Advertisement

Cigarette Smoking and Advice to Quit in a National Sample of Homeless Adults

  • Travis P. Baggett
    Correspondence
    Address correspondence to: Travis P. Baggett, MD, MPH, 50 Staniford Street, 9th Floor, Boston MA 02114
    Affiliations
    General Medicine Division, Massachusetts General Hospital, Boston, Massachusetts

    Department of Medicine, Harvard Medical School, Boston, Massachusetts
    Search for articles by this author
  • Nancy A. Rigotti
    Affiliations
    General Medicine Division, Massachusetts General Hospital, Boston, Massachusetts

    Tobacco Research and Treatment Center, Massachusetts General Hospital, Boston, Massachusetts

    Department of Medicine, Harvard Medical School, Boston, Massachusetts
    Search for articles by this author

      Background

      Cigarette smoking is common among homeless people, but its characteristics in this vulnerable population have not been studied at a national level. Whether homeless smokers receive advice to quit from healthcare providers is also unknown.

      Purpose

      To determine the prevalence and predictors of current cigarette smoking, smoking cessation, and receipt of clinician advice to quit in a national sample of homeless adults.

      Methods

      This study analyzed data from 966 adult respondents to the 2003 Health Care for the Homeless User Survey, representing more than 436,000 people nationally. Using multivariable logistic regression, the independent predictors of smoking, quitting, and receiving advice to quit were identified. Analyses were conducted in 2008–2009.

      Results

      The prevalence of current smoking was 73%. The lifetime quit rate among ever smokers was 9%. Among past-year smokers, 54% reported receiving clinician advice to quit. Factors independently associated with current smoking included out-of-home placement in childhood (AOR=2.79, 95% CI=1.03, 7.52); victimization while homeless (AOR=2.36, 95% CI=1.15, 4.83); past-year employment (AOR=2.52, 95% CI=1.13, 5.58); and prior illicit drug use (AOR=7.21, 95% CI=3.11, 16.7) or problem alcohol use (AOR=7.42, 95% CI=2.51, 21.9). Respondents with multiple homeless episodes had higher odds of receiving quit advice (AOR=2.51, 95% CI=1.30, 4.83) but lower odds of quitting (AOR=0.47, 95% CI=0.29, 0.78).

      Conclusions

      Compared to the general population, homeless people are far more likely to smoke and much less likely to quit, even though more than half of smokers received quit advice in the past year. Interventions for homeless smokers should address the unique comorbidities and vulnerabilities of this population.

      Introduction

      Cigarette smoking is common among homeless people,
      • Connor S.E.
      • Cook R.L.
      • Herbert M.I.
      • Neal S.M.
      • Williams J.T.
      Smoking cessation in a homeless population: there is a will, but is there a way?.
      • Snyder L.D.
      • Eisner M.D.
      Obstructive lung disease among the urban homeless.
      • Gelberg L.
      • Linn L.S.
      • Usatine R.P.
      • Smith M.H.
      Health, homelessness, and poverty A study of clinic users.
      • Lee T.C.
      • Hanlon J.G.
      • Ben-David J.
      • et al.
      Risk factors for cardiovascular disease in homeless adults.
      • Szerlip M.I.
      • Szerlip H.M.
      Identification of cardiovascular risk factors in homeless adults.
      • Gelberg L.
      • Linn L.S.
      Assessing the physical health of homeless adults.
      • Weinreb L.
      • Goldberg R.
      • Perloff J.
      Health characteristics and medical service use patterns of sheltered homeless and low-income housed mothers.
      contributing to the high prevalence of acute and chronic illness in this population.
      • Wright J.D.
      The health of homeless people: evidence from the National Health Care for the Homeless Program.
      • Burt M.R.
      Urban Institute
      Homelessness: programs and the people they serve: findings of the National Survey of Homeless Assistance Providers and Clients: technical report.
      • Breakey W.R.
      • Fischer P.J.
      • Kramer M.
      • et al.
      Health and mental health problems of homeless men and women in Baltimore.
      Obstructive lung disease is more than twice as prevalent in homeless people than in the general population,
      • Snyder L.D.
      • Eisner M.D.
      Obstructive lung disease among the urban homeless.
      and the rates of death from cardiovascular, pulmonary, and other smoking-related causes are substantial.
      • Hwang S.W.
      • Orav E.J.
      • O'Connell J.J.
      • Lebow J.M.
      • Brennan T.A.
      Causes of death in homeless adults in Boston.
      • Morrison D.S.
      Homelessness as an independent risk factor for mortality: results from a retrospective cohort study.
      • Hwang S.W.
      • Wilkins R.
      • Tjepkema M.
      • O'Campo P.J.
      • Dunn J.R.
      Mortality among residents of shelters, rooming houses, and hotels in Canada: 11 year follow-up study.
      Despite this, the characteristics and natural history of smoking behavior in a homeless population are not well described. The high prevalence of tobacco use among homeless people is commonly attributed to the disproportionate burden of substance abuse and mental illness, superimposed on a background of heightened personal stress and social chaos.
      • Scanlan B.C.
      • Brickner P.W.
      Clinical concerns in the care of homeless persons.
      • Okuyemi K.S.
      • Caldwell A.R.
      • Thomas J.L.
      • et al.
      Homelessness and smoking cessation: insights from focus groups.
      Yet the independent contributions of these and other factors unique to homelessness on smoking behavior have not been assessed.
      Listen to related Podcast at www.ajpm-online.net.
      Furthermore, the extent to which smoking is addressed by clinicians caring for this population is not known. Evidence-based guidelines direct healthcare providers to address tobacco use routinely when a smoker makes an office visit,
      U.S. Preventive Services Task Force
      Counseling and interventions to prevent tobacco use and tobacco-caused disease in adults and pregnant women: U.S. Preventive Services Task Force reaffirmation recommendation statement.
      because even brief advice improves cessation rates.
      • Fiore M.C.
      • Jaen C.R.
      • Baker T.B.
      Treating tobacco use and dependence: 2008 update Clinical practice guideline.
      • Lancaster T.
      • Stead L.
      Physician advice for smoking cessation.
      In the setting of homelessness, however, tobacco use may be overshadowed by more urgent medical, psychiatric, and social concerns at clinical encounters.
      • Scanlan B.C.
      • Brickner P.W.
      Clinical concerns in the care of homeless persons.
      Also, clinicians may be unaware that many homeless smokers are interested in quitting.
      • Arnsten J.H.
      • Reid K.
      • Bierer M.
      • Rigotti N.
      Smoking behavior and interest in quitting among homeless smokers.
      • Butler J.
      • Okuyemi K.S.
      • Jean S.
      • Nazir N.
      • Ahluwalia J.S.
      • Resnicow K.
      Smoking characteristics of a homeless population.
      Understanding the pattern of tobacco use among homeless individuals at a national level would provide valuable insight into what components might be needed for effective intervention strategies for this vulnerable population. To address this gap in evidence, data from a unique national survey of homeless adults were analyzed to determine the prevalence and correlates of current cigarette smoking, smoking cessation, and receipt of clinician advice to quit.

