Health Canada
Symbol of the Government of Canada
Skip to content | Skip to institutional links

Common menu bar links

Home > Health Concerns > Reports & Publications > Tobacco - Reports and Publications

Institutional links

Health Concerns

Workplace Smoking: Trends, Issues and Strategies

ISBN: 0-662-24401-X
Cat. No.: H39-370/1-1996E

Table of Contents

Abstract

This paper reviews the current state of knowledge about the exposure of non-smokers to environmental tobacco smoke (ETS) in the workplace. The paper's main objective is to provide information and analysis for planning actions to further reduce ETS exposure in the workplace. Arguing that more concerted action in the workplace is essential, the paper discusses the factors that will both inhibit and facilitate achieving this goal.

The goals of the National Strategy to Reduce Tobacco Use (NSTRTU) are 1) to cut overall smoking prevalence from 29% in 1993 to 24% by 2000 and 2) to reduce exposure to ETS in the workplace from 45% of workers in 1992 to 23% by 2000. While Canada has made considerable progress in reducing workplace ETS exposure, the NSTRTU goals have been achieved in only a few industries and occupations. Blue-collar occupations and huge areas of the service sector have smoking rates well above average. Small firms, which employ the majority of Canadian workers, remain a "black box" with regard to smoking regulations. The hospitality industry is slowly becoming the focus of regulatory action. The high smoking rates in this sector, coupled with its large proportion of teenage workers, demand more intensive intervention. Moreover, total bans on ETS are not yet the standard for regulation. Generally, workers comply with smoking legislation or policies. However, achieving greater reductions in ETS exposure will require enforcement strategies aimed at non-conforming groups.

An integrated ETS reduction strategy must involve government, employers, unions and public health organizations. Uniform minimum standards are also essential and should be established at the provincial level. However, most progress in achieving smoke-free workplaces is now occurring at the municipal level. The use of enhanced occupational health and safety legislation may also be a useful strategy to reduce ETS in high-exposure worksites. The paper concludes by recommending specific research and policy initiatives.

Introduction

In less than 10 years, Canada has achieved substantial reductions in tobacco consumption, partly due to a succession of increasingly comprehensive federal, provincial and municipal restrictions on workplace smoking. The early 1990s mark a watershed in the public health campaign against environmental tobacco smoke (ETS). Bolstered by the 1992 U.S. Environmental Protection Agency's (EPA) classification of ETS as a Class A carcinogen and growing public support for clean indoor air, total smoking bans in workplaces have become more common in both Canada and the United States.

Despite these advances, specific industries and occupations have smoking rates and involuntary ETS exposure well above the average. The growing hospitality industry, for example, has been identified as one area that requires immediate regulatory action. Small firms comprise the majority of Canadian workplaces, yet smoking restrictions are less common in them than in the corporate sector or government. Moreover, achieving full compliance with existing regulations is an elusive goal, mainly because of weak monitoring and enforcement mechanisms.

The goals for the National Strategy to Reduce Tobacco Use (NSTRTU) are 1) to cut overall smoking prevalence from 29% in 1993 to 24% by 2000 and 2) to reduce exposure to ETS in the workplace from 45% of workers in 1992 to 23% by 2000 (Health Canada, 1993, 1995a). Approximately 80% of a non-smoker's exposure to ETS occurs in the workplace (Repace and Lowrey, 1985). Coupled with this, many workplaces are more enclosed and less adequately ventilated than homes. Indeed, non-smoking employees consistently exposed to ETS at work may face greater risks than non-smoking spouses exposed to ETS in the home, although the epidemiological research on the health effects of ETS has until recently focused on the latter (Hammond, Sorensen, Youngstrom and Ockene, 1995).

This paper reviews the current state of knowledge about the exposure of non-smokers to ETS in the workplace. Evidence presented in the paper is drawn from an extensive survey of relevant medical and social science literature, supplemented with unpublished materials and data from recent national surveys on smoking behaviour. The paper's main objective is to provide information and analysis for planning actions to further reduce ETS exposure in Canadian workplaces. Arguing that concerted action in the workplace is essential, the paper discusses the factors that will both inhibit and facilitate achieving this goal. Specifically, the paper documents current trends in workplace smoking in Canada, workplace smoking restrictions, the impact of and public support for such restrictions, and the economics of smoking restrictions in the workplace. The paper concludes by identifying specific recommendations for policy action and future research that flow from the analysis.

