Implementing outside the box: Community-based social service provider experiences with using an alcohol screening and intervention

Abstract

Objective

The aim of this study is better understand perceptions of front-line social service workers who are not addiction specialists, but have to address addiction-related issues during their standard services.

Method

Six social service organizations implemented a validated alcohol assessment and brief education intervention. After a 3-month trial implementation period, a convenience sample of 64 front-line providers participated in six focus groups to examine barriers and facilitators to the implementation of an alcohol screening and brief intervention.

Results

Three themes emerged: (1) usefulness of the intervention, (2) intervention being an appropriate fit with the agency and client population, and (3) worker commitment and proper utilization during the implementation process.

Conclusions

A cross-cutting theme that emerged was the context in which the intervention was implemented, as this was central to each of the three primary themes identified from the focus groups (i.e., the usefulness and appropriateness of the intervention and the implementation process overall). Practitioner buy-in concerns also indicate the need for better addiction service training opportunities for those without addiction-specific educational backgrounds. Future research should assess whether targeted trainings increase addiction screening and education in social services settings.

INTRODUCTION

The last decade has seen a great deal of research describing barriers to and facilitators of empirically supported treatments in a wide variety of therapeutic practice settings (Dopson and Fitzgerald, 2006; Humphreys and McLellan, 2011; Patterson and Dulmus, 2012; Patterson et al., 2012; Patterson et al., 2013a). Human service organizations that serve clients with mental health disorders have been one setting of particular interest, with consistent findings suggesting that the primary barriers to implementation include lack of institutional and supervisor support, insufficient training and funding, and opposition from front-line staff (Rapp et al., 2010; Spear et al., 2009). Many studies have examined interventions that were specific to mental health symptoms or conditions such as depression and anxiety. Substance and alcohol abuse interventions have not been the focus of much mental health implementation research and are perhaps assumed by some researchers to be part of the repertoire of mental health workers. However, although mental health and substance abuse services overlap, they are not the same fields, and the nature of the work can be fairly distinct.

Twenty-five percent of caseloads in human service organizations consist of clients with alcohol use issues (Begun, 2004). In fact, high rates of alcohol consumption have been found to be comorbid with a range of different psychiatric disorders, including mood and anxiety disorders (Grant et al., 2004a; Grant et al., 2004b), schizophrenia, and even suicide (Reiger et al., 1990). Meanwhile, even though social service workers are in a prime position to intervene with problem drinkers, many fail to do so. Studies have shown that the main reason for failure to intervene is the lack of alcohol-specific courses and field placements available for those in social service programs. For instance, the great need for social workers to obtain training in addiction services is largely going unmet (Abbott, 2000; Abradinsky, 2004; Richardson, 2007; Straussner, 2004). In New York State, for example, 57% of universities offer master of social work (MSW) programs; however, only 11% require students to complete alcohol and substance abuse courses before graduation, and only 42% offer these courses as an elective (Richardson, 2007). In addition, only 40% of the MSW programs in New York State have students in addiction-specific field placements (Richardson, 2007).

The limited educational and training experiences received by students coming out of schools of social work may result in barriers to conducting assessments for clients with alcohol or other drug problems (Begun, 2004; Gilbert and Terrell, 2002). Lack of alcohol-related knowledge may contribute to social workers’ inability to identify problem drinking and their unwillingness to address problem drinking with clients (Bailey, 1970; Peyton et al., 1980; Richardson, 2007). To overcome this barrier, social workers need better training in best practice treatments for clients who are problem drinkers. One such treatment is a brief intervention for substance and alcohol abuse. Brief interventions for substance and alcohol abuse, which are fairly easy to learn, easy to train, and beneficial throughout the social services system, have been found to be highly effective. Most of these have been implemented in medical or primary care settings or in substance abuse–specific treatment settings, and not much is known about whether they can be incorporated into the practice of social workers or other social service professionals working in nonmedical settings (Benishek et al., 2010). Given the high rates of clients with comorbid alcohol dependence and other mental health disorders, the successful utilization of evidence-based practices in the medical field may also generalize to alternative clinical settings. However, more research is needed to explore this understudied issue.

