Abstract
Objectives. We examined the number and clinical needs of uninsured veterans, including those who will be eligible for the Medicaid expansion and health insurance exchanges in 2014.
Methods. We analyzed weighted data for 8710 veterans from the 2010 National Survey of Veterans, classifying it by veterans’ age, income, household size, and insurance status.
Results. Of 22 million veterans, about 7%, or more than 1.5 million, were uninsured and will need to obtain coverage by enrolling in US Department of Veterans Affairs (VA) care or the Medicaid expansion or by participating in the health insurance exchanges. Of those uninsured, 55%, or more than 800 000, are likely eligible for the Medicaid expansion if states implement it. Compared with veterans with any health coverage, those who were uninsured were younger and more likely to be single, Black, and low income and to have been deployed to Iraq and Afghanistan.
Conclusions. The Patient Protection and Affordable Care Act is likely to have a considerable impact on uninsured veterans, which may have implications for the VA, the Medicaid expansion, and the health insurance exchanges.
The Patient Protection and Affordable Care Act (ACA)1 represents one of the most significant overhauls of the US health care system and is expected to affect millions of uninsured people across the country. Military veterans constitute a particularly important segment of the population because of their service to the country, access to US Department of Veterans Affairs (VA) health care, and other special benefits after their service. However, little has been written on the potential impact of the ACA on the health and health care of veterans.2 Although the VA operates an integrated national health care system that offers free or low-cost services to eligible veterans, many veterans are not enrolled in VA health care, and some are ineligible. Enrollment in VA health care satisfies the ACA’s requirement for insurance coverage, but eligibility for VA health care is determined on the basis of a complex system of priorities, mostly based on service-connected disability, income, and age, and it generally requires a military service discharge that is other than dishonorable (i.e., honorable, general).
One study estimated that only 13% (3.6 million) of veterans report receiving some or all of their health care at the VA, and the vast majority (> 20 million) receive no health care from the VA.3 Most veterans thus rely on non-VA health care and are covered by various private or other public forms of health insurance, including Medicare and Medicaid. A small, albeit important, minority of veterans have no health insurance coverage. Estimates based on data from 1987 to 2004 showed that 7.7% of veterans were uninsured (including having no VA coverage), which equates to nearly 1.8 million veterans and represents 4.7% of all uninsured US residents.4
Lack of health insurance coverage is an important problem because it can hinder access to effective health care, including needed medical visits, preventive care, and other services, and it can ultimately lead to poor health, premature mortality, and high medical costs.5,6 Being uninsured is a growing problem in the United States that the ACA addresses by requiring virtually all legal US residents to have health insurance. The ACA includes various provisions to help US residents, including veterans, accomplish this.
One major provision that is optional for states to implement is the expansion of Medicaid coverage to all individuals aged 18 to 65 years with incomes at or below 138% of the federal poverty level. Although not all states will implement this expansion, and the number of participating states is currently unknown, many poor, uninsured adults will be able to obtain Medicaid coverage in states that implement the Medicaid expansion. Uninsured adults who have incomes above the Medicaid expansion limit or who live in states that do not implement the Medicaid expansion will have to purchase health insurance and may participate in the health insurance exchanges.
A second major provision of the ACA is the creation of health insurance exchanges in each state whereby individuals may purchase competitive health insurance plans that are eligible for federal subsidies, but those subsidies are only available to those with income above the federal poverty level. Both of these major ACA provisions are planned for implementation in 2014 and will introduce a variety of coverage options for US residents, including veterans.
There has been little study of uninsured veterans and no study of the potential impact of the ACA on veterans in general. Moreover, most data that exist on veterans are based on VA data, which only contain information about veterans who use VA health services and do not include information about those who are uninsured or not covered by VA health care. However, 1 population-based study7 has provided some evidence that a substantial number of veterans are uninsured (particularly those younger than 65 years) and that many uninsured veterans are in poor health, often forego needed health care because of costs, and have equal or worse access to health care than other uninsured adults in the general population. As the country moves toward a new era of health care with the ACA and continues to engage in conflicts in the Middle East, the impact of the ACA on the health care of veterans needs to be considered.
