Smoking continues to be the single most preventable cause of death in the United States (
1,
2). Smoking prevalence has declined recently among the general U.S. population (
2), but it remains high among individuals with substance use disorders (
3). The rate of smoking among adults with substance use disorders or mental illness is 94% higher than among adults without these disorders (
4). Population prevalence estimates of smoking among persons receiving treatment for substance use disorders range from 61% to 88% (
5–
10), arguably making tobacco dependence the most lethal and common addictive disorder. Yet smoking cessation (SC) is rarely a goal of addiction treatment programs (
11,
12) or providers, despite clinical evidence that concurrent treatment of all substance use disorders, including tobacco-related disorders, may improve abstinence from alcohol and illicit substances (
13–
19) and enhance SC (
17,
20–
23).
The Veterans Health Administration (VHA) is the nation’s largest single provider of mental health and substance use disorder treatment. Over the last decade, the VHA has made systematic efforts to increase access to and delivery of evidence-based SC treatment in order to reduce smoking prevalence among the general population of veterans receiving treatment at VHA facilities (
24,
25). VHA policy requires tobacco use screening and the availability of evidence-based SC treatment in specialty care settings, including substance use disorder programs (
26,
27); however, requirements alone do not ensure adequate and uniform implementation in a large national health care system.
The VHA has 65 substance use disorder residential treatment programs (SRTPs) nationwide. SRTPs are residential rehabilitation treatment programs or domiciliaries that provide a stable, supervised, and recovery-oriented environment for substance use disorder treatment and rehabilitation. Contrary to common beliefs, several studies have found that a substantial proportion of patients (up to 62%) in residential or inpatient substance use disorder programs are interested in SC (
10,
23,
28,
29). Recent findings indicate that 79% of veterans receiving treatment in a SRTP used tobacco, but only 33% had a documented tobacco dependence diagnosis, 34% received SC treatment, and 11% had a tobacco dependence diagnosis and received SC treatment (
30). Although prior studies have examined the scope of SC practices (
31) and factors associated with adoption of SC treatment (
32–
37) in addiction treatment settings, we found no studies that examined barriers and facilitators related to SC service delivery in VHA SRTPs. Understanding the current scope of SC treatments in VHA SRTPs, as well as barriers and facilitators related to integration, will help inform and target efforts to improve care in this setting.
This evaluation was guided by the Promoting Action on Research Implementation in Health Services (PARiHS) framework, which specifies that implementation of evidence-based practice is influenced by the interaction of context, evidence, and facilitation (
38–
41). Elements influencing implementation include providers’ perceptions of the evidence derived from a variety of sources, including scientific research, clinical experiences and patient preferences, and context features such as organizational culture and leadership (
42). This conceptualization is consistent with findings on SC practices in substance use disorder treatment settings (
32,
34). As part of a larger quality improvement initiative to support SC implementation within VHA SRTPs (
30), we assessed providers’ knowledge and attitudes about SC evidence, existing SC practices and strategies, and perceived barriers and facilitators to integrating SC in SRTPs.
Methods
Selection of Sites and Participants
We purposively sampled SRTPs (
43) with varying levels of resources and services. Data from the VHA 2010 Drug and Alcohol Program Survey (DAPS) were used to stratify programs by level of resources and services. Resource levels were determined by staff-to-bed ratio criteria outlined in the
VHA Mental Health Residential Rehabilitation Treatment Program Handbook (
44). Programs with staff-to-bed ratios that exceeded the handbook’s minimum recommendations by 20% were defined as having high resources. Programs with staff-to-bed ratios below minimum recommendations were defined as having low resources. Service levels were determined by the proportion of tobacco-using patients who received SC pharmacotherapy or counseling. Using 2010 DAPS data, we approximated the average proportion of tobacco-using patients who receive SC counseling and pharmacotherapy while in SRTPs. We then designated programs that provided counseling and pharmacotherapy to a higher than average proportion of patients as high-services programs, and we designated programs that provided no SC counseling or pharmacotherapy or that provided SC counseling or pharmacotherapy to a lower than average proportion of patients as low-services programs.
SRTPs within each quadrant—for example, high services–high resources—were contacted in random order to invite participation. Program leaders who agreed to let the program participate nominated staff who were directly involved in or knowledgeable about SC implementation efforts. Nominated staff were invited by e-mail and phone to participate in a phone interview and asked to identify additional program staff to interview. If leaders at a selected site declined to let the program participate, the program listed next within the randomly assigned program ranking was contacted. Program leaders at 27 SRTPs were contacted, and leaders at 15 SRTPs agreed to let the program participate; a total of 25 participants across the 15 study sites were interviewed during fiscal year 2010.