      Methods

      Data Source

      In 2008–2009, the authors performed a secondary analysis of the 2003 Health Care for the Homeless (HCH) User Survey, the first nationally representative survey of people using clinical services provided by the federally funded HCH program. The HCH User Survey was administered by Research Triangle Institute (RTI) International in collaboration with the Health Resources and Services Administration's Bureau of Primary HealthCare.

      Participants and Setting

      A three-stage sampling design was used to conduct the survey.
      • Greene J.
      • Fahrney K.
      • Byron M.
      Health Care for the Homeless User/Visit Surveys, RTI Project Number 07147.021.
      Of 131 HCH grantees that had been in operation for at least 1 year, 30 were sampled using a geographically stratified probability-proportional-to-size (PPS) technique. Interviews were conducted in person by RTI field staff at a PPS sample of 79 HCH clinic sites operated by the 30 grantees. The target population was defined as people receiving face-to-face services from an HCH provider. Individuals were eligible if they had received such services at least once in the year prior to the survey, because the reference period for many of the questions was 12 months. Participants were selected consecutively with a goal of 33 interviews per grantee. Respondents provided informed consent, and the study was approved by the RTI International IRB.
      Of 1444 selected people, 11 were subsequently found to be ineligible, and 416 refused or did not complete the survey. The total number of completed interviews was 1017, yielding a response rate of 70%. This analysis was confined to the 966 respondents who were aged ≥18 years, representing more than 436,000 adult HCH clinic users.

      Definitions of Smoking Behavior and Outcomes

      Three outcomes were analyzed: (1) current smoking; (2) smoking cessation; and (3) receipt of clinician advice to quit smoking. Smoking status was defined in response to the following two questions: Have you ever smoked at least 100 cigarettes in your entire life? and How long has it been since you last smoked a cigarette? Response categories for the second question were within the past 30 days, more than 30 days, but within the past 12 months, and more than 12 months ago. Respondents who had ever smoked at least 100 cigarettes were defined as ever smokers, and respondents who had never smoked 100 cigarettes were classified as never smokers.
      Cigarette smoking among adults—U.S., 2003.
      Among ever smokers, those who had smoked within the past 30 days were considered current smokers, whereas those who had not smoked within the past 30 days were considered former smokers. The quit ratio was defined as the percentage of ever smokers who were former smokers.
      • Giovino G.A.
      • Schooley M.W.
      • Zhu B.P.
      • et al.
      Surveillance for selected tobacco-use behaviors—U.S., 1900–1994.
      Receipt of past-year quit advice by an HCH provider was assessed with the following question: During the past 12 months, did anyone at the HCH Health Center advise you to stop smoking for more than one day?

      Covariates of Interest

      Sociodemographic variables included age (18–29, 30–44, ≥45 years); gender; self-reported race/ethnicity (white non-Hispanic, black non-Hispanic, other non-Hispanic, and Hispanic); veteran status; marital status (married/partnered versus not); educational attainment (high school diploma or higher versus less than high school diploma); and employment (any past-year work for pay).
      Features of homelessness included out-of-home placement as a minor (placement into a foster home, group home, or other institution before age 18 years); the lifetime number of homeless episodes lasting at least 30 days (0, 1, 2 or more); victimization history (physical or sexual assault while homeless); and food insufficiency (sometimes or often not getting enough food to eat
      • Alaimo K.
      • Briefel R.R.
      • Frongillo Jr, E.A.
      • Olson C.M.
      Food insufficiency exists in the U.S.: results from the third National Health and Nutrition Examination Survey (NHANES III).
      • Alaimo K.
      • Olson C.M.
      • Frongillo Jr, E.A.
      • Briefel R.R.
      Food insufficiency, family income, and health in U.S. preschool and school-aged children.
      • Rose D.
      • Oliveira V.
      Nutrient intakes of individuals from food-insufficient households in the U.S..
      • Vozoris N.T.
      • Tarasuk V.S.
      Household food insufficiency is associated with poorer health.
      • Briefel R.R.
      • Woteki C.E.
      Development of food sufficiency questions for the Third National Health and Nutrition Examination Survey.
      • Cristofar S.P.
      • Basiotis P.P.
      Dietary intake and selected characteristics of women age 19–50 years and their children age 1–5 years by reported perception of food sufficiency.
      ).
      Comorbid psychiatric and addictive disorders included a history of mental illness, illicit drug use, or problem alcohol use. Mental illness history was defined as any past inpatient or outpatient treatment for “emotional or mental problems.” Illicit drug use history was defined as any past or current use of illicit or nonprescribed controlled substances, or any history of drug treatment. Problem alcohol use history was defined as consuming ≥5 drinks on a single occasion at least once in the past month,
      National Institute on Alcohol Abuse and Alcoholism
      NIAAA Council approves definition of binge drinking NIAAA Newsletter (Winter).
      consuming ≥3 drinks on a typical day of drinking in the past month,
      • Babor T.F.
      • Higgins-Biddle J.C.
      • Saunders J.B.
      • Monteiro M.G.
      The alcohol use disorders test: guidelines for use in primary care.
      or any history of alcohol treatment.
      In the smoking-cessation and quit-advice analyses, the following smoking-attributable medical comorbidities were considered: obstructive lung disease (asthma, chronic bronchitis, and COPD/emphysema); cardiovascular disease (coronary artery disease or stroke); hypertension; and any cancer diagnosis. The duration of time followed by the HCH clinic site was included in the quit advice analysis to account for differences in opportunity for advice.

