VA-Community Dual Care: Veteran and Clinician Perspectives

Abstract

Many veterans receive care in both community settings and the VA. Recent legislation has increased veteran access to community providers, raising concerns about safety and coordination. This project aimed to understand the benefits and challenges of dual care from the perceptions of both the Veterans their clinicians. We conducted surveys and focus groups of veterans who use both VA and community care in VT and NH. We also conducted a web-based survey and a focus group involving primary care clinicians from both settings. The main measures included (1) reasons that veterans seek care in both settings; (2) problems faced by veterans and clinicians; (3) association of health status and ease of managing care with sites of primary care; and (4) association of veteran rurality with dual care experiences. The primary reasons veterans reported for using both VA and community care were (1) for convenience, (2) to access needed services, and (3) to get a second opinion. Veterans reported that community and VA providers were informed about the others’ care more than half the time. Veterans in isolated rural towns reported better overall health and ease of managing their care. VA and community primary care clinicians reported encountering systems problems with dual-care including communicating medication changes, sharing lab and imaging results, communicating with specialists, sharing discharge summaries and managing medication renewals. Both Veterans and their primary clinicians report substantial system issues in coordinating care between the VA and the community, raising the potential for significant patient safety and Veteran satisfaction concerns.

1. Background

Vermont (VT) and New Hampshire (NH) are home to more than 225,000 Veterans, with more than a third of these residing in rural or highly rural counties (Department of Veterans Affairs, 2017). Previous studies have shown that rural Veterans use more dual care (care provided in both the Department of Veterans Affairs (VA) and community)[1], are older, have worse physical functioning [2] have less continuity of care, and are more likely to be disabled. [3] The 2014 Veterans Access, Choice and Accountability Act [4] (the “VA Choice Act”) authorized Veterans to receive care by community providers if they lived a certain distance from a VA hospital (20 miles for NH residents, 40 miles for VT residents) or could not be served within 30 days. Recent experience demonstrates a positive impact [5] for Veterans (more timely access, shorter travel times, care by known community providers, etc.) even as implementation and care coordination challenges [6] continue. The 2018 “VA Maintaining Internal Systems and Strengthening Integrated Outside Networks Act”[7] (the “VA MISSION Act”) has increased both the eligibility for and scope of VA-funded care in the community.

Coordinating care across health systems can pose significant safety, patient satisfaction and operational challenges, especially when the systems represent the largest health system in the country paired with community providers, often in small rural (primary care) practices. In 2017 research showed that communication and scheduling problems with VA Choice may have delayed needed care. [8] Veterans who use both VA and Medicare services have been found to have increased rates of potentially unsafe prescribing when compared with VA-only users. [9] Data from 2017 suggests that New England Veterans are more dissatisfied with community care than Veterans nationally. [10] Both VA and non-VA providers have reported poor coordination of dual care and potential impact on patient outcomes. [6], [11]–[16] When Veterans receive care from two or more systems, they increase their risk for suboptimal care coordination[3], [17], [18] including risk of death from opioid overdose, [19] overuse of testing, [20] increased rates of potentially unsafe prescribing, [9] and increase in hospitalization for ambulatory sensitive conditions. [21], [22] And patients with increased disease severity have increased coordination challenges. [23]

2. Objective

Elucidating the patient experience and coordination elements of VA community care will help guide the design and improvement of Veteran community care policy and practice. This mixed methods exploratory study sought to better understand both Veterans’ and clinicians’ experiences when care is provided simultaneously in both VA and community settings. We hypothesized that providing dual care is more challenging for Veterans and clinicians in isolated rural communities.

3. Methods

This study was conducted in community-based and VA primary care (PC) practices, among Veterans and PC providers in both VT and NH. Veterans in VT and NH using both community and VA health care were enrolled through four community sites and six VA PC practices. Veterans experiences were studied with office-based surveys (administered between February-June 2018) and through community-based focus groups convened (April-May 2018) in partnership with the American Legion and Disabled American Veterans (DAV), Veteran service organizations not affiliated with the VA.

