Abstract
Among nonelderly adult U.S. residents (aged 25–61), uninsurance rates increased from 13.7% in 2000 to 16.0% in 2005. Despite the existence of public insurance programs, rates remained high for low-income individuals reporting serious health conditions (25% across years) or disabilities (15%). Residents of southern states had even higher rates (32% health conditions, 22% disabilities). Those who did not belong to a federally mandated Medicaid eligibility category were about twice as likely to be uninsured overall and experienced more rapid increases over time. These regional and categorical differences reflect gaps in current policy that pose challenges to incremental health insurance reform.
Introduction
The percentage of nonelderly adult U.S. residents without health insurance has increased steadily for more than two decades. According to the Employee Benefit Research Institute, uninsurance rates increased among nonelderly adults in all but four years between 1994 and 2006.1 Previous research has established that low-income workers, those facing more stringent Medicaid eligibility requirements, and persons employed by smaller firms are more likely to lack health insurance.2 More recent research showed that those with moderate disabilities also confront high uninsurance rates because they can “fall through the cracks” between private insurance and Medicare (for disabled workers) or Medicaid (for disabled and poor persons, among others).3
To address this growing problem, the President and leaders in Congress are working on health insurance reform legislation that, as of this writing (early July 2009), is just beginning to emerge in detail.4 A common feature of the major proposals at this point is that coverage would be expanded by building on existing arrangements rather than by creating a new, comprehensive program. This approach has the advantage of allowing people to keep their current insurance if they wish to do so, but the complexity of current programs makes it difficult to design a reform that covers the uninsured without duplicating or destabilizing existing arrangements. The Medicaid program is particularly complicated because it is jointly financed and operated by the federal and state governments and each state has implemented it differently.
Our goal is to illuminate the size of the uninsurance problem for key groups and to show how the broad contours of existing public health insurance programs affect it. We focus on working-age persons with low incomes and chronic health conditions or disabilities because these persons often have difficulty obtaining and paying for private insurance.5 Many persons with disabilities are either not in the labor force or unemployed and therefore have no opportunity for employer-sponsored coverage. June 2009 figures for persons age 16–64 years old from the U.S. Bureau of Labor Statistics found that: 38.8% of men with disabilities were in the labor force (with 14.7% unemployed), compared with 84.9% of nondisabled men (10.4% unemployed); and 33.5% of women with disabilities were in the labor force (with 16.2% unemployed), compared with 73.0% of nondisabled women (8.7% unemployed).6 As a result, variations in the availability of public insurance could strongly affect uninsurance rates for these individuals.
Persons with chronic health conditions are particularly important for health reform because they are growing in number and they can experience particularly negative health consequences when uninsured. Using data from the National Health Interview Survey, a recent report found that 46.0 million nonelderly U.S. adults (aged 18–64) reported having at least one of seven major chronic conditions in 1997; by 2006, that number had risen to 57.7 million.7 This and other studies document significantly lower access to care among uninsured persons with chronic conditions compared with insured individuals, including lower rates of having a usual source of care, fewer primary care and specialist visits, more frequent use of emergency departments for primary care, and difficulties affording services.8 These studies complement a growing body of research documenting poorer health outcomes among uninsured persons with chronic conditions.9 Providing health insurance can improve health, changing downward trajectories of functional declines.10
We build on this literature by examining patterns in uninsurance for persons with disabilities or chronic health conditions by geographic area and Medicaid eligibility category and by considering the implications of these patterns for reform. Our results indicate that although public insurance programs appear to have reduced uninsurance rates for these populations nationally, very significant gaps exist for persons in particular Medicaid eligibility categories and geographic regions. Legislation to expand categorical eligibility for Medicaid and set a higher national minimum income threshold could have a substantial impact on these vulnerable groups.