To address this research gap, and indeed a potential implementation gap, the aim of this study was to understand front-line social service workers’ experiences using a brief alcohol screening and educational intervention during their standard services. Although this intervention has been investigated previously for client drinking outcomes (Fleming et al., 2002), studies have not examined worker perceptions of implementation barriers and facilitators of such outcomes. Front-line professionals are perhaps best equipped to provide feedback on the barriers to implementation because of the nature of their positions within human service organizations. Compared to administrators and supervisors, these workers are more likely to utilize interventions and have higher levels of client contact (Brandon, 2005; Lipsky, 1980).

METHODS

The intervention

This 2013 study describes findings from six focus groups conducted with a convenience sample of 64 front-line workers (i.e., those employees having direct service contact with clients) from six community-based social service organizations in the East Coast and Midwest regions of the United States. Workers were invited to participate and were consented with university-approved Institutional Review Board protocols. These organizations were involved in a larger study examining barriers to and pathways of adopting the National Institute on Alcohol Abuse and Alcoholism’s (NIAAA) Helping Patients Who Drink Too Much: A Clinician’s Guide, a simple and effective brief intervention that was originally developed for a primary care setting (National Institute on Alcohol Abuse and Alcoholism, 2005). The guide is available online and involves a number of steps and resources, including questions that help to diagnose alcohol abuse or dependence. In this study, mental health clinicians were asked to use this intervention with all new clients. This information was presented in three one-page documents excerpted from the NIAAA guide.

The first step was to ask whether clients drink (e.g., “Do you sometimes drink beer, wine, or other alcoholic beverages?”). If the client answered in the affirmative, the clinician was then instructed to show an illustration of and provide information on “What’s a Standard Drink?,” which explains that, for instance, a 12-ounce can of beer and a 1.5-ounce “shot” of liquor are both standard drinks. Once the client gained an understanding of what constitutes a standard drink, he or she was asked how many drinks he or she has per day and how many days per week. This information was used to establish whether the client had a high-risk drinking pattern. The workers were also trained to measure a client’s excessive drinking with the Alcohol Use Disorders Identification Test (AUDIT; Bohn et al., 1995), an alternative to the NIAAA method. Either screening method was acceptable for this step of the intervention.

With the client’s drinking pattern established, the worker showed the second illustration, “U.S. Adult Drinking Patterns,” which aimed to help clients understand their patterns of drinking, including whether they exceed daily or weekly limits in relation to the adult U.S. population’s patterns of drinking. This also helps clients see the link between excessive drinking and alcohol disorders because, for instance, the illustration shows that half of people who exceed weekly and daily recommended alcohol limits have an alcohol use disorder. The final document included “Strategies for Cutting Down,” a list of tips for limiting the amount of alcohol consumption, such as keeping track of the number of drinks, eating food while drinking, and having a plan to handle urges to drink.

Sample and procedure

After implementing the intervention for a trial period of 3 months, the practitioners participated in a focus group, during which they were asked their thoughts about the intervention, including barriers to and facilitators of its implementation. Between 2011 and 2013, six total focus groups were conducted by two different moderators: one for each of the two agencies in the East Coast city and four agencies in the Midwestern city. Each focus group included 4 to 12 participants and ranged from 30 to 60 minutes in length. The majority of participants in this study were front-line workers. Of the six participating agencies, two were affiliated with universities and counseled college students (referred to herein as “College Agencies 1 and 2”), one agency provided a broad range of social services to homeless (herein referred to as “Homeless Services”), one agency provided school-based services for adolescents (herein referred to as “School-Based”), one agency provided community-based services to adolescents (herein referred to as “Community-Based”), and one agency provided crisis services (herein referred to as “Crisis Services”).

We utilized a qualitative methodological approach because such methods are rapidly gaining ground as a “viable approach” for clinically applied fields (Padgett, 1998), especially as a strategy to investigate how workers implement evidence based practices (Anthony, Rogers & Forkas, 2003). Qualitative methods also will help us get inside the “black box” of programs implementing the alcohol education intervention. That is, it will help us distinguish the differences between process (what takes place during the intervention) and outcome (whether the programs goals were achieved), given the constraints of the organizational environment (Dehar, Casswell & Duignan, 1993). It will also help us capture the “lived experience” from the perspective of frontline workers implementing the intervention (Padgett, 1998). Most importantly, the use of qualitative methods will help us fill the gap of knowledge in this area by helping us gain new perspectives on phenomena which will help us question old, outdated or misinformed assumptions which we can later test quantitatively (Strauss & Corbin, 1990). Thus, qualitative methods were appropriate for this study due to the unique qualities of organizational and worker structures, as well as other variables that need to be identified that might later be tested quantitatively. The focus groups were designed to explore the following overarching question: What factors went into the decision to adopt, or not adopt, the new intervention?