Providers from six community practices were recruited from the Dartmouth CO-OP Practice-based Research Network (Dartmouth CO-OP PBRN), the nation’s oldest practice-based primary care research network, and the Bi-State Primary Care Association. VA PC practitioners were recruited from community-based outpatient clinics (CBOCs) as well as medical center-based primary care practices from the White River Junction (VT) and Manchester (NH) VA Medical Centers. Provider perceptions were sampled through a web-based survey and through a focus group.

This study was approved by the Veteran’s (VA) IRB of Northern New England and the Dartmouth College’s Committee for the Protection of Human Subjects. Documentation of written informed consent was waived. Veteran participants provided oral informed consent to participate in the voluntary survey and focus group processes. Informed consent to participate in the clinician web-based survey was accomplished by an opt-out provision and voluntary, confidential participation. Clinicians provided oral informed consent to participate in the voluntary focus group.

Veteran Surveys:

During the survey weeks, for each participating primary care practice, consecutive adult patients were screened and invited to participate. At community clinics, two screening questions were asked: 1) “Have you served in the military?” and 2) “Do you use both VA and community care?” Veterans who answered “Yes” to both screening questions were invited to take the survey until the target number of 20 participants for each practice was attained. At VA practices, consecutive patients were asked: “Have you used both VA and community health care in the past 3 months?” Veterans who answered “Yes” were recruited until the target number of 20 participants for each practice was attained. If eligible at either site, the Veteran was invited to complete a 16-question survey on use of VA and community services, benefits, challenges and the reason for dual use (survey available on request). Nine questions elicited information about reasons for using VA and community care, who manages most of the care, whether providers were informed and up-to-date about care, the kind of information shared, communication with specialists, and the benefits and challenges of using dual care. Four questions elicited demographic information. Three questions used previously validated items on 1) how hard or easy it was to manage their health care (0–10, Hard/Easy), 2) rating of health care quality (0–10, Worst Care Possible/Best Care Possible), and 3) overall health rating (1–5, Excellent/VG/Good/Fair/Poor).[24], [25] The only protected health information (PHI) collected from participants were age and zip code.

Veteran Focus Groups:

Two focus groups (n = 11 participants) were conducted, one in VT and one NH. The American Legion and DAV invited Veterans for the purpose of giving feedback about the benefits and challenges of using VA community care and coordinating care between the VA and community providers. No PHI was recorded.

Provider Surveys:

Primary care providers were surveyed via electronic mail utilizing REDCap [26] (survey available on request). Participation was voluntary and confidential and the only personally identifiable information (PII) collected was zip code. After one week, non-responders were emailed again and invited to complete survey.

Provider Focus Group:

Participants at the annual meeting of the Dartmouth COOP PBRN were invited to participate in a focus group (n=8) to elicit feedback about Veteran dual use. Participation was voluntary and no PHI was collected.

4. Main Measures

Health perception was dichotomized into Excellent/Very Good/Good vs Fair/Poor. To assess impact of rurality we used 2010 Rural Urban Community Area (RUCA) codes [27] dichotomized as Isolated Rural (category 10) and all else (categories 1–9). Initially, descriptive statistics and frequencies were evaluated. Categorical variables were analyzed using Chi Square and ANOVA. Student’s t-test was used for continuous variables. Survey data was collected in REDCap and analyzed in SPSS (IBM version 25).

Veteran free text comments elicited in the survey, and Veteran and clinician focus group notes were reviewed by a trained research associate and a principal investigator and consensus was reached on the key themes identified.

5. Results

Veteran Survey

One hundred eighty-seven Veterans completed the survey—81 from Vermont, 95 from New Hampshire, 4 from Maine, 1 from NY and 6 without location specified (see Table 1 for Veteran demographics). While all Veterans received care in both VA and community sites, 32.3% reported that most care was provided in the community, 43% reported that most care was provided in the VA, and 24.7% reported that care was balanced between VA and community.