We used a recent nationally representative survey of veterans to (1) describe the proportion and characteristics of veterans who are currently uninsured because they will likely be required to obtain coverage under the ACA; (2) determine, among those who are uninsured, who will likely be eligible for the Medicaid expansion; and (3) compare the sociodemographic and health characteristics of those who are uninsured and likely eligible for Medicaid expansion (LEME), those who are uninsured and not LEME, and those who currently have health insurance coverage. The results provide information about the number and health characteristics of veterans who will likely be affected by different provisions of the ACA and inform planning efforts for the VA and states that implement the Medicaid expansion and health insurance exchanges.
METHODS
The 2010 National Survey of Veterans is the sixth in a series of comprehensive nationwide surveys designed to help the VA plan future programs and service for veterans.8 The 2010 National Survey of Veterans was conducted by means of a mailed, self-administered questionnaire using address-based sampling. The sampling frame was all US postal addresses, and sampling occurred in 2 phases. In the first phase, a sample of addresses from all 50 US states was purchased from a commercial vendor, Marketing Systems Group (Horsham, PA), that used an implicit stratification resulting from sorting addresses by zip code, carrier route, and walking sequence. The sorted list was divided into equally sized continuous intervals and 1 address was selected from each interval at random. A short screening questionnaire was sent to these addresses to determine whether the household included members who were veterans. In the second phase, responding households with veteran members were then sent full surveys. Full details of the sampling frame construction and design have been described in the final report of the 2010 National Survey of Veterans.8
Of a total of 13 058 surveys distributed to veterans, 8710 veterans from all 50 states returned completed surveys (66.7% response rate). Survey data were weighted to represent the entire noninstitutionalized veteran population, and both total weights and relative weights were used. Total sampling weights were created for each veteran as the product of the final screener household weight (after nonresponse adjustment), a multiple-address adjustment factor, and a ratio adjustment factor to external population control so that the sums of the total sampling weights matched the known population.8 Relative weights included similar adjustments, but they maintained the sample size of the original sample so individuals were weighted in proportion to their representativeness in the population of inference, and their sums matched the original sample size. Nearly all the statistical analyses were conducted using total weights, except the analyses involving continuous variables, in which relative weights were used to preserve the standard errors of the original sample and not artificially depress them. Further details about the weighting procedures have been described elsewhere.9
Veterans were classified into 3 groups. First, we identified veterans who were uninsured and LEME as those younger than 65 years who reported no health insurance coverage (including VA coverage) and had a household income equal to or lower than 138% of the 2010 US Department of Health and Human Services federal poverty level, taking into account the number of people in their household (spouses were assumed to be living in the same household along with dependent children younger than 18 years). Veterans who did not meet all of these criteria were grouped into a second group—veterans who were uninsured and not LEME (who are likely to participate in the health insurance exchanges or enroll in VA coverage if eligible). A third group consisted of veterans who reported current health insurance coverage (including VA coverage, which will satisfy the requirements of the ACA).
Measures
Health insurance coverage was assessed by asking veterans to indicate whether they were currently covered by a list of different types of health insurance and health coverage plans, including insurance through a current or former employer, purchasing directly from insurance companies, Medicare, Medicaid, the VA, TRICARE (a form of health care for the armed forces), and the Indian Health Service. Veterans could mark multiple types of insurance and coverage plans, and we included a specific checkbox for “no health insurance” as well. VA service utilization data were cross-validated with reported insurance coverage; we considered veterans who reported using VA health services in the past 6 months as currently covered by the VA. Veterans were also asked whether they had ever been enrolled in VA health care in their lifetime.
Veterans self-reported background characteristics, including information about sociodemographic characteristics and military service. Of particular relevance for determining eligibility for the Medicaid expansion, household income was determined by asking veterans to report their income range (e.g., < $10 000, $10 000–29 999, $30 000–49 999), and household size was based on marital status (we assumed veterans lived with their spouses) and the number of dependent children younger than 18 years.
