Veteran Perspectives on Care Coordination Between Veterans Affairs and Community Providers: A Qualitative Analysis

Abstract

Objective:

To investigate veteran perspectives on challenges in care coordination between US Department of Veterans Affairs (VA) clinics and community providers in rural areas.

Methods:

We completed qualitative interviews with a geographically diverse sample of 51 veterans who had used both VA and community health care services. Interviews were audio-recorded and transcribed verbatim. We used directed content analysis (informed by previous work with VA and community staff) to elucidate findings, while remaining attentive to emergent themes.

Results:

We report results in 5 key domains related to interorganizational care coordination: organizational mechanisms; organizational culture; relational practices; contextual factors; and the role of the Third Party Administrators responsible for scheduling and payment for community services. Veterans described successes and challenges in interorganizational coordination across these domains, while also reporting a variety of workarounds and mitigation strategies.

Conclusions:

Veterans living in rural areas face myriad challenges when using health care services both within and outside of VA. In the absence of strong mechanisms for ensuring coordination and communication between health care providers at different institutions, veterans themselves may carry the primary burden for coordinating their care. Our results suggest the utility of both structural and relational approaches to enhancing interorganizational care coordination in these settings.

The US Department of Veterans Affairs (VA) is the largest integrated health care system in the country, with over 1,000 clinic locations nationally. Across many quality measures, VA delivers care that equals or exceeds the private sector,¹ but it has recently struggled with issues related to timeliness and access to care.² VA has developed several initiatives to meet these needs, including building capacity to deliver services via telehealth^3–7 and facilitating veterans’ receipt of health care outside of VA.^8–10

Veterans’ use of community health care services has grown rapidly following the Veterans Choice Act of 2014 and the VA MISSION Act of 2018.^11–13 These laws were intended to make it easier for veterans—especially those in rural areas^14,15—to receive care in the community. Under the Veterans Choice Act, scheduling and payment for these services was initially to be managed by 2 Third Party Administrators (TPAs), Health Net and TriWest.^8–10

Veterans’ simultaneous use of VA and community health care services can create commensurate problems with interorganizational care coordination—that is, the coordination of care among providers at different institutions.¹⁶ Deficits in care coordination can lead to duplicated, delayed, or contradictory medical services—or to patients falling out of care entirely.^17–19 To prevent these poor outcomes, VA has undertaken recent initiatives to improve interorganizational care coordination.^20,21 To explore the ways that interorganizational care coordination is proceeding, in previous work²² we reported results from qualitative interviews with VA and community providers regarding the delivery of shared care for rural veterans. In the current manuscript we expand on these findings, reporting results from similar interviews with veterans who have received both VA and community health care services in the same geographic areas as the providers interviewed in our prior work.

Methods

Overview

We conducted semi-structured interviews between May and August of 2019 with a purposive sample of veterans who had used both VA and community health care services. The VA Boston Institutional Review Board approved all study procedures. We used directed content analysis²³ to guide analyses, based on our earlier work,²² but we were also attentive to emergent themes brought up by our respondents.

Study Settings and Population

We recruited veterans who had used both VA and community health care services within the past 2 years. Our aim was to focus on veterans who had used these services in the same geographic areas (and during the same time frame) as the VA and community providers we interviewed in our previous work. We therefore focused on 8 distinct geographic areas—2 each in the US Northeast, Midwest, West, and South.

Sampling Plan

We used the Community Care portion of the Survey of Hospital Experiences of Care (SHEP)²⁴ to identify our sample. This survey is completed on a rolling basis by veterans with clinical encounters both within and outside of VA. We specifically identified veterans who had completed the Community Care portion of SHEP from the same 8 geographic areas in which we had conducted our previous interviews with providers.²² We further limited the sample to the subset of these veterans who had used community services in the same time period during which we conducted our previous interviews. This approach therefore qualifies as a stratified purposeful criterion sampling design.²⁵ A total of 276 veterans met these inclusion criteria.

Recruitment

We used opt-out methods to recruit our sample of interview participants.²⁶ First, we mailed potential participants a study packet containing a description of the study, noting that the research team would call them in about 2 weeks unless they opted out. To allow opt-out, the study packet included a phone number the potential participant could call, as well as an opt-out postcard they could return. The opt-out postcard included a study-specific ID code so that potential participants did not need to identify themselves on the postcard before returning it if they chose to opt out.

