Over 2 million service members have served in Iraq and Afghanistan in Operations Enduring Freedom, Iraqi Freedom, and New Dawn (OEF/OIF/OND). Thanks to improvements in trauma care, combat survival has improved dramatically since World War II, from 70% to 97%.1 But, when combined with long and frequent deployments,2 it is not surprising that OEF/OIF/OND veterans have higher rates of disability and medical needs than prior generations of veterans.3 Many return home with physical and mental health conditions requiring chronic care,4–6 including posttraumatic stress disorder (6%–11% met screening criteria for further evaluation),7,8 substance use disorders or unhealthy substance use such as frequent binge drinking (between 6%–8% and 16%–25%, respectively),8,9 depression (between 8%–21% met screening criteria for further evaluation),3,5 and traumatic brain injury (14%).8
The Veterans Health Administration (VHA) is often an attractive option for OEF/OIF/OND veterans, given broad eligibility criteria (Public Law 110-181), expertise in rehabilitation and mental health care, and low (if any) copayments. In total, 61% of OEF/OIF/OND veterans have enrolled in and utilized VHA,10 a high rate compared with other veteran groups. Historically, the VHA focused on providing care at its own facilities across the nation. Since 1957,11 VHA-enrolled Veterans have been able to access care in the community through a small purchased care program (“traditional Fee”), wherein VHA purchases private sector care for eligible veterans. Under traditional Fee, a veteran is eligible to use a community provider if VHA services are not feasibly available12 due to geographic inaccessibility or if VHA is unable to provide the service required.13 Care provided through traditional Fee must be preauthorized, except for situations in which preauthorized care is infeasible, such as emergency care. Consequently, traditional Fee was often used for medical specialty, emergency, and custodial care; primary care was rarely purchased. The demand for traditional Fee care has increased over time,14 accounting for ∼11% of the VHA budget in 2014.15
In 2014, after concerns surfaced about long delays and barriers to care at some VHA facilities, Congress enacted the Veterans Access, Choice and Accountability Act (VACAA; Section 101 of Public Law 113-146). As part of the law, Congress allocated an additional $10 billion to enable VHA to purchase more care in the community in fiscal years (FY) 2015–2017. As a point of reference, VHA’s authorized budget was ∼$59 billion in FY2014.16 The new purchased care Program has become known as the Veterans Choice program (VCP). Starting in FY2015, eligible VHA enrollees were able to receive care from non-VHA providers through VCP. Eligibility requirements for services included waiting >30 days from the clinically indicated date for an appointment, living over 40 miles driving distance from the closest VHA, or experiencing hardship (living 40 miles or less from the nearest VHA, but only being able to access that VHA by boat or plane, or facing other excessive travel burdens, such as heavy traffic, geographic barriers, hazardous weather, or medical challenges).17 VCP covers all services except emergency and long-term care.18 Eligible veterans must opt into VCP to receive care through VCP. Veterans who qualify for VCP due to wait-time are only authorized to receive services in the community related to the specific care for which they are waiting. Those who qualify for VCP due to mileage and/or hardship are permitted to access all of their care in the community.
Although VCP provides veterans with more health care options, there are concerns that it could also lead to care fragmentation (having providers both inside and outside VHA). Care fragmentation across delivery systems and providers can lead to communication and coordination errors from poor information sharing.19 Such errors can have meaningful consequences, such as provision of contraindicated care or lack of follow-up care, which can lead to reductions in quality of care and poorer health outcomes.20–26
As a first step in understanding the impact of VCP on OEF/OIF/OND veterans, the goal of our study was to determine pre-VCP and post-VCP use of outpatient care by veterans eligible for VCP. We used standardized methods to categorize outpatient care received at VHA facilities and developed a method to categorize care received outside of VHA through the traditional Fee program and through VCP. Overall, we hypothesized that there would be substitution of outpatient care, specifically a reduction in VHA care after VCP implementation. Furthermore, given greater access to community care for veterans who qualify for VCP due to mileage/hardship versus wait-time, we hypothesized that there would be more VCP use and a greater reduction in VHA and traditional Fee use by the mileage/hardship group than for the wait-time group.
