Family Involvement in PTSD Treatment: Perspectives from a Nationwide Sample of Veterans Health Administration Clinicians.

Social support is bidirectionally linked to symptoms of posttraumatic stress disorder (PTSD). Evidence suggests that family involvement in veterans' mental health treatment is desired by both veterans and family members, and that such involvement has the potential to improve treatment outcomes. However, rates of family involvement are low in the Veterans Health Administration (VHA). We sought to understand VHA clinicians' perspectives on family involvement in PTSD treatment by conducting qualitative interviews with 31 providers at 10 VHA facilities across the U.S. The i-PARIHS framework was used to guide the interviews and analysis, and several major themes were identified. All clinicians reported that they at least occasionally offered family-inclusive sessions, and they frequently referenced both the influence of family behaviors or attitudes on veterans' functioning, and also how veterans' symptoms could cause tremendous disruption in the family. Clinicians' past experience with supervised family- or couple-based work strongly influenced their current comfort with family-inclusive sessions. Multiple potential avenues exist to support increased family involvement in PTSD treatment in VHA.

Social relationships and the symptoms of posttraumatic stress disorder (PTSD) are reciprocally related (Galovski & Lyons, 2004; Solomon et al., 2008). PTSD is the most common diagnosis across the Veterans Health Administration (VHA) (e.g., Cohen et al., 2010), and lack of social support has been identified as a key risk factor for the development, maintenance, and severity of PTSD, especially in the veteran population (Brewin et al., 2000; Ozer et al., 2008). Veterans’ PTSD symptoms are inversely related to relationship satisfaction and social support over time (Koenen et al., 2008; Lunney & Schnurr, 2007). Family members are also negatively impacted by their loved one’s PTSD, reporting greater levels of conflict and distress and less satisfaction compared to family members of veterans without PTSD (Jordan et al., 1992; Solomon et al., 2008). Providing veterans with PTSD and their family members with support, skills, and a structure for enhanced communication about the disorder is therefore of great importance for reducing associated symptoms and enhancing their quality of life. Thus, enlisting family members as critical partners in the psychotherapeutic environment benefits veterans and those around them.

Even without formal involvement in treatment, family members’ behavior shapes veterans’ treatment experience. Encouragement from family members is a key motivation for treatment-seeking among many veterans (e.g., Sayer et al., 2009). Families can also have meaningful impacts on Veterans’ trajectories once they are in treatment. One study using nationwide VA administrative data found that Veterans with PTSD who had at least one family session experienced a statistically significant drop in symptoms following that session, and the positive effects were even more pronounced when they had higher numbers of family-inclusive sessions (Laws et al., 2018). Family member behaviors can even substantially buffer the risk of dropout from evidence-based protocols (EBPs) for PTSD, resulting in dropout rates that are 50% lower (Meis et al., 2019; Thompson-Hollands et al., 2021b). This is especially notable because veterans in PTSD treatment are at high risk of nonengagement and dropout (e.g., Hale et al., 2019). In contrast to these positive influences, family members may also engage in less adaptive behaviors such as symptom accommodation (i.e., colluding with avoidance), which run counter to the goals of PTSD treatment (Campbell et al., 2017; Fredman et al., 2014). Formally incorporating families into care can capitalize on the power of family members while also perhaps mitigating problematic patterns. In contrast, routinely failing to include family members in PTSD care means that a critical life domain may be neglected in treatment.

There exists a spectrum of effective family involvement in EBPs for PTSD. Relatively brief engagements with family members typically consist of providing psychoeducation and an orientation to the treatment rationale (Thompson-Hollands et al., 2021c). Targeted skills training can also be used to reduce accommodation or other unhelpful behaviors. These types of brief interventions serve to reduce family members’ concerns about trauma-focused treatment and bolster an attitude towards avoidance that aligns with treatment goals. More substantial family involvement could consist of a fully conjoint PTSD treatment, such as Cognitive Behavioral Conjoint Therapy for PTSD (CBCT; Monson & Fredman, 2012), where the relative is present for all sessions and there is a dual focus on both trauma and relationship issues. CBCT is one of VHA’s EBPs for PTSD and relationship difficulties and is offered as a roll-out through VHA Family Services. Ultimately, the effects of family-inclusive treatment across intensities and formats are tri-fold: (1) reduced PTSD symptoms for the patient (e.g., Monson et al., 2012); (2) reduced distress for the family member (e.g., Sautter et al., 2015); and (3) reduced relationship discord (e.g., Weissman et al., 2018). The differences in circumstances across Veterans and their families, including available time, other family demands, and level of distress or impairment, necessitates a spectrum of family involvement that can be tailored to the individual Veteran. This spectrum is consistent with VHA policy on family involvement in care (U.S. Department of Veterans Affairs, 2019; VA Directive 1163.04), which requires a range of services be made available to families and regular conversations with veterans to discuss the possibility of family involvement.