The groups’ protocol began with free-flowing, open-ended semi-structured questions. Examples of focus group questions included: “What are your thoughts on implementing the new intervention? What, if any, were barriers to the implementation process? How did this effort fit into your usual services? How did you incorporate the screening and educational process into your usual services?” Throughout the focus group, respondents were given the opportunity to discuss questions freely with the moderator, asking for clarifications or deeper understanding probes.

Focus groups were audio-recorded and transcribed. To develop a basic codebook structure, an initial reviewer read the first two transcripts and identified common themes. A second reviewer repeated the process, and together they agreed on overarching themes and subthemes. A third reviewer then reviewed the transcripts and independently identified themes. The first, agreed-on structure was then compared with the new structure as developed by the third reviewer, and a new, modified, basic codebook was developed with input from the other reviewers. The second reviewer independently coded the interviews with the existing codebook. This led to identifying new themes and to clarifying and consolidating existing ones. This process also allowed for clearer definitions of the heuristics and codes from the previous codebook. The reviewers refined the codebook until they reached 90% agreement on their coding. The final codebook was then used as a guide in reviewing and coding the four remaining focus groups. This codebook was modified with an iterative process in which coders independently identified, discussed, and then either added or integrated new themes into the codebook. Any interrater reliability rates that fell under 90% were flagged and discussed by the coders. NVivo version 10 was used for coding. After codes were finalized, first author identified themes and sought feedback from second and third author during debriefing sessions.

RESULTS

The sample of participants had a mean age of 38 years (range 23–71); 70% were female; and 74% self-identified as White, 15% as African American, and 5% or less for any other category (multiple categories were allowed). At the time this survey was administered, participants had worked in the human service field for an average of 9 years (range 0–37) and at their current agencies for an average of 3 years (range 0–30). About 2% had completed high school, 2% had earned an associate’s degree, 16% had received their bachelor’s degree, 60% had obtained their master’s degree, and 20% had earned a doctoral degree. When asked in which discipline they had earned a degree, a plurality checked “Other” (35%), followed by social work and psychology both at 27.5%, education at 6%, and medicine at 4%. The disciplines composing the category of “Other” could not be determined. In order to show how the intervention was adapted and utilized at the “ground” level and to better show the “lived” experiences of frontline workers, we sampled more direct service providers more heavily (94%) than upper management staff (6%).

Three themes were identified by first author, indicating critical barriers to and facilitators of implementing the assessment and brief intervention. They consisted of practitioner perceptions of (a) the usefulness of the intervention, (b) the intervention being an appropriate fit with the agency and client population, and (c) worker commitment and proper utilization during the process of implementing the alcohol intervention in a general social services program. These themes provided much information to better understand frontline service workers’ experience with a brief alcohol screening and educational intervention during their day-to-day routines and what factors influenced their willingness to adopt the intervention.

Usefulness

Workers found that the program facilitated the exchange of information. Utilizing the brief assessment at the start of the worker–client session helped workers gain important information from clients, especially about how much alcohol they actually consumed. The following statement, by a worker from School-Based, demonstrates how such information could change the course of treatment or catch substance abuse issues early:

One girl actually—she was the only one out of the kids I did it with, has a drink in the morning to stave off a hangover. I wouldn’t have asked that during an assessment, but that question is on the AUDIT. That told me more than I would’ve known normally.

The statement shows how the use of AUDIT within the already existing assessment routine helped to provide accurate information with which workers diagnosed possible hazardous alcohol use. The assessment also helped workers start a conversation and control the frame of the conversation in order to integrate new information:

It was easy to integrate it into the opening session. What I found particularly helpful was the handout that showed the different amounts or alcohol in one serving of alcohol. I would ask questions about how much do you drink or how often? One or two drinks of what type of alcohol [is important to understand].”