Our cohort included all persons on the OEF/OIF/OND Roster as of August 2015 who were eligible for VCP in FY2015 (n=214,449). VCP eligibility type (mileage, hardship, and/or wait-time) was determined from the VACAA tables in VA’s Corporate Data Warehouse. We excluded data for some veterans eligible for VCP (n=849) because they became ineligible for VCP during the study period, and it was not possible at the time to determine how they had qualified for VCP (ie, mileage/hardship or wait-time). The Palo Alto VA Research Office determined that this project did not meet the Federal definition of research, and therefore did not require review by the Stanford University Institutional Review Board.
Data and Measures
We categorized our cohort into 2 subgroups—“VCP eligibles” and “VCP users.” VCP eligibles include those OEF/OIF/OND veterans who: (1) were enrolled in VHA and were eligible for the VCP program; and (2) received care through any of the VHA, traditional Fee, or VCP channels, regardless of whether they used VCP care or not. VCP users is a subset of VCP eligibles, which only includes OEF/OIF/OND veterans who actually accessed VCP care. We stratified the 2 subgroups by their VCP eligibility type in FY2015. Only veterans who utilized care were included because we could not tell from claims data if 0 utilization represented a veteran who needed care and did not use it or a veteran who did not need the care in the first place.
We examined VCP eligibles’ and VCP users’ utilization of VHA and traditional Fee care in FY2012–2015 and their utilization of VCP care in FY2015. For both VCP eligibles and VCP users, when veterans enrolled in VHA mid-FY, we assumed in our calculations that their utilization pattern would have been the same for the full year. However, if the veteran died mid-FY, we inputted 0 utilization after death. We examined the utilization behavior of both VCP eligibles and VCP users because we wanted to see if utilization behavior in general was changing after VCP implementation for those eligible for VCP, as well as if utilization behavior was specifically changing for those who chose to use VCP services. Just being eligible for VCP could have had an effect on utilization behavior in FY2015 (eg, shying away from VHA services because of bad media attention)—and this effect could be more or less present than for those who chose to use VCP services. For traditional Fee and VCP, since we examined claims data until October 2015, we likely did not capture all services provided in FY2015 as there can be a lag in when claims are submitted. However, we should have captured the vast majority of services provided as ∼90% of claims for purchased care are processed within 30 days of receipt.27
We categorized type of outpatient care for VHA, traditional Fee, and VCP. For traditional Fee and VCP outpatient care, we collapsed the “Category of Care” variable in the Fee data into the following groups: alternative medicine, compensation and pension, dental, dialysis, emergency/urgent, end of life, Fee ID card (renewals or updates to existing Fee cards), home health, laboratory, medical specialty care, mental health, other, pharmacy (VCP requires that prescriptions are filled in VHA pharmacies, with few exceptions), primary care (not typically available through purchased care until August 2014), radiology, rehabilitation, respite, skilled nursing facility, telephone and telehealth, or unknown. For VHA outpatient care, we used VHA clinic codes (primary stop codes) to inform our categorization of VHA care.
We also evaluated veterans’ characteristics to compare VCP users to VCP nonusers. We collected this information from the OEF/OIF/OND Roster and VHA utilization data. Demographic information included age, sex, race/ethnicity, education level, and marital status. We used branch of service, component (Active Duty, National Guard, Reserve), rank, and deployment status (never, once, multiple times) to identify military service characteristics. Finally, for health care access characteristics, we tracked VHA priority level (which indicates enrollment priority into VHA based on service-connected disabilities and aspects of socioeconomic status), insurance status (eg, major medical), rurality (eg, rural), and driving distance to nearest VHA facility.
To assess changes in VHA and traditional Fee outpatient utilization from FY2012–2015, we examined per capita utilization for VCP eligibles and VCP users. We used Wilcoxon rank-sum tests to compare mean utilization between FYs.
We also compared characteristics of VCP eligibles who used any type of VCP care to VCP eligibles who did not use any VCP care. We used t tests and χ2 tests to compare VCP users (n=1061) versus nonusers (n=80,190) who qualified for VCP due to mileage/hardship as well as VCP users (n=2760) versus nonusers (n=130,438) who qualified for VCP due to wait-time.