Several questions assessed health status by asking participants to rate their general health on a 5-point scale; whether they needed assistance with a list of activities of daily living, which were summed for a total score; whether they needed the aid of an attendant; whether they were permanently housebound; and whether they were exposed to any military combat trauma (exposed to dead, dying, or wounded people during military service).
Health service use was assessed by asking veterans what, if any, VA and non-VA health services they used in the past 6 months, including medical, mental health, rehabilitation, and emergency department services. The survey also assessed whether Medicaid was specifically used to pay for any of these services.
Data Analysis
We calculated descriptive statistics of the total weighted sample (n = 22 172 806) to summarize the number of veterans with different types of health insurance and their service use. Then, we divided veterans into 3 groups on the basis of their health insurance, age, income level, and the number of people in their household: (1) veterans who had no health care coverage and are LEME, (2) veterans who had no health care coverage and are not LEME, and (3) veterans who currently have health care coverage. We used the total weighted sample to provide estimated numbers for the entire veteran population.
We compared these groups on sociodemographic characteristics, general health, and health service use. Analysis of variance and the χ2 test revealed that all differences between groups were significant at the P < .001 level given the large sample size (and resultant statistical power to identify very small differences), so we instead focused on effect sizes. We calculated the Cohen d for continuous variables and differences in percentages (Δ%) for categorical variables. Medium effect sizes (Cohen d > 0.5 or ±Δ%) were specially noted.
RESULTS
The majority of veterans had health insurance coverage through an employer, and only about 7% of veterans reported having no health insurance coverage (Table 1). Nearly a quarter of veterans had been enrolled in VA health care before, and about 21% reported currently being covered by VA health care. A small minority, fewer than 4%, reported Medicaid coverage, and a small proportion of veterans reported using Medicaid to pay for various health services in the past 6 months.
TABLE 1—
Characteristics of Total Weighted Sample of Veterans (n = 22 172 806): 2010 National Survey of Veterans, United States
| Characteristic | No. (%) |
| Ever enrolled in VA health care | 5 466 647 (24.65) |
| Ever used VA health care | 5 956 942 (26.87) |
| Current health insurance coverage | |
| No health care insurancea | 1 549 902 (6.99) |
| Directly from insurance companyb | 2 794 405 (12.60) |
| Through current or former employer | 11 605 463 (52.34) |
| Medicare | 8 558 807 (38.60) |
| Medicaid | 774 629 (3.49) |
| VA | 4 613 739 (20.81) |
| TRICARE or other military health care | 2 041 667 (9.21) |
| Other type of insurance | 662 654 (2.99) |
| Used any VA health services in past 6 mo | 4 014 803 (18.11) |
| Used Medicaid in the past 6 mo for | |
| Medical or surgical care | 288 116 (1.30) |
| Outpatient medical care | 709 069 (3.20) |
| Inpatient mental health treatment | 40 216 (0.18) |
| Outpatient mental health treatment | 93 494 (0.42) |
| Rehabilitation or nursing care facility | 92 536 (0.42) |
| Emergency department | 259 055 (1.17) |
Note. VA = US Department of Veterans Affairs. TRICARE is a form of health care for the armed forces.
Including no Veterans Affairs coverage.
Some veterans reported more than 1 type of insurance coverage, so the sum of the coverage categories was more than 22 million or 100%.
Of the total weighted sample who provided adequate data to determine eligibility for the Medicaid expansion (n = 20 479 200), 3.31% reported no insurance and were LEME, 3.97% reported no insurance and were not LEME, and 92.73% reported having current health insurance. Thus, 7.28% or 1 490 886 veterans reported no current health care coverage and, of those, 54.53% were LEME.
Of all veterans who reported current health care coverage (n = 18 447 871), 4.18% were LEME (1.34% were already covered by Medicaid). Of all veterans who reported currently being covered by the VA (n = 4 440 400), 15.79% were LEME (1.57% were already covered by Medicaid). Conversely, of all veterans who were LEME (n = 1 921 639), 36.49% reported currently being covered by the VA. Among all Operation Enduring Freedom–Operation Iraqi Freedom–Operation New Dawn (OEF–OIF–OND) veterans (n = 1 577 532), 19.61% reported no current health care coverage; of those, 43.31% were LEME.