      Statistical Analysis

      The authors calculated the prevalence of current smoking among all adults, the prevalence of smoking cessation among ever smokers (i.e., quit ratio), and the prevalence of past-year quit advice among ever smokers who had smoked at any time in the preceding year. The unadjusted relationships between these outcomes and the predictors of interest were examined using chi-square tests. Separate multivariable logistic regression models were then constructed to determine the factors independently associated with each outcome. In the multivariable model for current smoking, former smokers were excluded from the analysis in order to better compare current and never smokers.
      • Fagan P.
      • Shavers V.
      • Lawrence D.
      • Gibson J.T.
      • Ponder P.
      Cigarette smoking and quitting behaviors among unemployed adults in the U.S..
      In the smoking-cessation model, former smokers were compared to current smokers to determine the factors associated with quitting. In the quit-advice model, all past-year smokers were analyzed to determine the factors associated with receiving advice to stop smoking. For the current smoking and quit advice outcomes, all hypothesized variables were entered into the regression models. For the smoking-cessation outcome, the limited number of quit events restricted the number of predictors that could be entered into the regression model;
      • Peduzzi P.
      • Concato J.
      • Kemper E.
      • Holford T.R.
      • Feinstein A.R.
      A simulation study of the number of events per variable in logistic regression analysis.
      therefore, only variables with univariate significance of p<0.05 were entered into the model to avoid overfitting. Respondents with missing data for a predictor or outcome were excluded from any analysis involving the variable. Smoking status was missing for three respondents. There were no missing data for quit advice. Item nonresponse for predictors of interest was generally minimal (range=0%–3.6%).
      Because of the complex sampling design of the survey, all analyses were performed on SAS-Callable SUDAAN, release 10.0, using weights developed by RTI that incorporated sampling probability, survey nonresponse, and poststratification adjustment to reduce bias in the study estimates and better reflect the target population.
      • Greene J.
      • Fahrney K.
      • Byron M.
      Health Care for the Homeless User/Visit Surveys, RTI Project Number 07147.021.
      Descriptive data are presented as unweighted numbers and weighted percentages. Results were considered significant at p <0.05.

      Results

      Respondent Characteristics

      Fifty-eight percent of respondents were male (Table 1). The median age was 41 years. Most participants were white (39%) or black (38%) non-Hispanics. In comparison to the 2003 U.S. general population, this sample of homeless adults had a greater proportion of non-Hispanic blacks and a lower proportion of high school graduates.
      U.S. Census Bureau
      2003 American Community Survey: summary tables.
      Table 1Characteristics of the study sample (n=966)
      Data source: 2003 Health Care for the Homeless User Survey, Health Resources and Services Administration
      Characteristicsn (%)
      Sample sizes are unweighted. Percentages are weighted to be nationally representative of adult Health Care for the Homeless clinic users, with sampling weights provided by Research Triangle Institute International. Percentages within categories may not total 100 because of rounding or item nonresponse.
      SOCIODEMOGRAPHIC CHARACTERISTICS
      Age (years)
       18–29113 (18.3)
       30–44395 (50.3)
       ≥45458 (31.5)
      Gender, male591 (58.2)
      Race/ethnicity
       White non-Hispanic334 (39.2)
       Black non-Hispanic382 (38.1)
       Other non-Hispanic66 (5.3)
       Hispanic172 (16.1)
      Veteran119 (11.5)
      Married/partnered127 (14.3)
      High school diploma580 (58.7)
      Employed, past year485 (54.5)
      FEATURES OF HOMELESSNESS
      Out-of-home placement as a minor
      Defined as placement into a foster home, group home, or other institution before age 18 years
      204 (23.8)
      Homeless episodes lasting ≥30 days
       0198 (16.8)
       1249 (26.3)
       ≥2498 (55.3)
      Food insufficiency
      Defined as sometimes or often not getting enough to eat
      235 (25.4)
      Physical/sexual assault history291 (32.8)
      PSYCHIATRIC/ADDICTIVE DISORDERS
      Mental illness history429 (47.6)
      Illicit drug use history
       Lifetime778 (82.6)
       Past year522 (56.3)
      Problem alcohol use history
       Lifetime502 (52.0)
       Past year383 (39.7)
      MEDICAL CONDITIONS
      Obstructive lung disease315 (36.4)
      Cardiovascular disease107 (8.8)
      Hypertension308 (28.5)
      Cancer54 (7.0)
      a Sample sizes are unweighted. Percentages are weighted to be nationally representative of adult Health Care for the Homeless clinic users, with sampling weights provided by Research Triangle Institute International. Percentages within categories may not total 100 because of rounding or item nonresponse.
      b Defined as placement into a foster home, group home, or other institution before age 18 years
      c Defined as sometimes or often not getting enough to eat
      More than half of respondents had worked in the past year. Nearly one fourth had a history of out-of-home placement in childhood. One in three reported being physically or sexually assaulted while homeless. Almost half had a history of mental illness. The lifetime prevalence of illicit drug use (83%) and problem alcohol use (52%) was substantial.