Data and Sample
Overseen by the federal Agency for Healthcare Research and Quality (AHRQ), the Medical Expenditure Panel Survey (MEPS) aims to provide on-going nationally-representative estimates of health care use, expenditures, payment sources, and insurance coverage for the civilian, noninstitutionalized U.S. population.11 Our study relied on person-level data from the household and medical provider components of MEPS, pooled across 2000–2005. Since MEPS has a rotating panel design with 2-year panels, our sample contains more than one observation per respondent. To account for this and other sources of potential bias, we employed survey weights and adjustments for complex survey design (sampling strata, primary sampling unit) as recommended in AHRQ’s MEPS documentation.12
We used MEPS because it reports detailed and frequent measures of health insurance status as well as a wide range of health and disability measures. Certainly, other surveys can provide counts and characteristics of uninsured populations, and results may differ slightly depending on which survey is used.13 Differences across surveys in estimates of the size of the uninsured population arise due to differences in instrument, survey timing, and sampling.14
Using various MEPS items, we defined our key analytical variables, including insurance coverage, health conditions, disability status, geographic region, and potential Medicaid eligibility under federally mandated categories. We used a definition of uninsurance that is common in the literature, considering persons to be uninsured if they lacked health insurance for an entire year. The MEPS questionnaire asks respondents about having ever been diagnosed with any of nine conditions: asthma, chronic obstructive pulmonary disease (COPD), high blood pressure, coronary heart disease, angina, heart attack, other heart disease, stroke, and diabetes. We considered persons to have a potentially serious health condition if they indicated having any of these nine conditions. An important limitation of MEPS is that the list of chronic conditions does not include some prominent diagnoses, including cancer. A more comprehensive list would have led to higher estimates of the share of the population with chronic conditions.
Our disability indicator relied on MEPS queries about functional abilities. We considered any of the following to indicate disability: persistent use of assistive devices, persistent upper body functional limitations, lower body functional limitations, cognitive impairment, blindness, or major hearing impairment. Importantly, MEPS does not ask questions about major mental health conditions, and therefore we had no specific measure of psychiatric disability (apart from conditions that might have caused cognitive impairments). As with chronic conditions, this limitation of MEPS probably led to lower estimates of the number of persons with disabilities.
Our functional definition of disability contrasts with research on uninsurance among persons with disabilities that defines disability based on ability to work.15 Although ability to work has the advantage of matching the definition used in determining Supplemental Security Income (SSI) or Social Security Disability Insurance (SSDI) eligibility, it combines functional considerations with strength of attachment to the labor force and could be affected by the outcome of SSI or SSDI determinations.16 Using our definition, 68% of persons with disabilities also had at least one health condition.
To describe geographic variation in uninsurance we were limited to the broad Census regions identified in MEPS public use files: Northeast, Midwest, South, and West. MEPS data do not identify individual states due to confidentiality concerns. Because of the breadth of these large geographic categories, we do not attempt to use geographic variation to separate the effects of local economic conditions from the effects of Medicaid policy. Instead, we explore Medicaid policy effects by looking at Medicaid eligibility requirements that apply across states.
Federal law specifies that adults with low incomes and assets can qualify for Medicaid if they belong to specific eligibility categories including: old age, blindness, disability, being pregnant, or having young children. States, in turn, establish their own income and asset thresholds, which can vary substantially. For example, although the federal income threshold for persons with disabilities in most states is approximately 74% of the federal poverty level (FPL), the Balanced Budget Act of 1997 authorized states to optionally enroll disabled individuals with family incomes up to 250% FPL.17 States also have limited power to expand eligibility to other categories of individuals through demonstration programs, provided they obtain federal approval first; these programs remain relatively small nationally.18 For our analyses, we defined a “low income” variable indicating incomes under 125% FPL. We selected this level to include persons with incomes moderately above the Medicaid eligibility threshold in most states because this group has high uninsurance rates and is a likely target of policy initiatives.
To investigate the potential importance of Medicaid eligibility for persons with disabilities or chronic conditions, we constructed a variable to identify those who are potentially Medicaid eligible based on the mandatory federal categories that are relevant to our study population and for which variables exist in MEPS (blind, disabled, families with children). Based on MEPS survey items, we considered persons to be in a federal Medicaid eligibility category if they: (a) were blind; (b) were a parent of a child age 18 years or younger; or (c) had zero wages, reported not working due to disability, and had an income source that suggests disability (i.e., SSI, SSDI, VA pension, or workers’ compensation).19 For brevity, we typically refer to such individuals as belonging to a “federal category.”
