Al-Anon Intensive Referral (AIR): A Qualitative Formative Evaluation for Implementation

Abstract

Introduction:

The Al-Anon mutual-help program helps concerned others (COs; e.g., families, friends) of persons with excessive alcohol use. Despite widespread availability of Al-Anon meetings, participation is limited and little is known about how to best facilitate engagement. Al-Anon Intensive Referral (AIR) was developed to facilitate COs’ engagement in Al-Anon and is being tested in a randomized controlled trial (RCT). Toward the end of the recruitment for the RCT, a qualitative formative evaluation was conducted to learn about facilitators, barriers, and recommendations for AIR implementation in substance use disorder (SUD) treatment clinics.

Methods:

Thirty-one directors and staff at ten VA and community SUD clinics were interviewed. Semi-structured interviews were based on the Consolidated Framework for Implementation Research and were thematically analyzed to identify facilitators, barriers, and recommendations for AIR implementation.

Results:

Perceived facilitators of AIR implementation included AIR’s face validity, adaptability, and alignment with staff values and skills, requiring only minimal training. Several interviewees in community settings thought AIR would fit with their current practices (e.g., family groups), and some clinics reported having sufficient staff available for delivering AIR. Perceived barriers included limited staff time, and VA clinics having limited resources for providing services to COs. Furthermore, many clients have no COs, or COs who are unwilling or unable to engage with them. Recommendations included fitting AIR within existing workflows and focusing on COs with highest readiness to receive support. Interviewees also thought AIR could be adapted to a website format or smartphone app, which may expand its reach while decreasing staff burden and cost; however, it may not be as effective and appealing to some demographic groups (e.g., older COs).

Conclusions:

AIR has strong potential for implementation in SUD treatment settings, but clinics vary on implementation capacity. Most clinics could implement AIR partially (e.g., case-by-case basis) while clinics with sufficient capacity (e.g., staff time) could implement it more fully. These findings can also inform implementation of other interventions for concerned others.

1. Introduction

An estimated 15 million individuals in the United States have an alcohol use disorder (AUD), the most prevalent substance use disorder (SUD; Substance Abuse and Mental Health Services Administration, 2020). In addition to those having AUD, at least another 15 million of their family members also likely experience negative consequences associated with the individual’s AUD (Orford et al., 2013), and counting friends and other “concerned others” (COs)¹ would add millions more (Timko et al., 2019). While there is no diagnosis for COs, the addiction of their loved ones affects them in many ways, contributing to the public health concerns for these populations. COs can have reduced quality of life (Birkeland et al., 2018; Casswell et al., 2011; Karriker-Jaffe et al., 2018), physical health (Casswell et al., 2011) and mental health (Karriker-Jaffe et al., 2018; Ray et al., 2007; Ray et al., 2009; Weisner et al., 2010), have strained relationships (Casswell et al., 2011), often suffer physical violence (Dawson et al., 2007), and endure higher healthcare costs (Ray et al., 2007; Ray et al., 2009; Weisner et al., 2010). COs require knowledge and skills to cope with problems and harms associated with someone else’s SUD (Copello & Orford, 2002; Copello et al., 2005; Hussaarts et al., 2012; Timko et al., 2013, 2019). Interventions to help COs can have significant barriers to their implementation in practice, such as lack of reimbursement (Miller, 2003) and providers’ focus on treating the individual with SUD (Copello & Orford, 2002). Accordingly, relatively brief referral interventions to existing and effective programs may have fewer implementation barriers and reach more COs for an overall bigger impact. One widely available and free program that can help COs is Al-Anon Family Groups (Al-Anon).

Al-Anon is a mutual support program for COs that developed in parallel to Alcoholics Anonymous, where members attend peer-led meetings, follow and work the 12 steps, find a sponsor, read Al-Anon literature, and engage in spiritual practices (e.g., prayer, meditation; Timko et al., 2012). Worldwide, there are over 24,000 Al-Anon groups (about half of them in the US) in 118 countries (Al-Anon Family Groups, n.d.). Al-Anon’s 2018 member survey indicates that most members are female (85%), married (60%), college graduates (66%), white (92%), and on average 61 years old (Al-Anon Family Groups, 2018). Al-Anon can improve COs’ outcomes, including mental health and overall wellbeing (Al-Anon Family Groups, 2018; Cutter & Cutter, 1987; Dittrich & Trapold, 1984; Keinz et al., 1995; Miller et al., 1999), coping (Gorman & Rooney, 1979; McGregor, 1990; O’Farrell & Fals-Stewart, 2003), and relationships (Timko et al., 2013).