Outpatient Utilization for VCP Eligibles
In Table 1, we present the number of VCP eligibles in FY2015 (n=214,449) who were enrolled in VHA in previous years stratified by their VCP eligibility type; these numbers increased over time as more veterans enrolled in VHA. There was limited use of VCP care by VCP eligibles in FY2015. The number of VCP users and visits increased in FY2015, with the highest number recorded in July (1078 unique VCP users and 2332 VCP outpatient visits). Across all months in FY2015, there were 3821 unique VCP users who qualified due to wait-time or mileage/hardship. There was no VCP use in FY2015 for dental, dialysis, end of life, Fee ID card, home health, laboratory, pharmacy, respite, skilled nursing facility, or telephone and telehealth care, and minimal use (number of unique VCP users) of compensation and pension (n=54), unknown (n=8), emergency (n=3), and other (n=1).
The top 6 groups of care (by number of unique VCP users) were medical specialty care (n=1848), rehabilitation (n=758), primary care (n=555), alternative medicine (n=478), mental health (n=217), and radiology (n=123). We focused our utilization analyses on these 6 groups of care. In Table 2, we present pre/post-VCP outpatient utilization for OEF/OIF/OND veterans eligible for VCP who used these 6 categories of care. In all of these groups of care, the lowest VHA utilization per user occurred in FY2015, after VCP implementation, compared with before VCP implementation (FY2012–2014). We display this in Figure 1, which illustrates VHA utilization over time, normalized to the average of the pre-VCP years. Thus, a value >1 indicates that utilization for that year is greater than utilization for the pre-VCP years on average, and a value <1 indicates that utilization for that year is less than utilization for the pre-VCP years on average. For example, a value of 0.90 means that that year’s VHA utilization was 90% as much as the pre-VCP VHA average utilization. Of note, the decline in VHA visits per user from FY2014 to FY2015—after VCP implementation—was statistically significant in all 6 categories and for both VCP eligibility types (Table 2). There was less of a clear pattern over time for traditional Fee care (Table 2 and Figure 1, Supplemental Digital Content 1, http://links.lww.com/MLR/B331 which displays Fee utilization over time, normalized to the average of the pre-VCP years for VCP eligibles). FY2015 per capita VCP utilization (mean number of visits) was higher for the mileage/hardship group than for the wait-time group for mental health (4.5 vs. 3.6), primary care (1.4 vs. 1.3), and rehabilitation (6.0 vs. 5.8); lower for alternative medicine (5.0 vs. 6.1); and equivalent for medical specialty care (1.5) and radiology (1.1) (Table 2).
Outpatient Utilization for VCP Users
For comparison, we calculated outpatient utilization before and after VCP implementation for OEF/OIF/OND veterans who utilized each type of VCP outpatient care in FY2015. Compared with all VCP eligibles who used any services (Table 2), per capita VHA utilization in FY2015 for VCP users (Table 3) was lower for primary care, mental health, and rehabilitation; VHA utilization was higher for medical specialty care and radiology for those in the wait-time category, and equivalent for alternative medicine (no use) and radiology for those in the mileage/hardship category. Unlike all VCP eligibles, per capita VHA utilization was not necessarily lowest in these categories after VCP implementation in FY2015; it was lower in all cases except for radiology, wait-time medical specialty care, and mileage/hardship rehabilitation. We display this in Figure 2, which illustrates VHA utilization over time, normalized to the average of the pre-VCP years for VCP users. The decline in VHA visits per user from FY2014 to FY2015—after VCP implementation—was statistically significant for the medical specialty care mileage/hardship, both primary care, and both mental health groups (Table 3). Traditional Fee utilization was minimal and did not have a distinct increase or decrease pattern pre/post VCP implementation (Table 3 and Figure 2, Supplemental Digital Content 2, http://links.lww.com/MLR/B332 which displays Fee utilization over time, normalized to the average of the pre-VCP years for VCP users).