Table 1:

Participant Demographics

Veteran Demographics
		N	%
Gender	Male	169	90.4%
	Female	17	9.1%
Ethnicity	Hispanic	3	1.6%
	Non-hispanic	184	98.4%
Race	White	169	96.6%
	Other	6	3.4%
Age	Over 64 yr.	71	38.0%
Rurality
Isolated Rural		29	15.5%
Metropolitan, suburban or small towns		152	81.2%
Clinician Demographics
		Community Clinicians		VA Clinicians		Total
		N	%	N	%	N
Specialty	Family Practice	15	75.0%	6	31.6%	21
	Internal Medicine	5	25.0%	13	68.4%	18
Degree	MD/DO	17	85.0%	12	63.2%	29
	ARNP/PA	3	15.0%	7	36.8%	10
Gender	Female	13	65.0%	8	42.1%	21
	Male	7	35.0%	11	57.9%	18
		Year		Year
Year finished training	Median	1993		1993
	Range	1980–2016		1977–2013

Veterans provided multiple reasons for using dual care including convenience (43.3%), striving to get needed services (33.5%), preparing in case care might be needed in the future (22%), and to get a second opinion (22%). Additional reasons included cost, reduced travel time, access to specialized services (e.g. cardiac stress test, PTSD), and appointment timeliness.

About half the time Veterans reported both that community providers were informed of their VA care (56.7% Always or Very Often) and that VA providers were informed of their community care (55.1% Always or Very Often). Information sharing for both VA and community providers frequently involved labs, x-rays, medication information and office notes. A significant number of Veterans with specialist providers responded No or Unsure whether their community (39.5%) or VA provider (48.5%) communicated with the specialist(s).

Overall health was rated Fair/Poor by 27% of Veterans. Veterans who reported having the VA manage most of their care were more likely to report Fair/Poor health (42.5%) than those whose care was managed mostly by community providers (13.3%) or whose care was fairly balanced between VA and community (19.6%) (see Table 2, p<.001, Chi square). Veterans found it fairly easy to manage their health care (100-point scale, mean 78, median 84) and rated the quality of their healthcare highly (100-point scale, mean 81.6, median 84). These ratings did not differ by where most of care was managed.

Table 2.

Health Status by Site of Management of Most Care (n=186)

Site of Most Care
Health Status*	Community practice	Balanced Community and VA	VA practice	P Value
Fair/Poor	8/60 (13.3%)	9/46 (19.6%)	34/80 (42.5%)	<.001

^*Likert scale 1–5 (Excellent-Poor)

Care of veterans from isolated rural communities

Veterans who live in isolated rural areas were not different than non-rural veterans by most measures: health status, who primarily managed care, rating of health care quality, and communication between VA and community providers. Although rural veterans perceive that community PC providers communicate with specialists at a rate equal to non-rural veterans, they report that their VA PC providers communicate with specialists more frequently than non-rural veterans (65.5% vs 44.9%, p=.043, Chi square). They also report that their health care was easier to manage than non-rural veterans (mean rating 86.9 vs. 75.5, p<.01, T-test).

Veteran Qualitative Findings

Key themes of the qualitative comments from the Veteran survey and the Veteran Focus Groups are in Table 3. Mentioned often as advantages of the VA were cost and range of services. Frequently cited as advantages of Community Care were convenience and continuity. Quality was perceived by some veterans as better at the VA and by others as better in the community. Many veterans commented that challenges of duel care included communication between clinicians at the two sites and the logistics of eligibility and payment.

Table 3:

Themes from Veteran Survey Comments and Focus Groups

Category	Themes
Benefits of Community Care	• Being part of a community and known environment • Local doctors are better known • Perception that local doctors are better than at the VA • Long-term relationships with primary care and specialist physicians in the community are valuable • Travel is shorter and less expensive • Easier for veteran family members to accompany
Benefits of VA Care	• Unique clinical Services (e.g. PTSD, hearing) • Perception that VA doctors are better than in the Community • Medical depth and willingness to help • Services are less expensive • Extensive programs social services and special
Problems with Community Care	• More expensive • Staff may not understand veterans’ unique experiences and problems
Problems with VA Care	• Travel to the VA can be difficult • Establishing eligibility can be difficult
Hassles of Dual Care	• Billing, authorization and paperwork are complex and not well-understood at either end • Communication between community and VA providers is quite variable • Medications recommended by community doctors are not always available on the VA formulary