Research has shown that veterans and family members are interested in greater family involvement in mental health care (Batten et al., 2009; Thompson-Hollands et al., 2021a). Nevertheless, family-inclusive visits remain very low across VHA, accounting for less than 1% of mental health appointments among a sample of veterans with high rates of PTSD (Harper et al., 2022). Without understanding what drives or inhibits family involvement in PTSD treatment specifically, VHA facilities may continue to underutilize this approach to the detriment of veterans. Tapping into the perspectives of VHA clinicians, who can function as critical facilitators of family-inclusive treatment, is important for understanding the state of this care. In the only previous such examination of which we are aware, Sherman and Fischer (2012) conducted qualitative interviews with five clinicians at two VHA community-based outpatient clinics (CBOCs), as well as interviews with veterans and their family members. The interviews focused on the Support and Family Education (SAFE) program, a group family education program that has been widely implemented in VHA. Results from providers indicated that they generally supported the idea of family education, but that they lacked training in family-inclusive care and would need support and training from more experienced providers prior to delivering the SAFE program. The providers further expressed concern about adding programming to their already busy schedules and noted that their facilities sometimes lacked appropriate spaces for large group meetings. Finally, the providers also expected that the rural veterans and families that were served by the CBOC would be challenging to engage in a family program due to stigma and suspicion around mental health treatment. These findings provide a useful baseline, as the attitudes are not necessarily specific to the SAFE program innovation, but rather may reflect broader attitudes and concerns about engaging in family-inclusive care. However, given the small sample size and inclusion of just two rural CBOC facilities (Sherman & Fischer, 2012), an updated and expanded examination of family-engagement in PTSD treatment is warranted.

Study Aims

The present study aimed to examine the current attitudes and practices of VHA mental health clinicians regarding family involvement in PTSD treatment through semi-structured interviews. Guided by the integrated-Promoting Action on Research Implementation in Health Services (i-PARIHS; Harvey & Kitson, 2016) framework, we explored system, clinician, patient/veteran, and innovation factors that impacted uptake of family engagement.

Method Design

Qualitative methods are especially valuable when studying an issue where there is limited prior research and there are likely multiple dynamic processes of interest (Miles et al., 2014). We spoke with clinicians (N = 31) at 10 VHA facilities nationwide; half of the facilities were conducting family visits with a relatively higher proportion of veterans with PTSD (“positive deviants”), while the other half were conducting visits with a relatively lower proportion (“low engagement facilities”). The practice of identifying “positive deviants,” or organizations/sites that demonstrate comparatively strong outcomes in an area of interest (e.g., survival rates, wait times, clinical end points) (Bradley et al., 2009), is beneficial because these entities may have already discovered innovative solutions to problems which face that community. These strategies may be especially viable and sustainable, compared to strategies generated by outsiders, because they have been generated from within the existing system and presumably use existing resources.

Facility Selection and Recruitment Procedure

All study activities were approved by the VA Boston Health Care System IRB and a waiver of documentation of written informed consent was granted. Potential sites were identified and recruited using purposive sampling (Palinkas et al., 2015) targeting mental health clinicians at VHA facilities with high and low levels of family involvement in PTSD treatment. We extracted PTSD diagnosis and healthcare visit data from the VA Corporate Data Warehouse (CDW), a national repository of administrative data, to calculate rates of family involvement among veterans with PTSD at each VHA facility during fiscal years 2018 and 2019 (FY18-19). Specifically, for each VHA facility we determined the number of veterans who had a PTSD diagnosis associated with an encounter during FY18-19. We then determined the subset of those veterans at each facility who had attended ≥ 1 mental health-related visits in FY18-19. We considered this to be the pool of veterans with PTSD who were in mental health treatment (“treatment pool”). Then we calculated the subset of veterans in the treatment pool who had attended ≥ 1 family-inclusive visit in FY18-19 (denoted by Current Procedural Terminology [CPT] codes 90846, 90847, and 90849) (“family pool”). We calculated a ratio for each facility by dividing the family pool by the treatment pool. We then calculated a cut point for the top 10th percentile (“positive deviants”) and bottom 90th percentile (“low engagement facilities”) in terms of their ratio of family-inclusive care among veterans with PTSD in active mental health treatment. Notably, the overall rate of family involvement was low across all VHA facilities; even at the highest performing facility nationwide, less than 2% of veterans with PTSD in active mental health care had one or more family-inclusive sessions over 2 years, which is broadly consistent with an earlier study that have focused on the frequency of family involvement in VHA (Harper et al., 2022).