Comparing Characteristics of VCP Eligibles Categorized by VCP Usage
For both the wait-time and mileage/hardship groups, there were several statistically significant differences between VCP eligible veterans who used and did not use VCP services (Table 1, Supplemental Digital Content 3, http://links.lww.com/MLR/B336). In characteristics that showed differences for both the wait-time and mileage/hardship groups, VCP users were slightly older on average, had a modestly different racial/ethnic makeup, included more Active Duty versus Guard or Reserve service members, included more veterans with the highest priority level 1, and had different insurance status (more uninsured in the mileage/hardship group, and slightly more insured in the wait-time group) than nonusers. Other characteristics displayed statistically significant differences between VCP users and nonusers in only the mileage/hardship or wait-time group. For the mileage/hardship group, VCP users were more likely to be married, live in rural areas, and have a longer driving distance to a VHA facility. For the wait-time group, VCP users had higher education levels and had a different breakdown by branch of service (more in the Navy and less in the Army).
As VCP is a new program, the literature thus far has focused on the potential impact of this program.28–31 This is the first study to assess utilization before and after VCP implementation. We focused on recent veterans as they are of particular policy relevance and might have less established care patterns. We find that the program’s reach generally increased over time for OEF/OIF/OND veterans; specifically, the number of unique VCP users and visits generally increased monthly over the study period. Despite this trend toward increased uptake of the program, there was still low utilization – with only 3821 of the potential 214,449 (2%) OEF/OIF/OND veterans eligible for VCP through mileage/hardship and wait-time criteria using VCP. Low utilization thus far could be reflective of many things. First, VHA access could have been relatively adequate before VCP, and there might not have been much demand for care outside VHA. Second, veterans may prefer care in VHA, and decline to use services in the community even when they are eligible. Third, it may be easier to use another source of payment for care received outside VHA, such as public or private health insurance, and services may be rendered through those programs in lieu of VCP. Fourth, veterans may still be learning about the program and thus will be more likely to use VCP in the future. Fifth, imperfect program implementation (eg, between VHA and third party administrators) might have made it difficult for veterans to access care.
There was evidence that VCP implementation might have had an impact on VHA utilization amongst OEF/OIF/OND veterans. The top 6 groups of care (by number of unique VCP users) were medical specialty care, rehabilitation, primary care, alternative medicine, mental health, and radiology. In our cohort of VCP eligibles, per capita utilization of VHA care in those 6 groups of care was lowest after VCP implementation (FY2015) compared with before VCP implementation (FY2012–2014)—suggesting that VCP could have had an effect on VHA utilization. This pattern held for 8 of the 12 cases when examining the subset of OEF/OIF/OND veterans who utilized those particular types of VCP services. Together, these results suggest that the existence of VCP might have impacted both VCP eligibles’ and VCP users’ VHA utilization. There were no clear pre/post VCP utilization patterns for traditional Fee care. The decreases seen in VHA medical specialty care, rehabilitation, primary care, alternative medicine, and radiology utilization might be explained by their availability in the community. The decrease in mental health VHA utilization deserves additional attention, as it may be only a subset of VCP eligibles receiving care in the community, given VHA’s expertise in mental health. For example, veterans with depression might be more likely to seek VCP care, while those with posttraumatic stress disorder may continue to receive care in VHA.
Some of the differences found between those who used and did not use VCP care make intuitive sense and others require further investigation. For example, it is logical that mileage/hardship VCP users might be more likely to seek VCP care when they lack easy access to a VHA facility (ie, live in a rural area, have a longer driving distance to a VHA facility), but there is not an apparent reason for seeing differences by branch of service for wait-time users. The characteristics of VCP users and nonusers are important to note to understand these veteran populations, and also to inform outreach for VCP.
This analysis had some data limitations worth noting. First, we only had utilization data from the first year of VCP; data from the second year of the program is not yet available. As uptake of the program increases, future work can examine the longitudinal impact over multiple years of data, illuminating a more robust understanding of the impact of VCP. Second, it was unclear when a veteran enrolls in VCP. At the time of this study, available data only tracked VHA enrollment and did not record the date when a veteran switched between VCP eligibility categories. Future evaluations can coordinate with VHA operations partners to receive these data.
Although these results are preliminary, they are timely as VHA’s role as a payer for care in the community is increasing and consolidation of the traditional Fee programs and VCP is planned (Public Law 114-41). With an expanded care in the community program, future research should examine care coordination and duplication challenges that might arise with the public-private partnership.