Interviews and Analysis
A semistructured interview guide was developed on the basis of policy and practice guidelines, and an extensive literature review focused on SC in substance use disorder and mental health populations and treatment settings (
45–
47). Interview questions were reviewed by field experts and addressed interviewees’ knowledge and attitudes regarding SC, current SC activities, and perceived barriers and facilitators related to integrating SC into SRTPs. A doctoral-level psychologist with expertise in qualitative methods and evaluation of programs for substance use disorders conducted interviews by phone, with interviews lasting approximately 45 minutes. [The script of the questionnaire is available as an online
supplement to this article.] Informed consent was waived by the Stanford Institutional Review Board and the VHA Palo Alto Research and Development Committee. Saturation across major topics was reached, suggesting that these findings are a reasonably thorough summary of major issues. Interview notes were coded and analyzed by a qualitative research expert, who applied content analysis techniques to identify primary themes and their frequencies across content categories using Microsoft Excel (
48). The primary coder’s findings were refined by the lead evaluators and reviewed until consensus was reached.
Results
The SRTPs were distributed across the quadrants as follows: high services–high resources (N=3), high services–low resources (N=5), low services–high resources (N=5), and low services–low resources (N=2). The data were initially compared across quadrants to assess differences between SRTPs demonstrating higher implementation capacity and those struggling with sustained implementation; however, we found that the themes reported were consistent across settings, and thus findings are described in aggregate across programs.
Sample Characteristics
Table 1 presents characteristics of participants by licensure status. A slight majority (N=14, 56%) of participants were female, and most identified as Caucasian (N=17, 68%). As expected, the percentage of participants with a graduate degree was higher among licensed independent practitioners (LIPs) (N=9, 75%) than among nonlicensed staff (N=5, 38%), and LIPs had more experience in the field than nonlicensed staff (14.8 versus 11.6 years). Both groups reported similar smoking status, with approximately 44% indicating that they had never smoked, 40% identifying as former smokers, and 16% choosing not to disclose their smoking history.
Evidence: Knowledge, Attitudes, and Behaviors
Knowledge.
To assess knowledge regarding SC practices, participants were asked to describe all SC evidence-based practices they knew (
Table 2). If participants responded by naming a general category of interventions, they were asked to clarify and specify individual practices. Most participants identified broad categories of pharmacotherapy and counseling as evidence-based practices. Fewer participants identified combination therapy or screening, brief intervention, and referral (5 A’s). Although only four participants indicated they had received any SC training, 14 participants said they could apply their existing skills in substance use treatment and residential care to address smoking among SRTP patients. Unlicensed domiciliary staff spent the most time with SRTP patients, yet all participants were clear that domiciliary staff had no training on how to answer veterans’ questions regarding smoking; one domiciliary staff participant suggested that this was a missed opportunity.
Attitudes.
To assess attitudes, participants were asked whether SC should be provided for SRTP patients. All participants agreed that patients should stop smoking for health reasons; however, attitudes were mixed regarding whether SRTPs were good settings for SC treatment. Several participants indicated that SRTP was a good opportunity to provide SC treatment, given that patients are a “captive audience.” However, participants said that implementing SC interventions would require additional staffing and a plan for prioritizing multiple interventions. All participants agreed with VHA policy and guidelines that every patient should be screened for tobacco use and that clinicians should encourage patients to quit. However, participants stated concerns that “trying to force patients to quit” might make them leave the program. Some wondered whether attempting to provide SC treatment in SRTPs could jeopardize patients’ sobriety. Several participants indicated that because smoking was a “patient’s right,” it was inappropriate for staff to suggest quitting.
Behaviors.
To explore existing SC interventions within SRTPs, participants were asked to indicate whether their program offered key evidence-based SC practices.
Table 3 summarizes SC interventions offered by the SRTPs.