Smoking on the Job: Trends and Issues

Smoking Prevalence in the Labour Force

Variations by Region

Of the 12.6 million Canadians who were in the labour force (i.e., employed or unemployed and looking for work) in August 1994, 4.13 million, or 32.8%, are current (daily and non-daily combined) smokers. This is slightly higher than the national average of 30% that was documented in the August 1994 Survey on Smoking in Canada. Figure 1 uses data from this survey to examine current smoking prevalence by region and gender among the employed and unemployed. Ontario is considerably below the national average in labour force smoking prevalence (27.3% vs. 32.8%); British Columbia is slightly below the national average (31.0%). Quebec is the only region in which labour force smoking prevalence (41.7%) significantly exceeds the national average.

Figure 1: Current Smokers in the Labour Force, by Region and Sex, August 1994

Figure 1. Current Smokers in the Labour Force, by Region and Sex, August 1994

Source: Survey on Smoking in Canada, Cycle 2 (includes employed and unemployed but looking for work)

There are pronounced gender differences within regions. In both Ontario and the Prairie region, less than 25% of the female labour force smokes. In marked contrast, smoking prevalence in Ontario's male labour force is about 4 percentage points under the national average, while the opposite is true in the Prairie region. However, the most remarkable group is Quebec's female labour force, 46.2% of whom are smokers. This represents the largest gender gap (7.5 percentage points), yet it is a reversal of the national pattern, in which men are more likely to be smokers than women. Among all socio-demographic groups, current smoking prevalence is highest among the unemployed who are looking for work, at 45.6% overall and 51.7% for females (Survey on Smoking in Canada Cycle 2 data, special tabulations). Figure 1 shows that the entire female labour force in Quebec is approaching this level of prevalence.

Occupation and Industry Variations

Figures 2 and 3 use 1994-95 National Population Health Survey data to document the occupational and industrial distribution of current smokers. Turning first to Figure 2, we note that four occupational categories -- natural science, teaching, medicine and health, and social science and religion -- have a smoking rate below the NSTRTU overall prevalence target of 24% for the year 2000. Managerial and administrative occupations, with a smoking prevalence of 27.4%, are below the present national average.

Figure 2: Current Smokers by Occupation, Canada, 1994-95

Figure 2. Current Smokers by Occupation, Canada, 1994-95

Source: 1994-95 National Population Health Survey

In contrast, among occupations with the highest rates of smoking, four blue-collar, predominantly male occupations stand out. Over half (55.4%) of the workers in fishing, forestry and mining are current smokers --almost double the national average. Approximately 45% of workers in transportation and construction jobs are smokers. Two other blue-collar, predominantly male occupations (machining, processing) have smoking rates over 40%.

However, service occupations are close behind, with a smoking rate of 37.2%. There are several features of this occupational category that distinguish it from other occupations with a high prevalence of smoking. First, it comprises a diversity of white-collar jobs in areas such as food and beverage services, hairdressing, child care, tourism, hotels and motels, and protective services (policing, fire, security). Second, over half of the workers in this service category are female, and they comprise about 17% of the entire female labour force (Krahn and Lowe, 1993:74, 163). Services constitute the second largest occupation, next to clerical occupations, accounting for over 13% of the total labour force. Third, service workers tend to be younger than those in blue-collar occupations. In 1991, 29% of employed 15-19 year olds held service jobs, making up 14% of all service workers (Lindsay, Devereaux and Bergob, 1994:25). And fourth, there is a larger number of smokers in services (848,871) than in construction, transportation, and fishing/forestry/mining combined (743,794). In fact, service occupations have more smokers than any of the other occupations in Figure 2. It is interesting to note that clerical occupations rank second in the number of smokers.