      Current Smoking

      Overall, 73% of respondents were current smokers, 7% were former smokers, and 20% were never smokers (Table 2). Among current smokers (n=677), the mean number of cigarettes consumed per day was 14 (SE=0.8). Thirty-five percent (n=231) of current smokers consumed ≥20 cigarettes per day on average over the past month, and 84% (n=572) smoked every day of the past month.
      Table 2Smoking status and predictors of current smoking
      Data source: 2003 Health Care for the Homeless User Survey, Health Resources and Services Administration
      CharacteristicsSmoking status (n=966)
      Denominator is entire study sample (n=966). Current smoking = ever smoked 100 cigarettes and smoked in past 30 days; former smoking = ever smoked 100 cigarettes and has not smoked in past 30 days; never smoking = never smoked at least 100 cigarettes.
      Current vs never smoking (n=877)
      Denominator consists of current and never smokers (n=877); former smokers are excluded from this analysis. The multivariable model is adjusted for all variables displayed in the table using logistic regression.
      adjusted OR (95% CI)
      % current% former% never
      Overall72.77.419.7
      SOCIODEMOGRAPHIC CHARACTERISTICS
      Age (years)
       18–2968.92.128.21.00 (ref)
       30–4473.47.519.02.13 (0.92, 4.94)
       ≥4573.710.315.96.16 (2.30, 16.5)
      Gender
       Female61.57.031.11.00 (ref)
       Male80.37.811.83.00 (0.95, 9.49)
      Race/ethnicity
       White non-Hispanic80.86.112.71.00 (ref)
       Black non-Hispanic72.06.721.00.81 (0.34, 1.94)
       Other non-Hispanic72.014.913.10.34 (0.11, 1.09)
       Hispanic54.610.135.30.33 (0.07, 1.43)
      Veteran
       No72.77.819.31.00 (ref)
       Yes72.24.322.80.16 (0.06, 0.45)
      Married/partnered
       No75.57.616.61.00 (ref)
       Yes55.66.637.81.51 (0.72, 3.18)
      High school diploma
       No71.47.121.31.00 (ref)
       Yes73.57.618.60.57 (0.28, 1.15)
      Employed, past year
       No65.27.827.01.00 (ref)
       Yes78.87.113.72.52 (1.13, 5.58)
      FEATURES OF HOMELESSNESS
      Out-of-home placement as a minor
      Defined as placement into a foster home, group home, or other institution before age 18 years
       No68.96.524.31.00 (ref)
       Yes82.010.97.12.79 (1.03, 7.52)
      Homeless episodes lasting ≥30 days
       053.97.938.21.00 (ref)
       172.310.617.11.80 (0.67, 4.81)
       ≥278.55.815.40.94 (0.32, 2.78)
      Food insufficiency
      Defined as sometimes or often not getting enough to eat
       No72.16.920.81.00 (ref)
       Yes74.66.219.21.27 (0.67, 2.41)
      Physical/sexual assault history
       No68.56.424.91.00 (ref)
       Yes81.79.58.82.36 (1.15, 4.83)
      PSYCHIATRIC/ADDICTIVE DISORDERS
      Mental illness history
       No66.86.926.31.00 (ref)
       Yes79.38.112.42.02 (0.88, 4.66)
      Illicit drug use, lifetime
       No32.47.260.41.00 (ref)
       Yes81.27.411.27.21 (3.11, 16.7)
      Problem alcohol use, lifetime
       No56.48.135.21.00 (ref)
       Yes87.96.85.37.42 (2.51, 21.9)
      Note: Percentages are expressed as a function of row totals. Percentages are weighted to be nationally representative of adult HCH clinic users, with sampling weights provided by Research Triangle Institute International. Boldface indicates significance.
      a Denominator is entire study sample (n=966). Current smoking = ever smoked 100 cigarettes and smoked in past 30 days; former smoking = ever smoked 100 cigarettes and has not smoked in past 30 days; never smoking = never smoked at least 100 cigarettes.
      b Denominator consists of current and never smokers (n=877); former smokers are excluded from this analysis. The multivariable model is adjusted for all variables displayed in the table using logistic regression.
      c Defined as placement into a foster home, group home, or other institution before age 18 years
      d Defined as sometimes or often not getting enough to eat
      The prevalence of current smoking was higher in men than women and in whites than nonwhites (Table 2). Within these sociodemographic categories, the prevalence of smoking ranged from 84% among white men (n=223) to 25% among Hispanic women (n=96). Respondents with any past-year employment had a higher prevalence of current smoking than those who did not work. Current smoking prevalence increased with the number of homeless episodes, and was higher among respondents with a history of out-of-home placement in youth or victimization while homeless. The prevalence of smoking was also high among respondents with a mental illness history (79%); illicit drug use history (81%); and problem alcohol use history (88%). Of current smokers, 95% had a lifetime history of either illicit drug use or problem alcohol use, and 76% had abused any substance in the past year.
      In the multivariable analysis of current and never smokers, the factors independently associated with current smoking were age ≥45 years (AOR=6.16, 95% CI=2.30, 16.5); past-year employment (AOR=2.52, 95% CI=1.13, 5.58); out-of-home placement as a minor (AOR=2.79, 95% CI=1.03, 7.52); victimization while homeless (AOR=2.36, 95% CI=1.15, 4.83); illicit drug use history (AOR=7.21, 95% CI=3.11, 16.7); and problem alcohol use history (AOR=7.42, 95% CI=2.51, 21.9) (Table 2). After adjusting for confounders, veterans had lower odds of smoking than nonveterans (AOR=0.16, 95% CI=0.06, 0.45). Although significant in the unadjusted analysis, mental illness history was nonsignificant in the multivariable model (AOR=2.02, 95% CI=0.88, 4.66).

      Smoking Cessation

      Eighty-six of 763 ever smokers were former smokers, yielding a weighted quit ratio of 9% (Table 3). Among 700 past-year smokers, 4% (n=23) had quit. In the multivariable analysis, Hispanic respondents and individuals with hypertension had greater odds of quitting (Table 3). Respondents with two or more homeless episodes had lower odds of quitting (AOR=0.47, 95% CI=0.29, 0.78).
      Table 3Factors associated with smoking cessation among ever smokers (n=763)
      Data source: 2003 Health Care for the Homeless User Survey, Health Resources and Services Administration
      CharacteristicsCeased smoking (unadjusted %)
      Only variables related to the outcome at a level of p<0.05 are shown.
      Adjusted OR (95% CI) for cessation
      Adjusted for variables with univariate significance of p<0.05 as displayed in the table.
      Overall9.2
      SOCIODEMOGRAPHIC CHARACTERISTICS
      Race/ethnicity
       White non-Hispanic7.11.00 (ref)
       Black non-Hispanic8.51.06 (0.40, 2.79)
       Other non-Hispanic17.12.89 (0.65, 12.9)
       Hispanic15.72.45 (1.23, 4.86)
      FEATURES OF HOMELESSNESS
      Homeless episodes lasting ≥30 days
       012.71.00 (ref)
       112.80.83 (0.31, 2.23)
       ≥26.90.47 (0.29, 0.78)
      PSYCHIATRIC/ADDICTIVE DISORDERS
      Illicit drug use history, lifetime
       No18.21.00 (ref)
       Yes8.40.53 (0.23, 1.21)
      MEDICAL CONDITIONS
      Hypertension
       No7.91.00 (ref)
       Yes12.52.00 (1.36, 2.94)
      Note: Percentages are expressed as a function of row totals. Percentages are weighted to be nationally representative of adult Health Care for the Homeless clinic users, with sampling weights provided by Research Triangle Institute International. Boldface indicates significance.
      a Only variables related to the outcome at a level of p<0.05 are shown.
      b Adjusted for variables with univariate significance of p<0.05 as displayed in the table.

      Past-Year Quit Advice

      Overall, 54% of past-year smokers reported receiving advice to quit (see Appendix A, available online at www.ajpm-online.net). Unadjusted advice rates were highest among respondents with obstructive lung disease (63%) and respondents who had been followed by the HCH clinic site for more than 1 year (68%). In the multivariable model, participants with any number of homeless episodes lasting at least 30 days had higher odds of receiving advice, whereas other non-Hispanics and respondents followed by the HCH clinic site for the shortest period of time had lower odds of receiving advice (Appendix A).

