Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Am J Ind Med. Author manuscript; available in PMC 2007 December 1.
Published in final edited form as:
PMCID: PMC1945044

Smoking Behavior in Trucking Industry Workers

Nitin B Jain, MD, MSPH,1,2 Jaime E Hart, MS,1,3 Thomas J Smith, PhD,3 Eric Garshick, MD, MOH,1,4 and Francine Laden, ScD1,3,5,*



In retrospective occupational studies, the degree of confounding by smoking depends on variation in smoking among job-related exposure groups. We assessed the relationship between job title and smoking behavior as part of a study on occupational exposures and lung cancer.


A questionnaire on smoking was mailed to a sample of 11,986 trucking industry. Company records were used to gather other relevant information.


The response rate was 40.5%. Among white males, the age-adjusted prevalence of ever smoking was highest among longhaul truck drivers (67%) and lowest among clerks (44%). Smoking rates among workers with other job titles were similar.


Our results will be used to adjust for the differences in smoking among job-related exposure groups when assessing the association between particulate matter exposure and lung cancer mortality. Our study also suggests that an assessment of methods to control for smoking should be considered in the design of retrospective occupational health studies.

Keywords: smoking, occupational health, industry


In retrospective occupational cohort studies, the degree of confounding by smoking depends on variation in smoking behavior among exposure groups [Axelson, 1980, Blair et al., 1988, Steenland et al., 1984]. Usually, this information is not available in studies relying on work records. Since smoking behavior is expected to be similar among workers in a single occupational cohort, researchers commonly use an internal comparison group and assume that there is little variation in smoking behavior among exposure groups. There are few studies specifically assessing differences in smoking habits among workers within a single occupational cohort to test this assumption.

We have been collecting exposure and work history information in a large retrospective cohort study of unionized U.S. trucking industry workers to determine the association of exposure to diesel exhaust and other mobile source related particulate matter (PM) with lung cancer. Job title is closely linked to trucking industry exposures since job duties are well defined and have remained similar over time. Because cigarette smoking is one of the strongest known risk factors for lung cancer, we performed a survey on smoking habits in a sample of currently working and recently retired workers. The objective of our study was to assess determinants of smoking behavior in these workers.

Materials and Methods


The base population consisted of 57,852 unionized trucking industry employees working at three U.S. companies in 2002 or retired from these companies between 1997 and 2002. The three companies were members of the Motor Freight Carriers Association, and employees are members of the International Brotherhood of Teamsters. A questionnaire designed to obtain smoking history was first mailed in the summer of 2003 to a stratified sample of 11,986 workers, including all clerks and 9,730 workers randomly selected to represent the distribution of the remaining job titles. The study protocol was approved by the Brigham and Women’s Hospital and VA Boston Institutional Review Boards.

Smoking Questionnaire

The questionnaire used for our study was modeled after the American Thoracic Society (ATS) questionnaire [Ferris, 1978]. It contains questions on history of current and past cigarette smoking, age of first cigarette use, average number of cigarettes smoked per day, and age stopped smoking. Questions about occupational history prior to working at the current (or for retirees, last) company, year of joining the trucking industry, and educational status were also included. People not responding after two mailings were subsequently mailed a post-card with only three questions on ever and current smoking, and age of smoking cessation if a former smoker.

Company Records

Information on employee job title, region of residence (based on mailing address), date of birth, sex, race, and location and size of the most recent truck terminal was extracted from company records and merged with the data from the questionnaire and postcard. Age was calculated as of December 31, 2003.

Definition of Variables

Smoking Characteristics

Current smokers reported smoking within one month of answering the questionnaire. Never smokers were defined as those who smoked less than 20 packs of cigarettes in a lifetime or less than 1 cigarette a day for one year. Cumulative lifetime smoking (pack-years) was calculated.

