Although smoking prevalence in the United States has been declining in the general population, prevalence among the population with substance abuse problems has remained high and has generated considerable public health concern (
1). It is estimated that the smoking rate among those seeking substance abuse treatment is over 65% (
2), roughly three times the rate in the general population. In 2012, around 1.25 million people received treatment in substance abuse treatment facilities on an average day (
3). Therefore, these facilities are logical settings to provide smoking cessation treatment for smokers with substance use disorders.
Studies suggest that providing smoking cessation treatment during substance abuse treatment can improve smoking cessation rates (
4–
10) without negatively affecting the treatment of substance abuse problems (
4,
11–
14). There is a growing consensus that integration of smoking cessation services into substance abuse treatment is essential for reducing smoking prevalence among smokers with substance abuse problems (
4,
15–
18). However, the rate of adoption of smoking cessation services by substance abuse facilities has been low. It is estimated that only 43% of substance abuse inpatient treatment programs for adolescents and 41% of outpatient substance abuse treatment facilities provide smoking cessation coun
seling or pharmacotherapy (
19,
20). Similarly, although smoking bans are effective in promoting smoking cessation in the general population (
21,
22) and implementation of such bans in substance abuse treatment facilities is feasible (
23–
26), smoking bans have not been widely adopted in those settings. Concerns about implementing smoking bans include effects on smoking among staff, inadequate training to implement a ban, and the need to prioritize substance abuse treatment over smoking cessation (
27–
29).
Two studies have examined institutional factors associated with providing smoking cessation services in substance abuse treatment facilities (
19,
30). They found that the source of funding, breadth of service offerings, hospital affiliation, program size, and medical staffing were related to provision of smoking cessation services (
19,
30). However, these studies relied on small samples of facilities or facilities of a particular type (for example, outpatient or methadone treatment), limiting their generalizability. In addition, there has been no investigation of state-level factors that could have an impact on provision of smoking cessation services or implementation of smoke-free campuses at substance abuse treatment facilities.
The objective of this study was to examine the characteristics associated with smoking policies at substance abuse treatment facilities in the United States. Specifically, we analyzed data from a large facility-level survey to provide estimates of the delivery of smoking cessation services and the adoption of smoke-free policies. We also assessed facility-level and state-level factors associated with the provision of services and the adoption of smoking bans. Findings from this study could inform efforts to disseminate smoking cessation treatment and promote smoking bans at substance abuse treatment facilities.
Methods
Data Source
We obtained facility-level data from the 2012 National Survey of Substance Abuse Treatment Services (N-SSATS). Administered annually by the Substance Abuse and Mental Health Services Administration, the N-SSATS surveys all public and private substance abuse treatment facilities in the United States. The 2012 N-SSATS was provided to 16,114 eligible facilities and was completed by 14,995 facilities, a response rate of 93%. We excluded halfway houses, solo practitioners, facilities that provided treatment only for incarcerated persons, and facilities that did not provide substance abuse treatment services that year. The final sample included 13,094 substance abuse treatment institutions.
State-level characteristics related to smoke-free environments were linked to the 2012 N-SSATS by using state identifiers. State-level data were obtained from the Centers for Disease Control and Prevention (CDC) or the American Lung Association or were created by aggregating facility-level data from the 2012 N-SSATS.
Smoking Cessation Services and Smoke-Free Policies
Two binary outcome variables were analyzed: whether the facility provided any smoking cessation services and whether the facility banned smoking on the property. Facilities were coded as providing smoking cessation services if they offered nicotine replacement pharmacotherapies, non-nicotine smoking or tobacco cessation medications, or smoking cessation counseling. Facilities were coded as banning smoking if smoking was not permitted on the property or within any building.
Characteristics of Substance Abuse Treatment Facilities
Potential explanatory variables relevant to the outcomes were selected a priori and were grouped into four broad categories: service characteristics, financial characteristics, client characteristics, and general characteristics. Service characteristics included the primary focus of the facility, facility ownership, range of services, and type of services; financial characteristics included provision of free services to the poor and acceptance of government health insurance plans; client characteristics included type of treatment that clients were seeking and number of substance abuse admissions in the past 12 months; and general characteristics included facility location and licensing.
Tobacco Control Environment at the State Level
Characteristics of state-level tobacco control included whether the state had comprehensive smoke-free laws (
31), whether state spending on tobacco prevention and control exceeded 50% of the level recommended by the CDC (
32), and whether the proportion of smoke-free substance abuse treatment facilities in the state was greater than the national median of 30%. According to the CDC, a smoke-free law is considered comprehensive if it bans smoking in all indoor workplaces, restaurants, and bars, with no exceptions (
31).