There were 12 facilities in both the positive deviants category and in the low engagement category. (The first author) created a random-order list of both categories. We then proceeded sequentially down the lists for recruitment, with the aim of enrolling 2–4 providers per facility. We contacted the Chief of Psychology at each facility and asked for assistance in being connected to the director of the facility’s PTSD Clinical Team (PCT) or, if the facility did not have a PCT, to providers specializing in PTSD. We emailed PTSD providers information about the study (e.g., purpose, expected length of participation) and requested a reply if they were interested in participating. We recruited at a total of 10 facilities (5 positive deviants and 5 low engagement sites). The 10 facilities included representation from 9 different states/territories and eight Veterans Integrated Service Networks (a regional category used by VHA to geographically divide the country), providing a robust degree of geographical diversity. The number of recruitment sites was pre-specified prior to initiating the project, but the number of individual interviews was permitted to vary depending on when saturation was achieved.

Participants

A total of 31 mental health providers specializing in PTSD were interviewed across the 10 facilities, including 16 clinicians from positive deviant sites and 15 clinicians from low engagement sites. The mean number of interviews conducted at a given facility was 3.1 (SD = 0.74, range = 2–4). Table 1 displays the sample demographics.

Table 1

Participant demographics

Gender
Women	61.3%
Men	38.7%
Race
Caucasian	77.4%
Asian	9.7%
Black	6.5%
Other	3.2%
Ethnicity
Non-Hispanic	93.5%
Role
Psychologist	80.6%
Social worker	19.4%
VHA setting
PTSD specialty clinic	54.8%
General outpatient mental health clinic	35.5%
Family clinic	9.7%
Mean hours direct patient care per week	19.27 (SD = 7.43, range 4–30)
Mean years working as licensed VHA provider	8.29 (SD = 6.53, range 0–32)

Procedures

Participants took part in a one-time audio recorded phone interview. They were not compensated for their time. The interview guide was structured to assess three key constructs of the i-PARIHS framework: innovation characteristics (i.e., elements related to family involvement itself, such as the relative advantage of family involvement compared to individual-only care); recipients (i.e., elements related to the clinicians, such as their available time and resources, or their skills and knowledge); and context (i.e., elements related to inner and outer settings, such as leadership support for family involvement) (Harvey & Kitson, 2016). Prior to use, the draft guide was refined with input from clinician and Veteran stakeholders. Participants were asked about their attitudes towards and use of family-involvement, as well as factors that drive or discourage family involvement in PTSD treatment. Sample questions included: “How aware would you say you are about existing family-inclusive interventions for PTSD?”, “What are some ways that family involvement is initiated in this clinic?”, “In general, do you feel it changes the treatment when you incorporate family members? If so, in what ways?”, and “Do you feel you have the time necessary to incorporate family members as often as would be indicated?”

Interviews lasted an average of 34.45 min (SD = 8.27) and were conducted by (the first author), a licensed clinical psychologist with expertise in PTSD and family-inclusive treatments and training in qualitative methods. Interviews were audio recorded and later transcribed.

Data Analysis

Interviews were analyzed using a hybrid inductive-deductive rapid content analysis approach (Hamilton, 2013, 2020). A rapid approach to qualitative analysis has been compared directly to more traditional thematic analysis and demonstrated valid findings, while delivering results in less time (Taylor et al., 2018). This approach is relatively less interpretive, remaining “close to the data” and providing key actionable findings. The multi-step analysis approach followed common rapid analytic methods in implementation science (Hamilton & Finley, 2019). The research team first developed a summary template consisting of a neutral domain name corresponding to each section of the semi-structured interview guide, as well as space for key quotations and observations about emerging themes not captured by the other domains (Hamilton, 2013). Both inductive and deductive elements were of interest. Deductive themes that were particularly salient included: level of existing knowledge of/training in family-inclusive approaches; specific past positive and negative experiences of family involvement; and messages (overt or subtle) that providers received from their leadership regarding family involvement, among others. We generated summaries of each interview by entering minimally-interpretive descriptions of participant responding under the corresponding domain headers in the summary template. (The first author) completed summaries for all interviews, while 19.4% were summarized for a second time by a masters-level research coordinator (the second author) to ensure consistency across analysts. We resolved minor discrepancies by consensus. We then transferred the summary templates into two matrices, for clinicians from positive deviant sites and low engagement sites, respectively, to display themes within and across groups for each domain. Multiple team members reviewed the matrices and identified common/salient patterns, with a focus on factors that appeared to particularly drive or discourage the use of family involvement (Averill, 2002); the use of multiple analysts allowed for triangulation of findings. Guided by i-PARIHS (Harvey & Kitson, 2016), findings were organized by innovation, recipient, and contextual factors (see Table 2). Comparisons between positive deviant and low engagement sites were conducted within and across themes, with analysts comparing the content, relative tenor, and pervasiveness (within sites) of the perspectives expressed. Finally, an inquiry audit was conducted by a team member who had not been involved with the initial data collection and analysis. This team member reviewed the original transcripts, matrices, and summary sheets to confirm the dependability of the results.

