Abstract
Background. Tobacco-related chronic diseases contribute significantly to the increased morbidity and mortality observed in the homeless population. Few homeless service settings address tobacco use among their clients. Method. Directors and staff from emergency and transitional shelters in San Diego County completed a questionnaire on no-smoking policies and smoking cessation services and participated in in-depth, semistructured interviews to examine the barriers to and facilitators of implementing smoke-free policies and cessation services in their facilities. Results. Facilities differed in outdoor restrictions around smoking: 61.5% reported having an outdoor designated smoking zone, and 25% reported having a campus-wide ban on smoking. About one-third of the facilities offered on-site resources for smoking cessation. Although directors and staff supported smoke-free policies, they reported that the increased resources needed to “police” the policy created barriers to implementation. Almost all directors and staff expressed interest in developing an on-site tobacco control program, but they reported that lack of expertise among staff posed challenges to implementing such a program. Conclusion. Our findings suggest that for a tobacco control program to be effective in homeless shelters, it should include training and incentives for staff to implement smoke-free policies and cessation services.
Introduction
The prevalence of tobacco use in the homeless population is 70%—almost 3 to 4 times the general population (Baggett & Rigotti, 2010; Okuyemi et al., 2006). Tobacco-related chronic diseases are among the leading causes of morbidity and mortality in the homeless population (Baggett et al., 2015), highlighting a need for comprehensive tobacco control interventions among this population. In this study, we examined the barriers to and facilitators of implementing tobacco control programs in community-based organizations serving homeless adults in order to identify strategies to increase the feasibility of implementing such programs in homeless services settings.
Tobacco control programs that include smoke-free policies and cessation services have been effective in reducing tobacco use at the population level (Messer et al., 2007); however, these interventions are less common in homeless services settings. Homeless shelters may be an effective venue to introduce smoke-free policies and cessation services because most of their clientele are smokers (Arangua, McCarthy, Moskowitz, Gelberg, & Kuo, 2007), of whom the majority are interested in quitting smoking (Okuyemi et al., 2012; Vijayaraghavan & Pierce, 2015). Previous research on smoke-free policies in homeless shelters has shown that most shelters have indoor smoke-free policies (Arangua et al., 2007; Businelle et al., 2014; Vijayaraghavan & Pierce, 2015). However, whether shelters implement other types of tobacco control policies and interventions to support smoking cessation among their clients is unknown.
Similar to homeless shelters, mental health and substance use recovery programs serve populations that have high rates of comorbid substance use disorders, mental health disorders, and nicotine dependence. Research on tobacco control interventions in mental health and substance use recovery centers shows that there is wide variation in the types of policies implemented in these settings (El-Guebaly, Cathcart, Currie, Brown, & Gloster, 2002a; Glover et al., 2014). Some facilities have a permissive smoking culture, while others a smoke-free culture (Glover et al., 2014). Negative staff attitudes toward the smoke-free policy, staff smoking, and lack of knowledge among staff about nicotine addiction pose significant barriers to implementing and enforcing a smoke-free culture in mental health settings and substance use recovery centers (Fuller et al., 2007; Guydish et al., 2012; Schroeder & Morris, 2010). Similar factors may influence the feasibility of implementing tobacco control interventions in homeless shelters.
Among homeless shelters, a previous study showed that most transitional or long-term shelters were receptive to tobacco control efforts to reduce the prevalence of tobacco use among homeless adults (Arangua et al., 2007). No study has examined attitudes, beliefs, and practices around treating tobacco dependence among staff and management of homeless shelters.
In this study, we surveyed directors and staff of emergency and transitional shelters in San Diego County between July 2013 and June 2014 to examine the distribution of no-smoking policies and smoking cessation services in their facilities, and their attitudes toward treating tobacco dependence among their clients. We conducted in-depth, semistructured interviews with directors and staff of a subsample of participating shelters in order to examine barriers to and facilitators of implementing smoke-free policies and cessation services in their facilities.