Statistical Analysis
The institutional characteristics of facilities that provided smoking cessation services and those that did not were compared. Likewise, institutional characteristics of smoke-free facilities and facilities without smoke-free policies were compared. The correlation between providing smoking cessation services and having a smoke-free campus was analyzed by using a Pearson correlation. Separate multivariate logistic regressions examined the relation between institutional- and state-level factors and the facilities’ provision of smoking cessation services and use of smoke-free policies. No sampling weights were applied in this study because the N-SSATS samples all eligible facilities in the United States. Stata 12 was used to perform the statistical analysis. Results with p values <.05 were considered statistically significant.
Results
Of 13,094 substance abuse treatment facilities in the United States, 46.8% offered behavioral counseling, pharmacotherapy, or both services for smoking cessation in 2012. More facilities provided behavioral counseling (N=5,214, 40.8%) than pharmacotherapy (N=3,402, 26.0%). Approximately one-third of facilities (35.2% of 12,949 respondents) banned smoking on facility property. There was a significant correlation between the provision of smoking cessation services and banning smoking (r=.2, p<.001). In 2012, 26 states had comprehensive smoke-free laws; 13 states spent at least 50% of the CDC-recommended level on tobacco prevention and control; and 25 states ranked above the median (30%) in the proportion of substance abuse treatment facilities with complete smoking bans.
Table 1 contains descriptive information on institutional-level variables.
Table 2 presents results of the logistic regression for the availability of smoking cessation services and the adoption of complete smoking bans by institutional- and state-level characteristic. Facilities that focused primarily on substance abuse treatment (odds ratio [OR]=.8) and facilities that were licensed by a state substance abuse agency (OR=.8) or a state mental health agency (OR=.9) were less likely than facilities without these characteristics to provide smoking cessation services. Private nonprofit institutions (OR=.9) were less likely than private for-profit organizations to provide smoking cessation services. Public institutions (OR=1.3) were more likely than private for-profit institutions to offer smoking cessation treatment. Facilities that accepted government health insurance plans (OR=1.2), were located in a metropolitan area (OR=1.1), provided a greater number of ancillary services excluding smoking cessation counseling (OR=1.2), or offered inpatient or residential services (OR=11.3 and OR=1.4, respectively) were more likely than facilities without these characteristics to offer smoking cessation treatment.
Facilities with 101 to 500 admissions (OR=1.1) or more than 500 admissions (OR=1.5) in the past 12 months were more likely than facilities with fewer admissions to offer smoking cessation treatment. Facilities located in states that have comprehensive smoke-free laws (OR=1.2), spend at least half of the amount recommended by the CDC for tobacco prevention and control (OR=1.1), and are ranked above the median in the proportion of smoke-free substance abuse treatment facilities (OR=2.3) were significantly more likely to offer smoking cessation services than their counterparts. Offering outpatient substance abuse services and free treatment to the poor was not correlated with provision of smoking cessation service.
Facilities that provided a greater number of ancillary services besides smoking cessation counseling (OR=1.0), offered residential (OR=.5) or outpatient services (OR=.7), reported that more than half of clients with a substance use disorder sought alcohol abuse treatment only (OR=.7), or obtained a license from a state mental health department (OR=.9) were less likely than facilities without these characteristics to have banned smoking on their properties. Facilities with 101 to 500 or more than 500 admissions for substance abuse treatment in the past 12 months (OR=.8) were less likely to have banned smoking compared with facilities with fewer such admissions.
Private nonprofit (OR=2.3) and public institutions (OR=2.2) were associated with higher rates of banning smoking compared with private for-profit organizations. Facilities that offered inpatient services (OR=1.9), accepted government health insurance plans (OR=1.8), were located in a metropolitan area (OR=1.3), or obtained a license from a state substance abuse agency (OR=1.2) were more likely to have smoke-free policies compared with facilities without these characteristics. In addition, facilities in states with comprehensive smoke-free laws (OR=1.2) or in states that ranked above the median in the proportion of substance abuse treatment facilities with complete smoking bans (OR=2.7) were associated with higher rates of banning smoking. Primary focus of the facility, offering free treatment to the poor, and state spending on tobacco prevention and control did not predict the adoption of smoking bans.
Discussion
Using a large national facility-level survey, this study found that fewer than half of the substance abuse treatment facilities in the United States offered any smoking cessation services and approximately two-thirds of the facilities permitted smoking on the property in 2012. Although the rate of providing smoking cessation services has increased somewhat over the past decade (
20,
30), it is still low, according to these results. The proportion of facilities with a smoke-free campus was even lower, despite greater adoption of smoke-free campus policies in other health care settings, such as hospitals (
33). Currently, no federal regulations address smoking in substance abuse treatment facilities, although some state and local regulations do so (
23,
34). With increasing evidence suggesting that tobacco control in substance abuse treatment facilities is feasible and effective (
4–
10), factors that lead to greater availability of smoking cessation services and to smoke-free policies bear examination.