Table 2

i-PARIHS domains and associated themes

i-PARIHS domain	Theme from interviews
Innovation	Comparing family inclusive treatment to individual-only approaches
	Family involvement in EBPs for PTSD
	Positive and negative experiences with family involvement
	Family psychoeducational groups
Recipients	Knowledge
	Staff availability
Context	Veteran-driven, clinic-driven, or ad hoc
	Leadership promotion of family involvement
	Impact of COVID

Results

Descriptive Overview of Participants

Overall, interviewees were highly aware of the extent to which PTSD impacts and is impacted by the veteran’s family relationships. They frequently referenced both the power of family behaviors or attitudes on veterans’ functioning, and also how veterans’ symptoms could cause tremendous disruption and distress among family members. All providers in the sample described offering family-inclusive sessions at least some of the time, and certainly if the veteran requested such a session. By far the most common type of family involvement that participants described was a single joint session.

Across all interviews, the interviewee’s level of past training in family-inclusive treatments (especially including supervised experience during their graduate program, internship, or postdoctoral fellowship) clearly influenced their current use of and comfort with family-inclusive approaches. The majority of the sample had very little experience in providing family- or couples-focused interventions, and these clinicians frequently described feeling hesitant to take on family work. They often preferred to refer veterans and their loved ones to a specific provider at their location (in many cases there was only one) who had more formal training. Conversely, some interviewees had sought extensive training in family-inclusive approaches and considered family involvement to be an important clinical value. Differences in past or current experience with family-inclusive protocols did not correspond to whether or not the participant was from a positive deviant site.

What we were taught in my [graduate training] program was to become a family therapist based on the models that existed you really needed to go to a, like an LMFT program and or an MFT, whatever it is. And receive that specialized type of training …. So that was the model that I learned, which is a very complicated one. Like it’s a specialized area of competence. Some of my colleagues just moved over to the Vet Center and maybe now they offer family therapy. I would never do that, because to me it feels like this really specialized thing that I would definitely need to pursue additional training to do. (013).

Innovation Characteristics

Comparing Family Inclusive Treatment to Individual-Only Approaches

When asked how a family-inclusive approach compares to an individual approach, the most common response was that involving family members enhances veterans’ motivation to engage fully and stay in treatment. Interviewees also noted that they value the “richness” and comprehensiveness that a family perspective brings to their case conceptualization. Some interviewees felt that family involvement improved the effectiveness of the treatment directly; these providers tended to especially value the experience of having the veteran share information about the trauma as part of a joint session with the clinician and a loved one. Notably, comments about family-inclusive treatment being more effective than an individual approach were slightly more commonly expressed by interviewees from low engagement sites.

The ability for the veteran to just tell their story and really explore what the trauma means to them, with their partner—it’s so different than when they do it with me. It’s really hard to even put into words, but it’s just incredibly powerful to get that support and acceptance from the person whose opinion matters the most to them, to be heard by that person. (025).

Some providers perceived that family involvement did not impact the veteran’s treatment experience either positively or negatively, compared to a fully individual approach. Finally, a small number of providers expressed a view of family-inclusive treatment as potentially less effective than individual treatment. They felt that family involvement took the focus (and therefore, the responsibility) off the veteran. They also wondered if family-inclusive protocols were as emotionally rigorous as individual EBPs, potentially dampening treatment gains. These concerns were expressed primarily by interviewees at low engagement sites.

If you’re going to do trauma-focused care, which we don’t always do, actually, but if you are going to do that, I’m not sure where the family would be involved in that, because the vet needs to do the work. (001).

I’ve wondered, even though I haven’t done [CBCT] so what do I know, but I wondered if they go as deep within the context of the couple as they would on their own into the trauma. (013).