Beginning 2 weeks after the mailing, the study team called potential participants to describe the study in more detail and assess interest. Veterans who agreed to participate were given the opportunity to ask questions and either complete the interview immediately or schedule it for a future date. Our a priori goal was to recruit 40–60 participants, with plans to cease recruitment once (a) we had interviewed at least 40 participants; (b) we had obtained at least 1 respondent from each geographic area; and (c) we appeared to have achieved thematic saturation (ie, new interviews were similar in content to earlier interviews).

Qualitative Interviews

Our previous work with VA and community providers²² suggested 5 domains of importance regarding interorganizational coordination: external and veteran context; organizational mechanisms; organizational culture; relational practices; and the role of TPAs.^27–32 We developed a semi-structured interview guide based on these domains, along with open-ended questions about veterans’ experiences of receiving health services from VA and community providers. We conducted interviews over the phone after obtaining verbal informed consent. Each interview took 20–60 minutes to complete and was audio-recorded and transcribed verbatim; participants were reimbursed $40 each for their time.

Analysis Plan

We conducted a directed content analysis²³ of the interview transcripts. Coding proceeded through an iterative process using a codebook that focused on the same 5 domains that guided our interview questions. First, the 4 authors independently coded data from 1 interview, then discussed their coding as a group until consensus was reached on how the 5 codes (1 for each of the 5 domains) were to be interpreted and applied to data segments. Second, the authors coded 3 additional interviews each, then discussed these 12 interviews as a group to further specify characteristics of data segments for which each code was or was not to be applied. Third, each author was designated as the primary coder for approximately one quarter of the remaining interviews. Any data segments that were difficult for the primary coder to code were discussed as a group. We used NVivo Qualitative Data Analysis software (QSR International, Melbourne, Australia) to capture all coding activity.

We generated reports of data segments coded with each of the 5 codes. Each of the 5 reports was assigned to 1 of the authors to serve as the primary analyst, who analyzed the report to identify themes that aligned to those from our previous work²² and themes that uniquely emerged from the current data. For each code, these themes were documented by the primary analyst using a structured summary template. We then discussed as a group to reach consensus on whether to further clarify and/or consolidate the themes.

Results

Study Sample

Our sample included 51 veterans (18% response rate), of whom 39% were female (n=20). While we did not formally collect data on veteran age, interview results suggested that our sample included a broad age range, including veterans involved in recent conflicts in Iraq and Afghanistan as well as those who fought in the Vietnam War. We achieved reasonable geographic diversity with 20% of our sample (n=10) from the Northeast, 18% (n=9) from the South, 27% (n=14) from the Midwest, and 35% (n=18) from the Western United States.

Qualitative Results

Below we report results from each of 5 major domains: external and veteran context; organizational mechanisms; organizational culture; relational practices; and TPAs. In terms of emergent themes, many veterans spoke about the implications of breakdowns in interorganizational care coordination across these domains; we therefore report cross-cutting implications in a separate section. In each case, we begin with a brief description of the results we found in our prior work involving interviews with VA and community providers to provide context.

External and Veteran Context

Our previous work²² identified several ways that external context (eg, travel distance, unavailability of services) and veteran context (eg, age, specific diagnoses) impacted the process of receiving VA and community care. Specifically, providers from VA and community care organizations noted that long travel distances between sites made it difficult for care coordination staff to meet face-to-face. They reported difficulties recruiting and retaining staff in rural areas at VA and community sites alike. Staff respondents indicated that older rural veterans often struggled to manage multiple comorbid conditions. They also noted that areas with a shortage of VA health care services were likely to also suffer from a shortage of those same services in the community (especially substance use and mental health services).

Many of these same themes were also emphasized by veteran participants of our current study, including long travel distance to VA services, limited availability of desired services, and complex care needs. Long travel distance to VA services was a pivotal reason that veterans preferred non-VA services in our sample: several respondents noted that attending appointments at distant VA sites required at least half a day, while local community services might be only minutes away. Similarly, several respondents noted that they preferred community care for services that would have required long wait times at VA clinics.

Complex care needs were described by many participants as making care coordination challenging. In some cases this created “Catch-22” situations in which VA and community providers were reluctant to move forward with treatment until the other party addressed a different condition. One veteran described such a standoff: “I went back [to the non-VA clinic] and the [non-VA provider] said, ‘Don’t come back here anymore until the VA does something with the tumor in your head.’” Other veterans noted that it was particularly difficult to coordinate urgent care services (eg, emergency department visits) that by their nature could not be arranged beforehand.