Given our focus on working-age adults, our data included 93,839 observations from MEPS respondents who were between the ages of 25 and 61. We set our lower age cutoff at 25 years to eliminate college students who might have parental insurance. We set our upper age cutoff at 61 years because of a limitation of the MEPS income source response categories: MEPS indicates when respondents have Social Security income, but does not distinguish between disability and retirement income. Persons can start drawing Social Security income under old age provisions at age 62 years, so excluding persons over 61 years of age increases the accuracy of inferring that those reporting Social Security income have it through SSDI.20
Results
From 2000–2005, 14.8% of working-age U.S. residents lacked health insurance, including 11.6% of persons with disabilities and 11.1% of those with health conditions (Exhibit 1). Uninsurance rates were lower for those with disabilities and/or health conditions than for those without in all regional, income, and eligibility groups. This presumably reflected both the relative importance of health insurance to members of these groups and the availability of public insurance programs. Uninsurance rates were highest in the South (18.3%) and lowest in the Northeast (10.3%). Uninsurance rates were particularly high for low-income individuals (35.4% overall), reflecting the high cost of insurance relative to other needs for this population.
Exhibit 1.
Population and percent Uninsured by Health and Disability; Adults Aged 25–61, 2000–2005
| All | Disability | Health condition | Neither | |
|---|---|---|---|---|
| 100% | 4.5% | 30.5% | 68.0% | |
| Total population | (0%) | (.12%) | (.32%) | (.33%) |
| Percent uninsured | ||||
| 14.8% | 11.6% | 11.1% | 16.4% | |
| All | (.39%) | (.66%) | (.34%) | (.46%) |
| 10.3% | 5.0% | 6.7% | 12.0% | |
| Northeast | (.47%) | (.97%) | (.57%) | (.58%) |
| 11.2% | 10.0% | 8.8% | 12.2% | |
| Midwest | (.44%) | (1.52%) | (.54%) | (.61%) |
| 18.3% | 15.9% | 14.0% | 20.3% | |
| South | (.61%) | (.91%) | (.56%) | (.76%) |
| 16.5% | 10.2% | 12.0% | 18.4% | |
| West | (1.23%) | (1.78%) | (1.04%) | (1.35%) |
| 35.4% | 14.9% | 24.5% | 42.7% | |
| < 125% FPL | (.82%) | (1.36%) | (.78%) | (1.05%) |
| Not in federal | 14.9% | 13.9% | 11.1% | 16.8% |
| Medicaid category | (.38%) | (.87%) | (.45%) | (.48%) |
Source: Authors’ tabulations of MEPS data (standard errors in parentheses).
FPL: Federal Poverty Level.
Percent uninsured: Percent of person-years with no health insurance for whole year.
Disability and health condition categories are not mutually exclusive.
Disability: Persistent use of assistive devices, persistent upper body disability, lower body disability, cognitive impairment, blindness, or major hearing impairment.
Health condition: Ever diagnosed with asthma, hypertension, heart disease, angina, heart attack, stroke, pulmonary disease, or diabetes.
Federal Medicaid eligibility categories: blind; parent of child 18 or under; or zero wages and not working due to disability and income from SSI, SSDI, VA pension, and/or workers’ compensation.
Focusing on low-income persons with disabilities and health conditions, Exhibit 2 shows that uninsurance rates were very high in the South (21.5% for those with disabilities; 32.3% for those with health conditions) and much lower in the Northeast (4.6% and 12.2%). This is particularly significant because 39% of the total population and 46% of the uninsured lived in the South during this period (data not shown).
Exhibit 2.