Family Involvement in EBPs for PTSD

Across both positive deviant and low engagement facilities, many interviewees felt that family involvement was especially helpful when the veteran was engaged in Cognitive Processing Therapy (CPT) or Prolonged Exposure (PE) treatment. Many interviewees stated that they were particularly likely to involve families in a discussion prior to beginning a course of PE with a veteran, so that family members could be prepared for any temporary exacerbation of symptoms that the veteran might experience.

Because doing these evidence-based practices and addressing PTSD can be very activating and…it can be really a challenge for the veteran to maintain his composure and coping skills and emotion during some of these processes. Particularly in the beginning. And so I like the spouse to know what’s going on….[A]lmost to make the spouse have informed consent. (010).

In contrast, many interviewees stated that they were not willing to have a family member attend an actual PE session, particularly once the veteran had begun imaginal exposures, because they did not want to expose the family member to traumatic material. Some felt that it was generally challenging to integrate family-inclusive work into a busy EBP session, or worried that incorporating families (even briefly) could reduce fidelity to the protocol and therefore negatively impact effectiveness.

Positive and Negative Experiences with Family Involvement

We asked clinicians to share two cases that came to mind where family involvement in treatment was (1) helpful or even critical to the success of treatment, and (2) problematic or interfered with treatment. In terms of the positive anecdotes, clinicians mentioned cases where they were able to facilitate better communication within the dyad by having a family member participate in a session, or when psychoeducation by the clinician had helped a family member act in accordance with treatment goals (i.e., engage in less symptom accommodation) or become more emotionally supportive of the veteran. Clinicians felt that these instances supported the veterans’ individual treatment goals and retention in treatment by providing a more positive home environment. Many interviewees also mentioned situations where a veteran’s suicidal ideation increased during treatment and involving a family member for risk management allowed the veteran to remain safely in outpatient care rather than needing to be hospitalized. Family members had also helped to directly facilitate the veteran’s progress in an EBP, such as by reminding the veteran to complete homework assignments or by cuing the veteran to use specific treatment skills (e.g., cognitive reappraisal). Finally, many interviewees stressed that family encouragement, and in some cases ultimatums, were powerful motivators for veterans to seek and remain in treatment; in these cases the clinician often had not personally had contact with the family member, but the veteran’s report made it clear that their family’s wishes were a significant factor in their treatment decisions.

The vast majority of reported negative experiences involved family members who were engaging in problematic behavior outside of treatment (e.g., extensive accommodation of PTSD symptoms, negative commentary about treatment generally). In some of these cases the clinician had not had any direct contact with the family member but heard about areas of conflict or concern via the veteran. In other cases, the clinician had tried to address the issue by bringing the family member into session for psychoeducation but was ultimately unsuccessful in encouraging the family member to reduce the negative behaviors. Except in a single case when the clinician described inadvertently disclosing too much detail about the trauma during a joint session, there were no examples of family members being brought into treatment and subsequently causing more problems for the veteran or the treatment plan than previously.

A handful of clinicians raised the issue of the Caregiver Support Program (formally titled the “Program of Comprehensive Assistance for Family Caregivers” (PCAFC)) as particularly challenging to navigate with family members. The PCAFC is a benefit program through which family members can receive monthly financial stipends for supporting a veteran’s activities of daily living or for providing necessary “supervision, protection, or instruction” for the veteran (http://caregiver.va.gov). The clinicians who mentioned family members participating in the PCAFC to support veterans with PTSD (as opposed to supporting veterans with physical disabilities, or cognitive deficits resulting from a TBI, for example) almost universally expressed that they found the program frustrating because it may create a financial incentive for the veteran’s functioning not to improve. For example, a provider shared "I believe that [the Caregiver Support Program] was started with the best of intentions. And I think a lot of times it fosters illness behavior, frankly.” (016).

Family Psychoeducational Groups

Five of the facilities (50.0%) were currently running, or had run in the past, psychoeducational or supportive groups for family members; three of these were positive deviant sites and two were low engagement sites. The details of the group structure varied, but they were generally delivered monthly at most and did not have a standard protocol guiding the content. Facilities had often tried offering the group at different times throughout the week, including in some cases evening or weekend times. Interviewees described attendance at the family groups as “a constant challenge.” Some clinics had discontinued their groups due to ongoing lack of interest from families.

[W]e used to have…like a once a month family session where couples could come in with the veterans just to talk about family issues related to PTSD. At the end, you know, there were just like one or two couples who were continuing to do it consistently. We found people weren’t really using it as much. Unfortunately we actually found, the second time we tried to do this as well—a couple years ago we had an intern who tried to do a family psychoeducation group…and we really advertised it heavily to our veterans…but again, we did not get a lot of interest in that. (020).

