Expansion States Who Reported Having a Usual Place of Care, by Insurance Status, 2016 Uninsured Medicaid Private health insurance Expansion Non-expansion Expansion Non-expansion Expansion Non-expansion states (standard states (standard states (standard states states (standard states (standard error) error) error) (standard error) error) error) N = 1,834,000 N = 3,382,000 N = 7,671,000 N = 2,160,000 N = 5,408,000 N = 4,085,000 Had a usual 45.3 (4.20)a 46.0 (2.91)b 87.5 (1.54)a, b 83.2 (3.38)a, b 85.5 (1.69)a, b, c 78.2 (2.16)a, b, c place of care Source: GAO summary of the 2016 NHIS estimates produced by the National Center for Health Statistics. │GAO-18-607 Notes: Estimates are based on a selected survey question, which was from the Health Behaviors section of the Sample Adult Core component of the 2016 National Health Interview Survey (NHIS). For the purposes of these NHIS analyses, low-income adults are individuals ages 19 through 64, with family incomes that did not exceed 138 percent of the federal poverty level. These analyses excluded low-income adults who were noncitizens, were covered by Medicare, only received health care services through military health care or through the Indian Health Service, or had Supplemental Social Security Income. We also excluded adult females from the Sample Adult file who reported being pregnant. Low-income adults who reported that in the past 12 months the emergency department was their usual place of care were considered to not have a usual place of care. For the purposes of these NHIS analyses, the 30 states and the District of Columbia that expanded their Medicaid programs prior to July 1, 2016, were classified as expansion states. The remaining 20 states were classified as non-expansion states. Health insurance classification was based on a hierarchy of mutually exclusive categories in the following order: private health insurance, Medicaid, other coverage, and uninsured. Low-income adults with more than one coverage type were assigned the first appropriate category in the hierarchy and those with other coverage were excluded from these analyses. See R.A. Cohen and E.P. Zammitti, Coverage, Access, and Utilization by Medicaid Expansion Status: Estimates from the National Health Interview Survey, United States, 2016, (Hyattsville, Md.: National Center for Health Statistics, December 2017), accessed December 12, 2017, https://www.cdc.gov/nchs/health_policy/coverage_and_access.htm. a Difference between low-income adults who were uninsured in expansion states and low-income adults in expansion or non-expansion states who were insured—Medicaid or private health insurance—was statistically significant at p < 0.05. b Difference between low-income adults who were uninsured in non-expansion states and low-income adults in expansion or non-expansion states who were insured—Medicaid or private health insurance—was statistically significant at p < 0.05. c Difference between low-income adults who had private health insurance in expansion and non- expansion states was statistically significant at p < 0.05. Page 56 GAO-18-607 Medicaid Expansion and Access to Care Appendix VI: Estimates on Place of Care and Services in Expansion and Non-Expansion States and by Insurance Status Table 18: Percentage of Low-Income Adults in Expansion and Non-Expansion States Who Had a Usual Place of Care by the Type of Place They Usually Went for Care, 2016 Usual place of care Expansion states Non-expansion states (standard error) (standard error) N = 12,164,000 N = 6,554,000 Clinic or health center 30.7 (1.58)a 42.7 (2.13)a Doctor’s office or health 64.9 (1.69)a 52.8 (2.21)a maintenance organization Other 4.5 (0.84) 4.5 (0.88) Source: GAO summary of the 2016 NHIS estimates produced by the National Center for Health Statistics. │GAO-18-607 Notes: Estimates are based on a selected survey question, which was from the Access to Health Care and Utilization section of the Sample Adult Core component of the 2016 National Health Interview Survey (NHIS). For the purposes of these NHIS analyses, low-income adults are individuals ages 19 through 64, with family incomes that did not exceed 138 percent of the federal poverty level. These analyses excluded low-income adults who were noncitizens, were covered by Medicare, only received health care services through military health care or through the Indian Health Service, or had Supplemental Social Security Income. Low-income adults who reported that in the past 12 months the emergency department was their usual place of care were considered to not have a usual place of care. For the purposes of these NHIS analyses, the 30 states and the District of Columbia that expanded their Medicaid programs prior to July 1, 2016 and were classified