Some workers did not find the assessment or the educational materials to be useful. The worker’s attitude toward any intervention is important to understand as it can result in barriers to implementation (Aarons, 2004; Patterson et al., 2014) Workers who did not find the brief interview intervention to be useful often found that the intervention overlapped with the assessment tools they already utilized. They felt that the assessment was only useful when they had a hunch that a client might be at high risk. Otherwise, existing assessments, which were often briefer, provided all the information they needed.

Appropriateness

The appropriateness of the brief interview intervention was also a major theme related to utilization as has been found in earlier studies (Forman, Bovasso, & Woody, 2001). Many workers were concerned with whether the intervention was appropriate for the populations they served (McGovern, Fox, Xie, & Drake, 2004). The following statement shows that worker responses to the appropriateness of the intervention vary based on the workers’ clientele:

One of the first questions [on the new screening tool] is how many times a year have you had five or more drinks. One kid said, “I have five or more drinks every weekend.” So we multiplied how many in 52 weeks. So there’s 52 occasions where he has. We did talk about binge drinking and fatty liver and hepatitis and everything; but here’s a senior. He’s graduating, he’s like, I don’t really see it as a problem.

As the preceding statement implies, workers providing services for youths found questions designed for adults ill-suited to their youth populations. Workers providing services for college-aged populations had similar problems translating assessment and literature material over to their clients:

I had at least one person if not more question statistics [in the new educational hand-out] because they are for 18 years and older U.S. adults, not necessarily college kids. People were like, well but is that for college or for after college, because it’s different. The materials were not geared towards college-aged kids and so the college kids were more apt to dismiss them.

As the statement implies, this led to many situations in which staff observed that clients dismissed the educational piece of the intervention as a scare tactic. Materials targeted to prevent alcohol abuse and dependence were also confusing to clients with severe alcohol and drug dependence. Workers at the Homeless Services, an agency which served many clients with severe drug and alcohol issues, found that utilization of the brief intervention—which takes a preventative public health approach—can confuse clients and obstruct treatment goals:

I thought the information [statistics in educational hand-out] was really interesting. At the beginning they had a lot of interest in the amount they drank. But once you get to the last page I had mixed group, people who had been in treatment [and not]. Had been told abstinence was the option, not harm reduction. Don’t believe in the [total abstinence approach].

I think it has a lot to do with their experience in treatment. They found the first couple of pages interesting but if they had been in treatment they were taught abstinence [only, rather than try to decrease drinking].

For workers at the Homeless Services, resolving tensions between the public health and abstinence models was difficult, and they often found that they lacked confidence in the outcomes resulting from presenting educational materials. Workers also found that the assessment was not appropriate for clients in crisis or who had endured high levels of trauma:

If someone was there for a traumatic reason and emotionally distraught, it can be jarring to do an alcohol intake. It was out of context. If they came in with a specific issue and they were traumatized at the time, it was kind of out of context.

I had a similar experience where I work with a lot of kids. I didn’t get a lot of opportunities to use it. When I did have a new adult I know one of them had such a significant trauma history it didn’t really seem relevant at the moment or appropriate to bring up that particular thing [alcohol use] during our first session because of the weight of what she brought up—it almost seemed like I would’ve been taking away from the assessment.

As these workers explain, traumatized clients require cautious responses. Questions and education related to substance use may be inappropriate at a time when clients are in crisis and especially vulnerable to judgment by workers. These observations brought into question why clients were coming for services in the first place:

I’m not sure exactly why—I think that probably in the context of “okay, I’m here to cut down on my drinking behavior” it probably would’ve been better received. I don’t think I had any clients that were here as a presenting problem to cut down on their drinking behavior. They all had other things they were concerned about.

Implementation process and practitioner buy-in

Workers at the six agencies had significant autonomy when it came to implementing the screening and brief intervention into their client assessment routines because interactions involving this implementation process were often unsupervised. For this reason, worker buy-in was extremely important and determined whether the interventions were conducted and whether they were conducted properly (Obert et al., 2005). Worker buy-in and utilization tended to be related to organizational culture and workers’ task complexity.