      Discussion

      To the authors' knowledge, this is the first study to describe the smoking characteristics of a national sample of homeless adults. The 73% prevalence of current smoking in this study was 3.5 times higher than the 21% prevalence of current smoking in the adult U.S. general population the same year.
      Cigarette smoking among adults—U.S., 2003.
      Conversely, the 9% quit ratio among homeless smokers was much lower than the 50% quit ratio seen in the general population of ever smokers nationally.
      Cigarette smoking among adults—U.S., 2003.
      The smoking prevalence in this 2003 nationwide sample of homeless clinic users was identical to that seen in a 1987 clinic-based study of homeless people in Los Angeles.
      • Gelberg L.
      • Linn L.S.
      • Usatine R.P.
      • Smith M.H.
      Health, homelessness, and poverty A study of clinic users.
      However, during the same period of time, the prevalence of current smoking decreased by 25% in the adult U.S. population.
      Cigarette smoking among adults—U.S., 2003.
      CDC
      Tobacco use by adults—U.S., 1987.
      In identifying the factors associated with smoking in the setting of homelessness, this study provides insight into why these disparities in smoking behavior may exist and how future intervention programs might address this problem.
      Adverse childhood histories, victimization, and substance abuse were key predictors of smoking among homeless people in this study. Such risk factors are known to be more common among homeless people than in non-homeless populations,
      • Herman D.B.
      • Susser E.S.
      • Struening E.L.
      • Link B.L.
      Adverse childhood experiences: are they risk factors for adult homelessness?.
      • Koegel P.
      • Melamid E.
      • Burnam A.
      Childhood risk factors for homelessness among homeless adults.
      • Winkleby M.A.
      • Rockhill B.
      • Jatulis D.
      • Fortmann S.P.
      The medical origins of homelessness.
      • Linn L.S.
      • Gelberg L.
      • Leake B.
      Substance abuse and mental health status of homeless and domiciled low-income users of a medical clinic.
      • Johnson T.P.
      • Fendrich M.
      Homelessness and drug use: evidence from a community sample.
      and this study reinforces the high prevalence of these characteristics among homeless adults. Respondents placed out of home in youth had higher odds of current smoking even after adjusting for potential confounders such as employment status, mental illness, substance abuse, and victimization.
      • Tweddle A.
      Youth leaving care: how do they fare?.
      • Courtney M.E.
      • Heuring D.H.
      The transition to adulthood for youth “aging out” of the foster care system.
      • Herman D.B.
      • Susser E.S.
      • Struening E.L.
      Childhood out-of-home care and current depressive symptoms among homeless adults.
      • Viner R.M.
      • Taylor B.
      Adult health and social outcomes of children who have been in public care: population-based study.
      Adverse childhood experiences have been linked with smoking in non-homeless populations
      • Anda R.F.
      • Croft J.B.
      • Felitti V.J.
      • et al.
      Adverse childhood experiences and smoking during adolescence and adulthood.
      ; the current study extends this research in documenting the relationship between childhood out-of-home placement and smoking among homeless adults. Prior research has also demonstrated a higher prevalence of smoking among people with a history of physical or sexual assault,
      • Acierno R.A.
      • Kilpatrick D.G.
      • Resnick H.S.
      • Saunders B.E.
      • Best C.L.
      Violent assault, posttraumatic stress disorder, and depression Risk factors for cigarette use among adult women.
      • Lemon S.C.
      • Verhoek-Oftedahl W.
      • Donnelly E.F.
      Preventive healthcare use, smoking, and alcohol use among Rhode Island women experiencing intimate partner violence.
      • Weinbaum Z.
      • Stratton T.L.
      • Chavez G.
      • Motylewski-Link C.
      • Barrera N.
      • Courtney J.G.
      Female victims of intimate partner physical domestic violence (IPP-DV), California 1998.
      and a similar association was observed in the homeless population. Post-traumatic stress disorder (PTSD) may mediate this relationship. Individuals with this disorder may use smoking to reduce the burden of negative mood symptoms,
      • Feldner M.T.
      • Babson K.A.
      • Zvolensky M.J.
      • et al.
      Posttraumatic stress symptoms and smoking to reduce negative affect: an investigation of trauma-exposed daily smokers.
      suggesting that identifying and treating PTSD and other psychological sequelae of victimization may need to be part of intervention strategies for homeless smokers.
      Drug and alcohol use conferred high odds of current smoking in this homeless population, confirming a relationship that has also been observed in the general population.
      • Kahler C.W.
      • Strong D.R.
      • Papandonatos G.D.
      • et al.
      Cigarette smoking and the lifetime alcohol involvement continuum.
      • Stark M.J.
      • Campbell B.K.
      Drug use and cigarette smoking in applicants for drug abuse treatment.
      • Giovino G.A.
      • Henningfield J.E.
      • Tomar S.L.
      • Escobedo L.G.
      • Slade J.
      Epidemiology of tobacco use and dependence.
      Almost all (95%) of the homeless smokers in this national sample had a lifetime history of illicit drug use or problem alcohol use. Smoking-cessation programs for homeless populations may need to acknowledge and address these comorbidities. The integration of smoking-cessation services into substance abuse treatment programming may be an effective way to address tobacco use among homeless people with substance abuse problems.
      • Hwang S.W.
      • Tolomiczenko G.
      • Kouyoumdjian F.G.
      • Garner R.E.
      Interventions to improve the health of the homeless: a systematic review.
      • Burling T.A.
      • Burling A.S.
      • Latini D.
      A controlled smoking cessation trial for substance-dependent inpatients.
      • Reid M.S.
      • Fallon B.
      • Sonne S.
      • et al.
      Smoking cessation treatment in community-based substance abuse rehabilitation programs.
      • Irving L.M.
      • Seidner A.L.
      • Burling T.A.
      • Thomas R.G.
      • Brenner G.F.
      Drug and alcohol abuse inpatients' attitudes about smoking cessation.
      Mental illness was not an independently significant predictor of current smoking in the present study, despite having a significant unadjusted relationship with this outcome. Illicit drug use, problem alcohol use, out-of-home placement in youth, and victimization appeared to be important confounders; all of these factors were seen with greater frequency among those with mental illness. After adjusting for these differences, the independent predictive significance of mental illness was lost, although the point estimate remained positive.
      Homeless individuals who had worked in the past year were more likely to be smokers. This differs from the relationship observed in the general population, where the prevalence of smoking among employed people is typically lower than it is among the unemployed.
      • Waldron I.
      • Lye D.
      Employment, unemployment, occupation, and smoking.
      Three explanations may account for this. First, homeless people with past-year employment may have more money for purchasing cigarettes. Second, the blue-collar, service industry, or labor-oriented jobs often obtained by homeless people are associated with a higher prevalence of smoking in comparison to other forms of employment.
      • Lee D.J.
      • LeBlanc W.
      • Fleming L.E.
      • Gomez-Marin O.
      • Pitman T.
      Trends in U.S. smoking rates in occupational groups: the National Health Interview Survey 1987–1994.
      • Lawrence D.
      • Fagan P.
      • Backinger C.L.
      • Gibson J.T.
      • Hartman A.
      Cigarette smoking patterns among young adults aged 18–24 years in the U.S..
      • Bang K.M.
      • Kim J.H.
      Prevalence of cigarette smoking by occupation and industry in the U.S..
      • Gaudette L.A.
      • Richardson A.
      • Huang S.
      Which workers smoke?.
      • Nelson D.E.
      • Emont S.L.
      • Brackbill R.M.
      • Cameron L.L.
      • Peddicord J.
      • Fiore M.C.
      Cigarette smoking prevalence by occupation in the U.S. A comparison between 1978 to 1980 and 1987 to 1990.
      • Covey L.S.
      • Zang E.A.
      • Wynder E.L.
      Cigarette smoking and occupational status: 1977 to 1990.
      Last, the comparator group without past-year employment may be composed of individuals more accurately described as out of the labor force (i.e., not seeking work); such people appear to have a lower prevalence of smoking than those who are in the labor force but unemployed.
      • Lawrence D.
      • Fagan P.
      • Backinger C.L.
      • Gibson J.T.
      • Hartman A.
      Cigarette smoking patterns among young adults aged 18–24 years in the U.S..
      • Bang K.M.
      • Kim J.H.
      Prevalence of cigarette smoking by occupation and industry in the U.S..
      However, this level of detail in employment history could not be determined. Given the novelty of this finding, future studies should attempt to better elucidate the relationship between employment and smoking among homeless people, because distinct policy implications would follow from determining whether this association is driven more by financial factors or by the workplace environments often encountered by homeless workers.
      The percentage of those receiving past-year quit advice was higher than expected. The 54% receiving quit advice compared favorably to the 62% receiving quit advice in the general population of smokers.
      • Cokkinides V.E.
      • Ward E.
      • Jemal A.
      • Thun M.J.
      Under-use of smoking-cessation treatments: results from the National Health Interview Survey, 2000.
      Despite this, few homeless smokers quit. This highlights the need to better understand the way in which advice affects quitting, and the extent to which homeless smokers have access to additional tobacco counseling services and pharmacologic cessation aids. Although there was limited statistical power to explore the correlates of quitting, results indicated that individuals with two or more episodes of homelessness were less likely to be former smokers. This suggests that smoking cessation may be difficult to achieve when attempting to meet basic subsistence needs for shelter and safety.