Top of Page

Figure 3 analyzes smoking prevalence by industry. To interpret Figures 2 and 3, it is useful to keep in mind that occupation refers to the kind of work performed, whereas industry refers to what is being produced. However, the classification schemes used by Statistics Canada for occupation and industry overlap in some areas (e.g., agriculture, transportation, construction). These conceptual distinctions should assist in identifying specific industrial sectors or occupational groups as targets for ETS policy development.

Figure 3: Current Smokers by Industry, 1994-95

Figure 3. Current Smokers by Industry, 1994-95

Source: 1994-95 National Population Health Survey

The lowest prevalence of smoking (24.5%) is found in community services (education, health, social welfare). This is the only industrial sector to have come close to meeting the NSTRTU overall smoking prevalence target of 24% for the year 2000. With the exception of agriculture, the other industrial sectors that do not exceed the national rate of smoking are both services: public administration and finance. Construction and resources (e.g., mining, energy, forestry) have the highest smoking prevalence rates, both over 45%. In these sectors, much of the work activity occurs outdoors, which makes smoking regulations difficult to implement and enforce. Categories with the next highest rates are transportation and personal services.

Again, it is important to focus attention on the absolute number of smokers in personal services (609,478), which is greater than in the primary (i.e., fishing and trapping; logging and forestry; mining, quarrying, and oil wells) and construction industries combined. This figure reflects the dominance of service industries in the economy, which explains an apparent paradox in Figure 3: the community services sector, despite having the lowest prevalence of smoking, has the greatest number of smokers.

Furthermore, the three employment characteristics reported in Figure 4 indicate that smokers are far more likely than non-smokers to regularly work weekends or evening/night shifts or to be employed in small (fewer than 20 employees) firms. These are conditions of work typically found in the lower tier of the service sector.

Figure 4: Current Smokers by Selected Work Characteristics, Canada, 1990

Figure 4. Current Smokers by Selected Work Characteristics, Canada, 1990

Source: 1990 Health Promotion Survey

ETS Policy Implications

The different demographic profiles and working conditions of smokers in occupations and industries with a high prevalence of smoking necessitate carefully targeted strategies. To be successful, smoking reduction campaigns and workplace smoking bans among truckers or miners will require different content and tactics than those among restaurant workers or clerks. However, the use of aggregate industry and occupation categories in this analysis can reveal only broad trends. More detailed breakdowns, along with case studies of particular occupations or industries (e.g., O'Connor and Harrison, 1992), would enable policy-makers to target specific groups of workers.

While the literature points to blue-collar workers as a high-priority group for research and regulation, this generalization is based on what are quickly becoming obsolete assumptions about the composition of the labour force. It is important to recognize that now over 70% of all Canadian workers are in service industries and that 44% of all workers are women. The fact that the lower tier of the service sector has such a high smoking prevalence and the largest concentration of smokers -- and therefore the highest number of non-smokers exposed to ETS -- must inform future policy development. Those employees with the least ETS regulation in the service sector work in restaurants, bars, lounges, hotels and motels, casinos, and other related businesses.

Furthermore, blue-collar employment has been steadily declining as white-collar jobs proliferate. Manufacturing work areas may actually pose less of a health risk in terms of ETS exposure than office areas, where a far greater proportion of the workforce is employed. A recent study that used passive monitors to sample weekly concentrations of ETS in 25 workplaces found that in workplaces with no restrictions on smoking, a higher percentage of office workers than manufacturing workers were exposed to ETS at levels posing a significant health risk (Hammond, Sorensen, Youngstrom and Ockene, 1995:958).

The hospitality sector also exposes employees to high levels of ETS. The level of ETS in restaurants is 1.6-2.0 times higher than in office workplaces and 1.5 times higher than in residences with at least one smoker (Siegel, 1993; see also Jarvis, Foulds and Feyerabend, 1992). Levels in bars are 3.9-6.1 times higher than in offices and 4.4-4.5 times higher than in residences with smokers. There may be a 50% increase in lung cancer risk among food service employees that can be partly attributed to ETS exposure, compared with the risk in the general population (Siegel, 1993). Given this kind of epidemiological evidence, the rapidly growing hospitality sector has become the target of recent smoking control legislation in Canada and especially in the United States --and will no doubt be the battleground with the tobacco industry for future gains in eliminating ETS exposure.