Method
To recruit eligible emergency and transitional homeless shelters, we obtained an inventory of shelters from the San Diego County Housing and Community Development. Emergency shelters offer short-term housing for up to 3 months, whereas transitional shelters offer supportive services and long-term housing for up to 24 months. All emergency and transitional shelters in San Diego County had indoor no-smoking policies as required by California state law. Of the 83 eligible emergency and transitional housing programs in San Diego County, 52 facilities participated in the study (16 facilities were unresponsive and 15 declined participation, response rate = 62.7%). Eligible participants were directors or staff who worked in the facilities, were years of age or older, and were able to provide informed consent. We obtained responses from 31 directors who represented the 52 eligible emergency and transitional shelters, and 17 staff who represented 11 of the 52 eligible shelters. We compensated participants with $5.00 for completing the questionnaire, and $5.00 for completing the in-depth, semistructured interview. The University of California, San Diego, Institutional Review Board approved all study procedures.
Study Procedures
We contacted directors of each of the eligible facilities and invited them to participate in the study. Trained research staff administered a questionnaire to directors on the no-smoking policies and smoking cessation services in their facilities, as well as their attitudes toward treating tobacco dependence. Directors who were interested in participating in a 45-minute in-depth, semistructured interview were contacted at a later date.
We asked directors of each facility to refer eligible staff who might be interested in participating in the study. Staff were case managers, residential floor managers, substance use counselors, or other types of counselors. Study staff administered a questionnaire to shelter staff and conducted a 45-minutes in-depth, semistructured interview on attitudes toward treating tobacco dependence.
In-depth, semistructured interview discussion guides were developed by the principal investigator and a multidisciplinary team of researchers familiar with tobacco use in the homeless population (Supplementary Table 1, available online). The social ecological model provided the theoretical framework for developing the discussion guide for the in-depth, semistructured interviews with directors and staff (McLeroy, Bibeau, Steckler, & Glanz, 1988). The model considers interactions among policy-, environmental-, organizational-, social-, and individual-level influences on health behavior. We explored these levels of influences in relation to tobacco use, and how they might influence the implementation of tobacco control programs in homeless shelters during the in-depth interviews with directors and staff. Questions were pilot-tested with representatives of the target population.
Director and Staff Measures
Directors reported information on shelter characteristics, including the type of housing (i.e., emergency or transitional), maximum length of stay, capacity, and whether their organization offered on-site behavioral health services, clinical services, or substance use treatment services. Directors reported the number of male and female adult clients in their facilities at the time of the survey.
Directors reported whether smoking was allowed or not allowed in the following areas of their property: indoors in living areas, indoors in shared areas, and outdoor areas of the property. We categorized facilities as having a campus-wide ban if they prohibited smoking in both indoor and outdoor areas of their property. Directors reported whether their facilities had a designated smoking zone, whether there was time-controlled access to the designated smoking zone, and whether they had enforcement policies. We asked directors whether they offered on-site smoking cessation services (i.e., structured smoking cessation class, informal smoking cessation counseling, or smoking cessation included in health and wellness initiatives) or referred outside for smoking cessation (i.e., telephone quit line, primary care provider, or other).
We explored attitudes toward and barriers to treating tobacco dependence among directors and staff by presenting statements on these topics, and obtained responses using a 5-point Likert-type scale (strongly agree, agree, neither agree nor disagree, disagree, and strongly disagree; Table 3). We also assessed practices related to treating tobacco dependence among staff using the same Likert-type scale. Survey items on attitudes and barriers were different for staff and directors to reflect their differing roles and intensity of interactions with clients. Statements around attitudes toward and organizational barriers to providing cessation care were directed toward directors. Because most staff had one-on-one interactions with clients, we asked them to respond to statements on practices related to treating tobacco dependence.
Tobacco Measures and Other Covariates
We categorized directors and staff who had smoked at least 100 cigarettes in their lifetimes as ever smokers, and ever smokers who smoked daily or some days as current smokers. We categorized former smoker who reported not smoking at all as recent former smokers (i.e., those who had quit in the past year) or long-term quitters (i.e., those who had quit more than a year ago). Current smokers reported their daily cigarette consumption, time to first cigarette after waking (<30 minutes vs. ≥30 minutes), intention to quit smoking (never expect to quit, may quit in the next 6 months, will quit in the next 6 months, and will quit in the next month), and whether they had made a quit attempt in the past year. Participants self-reported their age, sex, race/ethnicity (White, African American, Latino, Asian/mixed/other), and education (less than high school, high school or GED, some college, college, or professional training).