Most agencies provided workers with professional discretion when it came to utilizing the intervention or not utilizing it. For example, at College Agency 2, workers were told, “You’re a professional, use your judgment.” At such agencies, brief intervention was just another human service technology from which staff could draw to improve their practice. The role of directors was to try to convince workers to use the brief intervention. An agency director explained, “Like you presented, get people on board, and then you entrust everyone to take care of it.” However, there were issues using this implementation strategy, as one worker from the Homeless Services pointed out:

I think probably the difference would have been for me, and this is maybe kind of bad, is I knew in the back of my head if I don’t do this [new intervention] with somebody it’s not that big of a deal. But if [my supervisor] had come in and said you have to do it, I’d be like, Oh, I have to do it. If it had been on my evaluation, I wouldn’t have forgotten [to use it with all clients].

There was a delicate balance between forcing workers to utilize the intervention and relying on them to use their discretion in a way that will guarantee the effective use of the intervention. Another director explained his approach to promoting worker buy-in:

I think it’s a tricky balance. People don’t like feeling measured and checked up on, but at the same time, measuring and checking up has made us a little more productive and a little more thorough in some key areas. I think it’s a tricky balance, and sometimes people don’t like it, but it makes things go better.

Only the crisis agency, Crisis Services, required that workers report on the number of assessments they completed. Here, this requirement was reinforced because the agency tied the intervention into existing funding requirements. However, many of the workers felt resentful about the loss of their professional discretion:

I don’t know if it was sold [to workers] well, of the benefit. If you don’t have staff involvement with the whole setup, we don’t buy into things all that easily.

The rejection of the brief intervention by many of the workers at Crisis Services also was related to the complexity of relationships with clients. Crisis situations demand that workers make many different problem-solving decisions, which could have large consequences if done incorrectly. Because of the pace of worker–client interactions, workers did not want to spend a significant amount of time diagnosing hazardous alcohol use, especially when they could not relate it directly to the presenting issues of the clients:

Well, I think we normally—before this was implemented, we normally asked them if they use drugs or if they use alcohol, ‘cause we were trying to determine if their behaviors or symptoms are part of the mental illness or part of them using some type of substance—but it’s just sometimes not appropriate to use; like, they don’t wanna answer what day it is, so you can’t really go forward and be like, well, do you drink? They’re so incoherent they’re talking about stuff that’s not even making sense, so sometimes it’s really hard to even get to that point.

Workers at Crisis Services as well as other agencies also found it a challenge to combine assessment and educational aspects of the brief intervention. This was because many workers in the study had already adjusted to existing routines with clients in which assessment and education components were separated. The integration of the education aspect of the intervention into existing assessment routines led many workers to neglect the education piece. Others created methods to remember this aspect of the brief intervention and to better incorporate this type of interaction into the assessment and information-gathering routine with clients:

Every time I had a new client on my schedule, when I pulled the new file, I would get the sheets out and stick it with the folder so that I wouldn’t forget. It was always right there.

Workers would try to have the materials on their laps so they could refer to them without clients realizing that this was occurring. For these workers, having the educational materials ready helped to make the transition from assessment to education less awkward. Still, many workers found this transition to be uncomfortable and forced.

Task complexity also was related to the efforts required to get clients engaged in treatment. For Crisis Services workers, this was a significant challenge:

Sometimes they don’t even wanna talk to us. We have to get those questions in that we need to ask for our assessment, and those are kind of priority over, how many times do you drink? They only wanna answer a few questions. We gotta ask the important ones for our assessment.

This was by no means a problem that only Crisis Services workers encountered. There were time limitations on client assessments to which workers needed to adhere. In addition, many workers found gaining client trust and rapport to be a challenge as this is very important to client outcomes (Meier, Barrowclough, and Donmall, 2005). Introducing a rather stigmatized practice, such as alcohol use, during the early stages of the worker–client relationship brought tension to their relationships with clients, as the following worker explained:

For students that are already guarded, coming at them and asking them about something that they know they’d get in trouble for [drinking alcohol] especially when that’s their first meeting you?