Organizational Mechanisms

In our prior work, VA and community providers noted challenges in determining overall responsibility for care coordination (ie, who is primarily responsible for the patient), as well as challenges in determining responsibility for specific care coordination tasks.²² Providers also noted that medical record interoperability in the form of Health Information Exchanges (HIE) was not consistent, with community providers frequently lacking access to VA medical records for shared patients. Providers also discussed administrative solutions to these challenges, such as the establishment of care coordination contracts, Memoranda of Understanding (MOUs), and VA-based care coordination offices.

Several of these findings were echoed by our veteran respondents. Many veterans noted that complexities arose based on the sheer number of clinicians and staff involved in delivering and coordinating care across organizational lines. In many cases, these veterans reported being unable to keep track of the responsibilities or roles pursued by each of these different care coordination and clinical staff. One veteran described feeling he was simply being “pushed off from 1 person to the next” in the absence of a clear understanding of who was responsible for what. This issue was less problematic for veterans whose clinical conditions did not require extensive coordination (eg, those receiving dental care in the community and other health care services within VA). Even in those cases, however, the number of parties involved could still be challenging for some care process (eg, authorization, initial scheduling, and getting community prescriptions filled at VA pharmacies).

Some veterans noted that VA and community clinics had the capacity to share medical records electronically, but others noted a lack of HIE interoperability. One veteran noted that, in the absence of a shared medical record, their community clinic required records be faxed from the VA—but that fax transmissions were frequently lost or misdirected.

Unlike our previous provider respondents, veteran respondents from the current study did not describe administrative mechanisms for enhancing care coordination (eg, MOUs, contracts, VA-based care coordination offices). However, this is not necessarily surprising, as many of our respondents reported being unsure of what went on “behind the scenes” (ie, did not have ready access to the inner workings of the VA bureaucracy).

Organizational Culture

In our prior work, we found similarities and differences between the organizational cultures of VA and community clinics. These differences included perceptions that VA could be more bureaucratic and insular than community clinics, and that VA clinicians could at times struggle to establish therapeutic alliance. Several of these findings were also echoed in our current study.

First, many veterans in our sample—including those who reported struggling with care coordination difficulties—noted the dedication, commitment, and passion of their VA and community clinicians alike. These veterans noted that the organizational cultures of VA and community providers were aligned in prioritizing quality care for their patients.

With that said, several respondents noted bureaucratic challenges that appeared to stem from VA policies (eg, related to obtaining authorizations, getting paperwork sent to community clinics, clarifying reimbursement procedures for community care, and getting routed to the correct person within VA). Some respondents used terms like “VA bureaucracy,” “complacent,” and “very military-like” in describing VA culture, which could ultimately lead to delays in obtaining or reimbursing community care.

Second, some veteran respondents echoed findings from our prior work that VA could be perceived as more insular than their community counterparts. For example, VA clinicians were perceived to be less likely to reach out to community providers regarding treatment disposition for shared veteran patients.

Third, some veterans perceived that VA clinicians may have a harder time in developing therapeutic alliance and continuity of care with veterans, with 1 veteran stating “I like outside physicians better because I feel like I have a connection.” Some veterans reported feeling like “a number” or “the middleman” within VA rather than a valued member of the care team. This was in part attributed to the common presence within VA of trainees on time-limited rotations, contract clinicians, large caseloads (“giving too many patients to too few doctors”), and clinicians nearing retirement. One veteran respondent noted that this resulted in misdiagnosis within VA that was only corrected once she sought community care:

“In the end, because [my community provider] listened to me, it was determined that I had an injury from military service that I’d been complaining about for years going back and forth. And they discovered it…when I had been going to the VA complaining about the same thing the only thing they would say [was that] it was arthritis.”

Notably, however, several veteran respondents provided counterexamples in noting that they received more “personalized” and “sensitive” care within VA than in the community, such as through VA-based mental health care. Others noted improvement in VA’s organizational culture over time with VA “changing for the better,” and “listening more now.”

Relational Practices

In our prior work, we found that community providers appreciated communication with 2 types of VA staff. First, they desired access to VA clinicians who cared for shared veteran patients, and second, they valued having a single administrative contact at VA with intimate knowledge of care coordination policies and procedures. In some cases, they reported challenges related to timely communication with both groups.

In the current study, many veterans echoed the value of having open communication directly between their VA and community clinicians. For example, 1 veteran described a “seamless” pattern of communication and mutual decision-making between his VA and community clinicians related to an important surgical procedure.