Quantitative Data Analysis
We used means (SD) for continuous variables and proportions for categorical variables for all descriptive variables. We showed differences in shelter characteristics between emergency and transitional shelters as well as demographic and tobacco use characteristics between staff and directors using the chi-square statistic for categorical variables and the t test for continuous variables. We dichotomized responses to statements on attitudes toward, barriers to, and practices related to treating tobacco dependence as “agree or strongly agree” versus “disagree or strongly disagree or neither agree nor disagree.” To obtain an overall summary of response to each of the domains (e.g., attitudes, barriers, and practices), we calculated the average score for each domain for each participant. Scores ranged from 1 to 5, with 1 representing the least favorable attitude or highest number of barriers toward treating tobacco dependence and 5 representing the most favorable attitude or least number of barriers toward treating tobacco dependence. We reverse coded negatively worded items. We then presented a group-level (i.e., staff or directors) average for each domain on a scale of 1 to 5. Average group-level scores are not comparable for directors and staff because survey items differed for these groups. We reported p values for differences in overlapping items between staff and directors using logistic regression analysis, accounting for correlation of responses among staff from the same shelter. We conducted all analyses using Stata, Version 11.
Qualitative Data Analysis
The audiotaped in-depth, semistructured interviews were transcribed verbatim by a contracted professional transcription service, and transcribed texts were redacted of any personal identification data. We analyzed qualitative data using a directed content analysis approach (Hsieh & Shannon, 2005), where we used the theoretical framework from the social ecological model and prior research to help focus research questions and identify key concepts (Businelle et al., 2014; McLeroy et al., 1988; Vijayaraghavan & Pierce, 2015). The first author checked the transcripts for accuracy and coded the transcripts. We resolved disagreements in assignment or description of codes through discussion and consensus among the research team, and developed a final codebook. Through iterative processing of the transcripts, we further refined and reduced the number of overall codes by grouping them into a short list of inclusive categories and themes. Features of Microsoft Word and Microsoft Excel were used to organize the transcribed text and to facilitate efficient coding. We chose quotes that were illustrative of the themes we identified.
Results
No-Smoking Policies and Smoking Cessation Services in Emergency and Transitional Shelters in San Diego County
On average, shelters served mostly women (62% women vs. 38% men). Facilities differed in outdoor restrictions on smoking: 61.5% had an outdoor-designated smoking zone, 9.6% had time-restricted access to the smoking zone, and 25.0% had a campus-wide ban on smoking (Table 1). About a third of the facilities offered on-site resources for smoking cessation. A higher proportion of transitional shelters offered on-site behavioral, clinical, or substance use services than emergency shelters, but these results did not attain statistical significance (Table 2).
Facility Characteristics and Smoke-Free Policies and Smoking Cessation Services in Emergency and Transitional Shelters in San Diego County
Complete data are missing for four transitional shelters.
Demographic and Smoking Characteristics of Directors and Staff of Emergency and Transitional Shelters in San Diego County (N = 48)
Those who reported smoking at least 100 cigarettes in their lifetime. bThose who reported smoking some days or every day. cThose who reported smoking not at all.
Attitudes Toward Treating Tobacco Dependence Among Directors and Staff
The mean age of the 31 directors was 44.1 years, and 70.9% were White (Table 2). Among the directors, 45.2% were ever smokers, and most of these ever smokers had quit cigarette smoking (quit rate 61.2%). Almost all the directors agreed that smoke-free policies were important because they promoted a healthy working environment, and that support for smoking cessation should be part of the services they provide to their clients (Table 3). The majority of the directors reported that lack of expertise to provide smoking cessation services and clients’ other priorities posed barriers to addressing tobacco use in their facilities. The average scores were 4.0 (SD = 0.5, α coefficient = .7) and 2.7 (SD = 0.6, α coefficient = .6), respectively, for items describing attitudes toward and barriers to treating tobacco dependence among sheltered homeless clients.