Percent Uninsured by Region, Income, Health Conditions, and Disability; Adults Aged 25–61, 2000–2005
| All | North-east | Midwest | South | West | |
|---|---|---|---|---|---|
| Disability | |||||
| 11.6% | 5.0% | 10.0% | 15.9% | 10.2% | |
| All | (.66%) | (.97%) | (1.52%) | (.91%) | (1.78%) |
| 14.9% | 4.6% | 12.6% | 21.5% | 12.7% | |
| < 125% FPL | (1.36%) | (1.74%) | (2.75%) | (2.03%) | (2.71%) |
| 9.7% | 5.2% | 8.4% | 12.7% | 8.9% | |
| > 125% FPL | (.59%) | (1.00%) | (1.10%) | (.90%) | (1.82%) |
| Health condition | |||||
| 11.1% | 6.7% | 8.8% | 14.0% | 12.0% | |
| All | (.34%) | (5.7%) | (.54%) | (.56%) | (1.04%) |
| 24.5% | 12.2% | 21.0% | 32.3% | 23.3% | |
| < 125% FPL | (.78%) | (1.26%) | (1.78%) | (1.26%) | (1.94%) |
| 8.8% | 5.7% | 7.1% | 10.6% | 10.2% | |
| > 125% FPL | (.29%) | (.63%) | (.44%) | (.51%) | (.89%) |
| No disability or health condition | |||||
| 16.4% | 12.0% | 12.2% | 20.3% | 18.4% | |
| All | (.46%) | (.58%) | (.61%) | (.76%) | (1.35%) |
| 42.7% | 27.4% | 37.6% | 52.3% | 41.6% | |
| < 125% FPL | (1.05%) | (2.42%) | (2.16%) | (1.29%) | (2.17%) |
| 13.1% | 10.4% | 9.7% | 15.6% | 15.1% | |
| > 125% FPL | (.39%) | (.55%) | (.50%) | (.70%) | (1.19%) |
Source: Authors’ tabulations of MEPS data (standard errors in parentheses).
FPL: Federal Poverty Level.
Percent uninsured: Percent of person-years with no health insurance for whole year.
Disability and health condition categories are not mutually exclusive.
Disability: Persistent use of assistive devices, persistent upper body disability, lower body disability, cognitive impairment, blindness, or major hearing impairment.
Health condition: Ever diagnosed with asthma, hypertension, heart disease, angina, heart attack, stroke, pulmonary disease, or diabetes.
Exhibit 1 shows that the federal Medicaid eligibility categories did not make much difference in the overall uninsurance rate (14.9% uninsured among those not in a federal category vs. 14.8% overall). However, when the focus is narrowed to low-income persons with disabilities or health conditions, Exhibit 3 shows that federal Medicaid eligibility categories made an enormous difference. About 80% of low-income persons with disabilities or health conditions fell into federal categories (data not shown). Those who were not in federal categories represented a relatively small group, but they had very high uninsurance rates. Persons not in federal categories were approximately twice as likely to be uninsured (25.6% vs. 12.8% for low-income with disabilities and 37.2% vs. 19.9% for low-income with health conditions). This pattern held in each region as well as across the nation as a whole (data not shown). An example of such a person would be a childless adult with income near poverty and a chronic illness or perhaps a functional impairment who has not qualified for income support from SSI or SSDI.
Exhibit 3.
Percent Uninsured by Federal Medicaid Eligibility Categories, Income, Health, and Disability; Adults Aged 25–61, 2000–2005
| Federal categories | Not federal categories | Share of population | |
|---|---|---|---|
| Disability | |||
| 10.6% | 13.9% | 4.5% | |
| All | (.89%) | (.87%) | (.12%) |
| 12.8% | 25.6% | 1.6% | |
| < 125% FPL | (1.30%) | (4.00%) | (.07%) |
| 8.8% | 11.1% | 2.8% | |
| > 125% FPL | (.84%) | (1.02%) | (.09%) |
| Health condition | |||
| 11.0% | 11.1% | 30.5% | |
| All | (.42%) | (.45%) | (.32%) |
| 19.9% | 37.2% | 4.4% | |
| < 125% FPL | (.83%) | (2.04%) | (.14%) |
| 8.6% | 9.0% | 26.1% | |
| > 125% FPL | (.38%) | (.39%) | (.31%) |
| No disability or health condition | |||
| 16.1% | 16.8% | 6 8.0% | |
| All | (.55%) | (.48%) | (.33%) |
| 38.3% | 53.7% | 7.6% | |
| < 125% FPL | (1.18%) | (1.59%) | (.21%) |
| 12.3% | 14.0% | 60.4% | |
| > 125% FPL | (.48%) | (.42%) | (.37%) |
Source: Authors’ tabulations of MEPS data (standard errors in parentheses).