Perhaps most striking about the pattern of smoking cessation treatment at facilities that treat substance abuse is that those that focused primarily on treatment of substance abuse were significantly less likely than other facilities to provide smoking cessation services, despite the evidence that incorporating smoking cessation into substance abuse treatment improves smoking cessation while maintaining positive effects on substance abuse problems. Prior research identified that a major obstacle to offering smoking cessation services to the substance-abusing population is the concern that doing so would have negative effects on substance abuse treatment (
35). Studies that have assessed the true impact of such services during substance abuse treatment have found the opposite. Smokers who receive smoking cessation treatment are more likely to quit (
4–
10), but more important, they are more likely to have positive substance abuse outcomes (
4,
11–
14). Yet this information has not led to the adoption of a new service model among facilities dedicated to substance abuse treatment.
Facilities appeared to be influenced by the tobacco control norms in their area, both at the state level and the facility level. Although greater tobacco control spending in a state does not directly lead to increased funding for the facility, it suggests that smoking cessation is a state priority. Likewise, comprehensive smoke-free laws in the state and higher densities of smoke-free substance abuse treatment facilities reflect tobacco control norms and were positively associated with provision of smoking cessation services and smoke-free campus policies.
Perhaps not surprisingly, financial considerations appear to play a role in whether facilities offer smoking cessation services and whether they implement a complete smoking ban. Larger facilities and facilities that offered a greater range of services were more likely to offer smoking cessation programs. Perhaps, as has been suggested by others, large facilities have greater and more diversified resources, allowing them to absorb the risks associated with offering new services (
19). However, at the same time, such facilities were less likely to ban smoking on facility property. The reason for this is unclear, but it is possible that larger facilities had greater difficulty in implementing and enforcing a smoke-free environment. Consistent with previous research, private for-profit institutions and facilities that were less reliant on government health insurance plans were less likely to institute campuswide smoking bans (
18). An important consideration for such institutions is to generate a profit. Although there is no evidence that instituting these programs affects profitability, especially in states with comprehensive smoke-free laws, there might be a concern among these institutions that potential clients will choose facilities that allow smoking. With the expected increase in access to health insurance and the expanded coverage of smoking cessation treatment as a result of the passage of the Patient Protection and Affordable Care Act (
36), we anticipate that more facilities will start providing smoking cessation services and adopt smoking bans.
This study had several limitations. First, the analyses were based on a cross-sectional data set, which precludes interpretation of causal relationships and examination of the adoption and diffusion process. Second, the model was limited to the data available. State-level tobacco control data were merged with facility-level information, but other factors that might influence provision of services and policy adoption, such as detailed profiles of clients or facility staff, were unavailable. Third, we did not account for recent regulations, for example, in New York and New Jersey, that require smoking cessation services and smoking bans in substance abuse treatment facilities (
23,
34).
Despite its limitations, this study has policy implications regarding the integration of tobacco control and substance abuse treatment. Many characteristics at the service, financial, client, and state levels and general characteristics were positively associated with both use of smoking cessation services and smoke-free policies. As a result, some interventions may provide smoking cessation programs and ban smoking simultaneously and perhaps even have a synergistic effect on smoking cessation outcomes. For example, an intervention designed to frame the importance of integrating smoking cessation into substance abuse treatment at public facilities is likely to have favorable outcomes on both use of smoking cessation services and smoking ban policies. Likewise, providing greater support or incentives for private institutions to adopt smoking cessation services and smoking ban policies may have a positive effect on facilities in areas with weak tobacco control policies.
The fact that there was a positive relationship between offering smoking cessation services and implementing smoking bans suggests that some substance abuse programs have already taken a comprehensive approach to smoking in their facilities. New interventions should promote this comprehensive approach rather than targeting either cessation treatment or smoking bans alone. An intervention encouraging substance abuse programs to consolidate into larger programs with broader services and greater numbers of admissions, for example, might result in more programs offering smoking cessation treatment, given that larger programs were more likely to offer such programs. However, it could make the implementation of a smoking ban more problematic, given that larger programs were less likely to have such bans. It would be important to consider the differential impact of any intervention on these two outcomes to avoid solving one problem by exacerbating the other.
Conclusions
With increasing evidence suggesting that providing tobacco control services to the substance-abusing population is feasible and effective (
4–
10), the integration of smoking cessation efforts and substance abuse treatment in substance abuse treatment facilities is potentially an effective strategy to reduce the high smoking prevalence among substance-abusing patients. As of 2012, fewer than half of substance abuse treatment facilities in the United States provided smoking cessation services and two-thirds still permitted smoking on facility property. Institutional-level characteristics and state-level tobacco control environments were related to cessation service provision and smoke-free policies. These factors could be used to design comprehensive dissemination programs for smoking cessation treatment and promotion of smoking bans in substance abuse treatment facilities.