While Al-Anon is widely available and can help COs, engagement in Al-Anon is relatively low, and more than half of new members (57%) stop attending within 6 months (Timko et al., 2014). Best practices for facilitating Al-Anon engagement do not yet exist; however, the Al-Anon membership survey indicates that 84% of members began attending meetings because of a professional referral (e.g., counselor, therapist; Al-Anon Family Groups, 2018), indicating that this may be a promising avenue for facilitating engagement.

1.1. Al-Anon Intensive Referral

Al-Anon Intensive Referral (AIR) is an intervention developed for facilitating engagement in Al-Anon. It is based on prior studies of intensive referral that showed this brief intervention can successfully facilitate engagement in 12-step programs among people in treatment for substance use disorders (Grant et al., 2018; Timko et al., 2006, 2011). AIR consists of four sessions conducted by a Coach over a span of about three months. The Coach conducts the initial two sessions as in-person group meetings or individually by phone, and the two booster sessions individually by phone. In these sessions, the Coach introduces COs to Al-Anon (history, purpose, meeting locations and times), encourages Al-Anon meeting attendance, and discusses COs’ experiences attending the meetings in subsequent sessions. Through these discussions, the Coach tries to reduce ambivalence and address barriers to attending Al-Anon meetings, addresses questions or concerns, and connects COs with Al-Anon volunteers for further information and guidance.

AIR is currently being evaluated for effectiveness in a randomized controlled trial (RCT) at community and Veterans Affairs (VA) SUD treatment programs in Arkansas, California, and Nebraska. Study staff recruited persons in treatment for AUD and asked them to identify COs. Study staff then contacted COs and randomized those who agreed to participate to receive the AIR intervention delivered by study staff or “usual care” (family education groups).

To inform future adaptations and implementation of AIR in routine practice, where we expect it would be delivered by SUD clinic staff, we interviewed clinical directors and staff at SUD clinics to qualitatively examine their perceptions of barriers, facilitators, and recommendations for implementing AIR in practice. This paper focuses on the findings from these interviews.

2. Methods

2.1. Study design and sample

Consistent with the Hybrid Type I effectiveness-implementation design (Curran et al., 2012) and in conjunction with the Al-Anon Intensive Referral (AIR) RCT, this study used a qualitative formative evaluation design to learn about barriers, facilitators, and recommendations for implementing AIR (Stetler et al., 2006). To learn from a variety of perspectives, the study aimed to recruit one clinical director and two staff members at ten SUD clinics, for a total of 30 key informants. Eight of these clinics participated in the AIR trial (two thirds of the 12 clinics participating in the trial), and two were naïve clinics (i.e., clinics unfamiliar with AIR). Including naïve clinics allowed us to learn from directors and staff with no prior knowledge of AIR, which can help mitigate potential selection bias from clinics participating in the trial (who may be “early adopters” of innovations). The University of Arkansas for Medical Sciences and Central Arkansas Veterans Healthcare System IRBs determined that this study is not human subjects research.

2.2. Data collection

To recruit interviewees (key informants), study liaisons at clinics participating in the RCT and senior leadership at naïve clinics helped identify clinical directors and staff to participate in the interviews. At all stages of the formative evaluation process, we stressed to interviewees that their participation was voluntary, and that their responses would remain confidential and anonymous. Most interviews were conducted by telephone to facilitate scheduling and minimize work interruption. The interviews were, on average, 30 minutes long, and the study provided no participant compensation.

The semi-structured interview guide, based on the Consolidated Framework for Implementation Research (Damschroder et al., 2009), guided data collection on implementation determinants from five domains: intervention characteristics (e.g., relative advantage, cost), outer setting (e.g., client needs, policy environment), inner setting (e.g., learning climate, leadership support), staff characteristics (e.g., readiness for change, knowledge and beliefs about intervention), and process (e.g., planning, engaging change agents). We also asked the interviewees what they thought about adapting AIR as an online or smartphone app, which may have the potential to reach more people and at a lower cost. During interviews with key informants from naïve clinics, we also described the key elements of the AIR intervention (its purpose, number of sessions, format, and content) and answered any questions about them. JB, who at the time was a postdoctoral fellow with multiple years of experience in qualitative methods, conducted all interviews. All interviews were recorded (after obtaining verbal consent) and later transcribed verbatim before analyses.

