Despite a smoking prevalence approaching 70% among homeless adults in the US,1,2 little is known about ways to intervene on smoking behaviour in this marginalised population. Tobacco control advocates point to marketing by the tobacco industry and the pervasiveness and social acceptance of tobacco use in homeless settings as barriers to promoting effective cessation of smoking and smoke‐free environments in this vulnerable population.3 They maintain that homeless service providers continue to hold apparently common assumptions that tobacco is a resource and that their clients have higher priority needs than to quit smoking.3,4,5 Recent studies, however, have begun to challenge these assumptions.2,3,4,5,6 In a series of focus groups and interviews exploring tobacco use behaviours among the homeless,2 up to 76% of homeless persons interviewed reported an intention to quit smoking in the next 6 months. Recent investigations in tobacco, alcohol and drug addiction have also demonstrated that rehabilitation programmes for misusers are commonly more effective when risky health behaviours are addressed in a supportive, stable environment,7,8 suggesting that an optimal venue for targeting a tobacco control intervention to the homeless smoker might be in an established, protective environment such as long‐term transitional housing. In this communication, we present original data in support of this emerging viewpoint, examining institutional and clientele factors that may influence the receptivity of transitional shelters to environmental tobacco control interventions.
A telephone survey of all long‐term transitional shelters in Los Angeles County was conducted from March to June 2005. The 20 min institutional survey utilised a pretested 31‐item questionnaire to query administrative officials at each participating shelter about resident demographics, institutional characteristics (eg, size, average length of stay, etc), inventory of existing “no smoking” policies and the willingness of each facility to participate in environmental tobacco control efforts. The study inclusion criteria included that (1) each facility must be operational at the time of the survey; (2) each facility must offer at least a 12‐month length of stay to their homeless residents to qualify as a “long‐term” transitional shelter; and (3) each facility must be a transitional housing programme based on the continuum of care model for rehabilitating homeless individuals back to mainstream society.9
Bivariate and multivariate analyses were conducted using the STATA 9.0 statistical software package to evaluate the relationships between shelter receptivity to environmental tobacco control interventions (dependent variable) and institutional‐resident characteristics (independent variables). To understand whether tobacco control policies varied by size of facility, subgroup analyses were conducted by facility‐size categories based on the maximum number of beds that were available at each facility: small (<50 beds), medium (50–199 beds) or large (>200 beds). To estimate individual‐level statistics from these facility‐specific data, some sociodemographic measures were weighted by the number of beds reported for the facility; other statistics were computed directly from shelter‐level data.
In all, 76 transitional shelters met the study inclusion criteria and were contacted (table 1). Of these, 71 (93.4%) participated in the survey. Nearly a quarter (23%) had a programme that focused on helping individuals who were addicted to drugs or to alcohol to reduce their dependence on these substances. Another third (30%) had programmes that focused on helping mothers with small children transition to more permanent housing; eight (11%) were dedicated to housing the mentally ill; three (4%) focused on helping war veterans, emancipated youth or gay men find more permanent housing. A majority of the transitional shelters reported having an indoor “no smoking” policy (75%) and designated smoking areas (78%). A total of, 72% and 95%, respectively, reported that they would be open to adopting new anti‐smoking policies and environmental tobacco control interventions designed to reduce resident smoking.
Table 1 Clientele and shelter characteristics by size of the facility*.
| Characteristic(s) | Overall† (n = 71) | Small (n = 38) | Medium (n = 21) | Large (n = 12) | p Value‡ |
|---|---|---|---|---|---|
| Clientele–shelter residents | |||||
| Gender | % | % | % | % | |
| Female | 60 | 81 | 58 | 62 | <0.05 |
| Male | 40 | 19 | 42 | 38 | — |
| Age | |||||
| Emancipated youth (%) | 2 | — | 3 | 3 | 0.428 |
| Age 18–30 (%) | 29 | 29 | 26 | 30 | 0.898 |
| Age 31–50 (%) | 34 | 38 | 38 | 24 | 0.561 |
| Race/ethnicity | % | % | % | % | |
| Latino/Hispanic | 26 | 29 | 24 | 24 | 0.239 |
| Black | 47 | 38 | 51 | 55 | <0.05 |
| Facility | |||||
| No of beds and clients | |||||
| Maximum number of beds | 141 | 26 | 82 | 271 | <0.001 |
| Average number of clients per night | 129 | 24 | 71 | 213 | <0.001 |
| Length of stay | |||||
| Maximum length of stay (months) | 28 | 21 | 22 | 37 | 0.199 |
| Average length of stay (months) | 15 | 16 | 11 | 21 | <0.05 |
| Current institutional tobacco control policy | |||||
| Indoor no smoking” policy (% yes) | 75 | 95 | 100 | 57 | <0.001 |
| Designated no smoking area (% yes) | 78 | 82 | 86 | 50 | <0.001 |
| Attempt to limit smoking (% yes) | 63 | 59 | 50 | 83 | <0.001 |
| Receptivity to tobacco control | |||||
| Receptive (% yes) | 72 | 89 | 76 | 67 | 0.267 |
| Willing to participate in programme? (% yes) | 95 | 94 | 90 | 100 | 0.689 |
| Organisational readiness to change§ | |||||
| Policy open to staff input? | |||||
| Agree (%) | 64 | 59 | 64 | 57 | <0.01 |
| Strongly agree (%) | 27 | 38 | 14 | 14 | — |
*Shelter size is categorised as small (<50 beds), medium (50–199 beds) or large (>200 beds) based on the maximum number of beds available at each facility. The cut‐off points between the three categories are informed by a review of the available public data on transitional shelters in Los Angeles county, including careful considerations of the geographic distribution and size‐related characteristics of each facility (eg, types of special services offered and target population).