as expansion states. The remaining 20 states were classified as non-expansion states. See R.A. Cohen and E.P. Zammitti, Access and Utilization by Medicaid Expansion Status for Low-Income Adults Aged 19-64: Estimates from the National Health Interview Survey, United States, 2016, (Hyattsville, MD.: National Center for Health Statistics, January 2018), accessed January 17, 2018, https://www.cdc.gov/nchs/health_policy/coverage_and_access.htm. a Difference between low-income adults in expansion and non-expansion states was statistically significant at p < 0.05. Table 19: Percentage of Low-Income Adults in Expansion and Non-Expansion States with a Usual Place of Care by the Type of Place They Usually Went for Care, by Insurance Status, 2016 Place that low-income Uninsured Medicaid Private health insurance adults reported going to Expansion Non-expansion Expansion Non- Expansion Non- most often for care in the states states (standard states expansion states expansion past 12 (standard error) (standard states (standard states error) error) (standard error) (standard error) error) N = 831,000 N = 1,557,000 N = 6,709,000 N = 1,798,000 N = 4,626,000 N = 3,196,000 Clinic or health center 34.2 (5.86)a 59.6 (4.32)a, b 30.4 (2.16)b, c 41.2 (4.05)b, c 30.4 (2.55)b 35.4 (2.81)b Doctor’s office or health 52.1 (6.78)a 29.4 (3.99)a, b 65.6 (2.26)b 58.2 (4.05)b 66.1 (2.70)b 61.1 (2.96)b maintenance organization Other NA 11.0 (2.93)b 4.0 (1.08)b, c 0.6 (0.50)b, c NA 3.6 (0.98)b Legend: NA = Not available because of estimates did not meet the NCHS’s standards for accuracy and precision. Source: GAO summary of the 2016 NHIS estimates produced by NCHS. │GAO-18-607 Notes: Estimates are based on a selected survey question, which was from the Access to Health Care and Utilization section of the Sample Adult Core component of the 2016 National Health Page 57 GAO-18-607 Medicaid Expansion and Access to Care Appendix VI: Estimates on Place of Care and Services in Expansion and Non-Expansion States and by Insurance Status Interview Survey (NHIS). Estimates were not available when they did not meet the National Center for Health Statistics’ (NCHS) standards for accuracy and precision. For the purposes of these NHIS analyses, low-income adults are individuals ages 19 through 64, with family incomes that did not exceed 138 percent of the federal poverty level. These analyses excluded low-income adults who were noncitizens, were covered by Medicare, only received health care services through military health care or through the Indian Health Service, or had Supplemental Social Security Income. We also excluded adult females from the Sample Adult file who reported being pregnant. Low-income adults who reported that in the past 12 months the emergency department was their usual place of care were considered to not have a usual place of care. For the purposes of these NHIS analyses, the 30 states and the District of Columbia that expanded their Medicaid programs prior to July 1, 2016, were classified as expansion states. The remaining 20 states were classified as non-expansion states. Health insurance classification was based on a hierarchy of mutually exclusive categories in the following order: private health insurance, Medicaid, other coverage, and uninsured. Low-income adults with more than one coverage type were assigned the first appropriate category in the hierarchy and those with other coverage were excluded from these analyses. See R.A. Cohen and E.P. Zammitti, Coverage, Access, and Utilization by Medicaid Expansion Status: Estimates from the National Health Interview Survey, United States, 2016, (Hyattsville, Md.: National Center for Health Statistics, December 2017), accessed December 12, 2017, https://www.cdc.gov/nchs/health_policy/coverage_and_access.htm. a Difference between low-income adults who were uninsured in expansion and non-expansion states was statistically significant at p < 0.05. b Difference between low-income adults who were uninsured in non-expansion states and low-income adults in expansion or non-expansion states who were insured—Medicaid or private health insurance—was statistically significant at p < 0.05. c Difference between low-income adults who had Medicaid in expansion and non-expansion states was statistically significant at p < 0.05. Receiving selected health care services. Tables 20 and 21 present estimates and differences in estimates of survey questions that asked respondents whether they had received a blood cholesterol check, flu vaccine, or other selected services. Table 20: Percentage of Low-Income Adults in Expansion and Non-Expansion States Who Reported Receiving Selected Health Care Services, 2016 Health care service received in Expansion states Non-expansion states