2.3. Analysis

We thematically analyzed the interviews using both deductive (a priori codes, i.e. barriers, facilitators, recommendations, and domains of the Consolidated Framework for Implementation Research) and inductive (emergent codes) approaches (Ryan & Bernard, 2003). JB and GC first independently read six (~20%) randomly selected transcripts and coded them for barriers, facilitators, and recommendations. They then compared the coded content and resolved any differences in coding until 100% agreement was reached. After ensuring a shared understanding of codes, JB coded the remaining transcripts independently. Finally, JB examined the content of each theme to identify the most prominent response patterns, which he compared and grouped together into subthemes using comparative analyses (Gibbs, 2007), and organized them by Consolidated Framework for Implementation Research domains (Damschroder et al., 2009). Finally, JB examined whether there were any differences in responses by clinic and key informant characteristics (e.g., VA vs. community setting, RCT vs. naïve clinic, director vs. staff perceptions).

3. Results

We conducted interviews with 31 key informants at 10 SUD clinics between October 2018 and April 2019. Eight of the clinics were part of community and VA programs located in Arkansas, California, and Nebraska that participated in the Al-Anon Intensive Referral (AIR) RCT, and two community clinics in Arkansas had no prior knowledge of AIR (“naïve” clinics). Eight of the clinics offered residential services, and two offered intensive outpatient services. Key informants at each clinic included one clinical director and two or three staff in clinical roles (e.g., psychologists, counselors, peer support specialists, nurses) or administrative roles (e.g., marketing, intake coordinator).

The study identified several themes of facilitators, barriers, and recommendations, organized along the five domains of the Consolidated Framework for Implementation Research in the paragraphs below. To illustrate the themes, we included exemplar quotes in Table 1. While the themes were often inter-related, we presented them separately for clarity. There were no differences in themes between clinic types and key informant groups, except for some differences between community and VA clinics; we point out these differences for the themes in which they occurred.

Table 1:

Exemplar Quotes for the Themes Identified in Qualitative Interviews

Theme	Exemplar quote
Characteristi cs of the intervention: Al-Anon Intensive Referral’s structure and evidence	• I’ve been doing this a decade, a little over twelve years, and I’ve never heard an individual talking about what you’re talking about. And I know in my heart and spirit, we need it. What we’ve been doing is hoping people would go in there and go and do it. You’re trying to guide them into it, and it’s a benefit so we need to try to do it. ~Case Manager at Clinic 9 (Naïve) • I like the model, I think that this would work in the hands of the right person that has the amount of time to do it. I like how it’s laid out with the sessions and all, makes perfect sense to me. The motivated person is going to follow these steps with the right individual providing it. ~Director at Clinic 1 (RCT) • I know within the confines of your research project it had to be people with alcohol use disorders… But in the, in the real life implementation, you know, not having limitations like that [would work better]. You know, where it’s just open to anyone who is and willing rather than this group or that group. ~Director at Clinic 2 (RCT)
Outer setting: Concerned others’ needs, relationships, and readiness for change	• I think it [AIR] is needed, because a lot of time families don’t understand. And then you got some that understand but they don’t understand that they’re hurting, hurting their family member … especially with mothers and sons, and they still think they got to take care of them. ~Director at Clinic 9 (Naïve) • I find that like there’s family members that are like super involved and really willing to come in, or family members that are not at all willing. I think when family members are actually involved it’s typically because they care a lot and they want resources to be helpful. ~Social Worker at Clinic 5 (RCT) • I mean, once somebody has been in treatment seven times, mom and dad are probably like, “I’m tired of doing this.” So we’ve had clients that say, “Yeah, no one’s, they’re not coming because they’re tired of having to see me in places like this.” ~Director at Clinic 6 (RCT) • Another barrier, I see a lot of resentment. Yeah, actually just wanting to say “I’m done,” and not being willing or open to giving that person another chance. ~Counselor at Clinic 10 (Naïve)
Inner setting: Clinic cultures, practices, and staffing	• [We] try to be very kind of on the front edge of implementation of things that are innovative and best practices. ~Director at Clinic 2 (RCT) • I think it [AIR] is something that we can implement in the family group. ~Counselor at Clinic 4 (RCT) • I think that it would be great…We just don’t have the capacity to serve both veterans and their family members, which is unfortunate… [we could] implement it on a case-by-case basis. ~Social Worker at Clinic 5 (RCT) • We’re not all about not having time to do it, we’re all about making time to do it. ~Case Manager at Clinic 9 (Naïve) • We would write the procedures such that there was a requirement in a person’s job description. Because if, if we just say, “John Doe is going to be the champion of this,” that’s fine as long as John Doe is working in that position in that location and doesn’t get promoted or quit or something else. So we’ve got to have it where it’s not just a person by name. ~Director at Clinic 4 (RCT)
Characteristics of individuals: Staff attitudes, knowledge, skills, and status	• The [counselor] that we have doing Sunday afternoon [family education group]… she’s well educated, she understands recovery from all angles, she would be a dynamic person to do that [AIR]. ~Director at Clinic 1 (RCT) • If it was something that we’re like “That doesn’t fit,” maybe there’d be more pushback…But if it’s like a valuable resource and it’s identifying a need, I think it would, you wouldn’t get pushback. ~Social Worker at Clinic 5 (RCT) • We’ve got some staff members who themselves are very friendly to Al-Anon…and so I think I could point to one or two people who would probably want to champion it. ~Director at Clinic 7 (RCT) • There would probably need to be some training on and about Al-Anon. Everyone’s familiar with the term and what it’s supposed to do, but I don’t know that everybody has a perfect understanding of it. So being more educated about Al-Anon would be one thing I think that we would need. ~Therapist at Clinic 7 (RCT)
Process: Leadership support and logistics	•I think there’d be a number of sort of folks from leadership to be involved with that [AIR implementation]. You know, our nurse manager, medical director. ~Director at Clinic 3 (RCT) •I would be your best person to make it happen. I’m the [clinical director] and I can take it directly to the mental health leadership. And if they see the merit in it we can do it. And they look to me to be the one evaluating and presenting it. ~Director at Clinic 7 (RCT) •I think the materials that you guys were utilizing…having that be available for us to, at a minimum, print off, because, you know…having them in PDF form or whatever for us to be able to provide to people. That would be a good start because I found when starting programs or rolling out new aspects of the program, creating materials takes a whole lot of time. ~Director at Clinic 2 (RCT)

Notes: RCT = Clinic participated in the AIR trial; Naïve = Clinic unfamiliar with AIR

3.1. Characteristics of the intervention: Al-Anon Intensive Referral’s structure and evidence

Facilitators:

In general, interviewees perceived Al-Anon Intensive Referral (AIR) positively. While recognizing that AIR’s effectiveness is not yet known, interviewees thought that the AIR model makes sense (strong face validity). Many interviewees also positively perceived Al-Anon and other 12-step programs, recognized such programs as useful and widespread resources available to clients and their concerned others (COs), and welcomed additional efforts to educate and encourage COs to engage in these resources.

Barriers:

The time horizon to complete AIR (~3 months) is longer than some short-term residential stays (~1 month), which some interviewees thought could impede the clinics’ ability to deliver AIR fully (all four sessions) because they could not dedicate resources to serve COs of clients that have already completed treatment. Furthermore, while Al-Anon is generally well-perceived, it is not available everywhere, and some staff and COs may prefer other supportive resources available in the community. Some interviewees thought that AIR’s exclusive focus on AUD and Al-Anon limits its appeal and potential for a positive impact. Also, one VA clinical director thought that it may be legally challenging for VA clinics to “promote” one community resource above others.