FPL: Federal Poverty Level.
Percent uninsured: Percent of person-years with no health insurance for whole year.
Disability and health condition categories are not mutually exclusive.
Disability: Persistent use of assistive devices, persistent upper body disability, lower body disability, cognitive impairment, blindness, or major hearing impairment.
Health condition: Ever diagnosed with asthma, hypertension, heart disease, angina, heart attack, stroke, pulmonary disease, or diabetes.
Federal Medicaid eligibility categories: blind; parent of child 18 or under; or zero wages and not working due to disability and income from SSI, SSDI, VA pension, and/or workers’ compensation.
Finally, having found that uninsurance was concentrated by region, income, and federal Medicaid eligibility category, we explored how recent trends in uninsurance rates affected these groups. Exhibit 4 shows that uninsurance rates increased overall from 13.7% in 2000 to 16.0% in 2005. This trend (about a 17% increase) was evident in all four geographic regions. Roughly the same trend existed for all low-income persons without disability or health conditions. What is most interesting is that this pattern did not apply to low-income persons with disabilities or health conditions. Those who belonged to a federal eligibility category experienced only a 3.5% increase (from 19.8% to 20.5%) while those not in a federal category saw rates jump 18.6% (from 35.4% to 42.0%).
Exhibit 4.
Trends in Uninsurance by Region, Income, Health, Disability, and Federal Medicaid Eligibility Categories; Adults Aged 25–61, 2000–2005
| 2000 | 2001 | 2002 | 2003 | 2004 | 2005 | |
|---|---|---|---|---|---|---|
| 13.7% | 13.9% | 14.5% | 15.0% | 15.5% | 16.0% | |
| All | (.58%) | (.54%) | (.42%) | (.41%) | (.53%) | (.63%) |
| 9.9% | 10.7% | 10.6% | 9.8% | 9.3% | 11.6% | |
| Northeast | (.63%) | (.59%) | (.63%) | (.91%) | (.84%) | (.98%) |
| 9.2% | 10.7% | 10.9% | 11.1% | 12.5% | 12.7% | |
| Midwest | (.75%) | (.59%) | (.72%) | (.61%) | (1.06%) | (.80%) |
| 17.1% | 16.2% | 17.8% | 19.2% | 19.5% | 19.7% | |
| South | (.98%) | (1.03%) | (.64%) | (.77%) | (.87%) | (.82%) |
| 15.9% | 16.2% | 16.1% | 16.6% | 17.3% | 17.0% | |
| West | (1.54%) | (1.41%) | (1.19%) | (1.10%) | (1.43%) | (1.96%) |
| < 125 FPL with disability and/or health condition | ||||||
| In federal | 19.8% | 18.5% | 19.7% | 19.7% | 19.7% | 2 0.5% |
| categories | (2.19%) | (1.87%) | (1.64%) | (1.54%) | (1.53%) | (2.20%) |
| Not in | ||||||
| federal | 35.4% | 30.5% | 34.5% | 35.1% | 40.3% | 42.0% |
| categories | (6.35%) | (3.71%) | (4.20%) | (3.39%) | (3.87%) | (3.95%) |
| < 125 FPL without disability or health condition | ||||||
| In federal | 36.6% | 38.8% | 37.6% | 37.4% | 38.6% | 40.6% |
| categories | (2.35%) | (2.09%) | (1.96%) | (1.63%) | (1.92%) | (1.83%) |
| Not in | ||||||
| federal | 50.8% | 50.0% | 51.0% | 53.3% | 59.0% | 56.0% |
| categories | (5.03%) | (3.55%) | (2.18%) | (3.28%) | (3.35%) | (2.49%) |
Source: Authors’ tabulations of MEPS data (standard errors in parentheses).