In other cases, however, veterans reported frustration at an apparent lack of communication. For example, 1 veteran felt like the VA had “washed their hands of [his] case” once the referral to community care was made. Another veteran was frustrated that VA staff insisted on using fax machines—rather than more efficient electronic methods—to reach community staff. Some veteran respondents noted that the consequences of this lack of communication could include delays in care, redundant health care services (eg, multiple unnecessary MRIs), or an apparent “tug of war” between VA and community staff. Among our respondents, veterans attributed a paucity of communication to their community and VA clinicians, with 1 veteran reporting that it took direct intervention from him to connect them:

“I was at a total loss to try to figure out what [my community clinician] had prescribed. In fact, I got [both my VA and community clinician] on the phone together and they hashed out what they could and couldn’t do. So with some coordination by yours truly, they decided that each other’s medications wouldn’t give me a heart attack.”

In contrast, veterans appeared to deemphasize the importance of direct clinician-to-clinician communication when they had straightforward care needs that did not require ongoing consultation (eg, between primary care and dental services).

Veteran respondents also echoed community providers’ viewpoints on the importance of having a contact within VA—above and beyond their frontline clinicians—who could help coordinate VA and community health care across systems. Such coordination was seen as especially important for procuring medical equipment (eg, insulin pumps, walkers for post-surgical recovery) or prescriptions (eg, those written by community staff to be filled by VA pharmacies). Likewise, veteran respondents noted the utility of having at least 1 staff person at their community clinic with knowledge of VA procedures related to shared care.

Third Party Administrators (TPAs)

In our prior work, we noted numerous challenges in scheduling and reimbursement of community care for veterans through the TPAs. In the current study, our veteran respondents echoed many of these findings. First, many veteran respondents found TPA procedures to be overly complex and challenging to navigate. Some veterans were unsure of who to contact at the TPA, how to schedule appointments, and who was responsible for selecting community providers (TPA, VA, or veteran). Many veteran respondents reported having difficulty remembering or distinguishing whether they had worked with TPA or VA representatives to schedule appointments or discuss billing concerns. Several participants acknowledged the confusion and challenges in having the TPA as the “middleman” between VA and community providers:

“I know they kind of ended up the middleman but the process overall was very confusing. Who do I wait to call me? Do I call someone and schedule and then let them know? Once they made an appointment, can I change that appointment? [Sometimes the TPA would] make an appointment that just made zero sense.”

Additionally, several veterans described their frustrations with the complexity in the eligibility or re-authorization process. We found examples where veterans had to pay out-of-pocket for the services received in the community or had to stop treatment for several months while awaiting re-authorization.

Second, similar to our prior work, veterans perceived communication procedures with TPAs as inefficient, delayed or inaccurate. These issues could be exacerbated by veterans frequently experiencing long hold times before being rerouted to different TPA representatives (ie, being “hustled around from person to person”).

Third, veterans also perceived the effectiveness of TPAs as being hampered by their distance, both geographically and organizationally, from the clinics for which they were scheduling appointments. TPA staff had scant knowledge of local conditions and health care resources to connect veterans to their preferred services:

“[T]hen I had 1 [TPA representative] from Phoenix. Then I had some [TPA representative] from Pittsburgh…but they’re not in this area so they don’t know what the traffic situation is. And I’m 71, you know. They were getting me in a lot of heavy traffic. Of course, I will [drive to my appointments in heavy traffic] if I have to, but I’d rather not if I don’t have to. And they just don’t know the situation and it would just be better if [this city] had their [TPA] people there and they called you for this area.”

Fourth, we found that veterans experienced challenges related to payment/billing due to a lack of coordination between the VA, the community clinic, and the TPA, as well as when a veteran had dual coverage (eg, both VA and Medicare). As a result, veterans, while managing their illness, had to contact the TPA multiple times to attempt to resolve these billing issues or had to pay out-of-pocket. One veteran reported that miscommunications from the TPA resulted in his community provider reporting his bill to a collection agency; at the time of our interview the situation had not been resolved over a year after the appointment occurred.

Despite these challenges, several veterans reported that they had seen improvements over the past 2 years in the communication, procedures, and culture of the TPAs. Some veterans reported that TPA representatives “listen now” and have become “very friendly,” “eager,” “more pleasant,” “polite,” and “treat you like you’re a human being instead of not paying attention to your needs.”