      Limitations

      This study has certain limitations. The data analyzed were cross-sectional in nature, so causality cannot be definitively determined. All measures were self-reported and may be subject to recall and social desirability biases, particularly with respect to stigmatized behaviors and sensitive issues. The survey was conducted among people with at least one prior visit to a HCH clinic site, so the findings may not be generalizable to the homeless population as a whole, particularly those who do not routinely seek medical care. In the analysis of quit advice, the contribution of provider-level characteristics could not be determined; such information would further enhance understanding of how and when advice is given to homeless smokers. Despite these limitations, the study findings are notable for several reasons. This national study of smoking among homeless people confirms the high prevalence of smoking seen in prior single-city studies, describes the unique vulnerabilities and comorbidities of homeless smokers, and offers the first glimpse of quit advice in this population.

      Conclusion

      As the prevalence of smoking declines in the general population, the residual burden of tobacco use is falling disproportionately on our nation's most vulnerable people. This study provides substantial evidence that the high prevalence of smoking among homeless people is not due primarily to healthcare providers neglecting to address the topic, but more likely due to the considerable comorbidities and barriers to quitting faced by homeless smokers. Findings suggest that interventions for homeless smokers will need to be intensive and tailored to address the unique needs of this population. Expanded addiction therapy will likely be an important component for those with comorbid substance abuse, and additional counseling may be needed for homeless adults who use smoking to cope with the long-lasting effects of adverse childhood experiences and victimization.
      TPB is funded by grant number T32HP10251 from the Health Resources and Services Administration of the USDHHS to support the Harvard Medical School Fellowship in General Medicine and Primary Care. The views expressed in this publication are the opinions of the authors and do not necessarily reflect the official policies of the USDHHS or the Health Resources and Services Administration, nor does mention of the department or agency imply endorsement by the U.S. government.
      The Health Resources and Services Administration sponsored and assisted in designing the 2003 Health Care for the Homeless User Survey, but had no role in the design and conduct of this study or in the collection, management, analysis, and interpretation of the data presented. The Health Resources and Services Administration reviewed a final draft of this manuscript but was not involved in the manuscript preparation or revision process; the authors alone exercised final judgment in determining the form and content of the manuscript.
      NAR is funded by grant K24 HL4440 from the National Heart, Lung, and Blood Institute of the NIH . The NHLBI had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; or preparation, review, or approval of the manuscript.
      The authors thank E. John Orav, PhD, at the Harvard School of Public Health, for his input on the data analysis, and Charles Daly and colleagues at the Health Resources and Services Administration's Bureau of Primary Health Care for their assistance in accessing the 2003 Health Care for the Homeless User Survey data. The authors also thank Jody Greene at Research Triangle Institute, International, for her assistance in accessing the survey data and for providing supporting technical documents regarding the survey design. This manuscript does not represent the views or work of these individuals or their respective institutions.
      A preliminary summary of these findings was presented as a scientific abstract at the 32nd Annual Meeting of the Society of General Internal Medicine.
      NAR has received research grant funding from Pfizer , Sanofi-Aventis , and Nabi Biopharmaceuticals for the study of investigational and/or marketed smoking-cessation products, and has consulted for Pfizer and Free & Clear, Inc., about smoking cessation.
      No other financial disclosures were reported by the authors of this paper.