A youth worker from School-Based said the following about the complexity of gaining trust from her clients:

I think the forum itself is good, but the trust building has to happen first, the rapport. I found it difficult, because I know I had some kids come in and then I’d be like, oh, shoot. I gotta do that [new intervention]. I gotta do the AUDIT.

Part of this complexity is related to the institutions in which workers and clients are embedded. As some workers and clients were embedded in the same institutions (e.g., college) and often had different agendas, client engagement and trust were challenging. Youth workers from School-Based discussed the challenge of engaging youths within an educational institution:

Well, you know what else is different about that? This may be a different conversation somewhere else, but they’re coming to us in school, and their parents aren’t necessarily on board yet. Their parents are taking them to the doctor or the psychiatrist. Maybe it has something to do with that title of going to see a doctor.

The youth workers’ discussion illustrates how interactions between workers and clients in these settings take place over the long term, and workers are often connected to the institutions in which students must seek social services. The placement of doctors in separate institutions (e.g., institutions where clients were not being evaluated or were not seeking aid and social services) may be related to increased levels of cooperation on the part of clients, or institutional context may make it easier for workers to bring up client substance use. A worker from Crisis Services commented on this challenge:

Well, I think in the beginning, when you’re showing it in the doctor’s office and in that type of setting, it probably goes a little bit better. Even the people we see a lot, you wouldn’t say we have a working relationship with them. We show up to their house or wherever they are in the street or whatever. I think there’s a difference [between doing the intervention in an office compared to someone’s home], and that should sort of be addressed in the program and the training.

DISCUSSION

As expected, the majority of workers were largely unsupervised and therefore, their buy-in was vital to the implementation of the brief alcohol intervention. Workers were more willing to utilize the brief intervention when they deemed it useful and appropriate to the population they served. Task complexity, or the challenges that workers face establishing worker-client relationships, also was an important factor when it came to utilization. In other words, workers who found the brief alcohol intervention useful found that this intervention helped them better diagnose alcohol use problems in their clients or provide them useful information about their alcohol use. However, workers had to be careful about when to, and when not to, provide this information. More age-appropriate interventions (such as interventions specifically tailored to college-aged populations) and different types of assessment mechanisms (such as ones more appropriate for crisis situations) are also needed.

The preceding results indicate that the most salient issues related to the implementation of the screening and brief intervention program were the perceived usefulness of the intervention, its appropriateness for the client population, and the process of implementation and obtaining practitioner buy-in. These findings align closely with the Consolidated Framework for Implementation Research (CFIR), a well-established conceptual framework theorizing that Intervention Characteristics and the Process of Implementation are two core domains influencing the implementation of evidence-based practices in real-world settings (Damschroder et al., 2009). The results of this study are of critical significance to the field of implementation science, which aims to understand the barriers to, facilitators of, and potential strategies for optimizing the adoption and integration of evidence-based interventions into practice. For social service workers implementing addiction services, utilizing interventions that are not only effective but also appropriate for patient needs and the organizational context is of paramount importance.

Further, perhaps due to the inherent challenges of implementing addiction services in non-specialty care settings, extra attention must be given to each phase of the implementation process, including 1) planning efforts, 2) stakeholder engagement, 3) execution of the intervention, and 4) reflecting and evaluating the delivery of the intervention. Organizational leaders can preempt debilitating implementation challenges by focusing on selection of the proper intervention and attending carefully to each stage of the implementation process.

Another cross-cutting theme for all of the identified issues is implementation context (Patterson, 2014). Owing to the highly variable attitudes, agendas, and needs of stakeholders across different organizations, a given intervention is unlikely to be deemed equally useful and appropriate in all contexts and the implementation and buy-in process may vary between agencies. Organizational decision makers carry the responsibility of finding evidence-based interventions that fit particularly well with their agency structure and needs. This approach is likely to limit barriers related to a lack of usefulness of the intervention, inappropriateness for the setting, or implementation and practitioner buy-in processes (Obert et al., 2005).