To estimate the number of uninsured veterans who were LEME and living in 1 of the 15 states (Alabama, Alaska, Georgia, Idaho, Kansas, Louisiana, Maine, Mississippi, Montana, Nebraska, North Carolina, South Carolina, Texas, Wisconsin, and Wyoming) that had decided not to participate in the Medicaid expansion as of October 2013, we applied state-level proportions of uninsured veterans who were LEME provided in a report by the Robert Wood Johnson Foundation and the Urban Institute10 to the data. Approximately 31.45% of all uninsured veterans lived in states that do not plan to participate in the Medicaid expansion, which equates to approximately 487 444 uninsured veterans in this study. Of those uninsured veterans living in states that do not plan to participate in the Medicaid expansion, an estimated 40.82% were LEME or, stated another way, 198 975 veterans in this study will likely not have Medicaid coverage because they live in states that do not plan to participate in the Medicaid expansion.
Table 2 shows the sociodemographic and health characteristics of the total sample on the basis of insurance status and likely eligibility for the Medicaid expansion. First, we found sizable differences (d ≥ ±0.5 or Δ% > 5%) between veterans who were uninsured and those who had health insurance coverage; veterans who were uninsured appeared to be younger, less educated, lower income, and likely to be Black, unmarried, and to have served in OEF–OIF–OND.
TABLE 2—
Background Characteristics of Uninsured Veterans Who Will Likely Be Eligible for Medicaid in 2014 Compared with Those Who Will Likely Not Be Eligible and Those Currently with Other Health Care Coverage: 2010 National Survey of Veterans, United States
| Characteristic | 1. Currently Uninsured and LEME (n = 677 450), No. (%) or Mean ±SE | 2. Currently Uninsured and Not LEME (n = 811 736), No. (%) or Mean ±SE | 3. Currently Insured (n = 18 990 014), No. (%) or Mean ±SE | Effect Size Difference Between 1 and 2, d or Δ% | Effect Size Difference Between 1 and 3, d or Δ% |
| Age, y | 44.27 ±12.14 | 47.70 ±15.28 | 62.73 ±15.23 | −0.25 | −1.34 |
| Gender, male | 559 267 (89.24) | 693 917 (92.28) | 15 485 955 (92.36) | −3.04 | −3.12 |
| Education | |||||
| < high school | 38 710 (5.71) | 46 606 (5.75) | 952 411 (5.10) | −0.04 | 0.61 |
| High school or GED | 290 533 (42.89) | 332 731 (41.07) | 4 543 870 (24.31) | 1.82 | 18.58 |
| Some college | 269 963 (39.85) | 281 541 (34.75) | 5 429 996 (29.05) | 5.10 | 10.80 |
| Associate’s or bachelor’s degree | 71 036 (10.49) | 117 643 (14.52) | 5 327 562 (28.50) | −4.03 | −18.01 |
| Advanced degree | 7207 (1.06) | 31 676 (3.91) | 2 436 288 (13.04) | −2.85 | −11.98 |
| Race | |||||
| White | 369 906 (55.43) | 565 347 (72.14) | 15 989 593 (87.03) | −16.71 | −31.60 |
| Black | 252 554 (37.84) | 155 442 (21.24) | 1 707 473 (9.29) | 16.60 | 28.55 |
| American Indian/Alaska Native | 44 910 (6.73) | 42 526 (5.43) | 396 359 (2.16) | 1.30 | 4.57 |
| Asian/Pacific Islander | 0 (0.00) | 9387 (1.20) | 278 286 (1.51) | −1.20 | −1.51 |
| Household income, $ | |||||