past 12 months (standard error) (standard error) N = 14,913,000 N = 9,627,000 Blood cholesterol check 49.2 (1.72)a 41.7 (1.93)a Blood pressure check 77.5 (1.51)a 71.8 (1.67)a Blood sugar check 34.2 (1.68) 32.0 (1.81) Flu vaccineb 30.9 (1.47)a 23.9 (1.43)a Colon cancer testc 19.1 (2.54) 17.9 (2.74) d Mammogram 46.3 (3.89) 40.6 (5.10) Overnight hospitalization 8.9 (0.88) 8.8 (0.96) Hospital emergency department 27.1 (1.35) 27.9 (1.66) visit Source: GAO summary of the 2016 NHIS estimates produced by the National Center for Health Statistics. │GAO-18-607 Page 58 GAO-18-607 Medicaid Expansion and Access to Care Appendix VI: Estimates on Place of Care and Services in Expansion and Non-Expansion States and by Insurance Status Notes: Estimates are based on selected survey questions, which were from the Access to Health Care and Utilization sections of the Family and Sample Adult Core components of the 2016 National Health Interview Survey (NHIS). For the purposes of these NHIS analyses, low-income adults are individuals ages 19 through 64, with family incomes that did not exceed 138 percent of the federal poverty level. These analyses excluded low-income adults who were noncitizens, were covered by Medicare, only received health care services through military health care or through the Indian Health Service, or had Supplemental Social Security Income. We also excluded adult females from the Sample Adult file who reported being pregnant. For the purposes of these NHIS analyses, the 30 states and the District of Columbia that expanded their Medicaid programs prior to July 1, 2016, were classified as expansion states. The remaining 20 states were classified as non-expansion states. See R.A. Cohen and E.P. Zammitti, Access and Utilization by Medicaid Expansion Status for Low-Income Adults Aged 19-64: Estimates from the National Health Interview Survey, United States, 2016, (Hyattsville, MD.: National Center for Health Statistics, January 2018), accessed January 17, 2018, https://www.cdc.gov/nchs/health_policy/coverage_and_access.htm. a Difference between low-income adults in expansion and non-expansion states was statistically significant at p < 0.05. b Includes vaccines by both shot and nasal spray. c Limited to adults ages 50 to 64 years. d Limited to female adults ages 50 to 64 years. Table 21: Percentage of Low-Income Adults in Expansion and Non-Expansion States Who Reported Receiving Selected Health Care Services, by Insurance Status, 2016 Health care Uninsured Medicaid Private health insurance service received Expansion Non-expansion Expansion Non-expansion Expansion Non-expansion in past 12 states (standard states (standard states (standard states (standard states (standard states (standard months error) error) error) error) error) error) N = 1,834,000 N = 3,382,000 N = 7,671,000 N = 2,160,000 N = 5,408,000 N = 4,085,000 Blood cholesterol 23.3 (4.09)a 27.6 (2.84)b 56.0 (2.44)a,b 61.4 (4.07)a, b 47.9 (2.81)a, b 43.2 (2.91)a, b check Blood pressure 51.2 (4.23)a 54.1 (3.25)b 82.0 (2.01)a, b 87.8 (2.25)a, b 80.1 (2.26)a, b 77.8 (2.13)a, b check Blood sugar check 15.3 (3.52)a 23.8 (3.00)b 39.2 (2.22)a, b, c 52.7 (4.04)a, b, c 33.8 (2.78)a, b 27.9 (2.37)a Flu vaccined 11.8 (2.74)a 14.0 (1.96)b 33.5 (2.19)a, b 35.3 (3.93)a, b 33.8 (2.37)a, b, e 25.9 (2.22)a, b, e f Colon cancer test NA NA 24.3 (4.09) NA 18.1 (4.40) 24.0 (5.06) Mammogramg NA 21.2 (6.12)b 45.5 (5.43)b NA 60.2 (7.00)b 51.5 (6.67)b Overnight NA 7.3 (1.41)b 11.6 (1.36)b 16.8 (2.58)b 7.2 (1.30) 5.9 (1.28) hospitalization Hospital 20.9 (3.65)a 28.2 (2.98)b 35.2 (2.14)a 43.4 (3.65)a, b 17.6 (1.77)b 19.6 (2.12)b emergency department visit Legend: NA = Not available because of estimates did not meet the NCHS’s standards for accuracy and precision. Source: GAO summary of the 2016 NHIS estimates produced by NCHS. │GAO-18-607 Notes: Estimates are based on selected survey questions, which were from the Access to Health Care and Utilization sections of the Family and Sample Adult Core components of the 2016 National Health Interview Survey (NHIS). Estimates were not available when they did not meet the National Center for Health Statistics’ (NCHS) standards for accuracy and precision. For the purposes of these NHIS analyses, low-income adults are individuals ages 19 through 64, with family incomes that did Page 59 GAO-18-607 Medicaid Expansion and Access to Care Appendix VI: Estimates on Place of Care and Services in Expansion and Non-Expansion States and by Insurance Status not exceed 138 percent of the federal poverty level. These analyses excluded low-income adults who were noncitizens, were covered by Medicare, only received health care services through military health care or through the Indian Health Service, or had Supplemental