In the United States, veterans of the Armed Forces die by suicide at an age- and sex-adjusted rate 71.8% greater than non-veteran adults (1). Veteran suicide prevention is a top priority of the United States government including the Department of Defense, Veterans Affairs, and Health and Human Services (2). The Department of Veterans Affairs has a multicomponent strategy to prevent veteran suicide, including promoting effective treatments for veterans identified to be at risk (2, 3). Many different mental health disorders are associated with increased suicide risk among veterans (4), suggesting a transdiagnostic approach focusing on cross-cutting risk factors is needed. Emotion dysregulation, or difficulties regulating emotions, occurs across mental health diagnoses and has been identified as a transdiagnostic treatment target (5). Emotion dysregulation is associated with suicide attempt frequency among those at risk for suicide (6), even when controlling for other risk factors (7, 8). Skills training in emotion regulation is a critical component of effective treatments for reducing suicide attempt (9).

Dialectical Behavior Therapy (DBT) is an intervention focusing on emotion regulation skills training (10) that is well supported for reducing self-directed violence, including suicide attempt (11). While DBT was originally studied in individuals with borderline personality disorder (BPD), a disorder characterized by high levels of emotion dysregulation (12), DBT has increasingly been utilized with a broad range of individuals at elevated risk for suicidal behavior (e.g., 13, 14), including veterans (15). Comprehensive DBT is a multi-component intervention including individual therapy, a skills training group providing instruction in emotion regulation and other skills, phone coaching between sessions, and a therapist consultation team (10). In 2020, VHA launched the “Suicide Prevention 2.0” program, with both community- and clinical-based intervention strategies (2). Since 2021, the clinical arm of this initiative, SP 2.0 Clinical Telehealth, has partnered with the VHA regional Clinical Resource Hubs, which provide telehealth to veterans, to provide evidence-based treatment for suicide prevention to veterans with a history of suicidal self-directed violence. Comprehensive DBT is one of the four treatments offered, specifically for veterans who are also living with BPD (16).

DBT is a complex intervention that requires some unique treatment elements, such as two group leaders for each skills group session, weekly consultation team for therapists, and access to therapists between sessions (17). Unpaid or unique elements of DBT may conflict with existing healthcare system structures or clinicians’ schedules (17). Implementation barriers to comprehensive DBT in public health settings are well documented in qualitative or mixed-method studies, and include concern about the fit of the intervention with the clinic’s pre-existing practices or structure (18), limited staff time or competing productivity demands (18–21), insufficient administrative support (19), absence of or limited administrator buy-in (19, 20), low resource availability (20, 22), challenges related to phone coaching (18, 21), the perceived difficulty of implementing DBT (22), and turnover of trained practitioners (18–20, 22). Concerns about the degree of fit between DBT and the healthcare system’s current practices may contribute to only some modes of DBT being implemented (17, 23), or even to healthcare settings stopping DBT implementation (23).

Newer data suggest the DBT skills group, supported by therapist consultation team, may be the treatment’s active ingredient. In adults with BPD at high risk for suicide, DBT skills group with consultation team (DBT-SG) and rigorous risk management and individual manualized case management services reduced suicide attempt and ideation with the same efficacy as comprehensive DBT (24). Similarly, among individuals with BPD, DBT-SG resulted in reduced suicidal and nonsuicidal self-injurious behavior and reduced affect instability (24). Additionally, DBT-SG increased coping or distress tolerance skills (25, 26) among BPD and transdiagnostic samples. DBT-SG has been applied to a wide variety of diagnoses, suggesting its utility well beyond the BPD diagnosis (27). While the DBT-SG innovation has high potential to reduce suicide attempt and emotion dysregulation across a range of mental health disorders, its implementation determinants have not been prospectively studied.