| < 10 000 | 314 101 (46.37) | 9963 (1.24) | 446 846 (2.52) | 45.13 | 43.85 |
| 10 000–29 999 | 348 653 (51.47) | 301 525 (37.67) | 3 445 306 (19.43) | 13.80 | 32.04 |
| 30 000–49 999 | 14 697 (2.17) | 253 323 (31.65) | 4 675 164 (24.03) | −29.48 | −21.86 |
| 50 000–74 999 | 0 (0.00) | 174 988 (21.86) | 4 046 252 (22.82) | −21.86 | −22.82 |
| 75 000–99 999 | 0 (0.00) | 28 198 (3.52) | 2 277 097 (12.84) | −3.52 | −12.84 |
| ≥ 100 000 | 0 (0.00) | 32 437 (4.05) | 3 253 858 (18.35) | −4.05 | −18.35 |
| Marital status | |||||
| Married or civil union | 295 975 (43.83) | 350 732 (43.37%) | 13 969 554 (74.74) | 0.46 | −30.91 |
| Widowed, divorced, or separated | 239 958 (35.53) | 273 905 (33.87%) | 3 553 962 (19.02) | 1.66 | 16.51 |
| Never married | 139 345 (20.64) | 183 975 (22.75) | 1 166 118 (6.24) | −2.11 | 14.40 |
| No. of dependent children < 18 y | |||||
| 0 | 289 266 (46.39) | 506 696 (69.15) | 12 205 008 (78.01) | −22.76 | −31.62 |
| 1 | 173 550 (27.83) | 129 398 (17.66) | 1 645 320 (10.52) | 10.17 | 17.31 |
| 2 | 73 939 (11.86) | 53 193 (7.26) | 1 249 046 (7.98) | 4.60 | 3.88 |
| 3 | 49 165 (7.88) | 41 871 (5.71) | 362 622 (2.32) | 2.17 | 5.56 |
| ≥ 4 | 37 672 (6.04) | 1562 (0.21) | 182 692 (1.17) | 5.83 | 4.87 |
| Year first entered active duty | 1985.70 ±12.18 | 1982.62 ±15.39 | 1967.79 ±15.48 | 0.22 | 1.29 |
| OEF–OIF–OND | 133 955 (21.07) | 175 344 (22.30) | 1 268 233 (6.96) | −1.23 | 14.11 |
| Employed | 196 936 (29.64) | 394 197 (50.86) | 8 425 638 (46.89) | −21.22 | −17.25 |
Note. GED = general equivalency diploma; LEME = likely eligible for Medicaid expansion; OEF–OIF–OND = Operation Enduring Freedom–Operation Iraqi Freedom–Operation New Dawn. All analyses were based on total weights, except relative weights (which maintain the original total sample size of 8710) were used in analyses of continuous variables (i.e., age and year first entered active duty) to preserve the standard errors.
Then, specifically comparing veterans who were uninsured and LEME with those who had health insurance coverage, we found that veterans who were uninsured and LEME were younger, had lower education and lower income, and were more likely to be Black, unmarried, and not employed; to have dependent children; and to have served in OEF–OIF–OND (d ≥ ±0.5 or Δ% > 5%). Finally, comparing both groups of veterans who were uninsured, we found that those who are LEME had lower income and were more likely to be Black, with dependent children, and unemployed than those who are not LEME.
Table 3 shows the general health characteristics and service use of veterans in these groups. The only sizable difference (d ≥ ±0.5 or Δ% > 5%) between veterans who were uninsured and those who had health insurance coverage was that veterans who were uninsured were less likely to use outpatient medical care. Comparing the 2 groups who were uninsured, we noted that those who are LEME reported poorer overall health and were more likely to need attendant care and to have used emergency department services in the past 6 months, but were less likely to report military combat trauma exposure than those who are not LEME.