Social Security Income. We also excluded adult females from the Sample Adult file who reported being pregnant. For the purposes of these NHIS analyses, the 30 states and the District of Columbia that expanded their Medicaid programs prior to July 1, 2016, were classified as expansion states. The remaining 20 states were classified as non-expansion states. Health insurance classification was based on a hierarchy of mutually exclusive categories in the following order: private health insurance, Medicaid, other coverage, and uninsured. Low-income adults with more than one coverage type were assigned the first appropriate category in the hierarchy and those with other coverage were excluded from these analyses. See R.A. Cohen and E.P. Zammitti, Coverage, Access, and Utilization by Medicaid Expansion Status: Estimates from the National Health Interview Survey, United States, 2016, (Hyattsville, Md.: National Center for Health Statistics, December 2017), accessed December 12, 2017, https://www.cdc.gov/nchs/health_policy/coverage_and_access.htm. a Difference between low-income adults who were uninsured in expansion states and low-income adults in expansion or non-expansion states who were insured—Medicaid or private health insurance—was statistically significant at p < 0.05. b Difference between low-income adults who were uninsured in non-expansion states and low-income adults in expansion or non-expansion states who were insured—Medicaid or private health insurance—was statistically significant at p < 0.05. c Difference between low-income adults who had Medicaid in expansion and non-expansion states was statistically significant at p < 0.05. d Includes vaccines by both shot and nasal spray. e Difference between low-income adults who had private health insurance in expansion and non- expansion states was statistically significant at p < 0.05. f Limited to adults ages 50 to 64 years. g Limited to female adults ages 50 to 64 years. Contacting health care professionals. Tables 22 and 23 present estimates and differences in estimates of survey questions that asked respondents whether they had visited or spoken to a general doctor, specialist, or other health care professionals about their health in the past 12 months. Table 22: Percentage of Low-Income Adults in Expansion and Non-Expansion States Who Reported Visiting or Speaking to a Health Care Professional about their Health, 2016 Type of health care Expansion states Non-expansion states professional visited or spoken (standard error) (standard error) to in past 12 months N = 14,913,000 N = 9,627,000 Any 79.0 (1.43)a 73.0 (1.47)a General doctorb 64.3 (1.66)a 50.9 (1.98)a Nurse practitioner, physician 23.7 (1.35)a 17.9 (1.36)a assistant, or midwife Medical specialistc 18.2 (1.11)a 13.2 (1.14)a Source: GAO summary of the 2016 NHIS estimates produced by the National Center for Health Statistics. │GAO-18-607 Page 60 GAO-18-607 Medicaid Expansion and Access to Care Appendix VI: Estimates on Place of Care and Services in Expansion and Non-Expansion States and by Insurance Status Notes: Estimates are based on selected survey questions, which were from the Access to Health Care and Utilization section of the Sample Adult Core components of the 2016 National Health Interview Survey (NHIS). For the purposes of these NHIS analyses, low-income adults are individuals ages 19 through 64, with family incomes that did not exceed 138 percent of the federal poverty level. These analyses excluded low-income adults who were noncitizens, were covered by Medicare, only received health care services through military health care or through the Indian Health Service, or had Supplemental Social Security Income. We also excluded adult females from the Sample Adult file who reported being pregnant. For the purposes of these NHIS analyses, the 30 states and the District of Columbia that expanded their Medicaid programs prior to July 1, 2016, were classified as expansion states; the remaining 20 states were classified as non-expansion states. See R.A. Cohen and E.P. Zammitti, Access and Utilization by Medicaid Expansion Status for Low-Income Adults Aged 19-64: Estimates from the National Health Interview Survey, United States, 2016, (Hyattsville, MD.: National Center for Health Statistics, January 2018), accessed January 17, 2018, https://www.cdc.gov/nchs/health_policy/coverage_and_access.htm. a Difference between low-income adults in expansion and non-expansion states was statistically significant at p < 0.05. b General doctors are general practitioners, family doctors, and internists. c Medical specialists exclude obstetricians, gynecologists, psychiatrists, and ophthalmologists. Table 23: Percentage of Low-Income Adults in Expansion and Non-Expansion States