Recommendations:

Several interviewees recommended expanding the AIR model to include other community resources, including other 12-step programs for COs (e.g., Nar-Anon, Celebrate Recovery), particularly where Al-Anon is not available in the community. Furthermore, while the evidence base for AIR is still developing, interviewees highlighted several types of evidence that they would find helpful and persuasive when deciding to implement AIR, from robust (and ideally replicated) findings on improved behaviors and outcomes (e.g., improved functioning and well-being of COs, improved client abstinence) to highlighting cases of success stories. Finally, several interviewees emphasized that adapting AIR for online websites or smartphone apps could enable unique ways of fostering engagement in AIR, including using interactive content (e.g., videos, personal stories, expert opinions) and tailoring of the content based on location and other demographic data (e.g., to resources available in each community). While some interviewees thought that without personal contact, apps can feel “cold and indifferent,” and that fostering group support in online/virtual settings is particularly challenging, others thought that integrating social media could help provide elements of group support in a virtual environment.

3.2. Outer setting: Concerned others’ needs, relationships, and readiness for change Facilitators:

Most interviewees recognized that COs would benefit from participation in Al-Anon, and thus that AIR would fill an important service gap. Interviewees thought that several CO groups may be particularly responsive to AIR and more likely to engage in Al-Anon, including close female family members (e.g., wives, mothers), those with more education, and COs of clients who are seeking treatment for the first or second time.

Barriers:

Several interviewees expressed the concern that some clients have poor or non-existent relationships with COs (e.g., they “burnt their bridges”, are homeless), or may simply not want their COs involved or contacted. Furthermore, some COs may lack readiness to seek support, due to either lack of information (e.g., not knowing what Al-Anon is and how it could help) or unwillingness to engage, particularly if they do not perceive themselves as having any problems. Some COs may also experience access barriers, due to lacking transportation or time, or both. These barriers may be particularly relevant for COs living in rural areas, who may have to travel long distances to attend Al-Anon meetings. Finally, due to the VA policy to provide services to veterans, VA providers may be limited in their ability to provide AIR to non-veteran COs.

Recommendations:

While interviewees recognized that identifying and engaging COs may be difficult, they identified several potentially helpful strategies. Targeting COs with highest readiness to engage in supportive services may provide a good starting point. Furthermore, approaching COs at different points in the treatment process or even after treatment may help reach them at the time when they exhibit highest levels of readiness (e.g., some COs may be most receptive at intake, while others may need some time away from the client). Sharing information and resources (particularly early on) may set the stage for increasing readiness over time (i.e. “planting the seed”). Hosting Al-Anon meetings (at the SUD clinic’s facility) may also help COs who are reluctant to go to unfamiliar places. Finally, most interviewees thought online or smartphone apps could help reach and engage certain groups, particularly those younger and more tech-savvy. Adapting AIR to a mobile app could also help overcome access barriers such as time and transportation issues, and allow individuals to explore content on their own at a time convenient for them.

3.3. Inner setting: Clinic cultures, practices, and staffing

Facilitators:

Many interviewees described their clinics as having an organizational culture that is either based on or aligned with the 12-step philosophy. Many clinics’ current services include activities ranging from disseminating information about 12-step programs to hosting meetings on site (including Al-Anon). Furthermore, several interviewees described their clinics having a learning culture, where they explicitly aim to implement EBPs as much as possible and may have routine processes in place to guide them (e.g., “rapid cycle implementation model”). Community clinics also reported conducting regular family education groups, indicating their capacity for AIR implementation. Some clinics also routinely conduct client follow-up calls, providing additional capacity for AIR implementation. Finally, several clinics indicated that they have sufficient staffing (both in numbers and their time/availability) to conduct AIR sessions.

Barriers:

In contrast, interviewees at several other clinics reported not having sufficient staffing for delivering AIR, limiting their ability to provide it on a routine basis. Furthermore, several interviewees thought that high rates of staff turnover typical for SUD settings could undermine continuity and sustainability of AIR in routine practice. Finally, some interviewees indicated that they rarely interact with clients’ COs (e.g., when clients arrive from other facilities), limiting their opportunities to recruit and engage them in AIR.

Recommendations:

Interviewees suggested that leveraging existing capacity (services, staffing) would facilitate AIR implementation, while also reducing the staff burden and time commitment. Some interviewees suggested that integrating AIR into staff workflows and institutionalizing it in organizational policies and procedures (e.g., staff job descriptions, annual reviews) could further strengthen its implementation, reliable delivery, and sustainability. Finally, interviewees also stressed that sharing a website address or a smartphone app with COs (either directly or through clients) would be easier and less burdensome for their staff.