Job Titles

Job categories and duties are similar across the unionized trucking industry, with only minor differences in job titles between companies. Long-haul drivers operate heavy-duty tractor-trailer trucks between cities. Pick-up and Delivery (P&D) truck drivers operate tractors and smaller trucks within cities or rural areas and deliver cargo between terminal docks and consumers. Dock workers load/unload cargo and operate forklifts. Combination workers perform duties of both P&D drivers and dock-workers and are more frequently employed at smaller terminals. Mechanics repair, maintain, and fuel tractors. Hostlers drive a small, specialized tractor unit to move trailers within the terminal yard. Clerks include cashiers, dispatchers, customer service representatives, and other workers in the terminal office.

Statistical Analysis

We used descriptive statistics (frequencies, proportions, and means) to examine response rates and smoking habits by job titles, race, sex, and other characteristics. Smoking characteristics of white men were described by direct standardization to the age distribution of the analysis cohort. We used logistic regression to determine the adjusted and unadjusted association between smoking behavior and characteristics of trucking industry workers. A linear regression model was used to assess the association of various characteristics with pack-years. Since we did not have information on education level of those responding only to the postcards, an indicator variable was used for missing values of education in the regression models. Intercooled STATA for UNIX (version 9.0), Stata Corporation, (College Station, TX) was used for all analyses.


The mailing sample included 3,000 longhaul drivers, 1,104 P&D drivers, 2,638 combination workers, 400 hostlers, 2,258 dock workers, 299 mechanics, 2,256 clerks, 21 janitors, and 10 managers. The overall response rate among workers in the remaining sample was 40.5%, omitting the 632 questionnaires returned either due to an incorrect mailing address or because the employee was deceased. The distribution of job titles, gender, region of residence, and terminal size and location among responders and non-responders was similar (Table I). However, the response rate among Whites (44%) was higher than among Blacks (25%) and Hispanics (28%). Also, responders (mean age=53.0 years) were older than the non-responders (mean age=49.9 years).

Table I
Characteristics of Responders and Non-Responders to a Smoking Survey in Trucking Industry Workers

Due to small numbers of females and non-white employees, we restricted this analysis to white males. We further excluded 36 responders with missing information on smoking and 3 janitors and 1 manager. Therefore, there were a total of 3,362 individuals available for analysis.

Characteristics by job title are presented in Table II. Longhaul drivers and clerks were older than other workers. Combination workers and P&D drivers worked in smaller terminals. Education status was similar across groups.

Table II
Characteristics of White Men in the Trucking Industry by Job Categories

Age-standardized smoking rates and pack-years smoked were determined by job titles, education, region of residence, terminal size, and terminal location (Table III). Longhaul drivers had the highest prevalence of smoking (18% current smokers and 49% ex-smokers), followed by hostlers (16% current smokers and 49% ex-smokers) and P&D drivers (8% current smokers and 55% ex-smokers). There was only minor variation in never smoking rates between non-clerk job titles. Similarly, although smoking rates were higher among workers in the Midwest, the variation by region of residence was relatively small. Smoking rates were also higher in workers with less than high school education, and varied little by terminal location and size.

Table III
Age-Standardized* Smoking Characteristics of White Men in the Trucking Industry

After adjusting for age, education, region of residence, terminal size, and terminal location, the long-haul drivers were more likely to smoke than the workers in other job categories (Table IV). However, these differences were small, with the exception of comparison to the clerks. The likelihood of ever smoking increased statistically significantly with increasing age. Workers in the South and West were significantly less likely to be ever smokers as compared with those in the Midwest. Among ever smokers, P&D drivers were significantly more likely to have quit smoking as compared with long-haul drivers, but there were only minor differences among other job titles (Table V). The likelihood of quitting smoking also increased with increasing age. When pack-years was used as the outcome in linear regression models, employment as a longhaul truck driver, increasing age, and terminal location in urban areas were significantly associated with greater lifetime smoking (pack-years) (data not shown).

Table IV
Unadjusted and Adjusted Likelihood of Smoking among 3,362 White Men in the Trucking Industry
Table V
Unadjusted and Adjusted Likelihood of Quitting Smoking (among Ever Smokers) for 2,083 White Men in the Trucking Industry

Similar results were obtained when regression analyses were conducted after excluding people who responded to the shorter personal history questionnaire on postcards (data not shown). Results were also similar if educational status (which had missing values) was dropped from the regression models.