Workplace Size

Workplace size is directly related to smoking prevalence among employees. Figure 4 documents that in firms with fewer than 20 employees, 37.4% are current smokers.

The prevalence of smoking declines steadily as firm size increases, falling to 30.5% in large organizations (500 or more employees). The largest drop in smoking prevalence occurs at the 100-employee mark. Even more telling is the finding in Figure 4 that smokers in small firms outnumber those in the largest firms. Furthermore, small workplaces are far less likely than larger workplaces to have any kind of smoking policy.

Almost half (47.4%; 1990 Health Promotion Survey special tabulation) of Canadian workers are employed in workplaces with fewer than 100 employees, and small service sector firms have for some time been the main source of job creation. In fact, the smallest firms (fewer than 20 employees) account for about 25% of all hours worked in full-time jobs (Morissette, 1991:40). This suggests that ETS exposure for non-smoking workers in this sector is substantial. Small firms are concentrated in construction and consumer services; they are far less common in forestry and mining, manufacturing, distributive services, and business services. Workers in small firms, compared with those in large ones (500 or more employees), tend to have lower levels of education, less work experience and higher job turnover and tend to earn less money (Morissette, 1991).

ETS Policy Implications

These observations about firm size have major consequences for ETS policy-making. Policy inaction in the small firm sector further disadvantages an already disadvantaged group of workers. Existing socio-economic differences in health status could be exacerbated, because non-smokers in large firms, where working conditions are significantly better, have been the main beneficiaries of smoking restrictions. Furthermore, by not targeting small firms, where younger workers often enter the workforce, policy-makers run the risk of leaving teenage and young adult workers exposed to the socialization of workplace smoking.

Top of Page

Looking Beyond Smoking Prevalence Trends

Interactions with Workplace Hazards

A more comprehensive understanding of what underlies occupational and industrial variations in smoking prevalence is required to design a successful smoking reduction strategy. While occupational status is used as a predictor or control variable in some studies of smoking, it has received far less attention than other socio-demographic factors such as age, gender or education. Indeed, Sterling and Weinkam (1990:457) argue that epidemiological studies of smoking tend to ignore occupation. Consequently, the confounding effects of smoking with other occupation-specific hazards are often missed. These researchers use U.S. National Health Interview Survey data to document that industries and occupations with high rates of exposure to toxic fumes and dust also have the highest proportions of smokers and the lowest rates of smoking cessation.

The synergistic effects of tobacco smoke with other workplace hazards are well known (Douville, 1990). For example, chronic bronchitis is associated with occupational exposure to coal, grain, silica, pollutants in welding environments and, to a lesser degree, sulphur dioxide and cement. Among exposed smokers, the rate of bronchitis is higher than that from smoking alone. Asbestos exposure increases the risk of lung cancer in smokers by a factor greater than the sum of the risks of independent exposure. Furthermore, there is evidence that the interactive effects of tobacco smoke and workplace pollutants contribute to lung cancer among coke oven workers and gas workers, to pancreatic cancer in refinery workers and to bladder cancer in groups such as janitors and cleaners, mechanics, mining machine operators, printing machine operators and workers in the metal mining industry. Rubber workers face an increased risk of pulmonary disability when exposed to occupational particulates and smoking combined. Compared with their non-smoking co-workers, cotton textile workers who smoke have an increased prevalence and severity of byssinosis and bronchitis.

Socio-economic Status and Smoking

Canadian research on the relationship between different occupations and the risk of smoking is limited. The research is narrow, describing only the prevalence of worksite smoking restrictions by using aggregate categories (e.g., Millar, 1988b; Krahn and van Roosmalen, 1991). However, these descriptive studies nonetheless reveal striking variations in smoking behaviour by socio-economic status. Occupation is a key indicator of socio-economic status. It is therefore essential to examine the complex set of economic, environmental, social and cultural factors that underlie smoking as a class-based behaviour. For example, Letourneay and Bujold's (1990:28-9) research on smoking among francophones in Quebec identifies blue-collar, less educated and low-income workers as high-risk groups that require specially tailored workplace intervention strategies. Generally, low socio-economic status, as measured by education and income, is one of the strongest predictors of smoking (Millar, 1988b; Millar and Hunter, 1990; Health Canada, 1994).