Who Reported Seeing or Speaking to a Health Care Provider About their Health, by Insurance Status, 2016 Type of health care Uninsured Medicaid Private health insurance professional visited or Expansion Non-expansion Expansion Non-expansion Expansion Non- spoken to in past 12 states states states states (standard states expansion months (standard error) (standard error) (standard error) error) (standard states error) (standard error) N = 1,834,000 N = 3,382,000 N = 7,671,000 N = 2,160,000 N = 5,408,000 N = 4,085,000 a b a, b, c a, b, c Any 46.0 (4.69) 53.9 (2.95) 83.9 (1.90) 90.8 (1.78) 83.5 (2.06)a, b 79.3 (2.20)a, b d a b a, b a, b a, b, e General doctor 35.0 (4.46) 33.0 (3.11) 68.9 (2.27) 62.2 (4.07) 67.8 (2.43) 59.7 (2.75)a, b, e Nurse practitioner, 13.0 (3.32)a 10.8 (1.97)b 23.4 (1.90)a, b 25.5 (3.19)a, b 27.8 (2.22)a, b, e 19.7 (1.98)b, e physician assistant, or midwife Medical specialistf NA 6.1 (1.28)b 19.7(1.58)b 24.1(3.15)b 20.5 (1.98)b, e 13.1 (1.79)b, e Legend: NA = Not available because of estimates did not meet the NCHS’s standards for accuracy and precision. Source: GAO summary of the 2016 NHIS estimates produced by NCHS. │GAO-18-607 Notes: Estimates are based on for selected survey questions, which were from the Access to Health Care and Utilization section of the Sample Adult Core components of the 2016 National Health Interview Survey (NHIS). Estimates were not available when they did not meet the National Center for Health Statistics’ (NCHS) standards for accuracy and precision. For the purposes of these NHIS analyses, low-income adults are individuals ages 19 through 64, with family incomes that did not exceed 138 percent of the federal poverty level. These analyses excluded low-income adults who were noncitizens, were covered by Medicare, only received health care services through military health care or through the Indian Health Service, or had Supplemental Social Security Income. We also excluded adult females from the Sample Adult file who reported being pregnant. For the purposes of these NHIS analyses, the 30 states and the District of Columbia that expanded their Medicaid programs prior to July 1, 2016, were classified as expansion states. The remaining 20 states were classified as non-expansion states. Page 61 GAO-18-607 Medicaid Expansion and Access to Care Appendix VI: Estimates on Place of Care and Services in Expansion and Non-Expansion States and by Insurance Status Health insurance classification was based on a hierarchy of mutually exclusive categories in the following order: private health insurance, Medicaid, other coverage, and uninsured. Low-income adults with more than one coverage type were assigned the first appropriate category in the hierarchy and those with other coverage were excluded from these analyses. See R.A. Cohen and E.P. Zammitti, Coverage, Access, and Utilization by Medicaid Expansion Status: Estimates from the National Health Interview Survey, United States, 2016, (Hyattsville, Md.: National Center for Health Statistics, December 2017), accessed December 12, 2017, https://www.cdc.gov/nchs/health_policy/coverage_and_access.htm. a Difference between low-income adults who were uninsured in expansion states and low-income adults in expansion or non-expansion states who were insured—Medicaid or private health insurance—was statistically significant at p < 0.05. b Difference between low-income adults who were uninsured in non-expansion states and low-income adults in expansion or non-expansion states who were insured—Medicaid or private health insurance—was statistically significant at p < 0.05. c Difference between low-income adults who had Medicaid in expansion and non-expansion states was statistically significant at p < 0.05. d General doctors are general practitioners, family doctors, and internists. e Difference between low-income adults who had private health insurance in expansion and non- expansion states was statistically significant at p < 0.05. f Medical specialists exclude obstetricians, gynecologists, psychiatrists, and ophthalmologists. Page 62 GAO-18-607 Medicaid Expansion and Access to Care Appendix VII: GAO Contact and Staff Appendix VII: GAO Contact and Staff Acknowledgments Acknowledgments Carolyn L. Yocom, (202) 512-7114 or yocomc@gao.gov GAO Contact In addition to the contact named above, Katherine M. Iritani (Director), Staff Tim Bushfield (Assistant Director), Deitra H. Lee (Analyst-in-Charge), Acknowledgments Kristin Ekelund, Laurie Pachter, Vikki Porter, Merrile Sing, and Emily Wilson made key contributions to this report. (102061) Page 63 GAO-18-607 Medicaid Expansion and Access to Care 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) 512-7700 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.