Recipients

Knowledge

Knowledge about family-inclusive interventions was mixed. Approximately half of the sample (58.8%) had a solid awareness of at least one family-inclusive intervention (mainly CBCT or Integrative Behavioral Couple Therapy), meaning that they were familiar with at least some of the protocol elements or the clinical issues for which the treatment would be indicated, even if they did not have training or experience (or interest) in delivering the intervention. Most of the remaining participants (38.7%) had only vague or no knowledge of family-inclusive interventions (knowledge level was not assessed or able to be inferred for 6.5% of interviewees). Levels of knowledge of family-inclusive approaches did not substantially vary for providers from positive deviant sites or low engagement sites.

Staff Availability

There was some indication that family involvement could be easily crowded out by more pressing concerns, stemming from system-level factors. Interviewees noted their full clinical schedules, the pressure to complete courses of EBPs with their patients, and insufficient staffing as factors that reduced their ability to engage family members. They noted that family involvement can sometimes displace a session that would otherwise be spent on EBP content, and that communicating and scheduling with multiple people created additional administrative burdens compared to purely individual work. Staffing concerns were most often mentioned by interviewees at low engagement facilities.

“We’re only three-quarters staffed right now….It’s been a rough time more recently, and that’s where as much as I’d love to consider new things [like family involvement], I’m like—Well, let’s keep the house from burning down first, and then we’ll think about the yard.” (003).

I feel like the answer for most of us would probably be that we do not have enough time [for family involvement] but we make the time. Like, we’ll all sacrifice certain things to get it done. But I do think that acts as a significant barrier at times when you just have no more time to give. (020).

Context

Veteran-Driven, Clinic-Driven, or Ad-hoc

We identified three avenues through which family contact tended to be initiated: Veteran-driven engagement, clinic-driven engagement, or ad hoc engagement. Veteran-driven engagement was initiated by the Veteran, often in the form of bringing a loved one to an initial intake appointment unprompted. Interviewees reported that this type of unexpected family session happened with a minority of their Veteran patients and was unobjectionable when it occurred. The clinicians would generally have the family member attend some portion of the visit and found it useful to gather some collateral information. Veteran-driven engagement could also arise later in the treatment episode, such as a Veteran requesting that their family member attend a session so that the dyad could discuss an issue jointly with the therapist’s assistance or requesting that the clinician provide psychoeducation to the family member. Some clinicians reported preferring Veteran-driven engagement because the clinician wanted family involvement to arise from the Veteran’s own interest, free from any pressure.

Clinic-driven engagement was the reverse of Veteran-driven engagement: rather than waiting for Veterans to suggest family involvement, clinics had developed their own procedures to somewhat standardize family inclusion. One PTSD Clinical Team (PCT, a common title for a trauma-focused clinic within a single VHA facility) had explicitly integrated families into treatment planning. Family involvement was not a requirement for involvement with the clinic, but Veterans were strongly encouraged to bring a loved one to the treatment planning session and this was framed as the default option that most Veterans would likely benefit from. This PCT had been established in the past year (the site had not previously had a PTSD specialty clinic) and the Program Manager who had helped to determine the clinic’s procedures around intake and treatment planning described having gone “out of my way” in graduate school to take courses in couple and family therapy. This facility was a positive deviant.

“There are different ways to approach inviting families in. I mean one way to approach it is you flip it and basically say the standard of care is that family members participate in treatment unless there’s a reason why they don’t.” (012).

In contrast to the clinic where family involvement was highly integrated in, we also conducted interviews with members of one PCT where family involvement was intentionally separated out. Interviewees from this clinic (also a positive deviant facility) noted that the expectation was that veterans would engage in either CPT or PE exclusively or else be transferred to another mental health clinic, and family involvement in treatment was not encouraged. Importantly, this PCT was located at a facility with a robust Family Clinic program, and it was not uncommon for veterans seen in the PCT to be engaged with the Family Clinic simultaneously. Providers in both the PTC and Family Clinic thought the system worked well.

By far the most common approach to family engagement was ad hoc, with clinicians evaluating the importance of offering family involvement based on the veteran’s overall clinical presentation. However, it was evident that the assessment of how relevant family/relationship functioning was to PTSD treatment varied substantially across the interviewees and was related to their overall orientation towards family engagement rather than to whether their facility was rated as a positive deviant or a low engagement site.

Leadership Promotion of Family Involvement

Interviewees universally indicated that family involvement was not an area of emphasis in terms of leadership guidance to them. They stated that they “never hear about it” from their own clinic leads, or from their Service Chief or Medical Center Director. Clinicians stated that other priorities, most prominently the extent to which veterans are or are not receiving EBPs for PTSD, are much more regularly discussed.