We examined smoking behavior of unionized trucking industry workers, primarily a blue-collar occupational group, based on job titles, age, education, region of residence, terminal size, and terminal location. Among white male workers, a greater likelihood of ever smoking was associated with employment as a longhaul truck driver, increasing age, residence in the Midwest, and educational attainment below high school. Clerks had the lowest prevalence of ever smoking; the other job titles were similar. Hence, there was minor variation in smoking behavior within trucking industry workers after adjusting for potential confounders.

Information on smoking is valuable to accurately associate the etiology of certain diseases with occupational exposures [Axelson, 1980, Blair, et al., 1988, Steenland, et al., 1984]. However, occupational cohort studies based on work or company records often lack information on smoking. The degree of confounding that can be attributed to smoking in such studies has been a matter of debate, and may be related to the variation in smoking behavior within a cohort. Some studies have reported that smoking only minimally confounds the risk estimates for the association between disease and occupational/environmental exposure [Siemiatycki et al., 1988]. Others have recommended a quantitative estimation of the impact of smoking on risk estimates [Axelson and Steenland, 1988]. Several direct and indirect methods to control for smoking in occupational health studies have also been discussed [Steenland, et al., 1984].

Smoking rates among adults have declined over the past decades in the United States (Table VI). Although rates are consistently higher than those in the general U.S. population, smoking rates have also steadily declined among blue-collar workers over the past decades. In our study, 15% of white male unionized trucking industry workers reported to be currently smoking, whereas 62% and 38% were ever and never smokers, respectively. Although recent estimates for unionized trucking industry workers are not available, the proportion of current smokers in our study was lower than historical rates reported in other blue-collar populations. This is unlikely to be due to the restriction to white males, since ever-smoking rates were lower in the females and non-whites who responded to the survey. However, these proportions may be attributed to a response bias by smoking status in our cohort. Our response rate was only 40.5%. Although this is low, it is not unexpected in an occupational cohort [Sorensen and Barbeau, 2004, Fortmann et al., 1984, Petitti et al., 1981]. The response was consistent across job title, but current smokers may have been less likely to respond than former and never smokers. The increased likelihood of non-responders being smokers [Winkleby et al., 1995], as well as the underreporting of smoking [Pechacek et al., 1984] in surveys has been previously described. However, many other studies have reported valid responses from smoking surveys [Fortmann, et al., 1984, Petitti, et al., 1981]. Although our study may underestimate current smoking rates in the trucking industry, a recent decline in smoking may be expected since the rate of strict smoking policies and smoking bans in workplaces has increased over the last decade [2000, Shopland et al., 2001, Sweeney et al., 2000]. Smoke-free workplaces are shown to encourage employees to quit smoking [Farkas et al., 1999, Fichtenberg and Glantz, 2002, Glasgow et al., 1997].

Table VI
Current and Former Smoking Rates from Selected Population and Occupational Surveys

The likelihood of ever smoking as well as quitting smoking increased with increasing age in our cohort. This is likely due to a birth cohort effect, where older workers started smoking when smoking rates in the U.S. were higher and quit later in life. A lower educational attainment has also been associated with higher smoking rates in the general population [2004, 2005], as was seen in our cohort. In addition, smoking rates varied significantly by geographic location of the trucking industry workers. Workers in the Midwest had higher smoking rates than those in other regions of the country. This is consistent with other reports in the general population and in blue-collar workers [2004, Shopland et al., 1996].

In a national study, data from 1992–93 showed that male blue-collar workers in the Midwest and South had higher rates of current smoking (38.8% and 40%, respectively) than those in the Northeast and West (34.5% and 32.4%) [Shopland, et al., 1996].