TABLE 3—
General Health and Service Use of Veterans Who Are Uninsured and Likely Eligible for Medicaid in 2014 Compared with Those Who Will Not Be Eligible or Those Currently With Health Care Coverage: 2010 National Survey of Veterans, United States
| Characteristic | 1. Currently Uninsured and LEME (n = 677 450), No. (%) | 2. Currently Uninsured and Not LEME (n = 811 736), No. (%) | 3. Currently Insured (n = 18 990 014), No. (%) | Effect Size Difference Between 1 and 2, d or Δ% | Effect Size Difference Between 1 and 3, d or Δ% |
| Military combat trauma exposure | 308 667 (45.56) | 441 202 (54.35) | 8 043 013 (42.35) | −8.79 | 3.21 |
| General rating of health | 2.68 (0.89) | 3.22 (1.01) | 3.13 (1.05) | 0.57 | −0.46 |
| Difficulties with ADLs score | 12.40 (4.64) | 12.16 (3.64) | 12.23 (4.68) | 0.06 | 0.04 |
| Need attendant care | 75 636 (11.21) | 27 798 (3.48) | 1 236 726 (6.61) | 7.73 | 4.60 |
| Permanently housebound | 18 544 (2.79) | 11 947 (1.48) | 294 834 (1.57) | 1.31 | 1.22 |
| Mental health service user | 55 842 (8.24) | 39 476 (4.86) | 1 266 558 (6.67) | 3.38 | 1.57 |
| Any health service use in the past 6 mo for | |||||
| Medical or surgical care | 70 254 (10.88) | 50 208 (6.62) | 2 302 265 (12.90) | 4.26 | −2.02 |
| Outpatient medical care | 237 238 (38.64) | 323 503 (41.95) | 13 858 627 (76.79) | −3.31 | −38.15 |
| Inpatient mental health treatment | 23 411 (3.73) | 3704 (0.47) | 161 169 (0.88) | 3.26 | 2.85 |
| Outpatient mental health treatment | 32 431 (4.80) | 39 476 (5.02) | 1 184 978 (6.40) | −0.22 | −1.60 |
| Rehabilitation/nursing care facility | 0 (0.00) | 3437 (0.43) | 458 519 (2.46) | −0.43 | −2.46 |
| Emergency department | 203 518 (30.23) | 127 251 (15.87) | 2 904 179 (15.58) | 14.36 | 14.65 |
Note. ADLs = activities of daily living; LEME = likely eligible for Medicaid expansion. All analyses were based on the total weights, except relative weights (which maintain the original total sample size of 8710) were used in analyses of continuous variables (i.e., general rating of health and difficulties with ADLs score) to preserve the standard errors.
DISCUSSION
Using a population-based sample of veterans, we found that about 7%, or more than 1.5 million veterans, reported no health insurance coverage in 2010, consistent with data from 1987 to 2004 showing that 7.7% of veterans had no health insurance coverage.4 These data suggest that the proportion of veterans who lack health insurance coverage has stayed roughly the same or decreased only slightly in more than 2 decades, whereas the number of uninsured adults in the general US population has risen from roughly 13% to 16% during the same time period.11,12 Therefore, veterans now constitute only about 3% of the 50 million uninsured US adults.
Thus, the ACA’s requirement that all legal US residents have health insurance may affect a modest proportion but a substantial number of veterans. One way these uninsured veterans may obtain coverage is to enroll in the VA if they are eligible; unfortunately, we could not determine who was ineligible for VA care. Two other ways for uninsured veterans to obtain insurance coverage is through the ACA’s Medicaid expansion or the health insurance exchanges. A little more than half of all veterans who were uninsured will likely be eligible for the Medicaid expansion in states that implement it in 2014, suggesting that Medicaid may represent a new form of coverage for many uninsured veterans. The rest of those who are uninsured and not eligible for the Medicaid expansion or who are living in states that do not implement the expansion may enroll in VA coverage if they are eligible or purchase coverage through the health insurance exchanges.
Given the fact that some uninsured veterans may not qualify for federal subsidies in the health insurance exchanges (i.e., they have income > 100% of the federal poverty level), states that do not implement the Medicaid expansion (which covers those with income ≤ 138% of the federal poverty level) may have many poor, uninsured veterans who are not able to afford coverage through the health insurance exchanges. These uninsured veterans may be more inclined to consider VA coverage as an option if they are eligible, but other options may be needed to ensure affordable coverage is available to them. Also, VA coverage provides a medical benefits package that includes an array of preventive care, outpatient and inpatient services, and medication and supplies for all enrolled veterans, but veterans with extensive health needs may need even more comprehensive services than VA coverage alone can provide.