      Supplementary data

      References

        • Connor S.E.
        • Cook R.L.
        • Herbert M.I.
        • Neal S.M.
        • Williams J.T.
        Smoking cessation in a homeless population: there is a will, but is there a way?.
        J Gen Intern Med. 2002; 17: 369-372
        • Snyder L.D.
        • Eisner M.D.
        Obstructive lung disease among the urban homeless.
        Chest. 2004; 125: 1719-1725
        • Gelberg L.
        • Linn L.S.
        • Usatine R.P.
        • Smith M.H.
        Health, homelessness, and poverty.
        Arch Intern Med. 1990; 150: 2325-2330
        • Lee T.C.
        • Hanlon J.G.
        • Ben-David J.
        • et al.
        Risk factors for cardiovascular disease in homeless adults.
        Circulation. 2005; 111: 2629-2635
        • Szerlip M.I.
        • Szerlip H.M.
        Identification of cardiovascular risk factors in homeless adults.
        Am J Med Sci. 2002; 324: 243-246
        • Gelberg L.
        • Linn L.S.
        Assessing the physical health of homeless adults.
        JAMA. 1989; 262: 1973-1979
        • Weinreb L.
        • Goldberg R.
        • Perloff J.
        Health characteristics and medical service use patterns of sheltered homeless and low-income housed mothers.
        J Gen Intern Med. 1998; 13: 389-397
        • Wright J.D.
        The health of homeless people: evidence from the National Health Care for the Homeless Program.
        in: Brickner P.W. Scharer L.K. Conanan B. Savarese M. Scanlan B.C. Under the safety net: the health and social welfare of the homeless in the U.S. Norton, New York1990: 15-31
        • Burt M.R.
        • Urban Institute
        Homelessness: programs and the people they serve: findings of the National Survey of Homeless Assistance Providers and Clients: technical report.
        U.S. Department of Housing and Urban Development, Office of Policy Development and Research, Washington DC1999
        • Breakey W.R.
        • Fischer P.J.
        • Kramer M.
        • et al.
        Health and mental health problems of homeless men and women in Baltimore.
        JAMA. 1989; 262: 1352-1357
        • Hwang S.W.
        • Orav E.J.
        • O'Connell J.J.
        • Lebow J.M.
        • Brennan T.A.
        Causes of death in homeless adults in Boston.
        Ann Intern Med. 1997; 126: 625-628
        • Morrison D.S.
        Homelessness as an independent risk factor for mortality: results from a retrospective cohort study.
        Int J Epidemiol. 2009; 38: 877-883
        • Hwang S.W.
        • Wilkins R.
        • Tjepkema M.
        • O'Campo P.J.
        • Dunn J.R.
        Mortality among residents of shelters, rooming houses, and hotels in Canada: 11 year follow-up study.
        BMJ. 2009; 339: b4036
        • Scanlan B.C.
        • Brickner P.W.
        Clinical concerns in the care of homeless persons.
        in: Brickner P.W. Scharer L.K. Conanan B. Savarese M. Scanlan B.C. Under the safety net: the health and social welfare of the homeless in the U.S. Norton, New York1990: 69-81
        • Okuyemi K.S.
        • Caldwell A.R.
        • Thomas J.L.
        • et al.
        Homelessness and smoking cessation: insights from focus groups.
        Nicotine Tob Res. 2006; 8: 287-296
        • U.S. Preventive Services Task Force
        Counseling and interventions to prevent tobacco use and tobacco-caused disease in adults and pregnant women: U.S. Preventive Services Task Force reaffirmation recommendation statement.
        Ann Intern Med. 2009; 150: 551-555
        • Fiore M.C.
        • Jaen C.R.
        • Baker T.B.
        Treating tobacco use and dependence: 2008 update.
        USDHHS. Public Health Service, Rockville MD2008
        • Lancaster T.
        • Stead L.
        Physician advice for smoking cessation.
        Cochrane Database Syst Rev. 2004; (CD000165)
        • Arnsten J.H.
        • Reid K.
        • Bierer M.
        • Rigotti N.
        Smoking behavior and interest in quitting among homeless smokers.
        Addict Behav. 2004; 29: 1155-1161
        • Butler J.
        • Okuyemi K.S.
        • Jean S.
        • Nazir N.
        • Ahluwalia J.S.
        • Resnicow K.
        Smoking characteristics of a homeless population.
        Subst Abus. 2002; 23: 223-231
        • Greene J.
        • Fahrney K.
        • Byron M.
        Health Care for the Homeless User/Visit Surveys, RTI Project Number 07147.021.
        RTI International, Research Triangle Park NC2004
      1. Cigarette smoking among adults—U.S., 2003.
        MMWR Morb Mortal Wkly Rep. 2005; 54: 509-513
        • Giovino G.A.
        • Schooley M.W.
        • Zhu B.P.
        • et al.
        Surveillance for selected tobacco-use behaviors—U.S., 1900–1994.
        MMWR CDC Surveill Summ. 1994; 43: 1-43
        • Alaimo K.
        • Briefel R.R.
        • Frongillo Jr, E.A.
        • Olson C.M.
        Food insufficiency exists in the U.S.: results from the third National Health and Nutrition Examination Survey (NHANES III).
        Am J Public Health. 1998; 88: 419-426
        • Alaimo K.
        • Olson C.M.
        • Frongillo Jr, E.A.
        • Briefel R.R.
        Food insufficiency, family income, and health in U.S. preschool and school-aged children.
        Am J Public Health. 2001; 91: 781-786
        • Rose D.
        • Oliveira V.
        Nutrient intakes of individuals from food-insufficient households in the U.S..
        Am J Public Health. 1997; 87: 1956-1961
        • Vozoris N.T.
        • Tarasuk V.S.
        Household food insufficiency is associated with poorer health.
        J Nutr. 2003; 133: 120-126
        • Briefel R.R.
        • Woteki C.E.
        Development of food sufficiency questions for the Third National Health and Nutrition Examination Survey.
        J Nutr Educ. 1992; 24: 24S-28S
        • Cristofar S.P.
        • Basiotis P.P.
        Dietary intake and selected characteristics of women age 19–50 years and their children age 1–5 years by reported perception of food sufficiency.
        J Nutr Educ. 1992; 24: 53-58
        • National Institute on Alcohol Abuse and Alcoholism
        NIAAA Council approves definition of binge drinking.
        USDHHS, Rockville MD2004 (3)
        • Babor T.F.
        • Higgins-Biddle J.C.
        • Saunders J.B.
        • Monteiro M.G.
        The alcohol use disorders test: guidelines for use in primary care.
        second edition. WHO, Geneva2001 (Document No. WHO/MSD/MSB/01.6a)
        • Fagan P.
        • Shavers V.
        • Lawrence D.
        • Gibson J.T.
        • Ponder P.