It is interesting to note that, just as VHA is expanding its network to include access to private providers, systems outside VHA are working to become more integrated through mechanisms such as accountable care organizations (ACOs). ACOs were a part of the Federal Affordable Care Act, and are defined by the Centers for Medicare and Medicaid Services as “groups of doctors, hospitals, and other health care providers, who come together voluntarily to give coordinated high quality care.”32 For veterans who use VHA, it is possible that care will become more fragmented if they also use VCP, whereas care for veterans in systems outside VHA might become less fragmented due to the establishment of ACOs and other efforts to coordinate care. Naturally, the VHA system and ACOs face different challenges. VCP encourages the use of services outside VHA for veterans with need or access deficits, whereas ACOs attempt to keep utilization within their system. Nonetheless, predictors of outside use in VCP may be of interest to ACOs.
As of yet, there has been low VCP utilization and it is difficult to understand what its long-term impact will be. If VCP utilization increases in the future, it is unclear whether improved access will be matched with better quality and reduced cost, vice versa, or some combination; we suspect this will likely vary based on type of care received. For example, there is evidence that mental health care provided in VHA is more culturally competent with respect to military issues than care provided in the community,33,34 and often of higher quality (eg, for depression) than care provided in the community.35,36 Thus, the value of mental health care in VHA versus other systems is likely higher, while it could be lower for other services that do not require military cultural competence (eg, dialysis). We do know that ACOs, which have been trying to decrease care fragmentation, have experienced mixed results with respect to improving quality and decreasing costs.37–50 What is clear in both the case of VCP and with ACOs is that accountability for care is a major issue, and serious efforts need to be made to improve the care experience for patients. This will become a greater concern if uptake of VCP increases over time.
1. General Accountability Office. VA and DOD health care: efforts to provide seamless transition of care for OEF and OIF service members and veterans
2006. Available at: www.gao.gov/assets/100/94286.pdf
. Accessed February 19, 2016.
3. Auto DH, Duggan M, Greenberg K, et al. The impact of disability benefits on labor supply: evidence from the VA’s disability compensation program (National Bureau of Economic Research). 2015. Available at: www.nber.org/papers/w21144.pdf
. Accessed February 19, 2016.
4. National Research Council. Returning home from Iraq and Afghanistan: assessment of readjustment needs of veterans
, service members, and their families. 2013. Available at: www.nap.edu/read/13499/chapter/1
. Accessed February 22, 2016.
6. Hoge CW, Auchterlonie JL, Milliken CS. Mental health problems, use of mental health services, and attrition from military service after returning from deployment to Iraq or Afghanistan. JAMA. 2006;295:1023–1032.
8. Larson MJ, Adams RS, Mohr BA, et al. Rationale and methods of the Substance Use and Psychological Injury Combat Study (SUPIC): a longitudinal study of Army service members returning from deployment in FY 2008-2011. Subst Use Misuse. 2013;48:863–879.
9. Adams RS, Larson MJ, Corrigan JD, et al. Frequent binge drinking after combat-acquired traumatic brain injury among active duty military personnel with a past year combat deployment. J Head Trauma Rehabil. 2012;27:349–360.
10. Epidemiology Program, Post-Deployment Health Group, Office of Public Health, Veterans
Health Administration, Department of Veterans
Affairs. Analysis of VA health hare utilization among Operation Enduring Freedom, Operation Iraqi Freedom, and Operation New Dawn veterans
, from 1st qtr FY 2002 through 1st qtr 2015. 2015. Available at: www.publichealth.va.gov/docs/epidemiology/healthcare-utilization-report-fy2015-qtr2.pdf
. Accessed February 22, 2016.
12. US Department of Veterans
Health Administration. Non-VA Medical Care Program (VHA Directive 1601). 2013.
13. Hospital care and medical services in non-VA facilities. 38 CFR 17.52. 2014.
19. Kizer KW. Veterans
and the affordable care act. JAMA. 2012;307:789–790.
21. Gellad WF, Zhao X, Thorpe CT, et al. Dual use of Department of Veterans
Affairs and Medicare benefits and use of test strips in veterans
with type 2 diabetes mellitus. JAMA Intern Med. 2015;175:26–34.
22. Helmer D, Sambamoorthi U, Shen Y, et al. Opting out of an integrated healthcare system: dual-system use is associated with poorer glycemic control in veterans
with diabetes. Prim Care Diabetes. 2008;2:73–80.