Successful implementation of an innovation requires understanding context-specific factors (17), including the availability of behavioral health providers (28) and their workload demands (29). DBT-SG innovation features, such as the therapist consultation team, may also affect implementation (17, 28) or require specific implementation strategies (30). To understand implementation determinants of DBT-SG among veterans at high risk for suicide attempt, we conducted a mixed-method formative evaluation, guided by the Integrated Promoting Action on Research Implementation in Health Services (i-PARIHS) framework (31), to identify how characteristics of the DBT-SG innovation, inner and outer context, and recipients may impede or assist DBT-SG implementation. This formative evaluation was conducted at outpatient mental health clinics across and within four VHA medical centers as a prelude to a multi-site effectiveness trial.

This mixed-method formative evaluation of implementation determinants to DBT-SG at four VHA sites found that quantitative and qualitative data both showed VHA staff had favorable views of the innovation overall. Furthermore, both quantitative and qualitative data converged to elucidate staff views about inner context barriers (i.e., site-level), such as provider shortage and concern for local resources. Qualitative data highlighted potential facilitators, such as local leadership support for implementing DBT-SG, and national policy drivers that prioritized suicide prevention interventions for high-risk veterans.

ORCA scores showed VHA staff agreed that DBT-SG was supported by evidence from research, patient preferences, and clinical experience. Qualitative data expanded on this to show VHA staff’s interest in suicide prevention interventions that can be transdiagnostic, consistent with reviews showing DBT skills interventions have been studied well beyond individuals with BPD (27). Previous clinical practice guidelines for VHA offered support for DBT for veterans at risk for suicide who have BPD (45), and the current Suicide Prevention 2.0 initiative extends DBT to veterans with history of suicidal self-directed violence who have BPD (16). While conserving resource-intensive treatments like comprehensive DBT for those who are most likely to benefit based on similarity to research findings is reasonable, our data suggest VHA staff see transdiagnostic utility in the DBT-SG intervention for veterans at risk for suicidal behavior, beyond the BPD diagnosis.

Qualitative data showed VHA staff awareness of the wide range of diversity that might influence how veterans interact with DBT-SG concepts. United States ethno-racial and sexual orientation minority groups are well represented in comprehensive DBT randomized controlled trials, supporting the use of comprehensive DBT for individuals from a wide range of demographic identities (46). To ensure DBT-SG at VHA is delivered with attention to veteran diversity, we recommend that DBT-SG providers in VHA use therapist consultation team to support delivering treatment with cultural competence and to ensure effective treatment of veterans who experience racism and other forms of minority stress (47). Regarding recommended DBT-SG therapist characteristics, our data are consistent with other DBT implementation literature emphasizing the importance of therapist qualities such as cognitive flexibility or open-mindedness (23, 30), assertiveness or interpersonal effectiveness (23, 30, 48), therapist interest or investment in the treatment model (18, 21, 23), and adequate training in the treatment (21, 23, 30).

Within each medical center (i.e., inner context), quantitative data ( Table 2 ) showed a belief that VHA staff have a personal commitment to improving veteran care and outcomes, which combined with staff perception of DBT-SG as having strong evidence as a suicide prevention treatment based on clinical trials, clinical experience, and patient preference, bodes well for implementing DBT-SG in VHA medical centers (49). However, there was less confidence in the adequacy of resources or leadership behavior to support DBT-SG implementation. This pattern of higher ORCA mean scores for the evidence of an innovation relative to contextual leadership support for implementing it is similar to prior studies using the ORCA within three quality improvement projects conducted within VHA medical settings (37), VHA addiction treatment programs adopting hepatitis prevention practices (49), and academic medical centers promoting medications for alcohol use disorder (50) and buprenorphine (51). Qualitative data showed that both providers and local administrators were invested in providing DBT-SG, but aware that limited staffing could be an insurmountable barrier. Unfortunately, high workload and insufficient staffing have been barriers to previous VHA provision of DBT (21) and other evidence-based psychotherapies (52, 53).

