Examining the link between working conditions and tobacco-smoking habits

Previous studies have been inconsistent in their findings about the link between working conditions and smoking tobacco. Some of the inconsistency may be due to the varying methods used to measure work environment exposures. It may also be due to other limitations of these studies, including small samples, low response rates, short follow-up time frames, or few follow-up time points.

This study aimed to help fill this gap by drawing on a large sample of Canadians who were surveyed on multiple occasions over a 16-year period.

The research team drew on the National Population Health Survey (NPHS), which Statistics Canada administered every two years from 1994 to 2010. The researchers looked at the survey responses of 5,461 people who worked at least 15 hours a week in 1994 and who provided answers about their smoking habits in at least two of the nine NPSH surveys administered from 1994 to 2010.

Working conditions were measured using responses to the Job Content Questionnaire, which was included in many of the NPHS surveys. It measured dimensions of skill discretion (e.g. variety in job tasks, opportunities for learning, skill requirements), decision authority (e.g. ability to make decisions about own job), psychological demands (e.g. conflicting requests, hectic pace), job insecurity, physical exertion and workplace social support.

Based on respondents’ answers about their daily cigarette use, the research team classified workers into four smoking groups:

• non-smokers (68 per cent of respondents);
• former smokers—those who smoked at least 14 cigarettes a day in 1994 but none in 2008-2010 (10 per cent);
• lighter smokers—those who smoked six to eight cigarettes throughout the study period (nine per cent); and
• heavy smokers—those who smoked 22 cigarettes or more a day in 1994 and who still smoked at least 14 cigarettes a day in 2010 (13 per cent).

Physically demanding jobs were linked with heavy smoking. People whose work involved high physical exertion were roughly 2.5 times more likely to be heavy smokers than people whose jobs required low physical exertion. People who initially worked in physically demanding jobs that became less physically demanding over time were 90 per cent more likely to be heavy smokers, and people whose work became more physically demanding over time were 50 per cent more likely to be heavy smokers—both compared to people whose jobs required low physical exertion throughout the period.

Those whose work involved low skill discretion were 67 per cent more likely than those whose work involved high skill discretion to be heavy smokers. Workers with high psychological job demands were 37 per cent more likely than those with low psychological job demands to be heavy smokers. And workers in environments with lower social support were 41 per cent more likely than those with high social support to be heavy smokers.

Work characteristics linked with heavy smoking include low social support, low skill discretion and high psychological demands. A number of potential reasons could explain this. Smoking may be a coping behaviour to deal with a poor working environment. As well, workers with heavy demands or less social support at work may be less receptive to anti-smoking and health promotion programs. Conversely, having greater social support and decision-making authority at work may influence an individual’s self-efficacy, providing the motivation to stop smoking.

Physically demanding work—a marker of blue-collar manual work—is strongly linked with smoking. Other studies have suggested that blue-collar workers are similarly motivated to quit smoking as white-collar workers; this study suggests that it may be worthwhile to explore specific characteristics related to physically demanding work that are linked to smoking behaviours in these environments.

A strength of the study was its use of a method called “trajectory modelling” that allowed the research team to draw on 85 per cent of an already large sample. This increased the generalizability of study results. Another strength of the study was its long follow-up period of 16 years, which provided a rich opportunity to explore changes in smoking behaviours.  

A limitation of the study was its reliance on self-reported data, which introduces the potential for bias. Also, as changes in smoking and work environment were explored simultaneously, conclusions about cause and effect cannot be drawn from these findings.