23. Wolinsky FD, Miller TR, An H, et al. Dual use of Medicare and the Veterans
Health Administration: are there adverse health outcomes? BMC Health Serv Res. 2006;6:1–11.
24. Jia H, Zheng Y, Reker DM, et al. Multiple system utilization and mortality for veterans
with stroke. Stroke. 2007;38:355–360.
25. Hester EJ, Cook DJ, Robbins LJ. The VA and Medicare HMOs—complementary or redundant? N Engl J Med. 2005;353:1302–1303.
26. Pizer SD, Gardner JA. Is fragmented financing bad for your health? Inquiry. 2011;48:109–122.
28. Dixon BE, Haggstrom DA, Weiner M. Implications for informatics given expanding access
to care for Veterans
and other populations. J Am Med Inform Assoc. 2015;0:1–4.
29. Gellad WF. The Veterans Choice
Act and dual health system use. J Gen Intern Med. 2016;31:153–154.
30. Giroir BP, Wilensky GR. Reforming the Veterans
Health Administration—beyond palliation of symptoms. N Engl J Med. 2015;373:1693–1695.
33. Burnam MA, Meredith LS, Taneilian T, et al. Mental health care for Iraq and Afghanistan war Veterans
. Health Aff. 2009;28:771–782.
35. Asch SM, McGlynn EA, Hogan MM, et al. Comparison of quality of care for patients in the Veterans
Health Administration and patients in a national sample. Ann Intern Med. 2004;141:938–945.
36. Kimerling R, Pavao J, Green L, et al. Access
to mental health care among women veterans
: is VA meeting women’s needs? Med Care. 2015;53:S97–S104.
37. McWilliams JM, Chernew ME, Landon BE, et al. Performance differences in year 1 of pioneer accountable care organizations. N Engl J Med. 2015;372:1927–1936.
39. Christensen EW, Payne NR. Effect of attribution length on the use and cost of health care for a pediatric Medicaid accountable care organization. JAMA Pediatr. 2016;170:148–154.
40. Claffey TF, Agostini JV, Collet EN, et al. Payer-provider collaboration in accountable care reduced use and improved quality in Maine managed care plan. Health Aff. 2012;31:2074–2083.
41. Colla CH, Goodney PP, Lewis VA, et al. Implementation of a pilot ACO payment model and the use of discretionary and non-discretionary cardiovascular care. Circulation. 2014;130:1954–1961.
42. Colla CH, Wennberg DE, Meara E, et al. Spending differences associated with the Medicare physician group practice demonstration. JAMA. 2012;308:1015–1023.
43. Hacker K, Santos P, Thompson D, et al. Early experience of a safety net provider reorganizing into an accountable care organization. J Health Polit Policy Law. 2014;39:901–917.
44. Kelleher KJ, Cooper J, Deans K, et al. Cost saving and quality of care in a pediatric accountable care organization. Pediatrics. 2015;135:e582–e589.
45. McWilliams JM, Landon BE, Chernew ME. Changes in health care spending and quality for Medicare beneficiaries associated with a commercial ACO contract. JAMA. 2013;301:829–836.
46. McWilliams JM, Chernew ME, Dalton JE, et al. Outpatient care patterns and organizational accountability in Medicare. JAMA Intern Med. 204;174:938–945.
47. Nyweide DJ, Lee W, Cuerdon TT, et al. Association of pioneer accountable care organizations vs traditional Medicare fee for service with spending, utilization, and patient experience. JAMA. 2015;313:2152–2161.
48. Sandberg SF, Erikson C, Owen R, et al. Hennepin health: a safety-net accountable care organization for the expanded Medicaid population. Health Aff. 2014;33:1975–1984.
49. Schwartz AL, Chernew ME, Landon BE, et al. Changes in low-value services in year 1 of the Medicare pioneer accountable care organization program. JAMA Intern Med. 2015;175:1815–1825.
50. Singh S, Khosla S, Sethi A. Comparison of healthcare quality outcomes between accountable care organizations and physician group practices. J Med Pract Manage. 2015;30:261–264.
access; veterans; choice
Supplemental Digital Content
Copyright © 2017 Wolters Kluwer Health, Inc. All rights reserved.