Impact of COVID

Several interviewees mentioned that the increase in video telehealth sessions in VHA following the COVID-19 pandemic (Jacobs et al., 2021) had made family involvement in treatment more feasible. Family members were able to arrange to join a video session for as little as 15 or 30 min, whereas previously attending a veteran’s session might have taken up to 3 h after accounting for travel time each way. The issue of travel time before the pandemic was noted across both highly rural and highly urban facilities. Additionally, clinicians reported that they had more incidental contact with family members during video sessions, largely in cases where the veteran was being helped by a more technologically-adept loved one.

Discussion

Our results indicate that clinicians felt positively about family involvement; they recognized the significant impact of PTSD on veterans’ social relationships and largely felt that attention to these relationships made sense in the context of PTSD treatment. Many felt that family involvement enhanced treatment, relative to a strictly individual approach, but a substantial number of clinicians lacked confidence in their personal ability to deliver family-inclusive care due to their minimal experience in this area. Many of our participants had little or no knowledge of family inclusive treatments, were not familiar with the research supporting family involvement, and had not been able to conduct family-focused clinical work under supervision. In contrast to their broad background in individual EBPs (every provider we spoke to was formally “VHA certified” in at least one, and frequently multiple, protocol-based treatments), they often felt uncertain about family-inclusive work, especially anything that would extend beyond pure psychoeducation. Nonetheless, some level of family involvement was often deemed to be especially important in the context of heavily trauma-focused approaches such as CPT and PE. Most clinicians were happy to hold a single family session to support the veteran’s efforts in an EBP or when a veteran expressed particular concern about the impact of PTSD on the family. Those providers who had been able to obtain formal training in family- or couple-focused approaches were enthusiastic about incorporating family members for one or more sessions.

Our results largely suggest that differences in family involvement are highly driven by clinician-level attitudes and past experiences, especially formal training and supervised clinical work. There were interviewees who were passionate about, or more hesitant about, family involvement in PTSD treatment at both positive deviant sites and low engagement facilities. For example, providers at low engagement sites were slightly more likely than providers from positive deviant sites to express that family-inclusive treatment was generally more effective than individual treatment, and also more likely to express concerns that family-inclusive work was not as “deep” as individual work. Clearly these comments came from different interviewees within the low engagement facilities, but these results indicate a wide range of opinions about family involvement in treatment at the facility level. The one area with the most consistent difference between positive deviant and low engagement sites was staffing; providers at low engagement facilities were more likely to mention being understaffed as a contextual factor that reduced their ability to conduct family sessions.

Lack of time and an abundance of competing priorities were prominent themes throughout the interviews. The clinicians had heavy caseloads full of complex patients, and frequently served in multiple administrative roles as well (e.g., internship training director, clinic director). They regularly received feedback about the importance of providing EBPs for PTSD to their veteran patients; in contrast, messages about family involvement were minimal or nonexistent. The emphasis on EBP provision reflects the underutilization of these individual trauma-focused treatments in VHA (Maguen et al., 2020), despite vigorous institutional efforts (Karlin et al., 2010). EBPs are the treatments that have the most evidence for meaningfully reducing PTSD symptoms (U.S. Department of Veteran Affairs and Department of Defense, 2017), and they must be prioritized. However, accumulating evidence suggests that certain family member behaviors (Meis et al., 2019; Sayer et al., 2017), or the incorporation of families into treatment (Thompson-Hollands et al., 2021b), enhances treatment retention in CPT and PE and therefore that these approaches may strengthen rather than detract from EBPs. For clinicians looking to increase their use of family involvement to support EBPs there are several lower-intensity protocols available, including Veteran-Centered Brief Family Consultation (U.S. Department of Veterans Affairs, 2018; VA Family Services) and the Brief Family Intervention for PTSD (Thompson-Hollands et al., 2021c; Thompson-Hollands et al., 2021b); both are specific to veterans and typically range from 1 to 3 sessions.

Clinicians described three “paths” to family involvement: veteran-driven, clinic-driven, or ad hoc. Ad hoc was most common, and the resulting patterns of family involvement for a given clinician (relatively high or relatively low, per the clinician’s own report) were highly linked to that individual’s past training and experience with dyadic or family treatments. When the choice to involve family members is ad hoc, there is some risk that family involvement will be offered according to veteran or clinician factors that are not empirically based; for example, assumptions or implicit biases about certain demographic groups. Indeed, an examination of VHA administrative data found that Black post-9/11 veterans were 59% less likely than White post-9/11 veterans to have a family-involved mental health visit (Harper et al., 2022), suggesting that unequal use (and therefore potentially outcomes) is possible. This is compounded by data indicating that historically marginalized groups have poorer rates of access to and utilization of PTSD treatment (e.g., Pole et al., 2008). There is currently no literature to guide clinicians definitively regarding for which patients under which circumstances family involvement would be beneficial, potentially making the implementation of family involvement in PTSD treatment particularly vulnerable to implicit bias or facilitation barriers.