3.4. Characteristics of individuals: Staff attitudes, knowledge, skills, and status

Facilitators:

Interviewees reported that they and their colleagues would generally be receptive to AIR, and that most of them have the skills and abilities to conduct it competently.

Barriers:

Several interviewees thought that some staff may be insufficiently informed (or even misinformed) about Al-Anon, what it does, and how it can help clients’ COs. Some interviewees also noted that some staff members may themselves have poor experience with Al-Anon and be reluctant to encourage COs to participate. Finally, some interviewees thought that clients and COs do not perceive all staff equally (e.g., differences in professional status, likability) and that some staff may thus be less successful in persuading COs to engage in AIR and Al-Anon.

Recommendations:

Most interviewees agreed that training would be required to ensure all staff understand the purpose and logic of the AIR model, the Al-Anon program, and how to facilitate attendance through open discussion (e.g., addressing attendance barriers). Finally, a careful assessment and selection of staff that would deliver AIR (i.e., their personal and professional characteristics) could help improve its effectiveness in practice.

3.5. Process: Leadership support and logistics

Facilitators:

Most interviewees agreed that involvement of leadership (clinical directors, as well as other administrative and other senior leadership) would be critical for AIR implementation. Some interviewees could also think of particular individual staff that could help “champion” AIR implementation (e.g., those with personal experience with Al-Anon).

Barriers:

Staff/time shortages that characterize the inner context of some clinics would also inhibit the implementation process itself (e.g., taking the time to attend AIR training, updating policies and procedures).

Recommendations:

To help decrease the time and burden of implementing AIR, interviewees suggested that our team adapt and share the materials developed for the study (e.g., training, brochures) with people implementing AIR. Finally, an app-based AIR would not only be easier to use, but would require less effort to implement, and would facilitate implementing it in other service settings (e.g., primary care).

4. Discussion

This formative evaluation study assessed facilitators, barriers, and recommendations for implementing Al-Anon Intensive Referral (AIR) in SUD treatment settings. We found that AIR has strong potential for implementation in SUD treatment settings. Clinic staff and leadership generally positively perceived AIR, which has the potential to address critical and unmet needs of concerned others (COs). However, SUD clinics have different levels of capacity and readiness to implement AIR fully and to offer it routinely. While clinics with existing services (e.g., family education groups) and sufficient staffing capacity could leverage these resources to implement AIR on a routine basis, clinics with limited capacity could try implementing AIR with a narrower scope and offering it on a case-by-case for COs with highest need and readiness. Furthermore, adapting AIR to facilitate connection with and engagement in a larger variety of community support programs (beyond Al-Anon) and/or adapting it as a website or smartphone app could expand its reach and reduce the burden on clinics and their staff. Importantly, the findings did not systematically differ between clinics that participated in the AIR trial and those that did not (naïve clinics), suggesting that these findings would likely generalize beyond early AIR adopters. These findings can inform implementation of the AIR intervention, as well as the development and implementation of its future adaptations and other similar interventions for concerned others (COs).

One possible strategy to implement AIR would be to first focus on COs with high-readiness to seek supportive services. Interviewees in our study noted that COs often do not perceive the need for or refuse to engage in any activities or programs for clients’ family and friends. Similarly, Kourgiantakis and colleagues (2018) found that family members of problem gambling individuals reported conflict and isolation as a barrier to their engagement in treatment. However, by approaching COs with highest readiness, SUD clinics could pilot-test AIR in their own settings and populations, while limiting the time and effort their staff spent engaging in AIR activities (particularly one-on-one calls with individual COs). If successful, these “small wins” could then be leveraged to reinforce further use of AIR (positive feedback loop). Clinics with limited capacity could maintain delivering AIR at this level, while those with higher capacity could scale up and deliver AIR to more COs. Future research needs to determine how to identify COs with high versus low readiness and what additional strategies may be needed to best engage COs with low readiness.

Furthermore, clinic leaders and staff perceived AIR as an adaptable intervention that can be tailored to different groups of COs, clinics, and locations. One adaptation suggested in interviews was simply expanding the menu of available community resources for COs beyond Al-Anon, based on what was available locally. While Al-Anon is a widely available program, it is not available everywhere, and its characteristics as a 12-step program (e.g., 12-step philosophy, group meeting format) may dissuade some from joining (Timko et al., 2014). Other community resources, if available, may be more acceptable and appropriate. While such adaptations of AIR would require further development and testing, they have the potential to help overcome low readiness among COs and facilitate reaching more individuals.