Compared with veterans with health insurance, those who were uninsured were more likely to be young, single, Black, and low income and to have served in OEF–OIF–OND; not surprisingly, given their lack of health insurance, they were also less likely to report use of outpatient medical care. This finding not only highlights the racial and socioeconomic disparities that exist in access to health care but also points to the veterans who will most likely be affected by the ACA, many of whom are recently returning veterans from the wars in the Middle East. Special outreach and attention to these uninsured veteran subgroups may be needed. It may be particularly important because some studies have shown that uninsured veterans have even poorer access to health care than other uninsured adults in the general population.7
Among veterans who were uninsured, those who will likely be eligible for the Medicaid expansion appeared to have more health needs and to be more likely to use emergency department services than those who will likely not be eligible. This finding has implications for states that decide to implement the Medicaid expansion because they may need to ramp up Medicaid-funded services to provide for the growing group who will become Medicaid eligible. It is also important to point out that eligibility does not always translate into enrollment; many who are currently eligible for Medicaid are not enrolled, so increased efforts at outreach and to facilitate Medicaid enrollment may be needed.13 As part of the ACA, the federal government will pay 100% of the coverage for individuals newly eligible for Medicaid under the expansion until 2016, which will be phased down to 90% of federal funding over time. States that do not implement the Medicaid expansion may experience greater costs and the burden of providing uncompensated care for those who are uninsured or underinsured.14,15 However, the situation around which states will and will not participate in the Medicaid expansion remains fluid16; currently, no time limit exists on participation, so some states may decide to participate later.
Because one fifth of veterans had enrolled in VA coverage, it is worth noting that about 16% of them will also likely be eligible for the Medicaid expansion and all will also be eligible to participate in the health insurance exchanges. Although what proportion of VA service users will enroll in other health coverage plans is unknown, dual or multiple enrollment may have its benefits and problems.2 Although it may increase health care access and options for veterans, it may also lead to more fragmented, lower quality care. These are important issues for the VA to plan for and possibly preempt.
Together, these findings suggest the ACA may have a considerable impact on veterans, particularly uninsured veterans. Given the provisions of the ACA, many veterans will have new health care options available to them in 2014, through either health insurance exchanges or the Medicaid expansion if they are eligible and live in states that implement the expansion. These increased health care options may also signal changes that states and the VA will experience. Some states will begin providing Medicaid-funded services for a larger group of eligible veterans, and other states that do not expand Medicaid may have to devise other ways to provide for poor, uninsured veterans. More veterans may begin enrolling in additional health coverage plans in addition to their VA coverage, and issues with coordinating care across providers may need to be considered. VA and state-level administrators should plan how they will best serve uninsured veterans who may now seek coverage under the ACA. Public health organizations and professionals can work to reach out to uninsured veterans and provide education about their coverage options.
Limitations of this study were as follows:
the cross-sectional design,
projected likelihood of Medicaid eligibility on the basis of current income,
lack of state-level data,
reliance on self-report measures including self-reported insurance coverage, and
crude measures of health status.
Future longitudinal research that tracks Medicaid enrollment as the ACA is fully implemented in each state will be important. Examination of the proportion of uninsured veterans who are eligible for VA care and the reasons they are not enrolled in VA care will also be informative. Nonetheless, the study’s weaknesses are counterbalanced by its strengths, which include use of a nationally representative sample of veterans (including non-VA service users); examination of multiple facets of health care such as insurance coverage, health status, and service utilization; and its relevance to the new era of health care the United States is entering under the ACA.
Acknowledgments
Data for the study were provided by the US Office of Policy, Planning, and Preparedness. This work was also supported by the US Department of Veterans Affairs, Veterans Health Administration, Office of Research and Development.
Note. The views presented here are those of the authors and do not necessarily represent the position of any federal agency or of the United States government.