Federal Medicaid eligibility categories: blind; parent of child 18 or under; or zero wages and not working due to disability and income from SSI, SSDI, VA pension, and/or workers’ compensation FPL: Federal Poverty Level.
Percent uninsured: Percent of person-years with no health insurance for whole year.
Disability and health condition categories are not mutually exclusive.
Disability: Persistent use of assistive devices, persistent upper body disability, lower body disability, cognitive impairment, blindness, or major hearing impairment.
Health condition: Ever diagnosed with asthma, hypertension, heart disease, angina, heart attack, stroke, pulmonary disease, or diabetes.
Discussion
Focusing on low-income persons with chronic health conditions or disabilities (about 6% of the working-age population), we found two large and growing gaps in public insurance programs. The first was regional: uninsurance rates were very high in the South and much lower in the Northeast. The second was categorical: uninsurance rates were approximately doubled across the country for individuals not in federally mandated Medicaid eligibility categories. Taken together, low-income persons with health conditions or disabilities who lived in the South and did not belong to a federal Medicaid eligibility category had an uninsurance rate of 49% across our study period. Given the scope of the current economic crisis, the rate of growth of these gaps has likely accelerated in the past year.21
Several elements of the health care reform bills currently (July 2009) before Congress are particularly well targeted for closing these gaps. Current legislative drafts would increase the minimum federal income eligibility threshold for Medicaid with thresholds that range from 115% of FPL (Senate Finance Committee) to 150% of FPL (Senate Health, Education Labor and Pensions Committee).22 These provisions would help low-income persons obtain Medicaid coverage and would probably disproportionately benefit those living in southern states where income thresholds tend to be low. For example, income thresholds for unemployed parents in 2009 were 21% of FPL in Florida, 29% in Georgia, and 13% in Texas. Corresponding income thresholds were typically much higher in northeastern states: 150% in New York, 90% in Ohio, and 133% in Massachusetts.23 On the other hand, the southern states may have limited capacity to raise additional revenues, so any legislation that expands Medicaid might have to feature larger federal match rates (probably more like CHIP than Medicaid in recent years).24
Current Senate and House bills also expand basic categorical eligibility for Medicaid to all income-qualifying persons, a change that clearly would benefit low-income persons not in federal categories as defined above. These bills also provide in some cases for higher thresholds for parents and pregnant women. Our results suggest that higher thresholds for persons with disabilities (broadly defined) or chronic health conditions also might be warranted. Alternatively, if budget pressures force cuts in the bills currently before Congress, our results indicate that preservation of eligibility expansions for these groups would target a population that is particularly vulnerable to uninsurance and its deleterious effects on health.
Notes
- 1.Fronstin P. Sources of Health Insurance and Characteristics of the Uninsured: Analysis of the March 2007 Current Population Survey EBRI Issue Brief No. 310. Washington: Employee Benefit Research Institute; 2007. October. [PubMed] [Google Scholar]; Fronstin reports that uninsurance rates for those under 65 grew from 15.9% in 1994 to 17.9% in 2006. Our figures are lower because we exclude those under 25 and 62–64.