Director and Staff Attitudes Toward, Barriers to, and Practices Related to Addressing Tobacco Dependence Among Clients of Emergency and Transitional Homeless Shelters in San Diego County
Survey items under each domain differ for directors and staff. bp values represent differences between directors and staff in responses to overlapping items under each domain. p values were obtained using logistic regression analysis accounting for correlation of responses by staff in the same facility. cAverage scores represent a group-level average for each domain. A higher score is indicative of a more favorable attitude toward treating tobacco dependence. Group-level averages cannot be compared between directors and staff because survey items are not comparable. dPractices were assessed only among staff.
The mean age of the 17 staff was 41.5 years, and 47.1% were White (Table 2). Among staff, 47.1% were ever smokers, and half of the ever smokers had quit cigarette smoking (quit rate 50.0%). Compared to directors, a higher proportion of staff were smokers (p < .01, Table 2). Similar to directors, almost all staff agreed that smoke-free policies were important because they promoted a healthy working environment (Table 3). Only a third of the staff reported being confident in providing cessation counseling, but almost all agreed to undergo further training to become a cessation counselor. About half of the staff had observed their colleagues smoke with clients, and one of the staff reported having smoked with a client to build a therapeutic relationship, despite all facilities having policies that restricted staff smoking with clients. Directors and staff achieved consensus on almost all of the overlapping items on attitudes toward and barriers to treating tobacco dependence. Directors had a nonsignificantly higher likelihood than staff to agree that strict smoke-free policies would reduce occupancy rates (p = .08; Table 3). A slightly higher proportion of staff than directors agreed that smoking cessation was not a feasible goal for their clients, although these results did not attain statistical significance (p = .07; Table 3). The average scores were 4.0 (SD = 0.6, α coefficient = .6), 3.0 (SD = 0.9, α coefficient = .5), and 3.6 (SD = 0.6, α coefficient = .4), respectively, for items describing attitudes, barriers, and practices related to treating tobacco dependence among sheltered homeless clients.
Qualitative Examination of Attitudes Toward Treating Tobacco Dependence Among Directors
Of the 31 directors, 15 agreed to participate in an in-depth semistructured interview. We identified the following major themes from director interviews: attitudes toward implementing smoke-free policies and organizational role in providing cessation services (see Table 4 for selected quotes).
Quotes Reflecting Major Themes
Attitudes Toward Implementing Smoke-Free Policies
The consensus on smoke-free policies was that the policies were important to protect nonsmokers and children from exposure to secondhand smoke. There were two subthemes that emerged from the interviews: barriers to implementing smoke-free policies and facilitators of implementing the policies. Some of the barriers to implementing smoke-free policies were the increased resources needed to “police” the policy and philosophical differences among staff on the importance of enforcing smoke-free policies. Directors reported that having to develop strategies to address violations was one of the other barriers to implementing policies. For example, enforcement policies that included clients being released from the facilities for violating smoking restrictions led to disagreements among staff and clients. Directors reported that clients’ interest in smoking cessation, restrictive social norms around smoking in California, and the positive effects of policies on the health of staff and clients were some of the facilitators to implementing smoke-free policies in their facilities.
Organizational Role in Providing Cessation Services
There was consensus among directors on the importance of addressing tobacco use among their clients; however, disagreements arose on the intensity with which shelters should take on the role as smoking cessation champions. The following subthemes emerged from the interviews: emphasis on recovery from drug and alcohol addiction, lack of resources to treat tobacco dependence, and competing priorities. While all directors reported asking for information on drugs and alcohol during their intake evaluation with clients, no directors reported asking about tobacco use. All directors acknowledged that although tobacco use was detrimental to their clients’ health, it was “legal” and did not affect clients’ progress in the program, and therefore tobacco cessation was not prioritized.