In our retrospective lung cancer mortality study we identified approximately 55,000 unionized trucking industry workers employed in 1985, and are assessing mortality through 2000. Job titles, which indirectly determine amount of PM exposure will be used to assign exposure groups. The results from the current study will be used to estimate and adjust for the confounding caused by smoking in the exposed and unexposed group. This indirect method of adjustment has been described previously to account for the interaction between smoking and occupational exposure in various other cohorts [Axelson and Steenland, 1988, Larkin et al., 2000, Siemiatycki et al., 1988]. Due to the small variation in smoking rates across job title, we expect that we will not likely see large effects of confounding by smoking in this population.

In summary, we assessed smoking behavior by various characteristics of trucking industry workers, primarily a blue-collar occupational group. We found that employment as a longhaul truck driver, increasing age, living in the Midwest, and an educational attainment below high school, were associated with a higher likelihood of ever smoking. Clerks had the lowest likelihood of ever smoking. These results will help in indirect adjustment for the effect of smoking on the relation between diesel exhaust and lung cancer. Our study also suggests that a careful assessment of the need and methods to control for smoking should be considered in the design of occupational health studies, even if for reassurance that confounding is minimal.


We would like to acknowledge Tim Lynch and Bill Rogers of the Motor Freight Carriers Association, LaMont Byrd and Scott Madar of the International Brotherhood of Teamsters Health and Safety Office, and all of the responders to our survey for their support of this study.