- 2.Cunningham PJ, Ginsburg PB. What Accounts for Differences in Uninsurance Rates Across Communities? Inquiry. 2001. Spring;38:6–21. [DOI] [PubMed] [Google Scholar]
- 3.Sommers AS. Access to Health Insurance, Barriers to Care, and Service Use Among Adults with Disabilities. Inquiry. 2006/2007. Winter;43:393–405. [DOI] [PubMed] [Google Scholar]
- 4.Health Reform [Internet] Menlo Park (CA): The Henry J. Kaiser Family Foundation; 1991. -. Side-by-Side Comparison of Major Health Care Reform Proposals [cited 2009 Jul 6]. Available from: http://www.kff.org/healthreform/upload/healthreform_sbs_full.pdf. [Google Scholar]
- 5.Roughly 30 states have established high-risk pools to make insurance available to otherwise uninsurable individuals, but the premiums are typically very high. See; Frakt AB, Pizer SD, Wrobel MV. High-Risk Pools for Uninsurable Individuals: Recent Growth, Future Prospects. Health Care Financ Rev. 2004/2005. Winter;26(2):73–87. [PMC free article] [PubMed] [Google Scholar]
- 6.Bureau of Labor Statistics [Internet], Washington: United States Department of Labor; 1995. -. Labor Force Statistics from the Current Population Survey [cited 2009 Jul 8]. Available from: http://www.bls.gov/cps/cpsdisability.htm. [Google Scholar]
- 7.Hoffman C, Schwartz K. Eroding Access Among Nonelderly U.S. Adults With Chronic Conditions: Ten Years of Change. Health Aff (Milwood). 2008. July;27(5):w340–w8. [DOI] [PubMed] [Google Scholar]
- 8.See also Wilper AP, Woolhandler S, Lasser KE, McCormick D, Bor DH, Himmelstein DU. A National Study of Chronic Disease Prevalence and Access to Care in Uninsured U.S. Adults. Ann Intern Med. 2008. August;149:170–6. [DOI] [PubMed] [Google Scholar]
- 9.Ayanian JZ, Kohler BA, Abe T, Epstein AM. The Relation Between Health Insurance Coverage and Clinical Outcomes Among Women with Breast Cancer. New Engl J Med. 1993;329 (5):326–31; [DOI] [PubMed] [Google Scholar]; McWilliams JM, Zaslavsky AM, Meara E, Ayanian JZ. Health Insurance Coverage and Mortality Among the Near-Elderly. Health Aff (Milwood). 2004. July;23(4):223–33; [DOI] [PubMed] [Google Scholar]; Ayanian JZ, Zaslavsky AM, Weissman JS, Schneider EC, Ginsburg JA. Undiagnosed Hypertension and Hypercholesterolemia Among Uninsured and Insured Adults in the Third National Health and Nutrition Examination Survey. Am J Public Health. 2003. December;93(12):2051–4; [DOI] [PMC free article] [PubMed] [Google Scholar]; Fowler-Brown A, Corbie-Smith G, Garrett J, Lurie N. Risk of Cardiovascular Events and Death—Does Insurance Matter? J Gen Intern Med. 2007. April;22(4):502–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.McWilliams JM, Meara E, Zaslavsky AM, Ayanian JZ. Health of Previously Uninsured Adults After Acquiring Medicare Coverage. JAMA. 2007;298(24):2886–94. [DOI] [PubMed] [Google Scholar]
- 11.Cohen J. Design and methods of the Medical Expenditure Panel Survey Household Component MEPS Methodology Report No. 1. Rockville (MD): Agency for Health Care Policy and Research; 1997. AHCPR Pub No: 97–0026. [Google Scholar]
- 12.Medical Expenditure Panel Survey [Internet]. Rockville (MD): Agency for Healthcare Research & Quality; (2007a); 1999 –. MEPS HC–036: 1996–2005 Pooled Estimation File [cited 2008 Jun 8]. Available from: http://www.meps.ahrq.gov/mepsweb/data_stats/download_data/pufs/h36/h36u06doc.pdf#xml=http://meps.ahrq.gov/cgi-bin/texis/webinator/search/pdfhi.txt?query=MEPS+HC-036&pr=MEPSFULLSITE&prox=page&rorder=500&rprox=500&rdfreq=500&rwfreq=500&rlead=500&sufs=0&order=r&cq=&id=4a5290db291. [Google Scholar]; Medical Expenditure Panel Survey [Internet]. Rockville (MD): Agency for Healthcare Research & Quality; (2007b); 1999 –. MEPS HC-036BRR: 1996–2005 Replicates for Calculating Variances File [cited 2008 Jun 6]. Available from: http://www.meps.ahrq.gov/mepsweb/data_stats/download_data/pufs/h36brr/h36b06doc.pdf#xml=http://meps.ahrq.gov/cgi-bin/texis/webinator/search/pdfhi.txt?query=HC-036BRR&pr=MEPSFULLSITE&prox=page&rorder=500&rprox=500&rdfreq=500&rwfreq=500&rlead=500&sufs=0&order=r&cq=&id=4a5290dc19 [Google Scholar]