Most programs screened for tobacco use status at intake, documented smoking status (although not diagnosis), advised smokers to quit (brief intervention), and provided referrals. However, screening for tobacco use was generally viewed as a singular event at intake rather than an opportunity for motivational intervention that should be repeated during a patient’s stay. Most programs used computerized clinical reminders that were triggered at intake or annually thereafter to prompt providers to conduct tobacco use screening. But the clinical reminder for tobacco screening was often viewed as the responsibility of another department or individual from outside the program, such as a primary care doctor or an SC specialist. There was also common discussion of how the reminders were not utilized effectively, for example, by “checking boxes” in an offhanded manner to “get through the screen.”
In many programs the brief-intervention component was missing or mitigated. Most participants expressed discomfort with advising uninterested patients to quit, indicating interventions were typically presented to only patients who explicitly asked for help. One participant described the process as, “If they say they don’t want to stop smoking, they get told about the dangers and that’s it.” There were exceptions to this pattern, however; in fact, staff at several programs were passionate about SC and provided active discussions with patients about considering quitting. Although a majority of programs (N=11, 85%) reported providing SC treatment referrals within their facility, almost no programs (N=1, 1%) conducted additional follow-up after the initial inquiry and referral. Many participants (N=16 of 20 respondents, 80%) indicated that they were willing to change procedures, but several respondents (N=9 of 20, 45%) indicated that the culture of the program would make it difficult to sustain changes.
All programs offered nicotine replacement therapy (NRT), primarily patches or gum. Access to NRT depended on the facility’s formulary and availability of prescribers, ranging from same-day access to substantial wait times if referral to a prescriber outside the SRTP was required. Bupropion was available at most programs, although fewer programs provided varenicline. Several participants indicated concern about potential side effects of varenicline. Staff follow-up after medication consults was limited.
SC counseling was generally provided outside the SRTP and rarely documented or prioritized in the treatment plan. Individual SC counseling was available at most programs, although it was rarely accessed. Several participants said that SC counseling was not in their job description and was outside their scope of work. When individual counseling occurred, it was often seen as the purview of an individual provider who was passionate about SC and not a function of the SRTP staff.
All participants supported the idea of making SC groups available to patients who were interested in quitting. SC group counseling was available to patients in most programs, although the proportion of patients receiving these services was low. Eight programs indicated that group treatment was available through facility-level SC programs, one indicated that groups were managed by a domiciliary outside the SRTP, and two participants said group treatment was available at their programs. Many participants indicated that veterans’ progress in the SC groups was not typically discussed during other care for substance use disorders.
Most participants indicated either that SC was not addressed in discharge planning or that it was addressed only if a patient requested SC at intake. No programs had made efforts to improve continuity of SC as part of discharge planning.
Context: Organizational Strategies
Participants were asked whether their organization had specific strategies related to SC.
Table 4 summarizes participant reports of organizational strategies to support SC treatment. The most commonly reported strategy was written guidelines limiting hours or locations for patient smoking. Limited hours for patient smoking meant that patients were unable to smoke when the SRTP was locked at night. Several participants indicated that patients were unhappy about this restriction and that limits on smoking hours were difficult to enforce. For example, patients complained about the rules and lined up early in the morning to have their first cigarette. In response, staff often opened the doors early or let patients stay out after curfews.
All programs indicated that patients had access to designated smoking areas within the facility, for example, “smoke shacks,” that provide shelter from inclement weather. Usually located in a central courtyard, main entrance, or other highly visible venue, these smoking areas tended to become social “hot spots,” adding to the difficulty of quitting.
Many participants indicated that staff who visibly carried cigarettes or smelled like smoke while on duty sent the wrong message to patients. However, only three programs restricted staff smoking to designated areas, two programs forbade staff from smoking with patients, and one program required staff to avoid the appearance of smoking.
Although several programs had written guidelines requiring staff to address patients’ smoking, participants reported that those programs usually referred patients who reported smoking to SC groups outside the SRTP. Program policies did not consistently require thorough documentation of smoking status in the clinical record, with only eight programs reporting a requirement to document smoking status in the clinical chart and five programs reporting a requirement to include smoking on the problem list.
Barriers and Facilitators
Table 5 summarizes organizational barriers and facilitators related to improving SC treatment.
Barriers.
Participants were asked whether each issue would be a barrier to integrating SC treatment in their program. The most common barrier was lack of time given current workloads. Several participants noted, however, that lack of time was not “a good excuse to not provide necessary treatment.” More than half of participants believed that some staff in their program would resist changes because of a belief that giving up smoking is “too much to ask” of SRTP patients.