†Weighted data; adjusted for size of facility.
‡p Values are derived from statistical analyses of each dependent variable with respect to size. Appropriate statistical procedures were performed based on the intent of the analysis and whether the variable distributions were normal or not normal. Statistical methods included: t test, χ2 analysis, Kruskal–Wallis (non‐parametric) methods and linear regression procedures.
§Based on the perception of the organisational official who participated in the survey.
Results from comparative analyses suggest that the size of the facility was inversely related to the likelihood that the facility would report having a current indoor no smoking policy (p<0.001), a designated no smoking area (p<0.001) and a policy to limit smoking among its residents (p<0.001) (table 1). The size of the facility and the service mission of the organisation (eg, those with substance abuse treatment programmes) also predicted greater receptivity towards environmental tobacco control interventions (p<0.001). These findings suggest that enforcement of a smoke‐free environment may be more difficult at larger facilities or in facilities without experience in treatment for substance misuse.
Contrary to the prevailing perception among tobacco control advocates that homeless service providers have casual attitudes towards client smoking behaviour,3 the present data showed that homeless service facilities, such as long‐term transitional housing, recognise the importance of intervening on this high‐risk behaviour; they are highly receptive to tobacco control efforts designed to reduce the prevalence of smoking in this population. This emerging viewpoint in the homeless community would suggest that greater investment of tobacco‐control resources, tailored to the unique needs of the homeless, would be justified and deserving of further investigation.3,10
Footnotes
Competing interests: None declared.
References
- 1.US Department of Health and Human Services (USDHHS). The health consequences of smoking: a report of the surgeon‐general Washington, DC: US Printing Office, 2004
- 2.Okuyemi K S, Caldwell A R, Thomas J L.et al Homelessness and smoking cessation: insights from focus groups. Nicotine Tob Res 20068287–296. [DOI] [PubMed] [Google Scholar]
- 3.Apollonio D E, Malone R E. Marketing to the marginalized: tobacco industry targeting of the homeless and mentally ill. Tob Control 200514409–415. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Arnsten J H, Reid K, Bierer M.et al Smoking behavior and interest in quitting among homeless smokers. Addict Behav 2004291155–1161. [DOI] [PubMed] [Google Scholar]
- 5.Butler J, Okuyemi K S, Jean S.et al Smoking characteristics of a homeless population. Subst Abuse 200223223–231. [DOI] [PubMed] [Google Scholar]
- 6.Connor S E, Cook R L, Herbert M I.et al Smoking cessation in a homeless population: there is a will, but is there a way? J Gen Intern Med 200217369–372. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Greene J M, Ennett S T, Ringwalt C L. Substance use among runaway and homeless youth in three national samples. Am J Public Health 199787229–235. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.McCarthy W J, Collins C, Hser Y I. Does cigarette smoking affect drug abuse treatment? J Drug Issues 20023261–79. [Google Scholar]
- 9.Robertson M J. The prevalence of mental disorder among homeless people. In: Jahiel R, eds. Homelessness: a prevention‐oriented approach Baltimore and London: Johns Hopkins University Press, 199257–86.
- 10.Rohrbach L A, Howard‐Pitney B, Unger J B.et al Independent evaluation of the California Tobacco Control Program: relationships between program exposure and outcomes, 1996–1998. Am J Pub Health 200292975–983. [DOI] [PMC free article] [PubMed] [Google Scholar]