- 13.Fronstin P. Counting the Uninsured: A Comparison of National Surveys EBRI Issue Brief No. 225. Washington: Employee Benefit Research Institute; 2000. September. [PubMed] [Google Scholar]
- 14.Short PF. Counting and Characterizing the Uninsured [Internet]. Ann Arbor (MI): Economic Research Initiative on the Uninsured, Working Paper 2; 2001. December 4 [cited 2005 Nov 4]. Available from: www.umich.edu/~eriu/pdf/wp2.pdf [Google Scholar]
- 15.Sommers AS. Access to Health Insurance, Barriers to Care, and Service Use Among Adults with Disabilities. Inquiry. 2006/2007. Winter;43(4):393–405. [DOI] [PubMed] [Google Scholar]
- 16.A determination of eligibility for SSI also qualifies the applicant for Medicaid in most states. Similarly, eligibility for SSDI benefits implies eligibility for Medicare, but only after a 24-month waiting period. For additional details see US House of Representatives Overview of Entitlement Programs, 2004.
- 17.US House of Representatives Overview of Entitlement Programs, 2004.; 74% of FPL corresponds to the SSI benefit level and is therefore the minimum income threshold for most states. States also may enroll individuals with higher incomes through “medically needy” programs. Such individuals must be members of one or more of the eligibility categories listed above.
- 18.Section 1115 of the Social Security Act and the Health Insurance Flexibility and Accountability (HIFA) Initiative enable states to extend eligibility under demonstration waivers explicitly designed to reduce the uninsurance rate in a state Demonstration waivers like these accounted for 8% of national Medicaid enrollment in 2000 (US House of Representatives Overview of Entitlement Programs, 2004). [Google Scholar]
- 19.There is no way to identify pregnant women in MEPS. This definition is consistent with rules summarized in documentation for the TRIM3 microsimulation model developed by the Urban Institute (TRIM3 project website, http://trim3.urban.org, downloaded on June 2, 2008).
- 20.80 percent of individuals 25–61 with OASDI income receive it due to disability (authors’ analysis of 2001 OASDI Public Use Microdata File, www.ssa.gov/policy/docs/microdata/mbr/). Our definition of working age (25–61) differs, for the reasons just provided, from that often considered in other studies (typically 18–64). We note that using the Current Population Survey, 17.7 percent of 18–65 year olds were uninsured in 2003. Fronstin P. Sources of Health Insurance and Characteristics of the Uninsured: Analysis of the March 2004 Current Population Survey. EBRI Issue Brief No. 276. Washington: Employee Benefit Research Institute; 2004 Dec. Using data from MEPS for the same year and age range we find that 16.3 percent were uninsured for the full year. One possible reason for the difference is that the CPS measure relies on respondent recall from only one interview to characterize full-year uninsurance, whereas MEPS interviews respondents several times each year, relying on a shorter recall period between interviews.
- 21.Rowland D. Health Care and Medicaid - Weathering the Recession. New Engl J Med. 2009. March;360(13):1273–6. [DOI] [PubMed] [Google Scholar]
- 22.Health Reform [Internet]. Menlo Park (CA): The Henry J. Kaiser Family Foundation; 1991. -. Side-by-Side Comparison of Major Health Care Reform Proposals [cited 2009 Jul 6]. Available from: http://www.kff.org/healthreform/upload/healthreform_sbs_full.pdf. [Google Scholar]
- 23.State Policy [Internet]. Menlo Park (CA): The Henry J. Kaiser Family Foundation; 1991. -. State Health Facts [cited 2009 Jul 7]. Available from: http://www.statehealthfacts.org The website has tables of Medicaid income thresholds by state. [Google Scholar]
- 24.Marquis MS, Long SH. Federalism and Health System Reform. Prospects for State Action. JAMA. 1997. August;278(6):514–7. [PubMed] [Google Scholar]