Implications of Breakdowns in Interorganizational Care Coordination

Across the domains above, our veteran respondents spoke about the consequences of breakdowns in interorganizational care coordination. First, several veterans noted that such breakdowns meant that they themselves became their own de facto care coordinator. One veteran, who was nearly 80 years old and had experienced an aneurysm alongside numerous comorbid health conditions, reported calling the VA every day for a month to get test results sent to his community provider. Another veteran reported “giving up” on expecting VA and community staff to coordinate care; he reported resigning himself to instead serving in the “middleman” role. Others reported experiencing gaps in care, being incorrectly billed for services, or enlisting the help of their clinicians to help manage care coordination breakdowns. One such veteran, who rallied a handful of his clinicians to help advocate for his care coordination needs, stated “I have no problem with the individuals treating me: it’s the system.” In some cases, veteran respondents noted that their community clinicians now required pre-authorization before treating veterans to avoid lengthy reimbursement delays. In extreme circumstances, some community clinicians dropped partnerships with VA entirely over these issues, with 1 veteran respondent summarizing, “They weren’t paid. Ultimately, I was actually dropped along with every VA patient.”

Discussion

Key Findings in Context

In this paper we present results from qualitative interviews with veterans regarding their personal experiences of interorganizational care between VA and community clinics. This study builds on our prior work,²² which involved similar interviews with VA and community providers in the same geographic areas.

Consistent with our previous work, our veteran respondents noted myriad challenges in accessing and coordinating care stemming from rural geography; staff recruitment and retention; and unavailability of services. Specifically, in many areas of the country, services that are unavailable in VA are also likely to be unavailable in the community (eg, timely mental health and substance abuse treatment). Our veteran respondents also noted that, even in places where care coordination mechanisms are in place, the veteran population may present unique challenges based on extensive comorbidities (both mental and physical³³) that complicate treatment and coordination alike—although the complicating effect of comorbidities is certainly not unique to veterans.³⁴

Veteran respondents also echoed our previous findings in reporting that they were unsure who “owned” their care—that is, who was ultimately responsible for coordinating care across organizations. This confusion was frequently compounded by a lack of communication between VA and community clinics. In our sample, this meant that the veteran or a family member frequently needed to step in as the de facto care coordinator. In other cases, our veteran respondents noted that a trusted clinician (eg, a mental health provider, nurse, or primary care provider) undertook this role. Other ongoing work is attempting to address this issue by comprehensively specifying care coordination roles and responsibilities.^21,35,36 Challenges in this domain were lessened in circumstances when the care being delivered within and outside of VA required little ongoing coordination (eg, primary care in 1 setting and dental care in the other). Veteran respondents noted variable access to HIE for their clinicians across institutions: faxed and hand-carried medical records were sometimes required in the absence of HIE options.

We heard disparate stories from veterans regarding the organizational cultures of their local VA and community clinics. In some cases, veterans echoed our previous findings that VA clinics may be perceived as less flexible, outward-facing, and patient-centered than their community counterparts. But many veterans praised their VA and community providers alike for their dedication to delivering personalized, high-quality care—and others noted that they preferred the personal attention they received at the VA. This is consistent with literature from outside of VA indicating the importance of patient-centeredness in health care delivery.³⁷ In many cases, our veteran respondents’ perceptions appeared to be driven primarily by their experiences with individual clinicians, emphasizing the importance of the therapeutic alliance^38,39 in shaping perceptions of health care institutions. Some veterans noted improvements over time in veteran-centeredness of VA care; this may be tied in part to recent VA initiatives aimed squarely at this domain (eg, Whole Health⁴⁰).

Veteran respondents noted the value of open communication with both their frontline clinicians and staff responsible for interorganizational care coordination at VA and community sites. These findings align closely with key components of Gittell’s theory of relational coordination,^28,31 which spells out the utility of communication that is frequent, timely, accurate, and solution-oriented in the pursuit of goals that are shared between parties.

Many veterans in our sample described challenges in interacting with TPAs (Health Net and TriWest) to schedule, administer, and reimburse community health care services. These findings are consistent with other recent research.⁹ In our sample, some veterans noted that the structure of the TPAs—distant bureaucracies serving as a “middleman” between VA and community care—was more likely to hamper than to facilitate interorganizational care coordination. Despite this, some respondents noted recent improvements in their interactions with the TPAs. Some TPA contracts were recently renewed, while others expired and were replaced by contracts with different TPAs, scheduled to run through at least March of 2021.⁴¹ Thus, while the intention is to eventually move scheduling and reimbursement activities completely within VA,⁴² the recent round of contracts suggest that these organizations will continue to play an important role in interorganizational care coordination for veterans.