        Cigarette smoking and quitting behaviors among unemployed adults in the U.S..
        Nicotine Tob Res. 2007; 9: 241-248
        • Peduzzi P.
        • Concato J.
        • Kemper E.
        • Holford T.R.
        • Feinstein A.R.
        A simulation study of the number of events per variable in logistic regression analysis.
        J Clin Epidemiol. 1996; 49: 1373-1379
        • U.S. Census Bureau
        2003 American Community Survey: summary tables.
        • CDC
        Tobacco use by adults—U.S., 1987.
        MMWR Morb Mortal Wkly Rep. 1989; 38: 685-687
        • Herman D.B.
        • Susser E.S.
        • Struening E.L.
        • Link B.L.
        Adverse childhood experiences: are they risk factors for adult homelessness?.
        Am J Public Health. 1997; 87: 249-255
        • Koegel P.
        • Melamid E.
        • Burnam A.
        Childhood risk factors for homelessness among homeless adults.
        Am J Public Health. 1995; 85: 1642-1649
        • Winkleby M.A.
        • Rockhill B.
        • Jatulis D.
        • Fortmann S.P.
        The medical origins of homelessness.
        Am J Public Health. 1992; 82: 1394-1398
        • Linn L.S.
        • Gelberg L.
        • Leake B.
        Substance abuse and mental health status of homeless and domiciled low-income users of a medical clinic.
        Hosp Community Psychiatry. 1990; 41: 306-310
        • Johnson T.P.
        • Fendrich M.
        Homelessness and drug use: evidence from a community sample.
        Am J Prev Med. 2007; 32: S211-S218
        • Tweddle A.
        Youth leaving care: how do they fare?.
        New Dir Youth Dev. 2007; (9–10): 15-31
        • Courtney M.E.
        • Heuring D.H.
        The transition to adulthood for youth “aging out” of the foster care system.
        in: Osgood D.W. Foster E.M. Flanagan C. Ruth G. On your own without a net: the transition to adulthood for vulnerable populations. University of Chicago Press, Chicago2005: 27-67
        • Herman D.B.
        • Susser E.S.
        • Struening E.L.
        Childhood out-of-home care and current depressive symptoms among homeless adults.
        Am J Public Health. 1994; 84: 1849-1851
        • Viner R.M.
        • Taylor B.
        Adult health and social outcomes of children who have been in public care: population-based study.
        Pediatrics. 2005; 115: 894-899
        • Anda R.F.
        • Croft J.B.
        • Felitti V.J.
        • et al.
        Adverse childhood experiences and smoking during adolescence and adulthood.
        JAMA. 1999; 282: 1652-1658
        • Acierno R.A.
        • Kilpatrick D.G.
        • Resnick H.S.
        • Saunders B.E.
        • Best C.L.
        Violent assault, posttraumatic stress disorder, and depression.
        Behav Modif. 1996; 20: 363-384
        • Lemon S.C.
        • Verhoek-Oftedahl W.
        • Donnelly E.F.
        Preventive healthcare use, smoking, and alcohol use among Rhode Island women experiencing intimate partner violence.
        J Womens Health Gend Based Med. 2002; 11: 555-562
        • Weinbaum Z.
        • Stratton T.L.
        • Chavez G.
        • Motylewski-Link C.
        • Barrera N.
        • Courtney J.G.
        Female victims of intimate partner physical domestic violence (IPP-DV), California 1998.
        Am J Prev Med. 2001; 21: 313-319
        • Feldner M.T.
        • Babson K.A.
        • Zvolensky M.J.
        • et al.
        Posttraumatic stress symptoms and smoking to reduce negative affect: an investigation of trauma-exposed daily smokers.
        Addict Behav. 2007; 32: 214-227
        • Kahler C.W.
        • Strong D.R.
        • Papandonatos G.D.
        • et al.
        Cigarette smoking and the lifetime alcohol involvement continuum.
        Drug Alcohol Depend. 2008; 93: 111-120
        • Stark M.J.
        • Campbell B.K.
        Drug use and cigarette smoking in applicants for drug abuse treatment.
        J Subst Abuse. 1993; 5: 175-181
        • Giovino G.A.
        • Henningfield J.E.
        • Tomar S.L.
        • Escobedo L.G.
        • Slade J.
        Epidemiology of tobacco use and dependence.
        Epidemiol Rev. 1995; 17: 48-65
        • Hwang S.W.
        • Tolomiczenko G.
        • Kouyoumdjian F.G.
        • Garner R.E.
        Interventions to improve the health of the homeless: a systematic review.
        Am J Prev Med. 2005; 29: 311-319
        • Burling T.A.
        • Burling A.S.
        • Latini D.
        A controlled smoking cessation trial for substance-dependent inpatients.
        J Consult Clin Psychol. 2001; 69: 295-304
        • Reid M.S.
        • Fallon B.
        • Sonne S.
        • et al.
        Smoking cessation treatment in community-based substance abuse rehabilitation programs.
        J Subst Abuse Treat. 2008; 35: 68-77
        • Irving L.M.
        • Seidner A.L.
        • Burling T.A.
        • Thomas R.G.
        • Brenner G.F.
        Drug and alcohol abuse inpatients' attitudes about smoking cessation.
        J Subst Abuse. 1994; 6: 267-278
        • Waldron I.
        • Lye D.
        Employment, unemployment, occupation, and smoking.
        Am J Prev Med. 1989; 5: 142-149
        • Lee D.J.
        • LeBlanc W.
        • Fleming L.E.
        • Gomez-Marin O.
        • Pitman T.
        Trends in U.S. smoking rates in occupational groups: the National Health Interview Survey 1987–1994.
        Occup Environ Med. 2004; 46: 538-548
        • Lawrence D.
        • Fagan P.
        • Backinger C.L.
        • Gibson J.T.
        • Hartman A.
        Cigarette smoking patterns among young adults aged 18–24 years in the U.S..
        Nicotine Tob Res. 2007; 9: 687-697
        • Bang K.M.
        • Kim J.H.
        Prevalence of cigarette smoking by occupation and industry in the U.S..
        Am J Ind Med. 2001; 40: 233-239
        • Gaudette L.A.
        • Richardson A.
        • Huang S.
        Which workers smoke?.
        Health Rep. 1998; 10 (ENG); 35–47 (FRE): 35-45
        • Nelson D.E.
        • Emont S.L.
        • Brackbill R.M.
        • Cameron L.L.
        • Peddicord J.
        • Fiore M.C.
        Cigarette smoking prevalence by occupation in the U.S..
        Occup Med. 1994; 36: 516-525
        • Covey L.S.
        • Zang E.A.
        • Wynder E.L.
        Cigarette smoking and occupational status: 1977 to 1990.
        Am J Public Health. 1992; 82: 1230-1234
        • Cokkinides V.E.
        • Ward E.
        • Jemal A.
        • Thun M.J.
        Under-use of smoking-cessation treatments: results from the National Health Interview Survey, 2000.
        Am J Prev Med. 2005; 28: 119-122