Human Participant Protection
Use and analysis of the data were approved by the institutional review board at the VA Connecticut Healthcare System.
References
- 1. Patient Protection and Affordable Care Act, Pub. L. No. 111–148, § 3502, 124 Stat. 119, 124.
- 2.Kizer KW. Veterans and the Affordable Care Act. JAMA. 2012;307(8):789–790. doi: 10.1001/jama.2012.196. [DOI] [PubMed] [Google Scholar]
- 3.Nelson KM, Starkebaum GA, Reiber GE. Veterans using and uninsured veterans not using Veterans Affairs (VA) health care. Public Health Rep. 2007;122(1):93–100. doi: 10.1177/003335490712200113. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Himmelstein DU, Lasser KE, McCormick D, Bor DH, Boyd JW, Woolhandler S. Lack of health coverage among US veterans from 1987 to 2004. Am J Public Health. 2007;97(12):2199–2203. doi: 10.2105/AJPH.2006.106302. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Baker DW, Sudano JJ, Albert JM, Borawski EA, Dor A. Lack of health insurance and decline in overall health in late middle age. N Engl J Med. 2001;345(15):1106–1112. doi: 10.1056/NEJMsa002887. [DOI] [PubMed] [Google Scholar]
- 6.Wilper AP, Woolhandler S, Lasser KE, McCormick D, Bor DH, Himmelstein DU. Health insurance and mortality in US adults. Am J Public Health. 2009;99(12):2289–2295. doi: 10.2105/AJPH.2008.157685. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Woolhandler S, Himmelstein DU, Distajo R et al. America’s neglected veterans: 1.7 million who served have no health coverage. Int J Health Serv. 2005;35(2):313–323. doi: 10.2190/UPBQ-C3RH-D367-5H9D. [DOI] [PubMed] [Google Scholar]
- 8.Westat. National Survey of Veterans, Active Duty Service Members, Demobilized National Guard and Reserve Members, Family Members, and Surviving Spouses. Rockville, MD: Westat and US Department of Veterans Affairs; 2010. [Google Scholar]
- 9.Westat. Appendix B: National Survey of Veterans Detailed Description of Weighting Procedures. Rockville, MD: Westat and US Department of Veterans Affairs; 2010. [Google Scholar]
- 10.Haley J, Kenney GM. Uninsured Veterans and Family Members: State and National Estimates of Expanded Medicaid Eligibility under the ACA. Washington, DC: Robert Wood Johnson Foundation and Urban Institute; 2013. [Google Scholar]
- 11.Moyer ME. A revised look at the number of uninsured Americans. Health Aff (Millwood) 1989;8(2):102–110. doi: 10.1377/hlthaff.8.2.102. [DOI] [PubMed] [Google Scholar]
- 12.US Census Bureau. Health insurance. 2013 Available at: http://www.census.gov/hhes/www/hlthins/data. Accessed May 20, 2013. [Google Scholar]
- 13.Sommers BD, Epstein AM. Medicaid expansion—the soft underbelly of health care reform? N Engl J Med. 2010;363(22):2085–2087. doi: 10.1056/NEJMp1010866. [DOI] [PubMed] [Google Scholar]
- 14.Tsai J, Rosenheck RA, Culhane D, Artiga S. Medicaid expansion: chronically homeless adults will need targeted enrollment and access to a broad range of services. Health Aff (Millwood) 2013;32(9):1552–1559. doi: 10.1377/hlthaff.2013.0228. [DOI] [PubMed] [Google Scholar]
- 15.Price CC, Eibner C. For states that opt out of Medicaid expansion: 3.6 million fewer insured and $8.4 billion less in federal payments. Health Aff (Millwood) 2013;32(6):1030–1036. doi: 10.1377/hlthaff.2012.1019. [DOI] [PubMed] [Google Scholar]
- 16.Advisory Board Company. Where the states stand on Medicaid. 2013 Available at: http://www.advisory.com/Daily-Briefing/Resources/Primers/MedicaidMap. Accessed October 25, 2013. [Google Scholar]