Limitations

First, our findings ultimately are based on a limited number of qualitative interviews across 8 geographically diverse sites, meaning that we may not have captured all issues relevant to interorganizational care coordination. Nonetheless, we aimed to enhance comprehensiveness and generalizability by conducting interviews with veterans (for the current study) as well as VA and community providers in the same geographic areas in prior work.²² Second, we conducted our qualitative interviews during a 4-month period (May through August of 2019), during which the Veterans Choice Act was superseded by the VA MISSION Act. Since then, VA and community care coordination has changed rapidly,⁴³ and it will be expected to evolve further during the COVID-19 pandemic. Thus, some study results may be less relevant than others moving forward. Third, our procedures did not involve speaking to representatives from the TPAs, potentially limiting generalizability.

Implications for Interorganizational Care Coordination Between VA and Community Providers

Taken together, we believe these findings may inform a fundamental policy question: under what circumstances, and how, should VA partner with community health care institutions to provide treatment for veterans under the MISSION Act? This question is even more complicated in the midst of the COVID-19 pandemic, as (a) face-to-face services may be contraindicated for many conditions; and (b) VA and other health care organizations undergo a massive shift toward telehealth-based care, which can remove geographic barriers to access in some circumstances.^3,4,6,7 While we cannot completely address these complex issues based solely on a limited number of qualitative interviews, Table 1 summarizes possible implications based on our current and prior results,²² as well as the broader literature on VA and community care,⁹ with an emphasis on contextual and structural considerations. Table 1 also includes implications that may generalize beyond VA, although we acknowledge that our findings are most likely to be applicable to VA and community care coordination given our sampling and data collection.

Table 1.

Implications for Interorganizational Care Coordination

Implications for VA and Community Interorganizational Care Coordination	Further Implications for Interorganizational Care Coordination Beyond VA
Before committing to referring veterans to community care, we recommend a thorough review of what health care services are available in the community. This review should also consider: Whether community providers are equipped and trained to provide treatment in the context of common comorbidities in the veteran population (eg, traumatic brain injury [TBI], posttraumatic stress disorder [PTSD])44 Whether local veterans have preferences for VA versus community care (eg, based on local transportation options) The extent to which services may be available via telehealth from other VA sites. For example, VA has funded a series of telehealth hubs4 and a National Telemental Health Center5 that can provide specialized services across state lines—services that may be easier to arrange than community care. The average wait times associated with those community services. In some cases, community care may involve longer wait times than would be experienced within VA.	Before committing to shared care between health care organizations we recommend careful consideration of: Potential differences between the patient populations traditionally treated by each organization (eg, comorbid conditions, symptom severity), and the implications of those differences for shared care Logistical considerations relevant to accessing care, such as transportation (for in-person visits) or hardware/software/connectivity (for telehealth) Variability in average wait times between the 2 organizations
We recommend considering the amount of coordination that will likely be required to maintain ongoing health care services across institutions. For example, referring veterans for dental care in the community may require little direct contact between clinicians at either institution. In contrast, if ongoing coordination is needed, we recommend explicitly establishing who within VA has primary responsibility for that coordination, and whom to contact at the community site. Succession plans for these key staff may minimize the impact of turnover or reassignment.45	We similarly recommend considering the amount of coordination that will likely be required to maintain ongoing health care services across organizations. These may include: Considering who within each organization has primary responsibility for coordinating the patient’s care at that institution Considering which organization overall has primary responsibility for the patient’s care, and how care coordination staff at that organization will communicate with their counterparts within the other organization(s) Developing succession plans for key care coordination staff to prevent loss of institutional knowledge
Veterans, VA staff, and community staff expressed a desire for increased medical record interoperability between VA and community providers. It is uncertain how the upcoming transition of the VA medical record system to the Cerner platform46 will impact such interoperability.	Health Information Exchanges (HIE) can increase the ease with which providers at different organizations can communicate regarding patient care

Ultimately, our analyses—rooted in first-hand reports from clinicians, administrators, and veteran patients—suggest that developing effective interorganizational care coordination between VA and community providers will require building effective coordination mechanisms and relationships.^28,31 Our results further suggest that such efforts may benefit from attending to (a) potential cultural differences between institutions and (b) the comorbidities and logistical challenges frequently faced by rural veterans. Efforts to achieve these goals may involve improving medical record interoperability; more comprehensively connecting VA and community care coordination staff; and leveraging telehealth technologies where appropriate. Ideally, future research will attend to these issues while developing and refining specific programs to enhance interorganizational care coordination between VA and community providers.