Several participants indicated that patients or staff were protective of the “right to smoke” and that it would be difficult to implement SC treatment in the current culture. As one participant stated, “The culture here and in the military is a culture of smoking.” Participants indicated that enforcement of smoking restrictions would be a significant barrier and suggested that provider and patient smokers would resist changes. Resistance by the employees’ union was cited as a barrier by several participants, with many describing specific instances in which the local union opposed restrictions on employee smoking. One participant felt that political influence would be a barrier, given a Congressional mandate (P.L. 102-686) requiring all VHA medical centers to provide smoking areas for veterans.
Facilitators.
When asked to reflect on facilitators of change, participants generally supported policies consistent with the goal of reducing all substance use disorders, including tobacco dependence, ensuring that policies do not inadvertently send mixed messages that encourage smoking. Many participants would prefer smoke-free facilities, in part to reduce behavior problems related to smoking.
Participants were asked to suggest ways to help staff understand that SC is a critical component of recovery; most indicated that SC training was critical and that in-person training and education regarding SC medications were particularly useful. However, as described by one participant and echoed by others, it is important to “make training mandatory [to] make sure supervisors allow their staff to go.” Along with using SC policies and training to help change the culture, participants suggested that a champion could assist programs with making incremental changes. However, many participants indicated that any changes would happen only with a specific local mandate and strong leadership support. As one participant indicated, “It’s got to come down from the bosses.”
Discussion
Although the participants generally supported SC treatment for veterans, the results indicate that there was variability in their willingness to provide SC treatment in SRTPs and explain why the SRTPs did not emphasize SC as an important part of substance use disorder recovery. As articulated by the PARiHS framework, context, evidence, and facilitation interact to influence implementation of evidence-based practice. In this case, general research evidence on SC was widely understood and endorsed, but specific evidence on this clinical population, setting, and patient experience was not. The perceived appropriateness of SC treatment was also limited by contextual challenges, including insufficient time and resources, staff misconceptions about SC, inadequate smoking-related policies, cultural norms that deemphasize SC, and perceived lack of training opportunities. These contextual challenges were consistent with previous studies in community addiction programs (
33,
36,
46,
49,
50).
Despite the challenges, these findings suggest actionable opportunities for improving SC integration in SRTPs, including providing training to enhance staff support and understanding, identifying clinical champions to advocate for and dispel misconceptions about SC treatment, and enacting policies and practices that support SC. Potential contextual changes include encouraging leaders to clarify the importance of SC for SRTP staff, facilitating training, and providing guidance for all staff to support SC, for example, through motivational enhancement. SRTPs should enforce policies that require consistent intervention and documentation regarding tobacco use, for example, by encouraging providers to add it to the problem list, offering brief interventions, and incorporating SC into treatment and discharge plans. Many of these changes require the active endorsement of local leadership to enforce national VHA policies on SC care and facilitate SC treatment implementation (
51).
Although the sites varied in terms of resources and services, these findings may not be representative of all SRTPs. Previous findings indicated that approximately 40% of staff in residential addiction settings smoke (
6), but no participants in this sample self-identified as current smokers. Smokers may have been unwilling to participate, or participants may not have felt comfortable revealing their smoking status. Further, program managers who did not respond to initial requests for participation may place less value on SC. Therefore, these results may not capture the full spectrum of issues or perspectives related to SC integration in VHA SRTPs.
Conclusions
SRTPs provide an important opportunity to address SC among veterans with substance use disorders and other psychosocial vulnerabilities (
30). Despite clinical practice guidelines, VHA policy, widespread knowledge of the dangers of smoking, and substantial evidence that SC treatment is appropriate and effective for patients in recovery from substance use disorders (
12,
15,
17,
52–
56), SRTP staff described both evidence-related and context-related barriers to SC implementation. Contextual factors such as lack of local leadership support and enforcement of local and national policies undermined accountability and uptake. Evidence factors such as inadequate SC knowledge and clinical experience perpetuated skepticism and ambivalence. These findings offer guidance for future efforts to facilitate SC treatment in SRTPs.
Acknowledgments
This project was supported by the Office of Tobacco and Health: Policy and Programs, U.S. Department of Veterans Affairs (VA) Office of Clinical Public Health and by Locally Initiated Project QLP 59-030 from the Substance Use Disorders Quality Enhancement Initiative Center. The views expressed are those of the authors and do not necessarily reflect the views of the VA or any other entity of the U.S. Government.
The authors report no financial relationships with commercial interests.