Most directors reported that it would be feasible to provide information on smoking cessation on-site, particularly if attending the programs was incentivized. Directors reported that financial incentives, credit toward meeting mandatory requirements for the program, or passes to leave the facility for outdoor excursions were the types of incentives that might be beneficial to this population. All directors reported that education on smoking cessation could be incorporated into support groups that were already offered to clients, and should be combined with access to pharmaceutical support for smoking cessation. However, while all directors reported interest in implementing a tobacco-counseling program, they also reported lack of knowledge and expertise on this topic. Most directors were willing to train their staff to provide cessation counseling, but some were unwilling to do so because of constraints on staff time. Only a few directors were of the opinion that smoking cessation should be “client-driven” and that organizations should not “overload [clients] with additional expectations of smoking cessation” because of competing priorities.
Qualitative Examination of Attitudes Toward Treating Tobacco Dependence Among Staff
Of the 17 staff, 13 agreed to participate in an in-depth, semistructured interview. The major themes that emerged from the staff interviews were attitudes toward smoke-free policies and staff roles in encouraging smoking cessation among clients.
Attitudes Toward Smoke-Free Policies
The consensus among staff was that smoke-free policies were important because they provided a healthier working environment and reduced exposure to “harmful effects of secondhand smoke.” Similar to directors, some of the staff reported that the time spent and resources needed to “police” the policy were some of the barriers to implementing facility no-smoking policies in their facilities. A few of the staff reported having seen their colleagues smoke with clients and reported that these “boundary issues” sent “mixed messages” to clients on the importance of smoke-free policies.
Staff Roles in Encouraging Smoking Cessation Among Clients
There was consensus among staff that educational initiatives to increase awareness about the harms of tobacco use should be built into the curriculum of their programs. Almost all staff including those who were smokers were amenable to taking on a role as smoking cessation counselor and expressed interest in being trained to become cessation counselors. Only one of the staff believed that smoking cessation counseling would not “fall within my responsibilities.” A few of the staff reported that disagreements and apathy among their colleagues on the importance of addressing tobacco use among clients were some of the barriers to building a tobacco cessation program in their facility, and that such initiatives would need support from the leadership and should include incentives for staff to be sustainable.
Discussion
In this study, we found that all participating emergency and transitional shelters in San Diego County had indoor no-smoking policies, two thirds had a partial outdoor ban on smoking, and one fourth prohibited smoking completely on their properties. In contrast, a study on smoke-free policies in transitional shelters in Los Angeles County from a decade ago showed that three fourths of the transitional shelters in Los Angeles County had indoor no-smoking policies and/or designated smoking zones, but none had campus-wide restrictions on smoking (Arangua et al., 2007). Our findings may be reflective of California’s antitobacco social norms, which may be stronger today compared to a decade ago. The majority of directors and staff were supportive of smoke-free policies and expressed positive attitudes toward implementing programs to address tobacco use among their clients, but they also reported significant barriers to doing so. Our findings highlight several key strategies to increase the feasibility of a tobacco control program that includes smoke-free policies and cessation services in homeless service settings.
Consistent with a previous study that explored service providers’ perspectives on providing cessation programs to homeless youth (Shadel, Tucker, Mullins, & Staplefoote, 2014), we found that lack of resources (e.g., money or personnel or time), staff training, and concerns about clients’ competing priorities posed significant barriers to addressing tobacco use in homeless service settings. Staff smoking with clients and philosophical differences among staff on the importance of addressing tobacco use were other barriers to implementing no-smoking policies and smoking cessation programs in emergency and transitional shelters in San Diego County. Among health care professionals providing services to homeless adults, 15% reported having used tobacco to engage or build a therapeutic relationship with patients (Baggett et al., 2012). In the same study, there were significant disagreements among providers on attitudes toward tobacco use in the homeless population, with some providers having a permissive attitude (Baggett et al., 2012). These barriers to implementing tobacco control programs have also been observed in substance use recovery programs and mental health services settings that serve populations that overlap significantly with the homeless population in their rates of smoking and comorbid mental health and substance use disorders (El-Guebaly, Cathcart, Currie, Brown, & Gloster, 2002b; Fuller et al., 2007).