Ethnic and cultural factors also influence smoking prevalence and can interact with socio-economic status. While it is beyond the scope of this report to examine such factors, suffice it to say that from the perspective of worksite smoking policy development, it is important to understand how the ethnic composition of a specific workforce may influence smoking attitudes and behaviour. For example, 59% of Aboriginal Canadians are regular smokers, compared with only 11% of Asian Canadians (Millar, 1992; see also Lando, Johnson, Graham-Tomasi, McGovern and Solberg, 1992).

ETS Policy Implications

For many industrial workers, tobacco smoking is but one of the serious health risks they face regularly on their jobs. This information could be used in occupational health campaigns designed to reduce smoking, but success will depend on more than just identifying smoking per se as the major problem. Unions representing such employees take the position that all workplace hazards must be addressed in any occupational health campaign, rather than singling out smoking. To focus on workers' smoking behaviour is to "blame the victim" and shift attention from the employers' responsibility to eliminate a host of other pollutants and toxins from the work environment.

What Is a Workplace?

Defining a Workplace

A major weakness in existing legislation is the lack of a consistent and comprehensive definition of "workplace." Historically, public places that are workplaces -- restaurants and bars being the main examples -- have been treated separately by clean indoor air legislation (Americans for Nonsmokers' Rights and Prospect Associates, 1994). Such workplaces are typically dealt with in "public place" local by-laws, on the assumption that it is the customer, rather than the employee, who requires protection from ETS.

Even in legislation restricting smoking on the job, "workplace" usually refers to indoor space only or is limited to offices. Recall that the highest rates of smoking are in blue-collar occupations, where much of the work is performed outdoors. Less common are more inclusive definitions of a workplace as any "place of employment." Some by-laws exclude private homes and independent contractors or partnerships or exempt private social functions held in otherwise restricted workplaces. Furthermore, virtually all legislation assumes that work is paid work. Volunteer work (e.g., unpaid work in voluntary organizations) is rarely mentioned explicitly.

Looking ahead, policy-makers should expand the definition of workplace to include homes. The computer revolution, coupled with the rise of "contracting out" as organizations downsize, has fuelled the rise of home-based businesses. Teleworking is also on the rise. A growing number of employees are performing their job functions at home or in other remote locations via computer, fax and modem. In 1991, over 500,000 Canadian employees worked at home, most in government, finance, computer and telecommunications industries (Frank, 1995:7). A 1991 Gallup Survey estimated that 23% of the workforce was engaged in home-based paid work (working at home full-time or part-time, as an employee or self-employed, at least one day per week) (Orser and Foster, 1992:69). These new forms of work are not yet well documented.

ETS Policy Implications

At issue, then, is the ability of legislation to extend to all places where individuals work. The distinction now common in Canadian laws between public places and workplaces has excluded a number of industries from the purview of workplace ETS restrictions. A more comprehensive definition of "workplace" is thus required if legislation is to benefit all non-smoking workers.

From a health policy perspective, we need to know the proportion of smokers involved in home-based businesses and teleworking. Will smokers who are prohibited from smoking at their place of employment be more likely to work at home if such arrangements are voluntary? To what extent will other household members, especially children, face exposure to elevated levels of ETS from smoking home-based workers? Home-based work is a popular option for women with young children. Thus, what family-friendly workplace advocates see as a solution could create added health risks for the children of smoking mothers who otherwise would be working outside the home.

Restrictions on Workplace Smoking

Variations in Workplace Smoking Restrictions

Variations by Region and Gender

Turning to Table 1, we find regional variations in workplace smoking policies for both male and female workers. As of August 1994, about 4 in 10 Canadian employees (or 4.5 million) reported that smoking was totally banned in their workplace. This ranged from a high of 46.8% in British Columbia to a low of 25.4% in Quebec. (A complete provincial breakdown, which cannot be reported in full when gender is controlled owing to high coefficients of variation, shows 52% in Newfoundland.) A similar proportion of employees reported that smoking was restricted to certain areas of their workplace. And one in five employees, or 2.29 million, reported no smoking restrictions at all.