In examining quantitative responses by roles, no differences emerged for the evidence support for DBT-SG based on whether the respondent was a study therapist or referring provider, or had a different role ( Table 3A ), or whether the respondent held a leadership role at the local VHA medical center ( Table 3B ). However, study therapists and referring providers, who would have the most direct relationship to providing or recommending DBT-SG, showed lower confidence that the organizational context (leadership culture, staff culture, leadership behavior, feedback from leadership, and resources) would support DBT-SG implementation, in comparison to respondents in leadership roles ( Table 3A ). Similar results on the resources subscale emerged in another pre-implementation study: prescribers who would directly prescribe medication for alcohol use disorder in an inpatient medical setting showed lower resource subscale scores (M = 2.8, 95% CI (2.5, 3.2), n = 26) than non-prescribers (M = 3.3, 95% CI (3.0, 3.6), n = 27 (50).

In our data, holding a leadership position at the local facility also differentiated some results in the ORCA context subscales: individuals who were in a leadership role (hospital director, administrative leader, clinical supervisor) showed greater confidence in leadership culture, leadership behavior, feedback from leadership, and resources ( Table 3B ), suggesting that leaders had a more positive perception of the implementation context for DBT-SG than those without leadership responsibilities, despite similar views on the evidence for DBT-SG. It is unclear which perception is more accurate. DBT clinicians have previously pointed out that organizational support is critical to implementation success (20, 21). In a prior VHA DBT program evaluation (21), DBT team leaders or point of contact identified barriers their site was working on or could not overcome included lack of individual therapists, lack of therapist availability to take on enough patients, inability to schedule a two-hour group, and inability to block time for consultation team. Similarly, in a large evaluation of DBT program sustainability in the UK, absence of organizational support was the most commonly identified implementation barrier, while presence of organizational support was the most commonly identified implementation facilitator (20).

Our data and prior literature suggest that implementing DBT-SG at trial sites may benefit from an implementation strategy that emphasizes the perspectives of DBT-SG clinicians and referring providers, such as implementation facilitation (31), a dynamic implementation strategy that includes problem-solving, enabling, and supporting implementation partners or stakeholders as they implement a clinical innovation into routine practice (54). Core implementation facilitation activities emphasize engaging stakeholders or partners to evaluate context, resources, and barriers, and then assisting them in tailoring or adapting a clinical innovation to their setting (55). Implementation facilitation requires targeting all groups and stakeholders who can impact implementation, including clinical providers who will implement the innovation, providers and staff who would refer to the innovation, and organizational leaders who could support, or inadvertently impede, implementation efforts (56).

Regarding outer context, in qualitative data VHA providers emphasized that their leaders are invested in suicide prevention, suggesting the outer context of policy and initiatives focused on suicide prevention (2) is evident to providers. However, VHA staff noted awareness that national leadership need to devote adequate resources to DBT-SG: as one local administrator said, “we have to prioritize this if it saves lives.” Across VHA, mental health clinical provider staffing levels were causally linked to lower probability of veteran suicide-related events (57), suggesting the need to increase staffing levels or improve the efficiency of currently available providers. In fiscal year 2024, VHA facilities surveyed identified severe staffing shortage in clinical mental health occupations including psychology (identified as a severe shortage by 61% facilities), psychiatry (47%), and social work (43%; 58). To further prioritize veteran suicide prevention, national leadership may need to consider prioritizing funding to support greater clinical VHA mental health provider staffing (57).

Strengths of this investigation include use of an established implementation science framework to guide study interview development and survey selection, mixed methods to examine the complementarity and expansion of data and fully capture staff views, data gathering across four sites and different categories of staff, and sampling to saturation for qualitative interviews. Limitations, as with other mixed method studies, include that these findings are not necessarily expected to generalize to other settings. Veteran experiences and views were not captured in these data and will be included in subsequent formative evaluation during the ongoing hybrid effectiveness-implementation trial.

Further mixed methods directions include qualitative interviews with veterans at high-risk for suicide attempt who participated in DBT-SG, examining the experiences of those who completed the treatment and those who dropped out (59, 60), using purposeful sampling to understand variation in participant experience (61). We will also gather interview- and ORCA-derived data about DBT-SG therapist, referring provider, and leadership perspectives on how to best facilitate the implementation of DBT-SG based on their views of the use of DBT-SG during our trial. This information will inform development of an implementation strategy for DBT-SG, should the treatment demonstrate effectiveness in improving emotion regulation of veterans and reducing their rates of suicide attempt.