Another adaptation would be to develop a website or smartphone app AIR to deliver AIR content. While most interviewees thought that such technologies typically do not appeal to older people, they also thought that technology approaches could be helpful for reaching younger and more tech-savvy individuals. A study of COs of individuals engaged in problem gambling found that study participants, who tended to be younger, chose web-based counseling for ease of access, privacy, anonymity, connection, and communication of emotions (Rodda et al., 2013). We can think about these technologies and how to test them in future studies in at least two ways. In the narrower sense, technologies such as websites or apps could serve as implementation strategies promoting in-person AIR by informing individuals about it and its availability. In the broader sense, these technologies would represent a new version of AIR that could reach more people, connect them to a wider array of resources (including other online resources available beyond their local communities), and potentially require less staff time.

Finally, we may need to view AIR, its adaptations, and other similar interventions through a prism of longer time horizons. Several interviewees reflected on the need to educate COs about their needs, resources available to them, and encouraging them to take care of themselves (despite lack of motivation or perceived need). While they recognized that some of these efforts may not produce tangible results in the short term, they may “plant the seed” and lead to a positive response at some later point in time. This observation is consistent with findings that engagement in Al-Anon is typically delayed, after accumulation of stressors (Timko et al., 2012). Two potential implications for additional research are that new implementation strategies could help shorten the time from exposure to AIR to engagement in Al-Anon, and that relatively frequent but low-dose and low-cost educational strategies targeted towards COs (e.g., use of printed and online materials) could be fruitful cumulatively over time, either by themselves, or in combination with AIR.

4.1. Limitations

Eight of the 10 clinics in the sample had agreed to participate in the AIR trial. They may be “early adopters” of novel and evidence-based practices and therefore may systematically differ from the wider population of SUD clinics where AIR could be implemented. Thus, findings may not generalize well to clinics that adopt innovations later. To some extent, we mitigated this limitation by recruiting a diverse sample of clinics and key informants, including individuals from two naïve community clinics that did not participate in the trial. However, no naïve VA clinics were included, limiting our ability to generalize to other VA clinics.

Furthermore, since our team both developed AIR and conducted the interviews, the interviewees may have been more positive and/or less critical of AIR. We mitigated this concern somewhat by the fact that the interviewer (JB) was not involved with intervention development and testing (RCT) and was unfamiliar to interviewees. Additionally, interviewees mostly had no or indirect experience with AIR, so their responses largely reflect perceived or potential facilitators and barriers. We also only learned about COs’ barriers and facilitators indirectly, through clinic leaders and staff, which may have provided a restricted representation. Nevertheless, leader and staff perceptions are important antecedents to behavioral implementation outcomes (e.g., adoption, penetration), and thus the findings in this study help contribute early information on the “implementability” of AIR and similar interventions.

Finally, one investigator performed most of the coding, which could bias the findings. However, we mitigated this concern by relying on the Consolidated Framework for Implementation Research to provide structure and guidance throughout the study (interview guide, codebook). And, to help ensure coding reliability and validity, a second investigator coded 20% of the transcripts and the two coders discussed coding differences until reaching agreement.

5. Conclusion

This study contributes early insights into barriers and facilitators of implementing AIR and other similar interventions focused on COs. If the trial shows AIR is an effective intervention to facilitate engagement in Al-Anon and improve CO outcomes, the findings presented in this paper could help inform future AIR implementation, in research and practice. If AIR is not an effective intervention, these findings may help explain why it was not effective (e.g., due to COs barriers to engagement) and can inform development of a revised AIR or a new intervention.

Highlights.

• SUD professionals reported the need for short interventions for concerned others

• Al-Anon Intensive Referral (AIR) was viewed as a potentially helpful intervention

• Some SUD clinics have the capacity (e.g., staffing) to implement AIR fully

• Other SUD clinics lack the capacity and could implement AIR on a case-by-case basis

• AIR adaptations (e.g., smartphone app) could widen its appeal and reach