Reference List

  • Cigarette smoking among adults--United States, 2001. MMWR Morb Mortal Wkly Rep. 2003;52:953–956. [PubMed]
  • State-specific prevalence of current cigarette smoking among adults--United States, 2002. MMWR Morb Mortal Wkly Rep. 2004b;52:1277–1280.
  • Cigarette smoking among adults--United States, 2003. MMWR Morb Mortal Wkly Rep. 2005;54:509–513. [PubMed]
  • State-specific prevalence of current cigarette smoking among adults and the proportion of adults who work in a smoke-free environment--United States, 1999. MMWR Morb Mortal Wkly Rep. 2000;49:978–982. [PubMed]
  • Cigarette smoking among adults--United States, 2002. MMWR Morb Mortal Wkly Rep. 2004a;53:427–431.
  • Axelson O. Aspects of confounding and effect modification in the assessment of occupational cancer risk. J Toxicol Environ Health. 1980;6:1127–1131. [PubMed]
  • Axelson O, Steenland K. Indirect methods of assessing the effects of tobacco use in occupational studies. Am J Ind Med. 1988;13:105–118. [PubMed]
  • Bang KM, Kim JH. Prevalence of cigarette smoking by occupation and industry in the United States. Am J Ind Med. 2001;40:233–239. [PubMed]
  • Blair A, Steenland K, Shy C, O'Berg M, Halperin W, Thomas T. Control of smoking in occupational epidemiologic studies: methods and needs. Am J Ind Med. 1988;13:3–4. [PubMed]
  • Centers for Disease Control and Prevention. Percentage of adults who were current, former, or never smokers, overall and by sex, race, Hispanic origin, age, and education, National Health Interview Surveys, selected years-United States, 1965–2000. 3-22-2005. Ref Type: Electronic Citation.
  • Farkas AJ, Gilpin EA, Distefan JM, Pierce JP. The effects of household and workplace smoking restrictions on quitting behaviours. Tob Control. 1999;8:261–265. [PMC free article] [PubMed]
  • Ferris BG. Epidemiology Standardization Project (American Thoracic Society) Am Rev Respir Dis. 1978;118:1–120. [PubMed]
  • Fichtenberg CM, Glantz SA. Effect of smoke-free workplaces on smoking behaviour: systematic review. BMJ. 2002;325:188. [PMC free article] [PubMed]
  • Fortmann SP, Rogers T, Vranizan K, Haskell WL, Solomon DS, Farquhar JW. Indirect measures of cigarette use: expired-air carbon monoxide versus plasma thiocyanate. Prev Med. 1984;13:127–135. [PubMed]
  • Glasgow RE, Cummings KM, Hyland A. Relationship of worksite smoking policy to changes in employee tobacco use: findings from COMMIT. Community Intervention Trial for Smoking Cessation. Tob Control. 1997;6(Suppl 2):S44–S48. [PMC free article] [PubMed]
  • Larkin EK, Smith TJ, Stayner L, Rosner B, Speizer FE, Garshick E. Diesel exhaust exposure and lung cancer: adjustment for the effect of smoking in a retrospective cohort study. Am J Ind Med. 2000;38:399–409. [PubMed]
  • Lee DJ, LeBlanc W, Fleming LE, Gomez-Marin O, Pitman T. Trends in US smoking rates in occupational groups: the National Health Interview Survey 1987–1994. J Occup Environ Med. 2004;46:538–548. [PubMed]
  • National Institute for Occupational Safety and Health. 2003. Work related lung disease surveillance report, 2002. NIOSH Publication No 2003-111241–246.
  • Nelson DE, Emont SL, Brackbill RM, Cameron LL, Peddicord J, Fiore MC. Cigarette smoking prevalence by occupation in the United States. A comparison between 1978 to 1980 and 1987 to 1990. J Occup Med. 1994;36:516–525. [PubMed]
  • Pechacek TF, Fox BH, Murray DM, Luepker RV. Review of techniques for measurement of smoking behavior. In: Matarazzo JD, Weiss SM, Herd JA, Miller NE, editors. A handbook of health enhancement and disease prevention. New York: John Wiley; 1984.
  • Petitti DB, Friedman GD, Kahn W. Accuracy of information on smoking habits provided on self-administered research questionnaires. Am J Public Health. 1981;71:308–311. [PubMed]
  • Shopland DR, Gerlach KK, Burns DM, Hartman AM, Gibson JT. State-specific trends in smoke-free workplace policy coverage: the current population survey tobacco use supplement, 1993 to 1999. J Occup Environ Med. 2001;43:680–686. [PubMed]
  • Shopland DR, Hartman AM, Gibson JT, Mueller MD, Kessler LG, Lynn WR. Cigarette smoking among U.S. adults by state and region: estimates from the current population survey. J Natl Cancer Inst. 1996;88:1748–1758. [PubMed]
  • Siemiatycki J, Wacholder S, Dewar R, Cardis E, Greenwood C, Richardson L. Degree of confounding bias related to smoking, ethnic group, and socioeconomic status in estimates of the associations between occupation and cancer. J Occup Med. 1988a;30:617–625. [PubMed]
  • Siemiatycki J, Wacholder S, Dewar R, Wald L, Begin D, Richardson L, Rosenman K, Gerin M. Smoking and degree of occupational exposure: are internal analyses in cohort studies likely to be confounded by smoking status? Am J Ind Med. 1988b;13:59–69. [PubMed]
  • Sorensen G, Barbeau E. Steps to a Healthier US Workforce: Integrating Occupational Health and Safety and Worksite Health Promotion: State of the Science. Boston, MA: NIOSH; 2004.
  • Steenland K, Beaumont J, Halperin W. Methods of control for smoking in occupational cohort mortality studies. Scand J Work Environ Health. 1984;10:143–149. [PubMed]
  • Stellman SD, Boffetta P, Garfinkel L. Smoking habits of 800,000 American men and women in relation to their occupations. Am J Ind Med. 1988;13:43–58. [PubMed]
  • Sweeney CT, Shopland DR, Hartman AM, Gibson JT, Anderson CM, Gower KB, Burns DM. Sex differences in workplace smoking policies: results from the current population survey. J Am Med Womens Assoc. 2000;55:311–315. [PubMed]
  • Weinkam JJ, Sterling TD. Changes in smoking characteristics by type of employment from 1970 to 1979/80. Am J Ind Med. 1987;11:539–561. [PubMed]
  • Winkleby MA, Schooler C, Kraemer HC, Lin J, Fortmann SP. Hispanic versus white smoking patterns by sex and level of education. Am J Epidemiol. 1995;142:410–418. [PubMed]