The reliability and strength of the AUDIT was questioned by workers who decided to use this tool. Workers expressed frustration regarding overlap between agency tools to assess alcohol and drug use and tools that have been statistically validated, such as the AUDIT. The AUDIT obviously provided many workers with more information about clients and could provide an opportunity for workers to start discussions with clients about their alcohol use. The use of these assessment tools could prove valuable to agencies struggling to get truthful answers from clients or with populations who have differing levels of education and comprehension. The AUDIT tool has been found to be valid and reliable among populations that varied race, gender, and age (Isaacson et al., 1994; Saunders et al., 1993a; Saunders et al., 1993b). Although other assessment tools could serve during a brief alcohol screening, the AUDIT is well studied and can be easily incorporated into a standard assessment process.

Agencies generally provided the intervention as a possible professional resource or, as scholars of organizational studies have referred to it, as “technology” (Sandfort, 2009, p. 269). This term fits here because it can be used like any other tool, such as computer systems, Web-based innovations, or other practices (evidence based and not evidence based) used to carry out the work of social service agencies. In fact, many of the same types of barriers identified in this study, including those related to organizational context and practitioner buy-in, have been documented in recent research on barriers to implementing technology-delivered interventions in behavioral health service settings (Ramsey et al., in press). However, because this technology serves clients with varying characteristics and motivations, its use requires professional judgment. For this reason, workers using this tool benefit from the mentorship of highly experienced supervisors (Brooks et al., 2012).

Worker–client interactions in social service agencies differ from doctor–patient interactions in hospitals. Usually, worker–client interactions take place over a longer period of time, involving different types of leverage and institutional legitimacy (Gilson and DePoy, 2008). Again, context appears to be a key element to the successful implementation of interventions (Aarons, 2005; Burns and Hoagwood, 2005; Nadler and Tushman, 1997; Patterson et al., 2013b; Rogers, 1995; Rousseau, 1997). Navigating the complexities associated with integrating new practices into highly variable organizations is a substantial challenge of which the health services field has little understanding. For these reasons, more research is needed to better inform implementation strategies for incorporating brief interviewing interventions into social service agencies. The pace and length of worker–client interactions as well as clinicians’ challenges with engaging specific populations are important factors in determining the proper utilization and promoting the success of brief interviewing interventions.

Conclusions and recommendations

Social service organizations and workers should determine which assessment tools are appropriate for their client populations. Although the AUDIT instrument has been studied and validated in multiple groups, there may be alcohol screening tools that are better targeted for specific types of clients. Results suggested that it is important to have workers use one appropriate screening tool without combining it with existing assessment mechanisms. Having a variety of screening tools for alcohol use is burdensome to workers and may decrease the likelihood that clients will be screened for alcohol use.

Because worker–client interactions differ from patient–health care provider interactions, brief alcohol interviewing mechanisms should be altered to allow workers to decide on appropriate moments for education. Social service workers who are required and trained to conduct biopsychosocial assessments might be reluctant to interrupt the flow of the assessment to provide education (Orlikowski, 2000). Changing the interaction from gathering information to providing information requires counseling techniques, experience, and a change in clinical routines (Gersick and Hackman, 1990; McGrath et al., 1984; Szulanski, 2000). Facilitating worker buy-in requires that workers and supervisors determine how the different parts of the intervention—the assessment and the brief educational intervention—should be incorporated and presented to clients.

Addiction issues are prevalent in all social service settings. As a result, addiction training should not be restricted to addiction programs only. The lack of education and training provided to social service workers who are not in an addiction-specialist track continues to be a problem. Increasing addiction training opportunities for those workers outside the addictions field would likely enhance workers’ ability to carry out alcohol-related screening and education. Future research should assess whether targeted trainings increase addiction screening and education in social service settings. The most efficient use of resources may be to focus the addiction-related training opportunities to defined organizational needs, rather than more universal, nonspecific training in addiction issues. These findings, however, should be interpreted with caution. Although we were able to compare multiple cases, via the use of focus group methodology, we mostly focused on frontline workers and therefore, these findings cannot be generalizable to experiences of upper level management. Future studies should take advantage of more triangulation methods – that is, including multiple stakeholders, including upper level management and consumers. Despite the need for continued research in this area, this study makes an important contribution to the field. By examining barriers and facilitators identified by social service workers implementing an alcohol screening and brief intervention, the current research highlights critical issues for social service agency leaders to focus on when implementing addiction services, namely selecting a proper intervention, attending to organizational context, and addressing the multi-stage process of implementation.