Veteran-driven and clinic-driven approaches were comparatively rare. Clinicians who described taking a largely veteran-driven approach to family involvement were concerned with not pressuring veterans into having a family member participate. This strategy likely produces relatively low rates of family engagement, as some veterans may not realize that family involvement is an option. Furthermore, this approach is not consistent with VHA guidelines, which explicitly require a conversation about family involvement at least annually. Finally, two clinics had developed a general stance towards family involvement (either actively encouraging it or somewhat discouraging it), effectively incorporating this attitude towards family into their clinic mission. The centrality and emphasis of this stance was much stronger at the site that was encouraging of family involvement, compared to the site that minimized family involvement. Sayer et al. (2017) found that clinic mission was a powerful explanatory factor in the extent of EBP “reach” across different VHA PCTs; teams that saw EBP provision as core to their identity were providing EBPs at a rate well above the national average in VHA. Both clinics that we identified as taking a “clinic-driven” approach were at positive deviant facilities; however, the PCT team that generally discouraged family involvement had an active family clinic at their facility which likely explained their high rates of family involvement overall.

Psychoeducational groups for family members were being conducted, or had been conducted in the past, at half of the facilities at which we interviewed. Unfortunately, these tended to be poorly attended by families and something of an afterthought for clinicians, despite good intentions on the part of the group leaders. Groups are often an appealing approach from the perspective of providers—they are efficient, potentially meeting the needs of many separate families with only 60–90 min of clinical time and are scheduled at the convenience of the clinic rather than conforming to the idiosyncratic availability of many different family units. It is possible that families themselves are less motivated to attend a group-based program (perhaps because it feels general rather than specific to their family dynamic, or perhaps because they are uncomfortable with the somewhat “public” nature of a group), or the designated schedule may be incompatible with the other demands on their time. Despite the administrative convenience of a group format, family involvement may perhaps be most effectively implemented through one-on-one sessions with families. Data from our interviews suggests that with the increased use of VVC appointments as a result of the COVID-19 pandemic, some scheduling barriers associated with such an individual approach can be more easily surmounted.

We gathered anecdotal reports of providers’ experiences with family involvement, both positive and negative. Specific positive examples that participants recalled tended to be crisis situations where the family member was able to provide additional support for the treatment or safety in response to a veteran’s suicidal ideation, though there were also several instances in which family involvement was beneficial for improving communication or connection in the dyad. The negative experiences raised by clinicians were not generally instances where deliberate involvement of the family member into treatment caused or exacerbated some problem. The large majority of the interviewees recounted situations where a family member had been difficult outside of the therapy context and either the clinician had not tried to directly involve the family member in the veteran’s treatment (perhaps based on clinical judgement that such involvement would be iatrogenic) or the clinician’s attempts at family involvement did not resolve the problem. While these instances may stand out as canonical examples of families being challenging (and virtually all clinicians reported having had this type of negative experience), very few of our interviewees recalled an instance where deliberate family engagement by the clinician had negatively affected the veteran’s care.

This study was limited by the reliance on administrative data to identify positive deviant and low engagement sites, which likely underestimates family involvement due to inaccurate coding of encounters. Furthermore, as noted we found low rates of family involvement in VHA generally, and thus similarities in findings across groups may reflect this lack of variability between positive deviant sites and low engagement sites at the facility level. It is possible that facility-level factors would become more clear differentiators if there was greater variability in family involvement across VHA sites. Finally, we acknowledge that this study addressed only the perspectives of clinicians and does not include interviews with veterans and family members.

Families remain an underutilized resource within VHA, despite clear indications that veterans desire family involvement (e.g., Batten et al., 2009) and preliminary evidence that family inclusive care may offer a bolstering effect to trauma-focused treatment (Thompson-Hollands et al., 2021b). Our results suggest that there are multiple avenues that could be pursued to more emphatically promote family involvement in care. These include offering providers more opportunities for supervised couple- and family-focused training, increasing staffing levels to increase provider flexibility, and more prominently emphasizing family-inclusive care in communication with providers. Future research should test the implementation of some or all of these strategies in enhancing uptake of family-inclusive visits, as well as any downstream effects on veteran outcomes and satisfaction with care.