Clinician Surveys

Of 195 primary care clinicians, 39 (20 community, 19 VA) completed surveys (response rate 20%). Clinician demographics are summarized in Table 1. Fifteen percent of community clinicians reported being enrolled in the VA Choice program and 30% routinely ask patients if they are Veterans. VA clinicians reported that they always ask Veteran patients whether they received dual care (100%) while community clinicians ask less frequently (42.1%). Among VA clinicians 27.8% reported they Always/Very often received notification of care from community providers, but only 5% of community clinicians reported Always/Very Often receiving notification of VA care. VA clinicians reported frequently sending information (44.4% Always/Very Often) but community clinicians reported sending information less often (25%). While information was shared in various ways (mail (17%), email (5.7%), fax (45.3%), phone (7.5%)), clinicians reported it was often (24.5%) the Veteran who transferred the documents. Most (51.6%) preferred to receive information via fax. Both VA and community clinician reported encountering many systems problems with dual care, including changes on medications, sharing lab and imaging results, communicating with specialists, discharge summaries and medication renewals (see Figure).

Figure:

Frequency of Clinician Reports of Systems Problems with Dual Care

Clinician Qualitative Findings

Clinicians reported in the focus group that having two clinicians can sometimes be better than one and that veterans may benefit from the best elements of both VA and community systems. However, a number of challenges were identified. Community clinicians often had little idea of the number of Veterans in their practice. They reported communications were often not clear, with confusion about which clinicians are taking care of particular Veteran problems— “information-wise it’s a black hole in space.” They often did not have the time to go through a scanned medical record to find old procedures or information. The Veteran may get medications from one place and get admitted someplace else. Their perception was that the VA is integrated with itself but “Balkanized” with respect to outside systems.

6. Conclusions

Our study found that primary care clinicians from both VA and community systems reported often not sharing important clinical information, and Veterans corroborated this finding. We found that Veterans primarily managed by the VA were sicker than patients primarily managed in the community. Veterans reported multiple issues with dual care, including communication gaps, getting care authorized, and travel. Both VA and community providers were informed of dual care only about half the time and reported many system problems including changes on medications, sharing lab and imaging results, communicating with specialist, discharge summaries and medication renewals. We did not find health status or care issues to differ significantly for Veterans in highly rural communities.

These findings validate critical patient safety concerns of care coordination in an increasingly complex dual care system. This study adds the perspective of rural Veterans and clinicians to a small, but growing literature on the challenges of providing safe, well-coordinated care in complex care systems.

Limitations of this study include modest sample size, limited female Veteran participation and limitation to just two states. Our results were collected in 2017 and reflect the collaboration systems available at that time. Care coordination policies and technologies associated with the VA MISSION Act implementation in June of 2019 may address some of the challenges we identified. While our data clearly demonstrate multiple care coordination issues from both the Veteran and clinician perspectives, systematic measurement of care integration using robust conceptual models with validated surveys over time would allow for more precise delineation of the most important elements of system design and its potential improvement. [28]–[32]

7. Discussion

Early assessment of the VA Choice program has revealed mixed results. Nationally, the top two drivers of overall Veteran satisfaction with community care were community care providers being up-to-date on VA care and VA providers being up-to-date on community care.[33] Access has improved for some Veterans but they also experience frustration, increased complexity, care fragmentation and poor coordination.[5], [34] The VA has consistently demonstrated care quality at least as good, if not better in some areas than private health care systems. [35]–[38] Veteran use of community care is projected to continue to increase. [39] This will likely increase costs and decrease quality of care. [40] At the same time, more effectively integrated care is associated with reduced emergency department visits, lower outpatient visit rates, [41] and improved patient care perceptions. [42]

Northern New England veterans and their clinicians – both VA and community - report substantial system issues in coordinating care between the VA and the community, raising the potential for significant patient safety and Veteran satisfaction concerns. In this exploratory study, isolated rural location was not a significant predictor of care coordination issues. While further research may facilitate finer delineation of the drivers of safety and satisfaction, much is already known about key elements of effective care coordination. As community care for Veterans continues to grow, effective system redesign and improvement, and concerted collaboration at all levels between VA and community care systems will be necessary to ensure the safe, high quality care all Veterans deserve.