Previous research has demonstrated the feasibility of capacity building interventions to reduce organizational barriers to implementing smoking cessation programs in substance use recovery programs and community-based organizations (Guydish et al., 2012; O’Brien et al., 2012). Findings from our study support these types of interventions that might include training staff to provide brief or more intensive cessation counseling (Guydish et al., 2012; O’Brien et al., 2012; Shadel et al., 2014) and improving knowledge among staff on clients’ interest in smoking cessation despite competing priorities (Baggett, Lebrun-Harris, & Rigotti, 2013; Connor, Cook, Herbert, Neal, & Williams, 2002). Although directors and staff achieved consensus on almost all overlapping items on attitudes toward and barriers to treating tobacco dependence, staff had a nonsignificantly higher likelihood than directors to agree that smoking cessation was not a feasible goal for their clients. These findings highlight a role for directors to provide leadership in incorporating tobacco cessation counseling as a formal service to shelter clientele and in modifying beliefs and attitudes among staff on the importance of addressing tobacco addiction (Guydish et al., 2012; O’Brien et al., 2012; Porter, Houston, Anderson, & Maryman, 2011; Shadel et al., 2014). Partnering with community-based organizations and health centers to provide cessation counseling and pharmaceutical support for smoking cessation and streamlining policies to incentivize efforts at smoking cessation among staff and clients may increase the feasibility of a tobacco control program in homeless service settings (Guydish et al., 2012; O’Brien et al., 2012; Shadel et al., 2014). Given the constraints on staff time, incentivizing staff to become smoking cessation champions will be crucial to the sustainability of a smoking cessation program in homeless shelters.
We found nonstatistically significant differences in the distribution of smoke-free policies and cessation services between emergency and transitional shelters. Whether these differences are associated with change in clientele smoking behaviors warrants further exploration. Strong smoke-free policies have the potential to change norms around smoking and encourage change in smoking behaviors among sheltered homeless adults (Vijayaraghavan & Pierce, 2015). Concerns that smoke-free policies may reduce occupancy rates in shelters are unfounded (Businelle et al., 2014; Vijayaraghavan & Pierce, 2015). The results of our study suggest that shelters should consider implementing smoke-free policies in indoor and outdoor areas of the property and discourage staff smoking with clients. California’s strong antitobacco norms were one of the primary motivators for some facilities to change their smoke-free policies. These findings highlight a role for local tobacco control organizations to provide tool kits to encourage community-based organizations to strengthen their smoke-free policies.
Limitations
Our study had several limitations. Emergency and transitional shelters that participated in this study were from a single county in Southern California. Although San Diego County has a large population of homeless adults (Regional Task Force on the Homeless, 2012), the findings from our study may not be reflective of perspectives of service providers from other parts of the United States. The state of California has been leading in efforts to denormalize tobacco use in the general population; thus, attitudes toward smoke-free policies may be different among service providers in other states where tobacco control has been less of a priority. While nonparticipating facilities may have had indoor smoke-free policies as required by California state law, we were unable to determine whether they had other no-smoking policies or smoking cessation programs. These characteristics may have influenced their decision to participate in our study, potentially leading to a selection bias. The perspectives of directors and staff who were interviewed in this study may not apply to other staff in the participating facility. Shelters serving homeless families and women were overrepresented in our study sample; thus, results of our study may not be representative of shelters in San Diego County serving single adult males.
Implications for Practice
Despite these limitations, these results provide a first glimpse of homeless services providers’ perspectives on implementing no-smoking policies and cessation programs for sheltered homeless adults. Our findings suggest that for a tobacco control program to be effective in homeless service settings, the program should include smoke-free policies and incentives for staff to undergo training to implement a tobacco control program and clients to engage in smoking cessation behaviors.
Footnotes
Acknowledgements
We would like to thank the staff and leadership of emergency and transitional shelters in San Diego County for their participation in our study. We would also like to thank Nicole Campbell, MA, for her help in conducting some of the study interviews and coding transcripts from in-depth, semistructured interviews.
This work was supported by the California Tobacco-Related Disease Research Program Grant 22XT-0020.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