Canada has made huge strides in reducing workers' exposure to ETS over the past eight years. In 1994, 80% of Canadian workers had smoking restrictions in their workplace. This is double the proportion of workers who had some form of smoking policy in 1986 (Millar and Bisch, 1989). While less than 10% of employees had total smoking bans in their workplace in 1986, total bans had increased four-fold by 1994.

Equally important are the gender differences in workplace ETS exposure. Table 1 confirms that women benefit more than men from total smoking bans. Half of employed women have such bans in their workplace, compared with just one third of men. The largest regional difference in total smoking bans for female workers is between Ontario (60.2%) and Quebec (34.1%). Regional variations for men are not as pronounced.

Top of Page
Table 1 Smoking Restrictions at Work, by Region and Sex, Canada, August 1994
  Population
estimate (000s)		 Total ban (%) Partial ban (%) No restrictions
Canada
Total 11,489 39.6 40.5 19.9
Male 6,922 33.1 41.1 25.8
Female 4,567 49.4 39.7 10.9
Atlantic Region
Total 909 44.4 35.9 19.6
Male 536 37.9 34.8 27.3
Female 373 53.7 37.5 .
Quebec
Total 2,788 25.4 44.2 30.4
Male 1,724 20.1 41.9 38.0
Female 1,065 34.1 47.9 18.0
Ontario
Total 4,213 48.2 38.2 13.6
Male 2,531 40.5 41.3 18.1
Female 1,682 60.2 33.4 6.4
Prairie Region
Total 2,057 33.8 47.4 18.8
Male 1,221 28.9 47.4 23.7
Female 835 41.0 47.3 11.7*
British Columbia
Total 1,521 46.8 33.6 19.6
Male 909 40.3 34.1 25.6
Female 612 56.4 42.9 10.7*

* High sampling variability
-- Data suppressed

Source: 1994-95 Survey on Smoking in Canada, Cycle 2

Nationally, 25.8% of male and 10.9% of female workers have unrestricted ETS exposure. The only region in which males and females are substantially above this national average is Quebec. There, 38.0% of males and 18.0% of females in the workforce face the risk of unregulated ETS exposure. Remarkably, only 6.4% of all Ontario female workers fall into this category.

Rural

Urban Differences. Few studies have examined rural-urban differences in smoking behaviour or workplace smoking restrictions. One exception is the 1990 Alberta Survey, which found province-wide receptivity to non-smoking policies, despite the fact that rural areas had higher smoking rates and a lower prevalence of workplace smoking restrictions (Krahn and van Roosmalen, 1991). The researchers pointed out that higher smoking rates in rural areas are concentrated among younger males. Generally, this demographic group is least likely to work in an organization with a no-smoking policy, mainly because they are over-represented in blue-collar jobs and live in towns, villages and rural areas where workplace smoking policies are less common.

Variations by Firm Size

Several studies have documented a relationship between firm size and workplace smoking restrictions. Perhaps the only Canadian study to address this issue is the assessment of the City of Toronto's 1993 public place and workplace smoking by-laws (Kendall, 1994). The study found that in small workplaces (fewer than 20 employees), a smoke-free environment was more likely to be provided if the owners or managers were not current smokers. This factor had no bearing on the presence of smoking policies in larger workplaces.

American research elaborates on this finding. An evaluation of California's Tobacco Control Program found that while the proportion of smoke-free workplaces almost doubled between 1990 and 1993, most of this gain was in large firms (Pierce, Evans, Farkas et al., 1994). Workplaces with fewer than 50 employees were less likely to have smoke-free policies, and proportionally fewer introduced new smoking restrictions between 1990 and 1992. The overall success of the California anti-smoking strategy is impressive, with 87% of indoor employees covered by workplace smoking prohibitions by 1993. Increasing this coverage will require initiatives within the small business sector, which, of course, necessitates action in the hospitality industry.

A study of employer compliance with a comprehensive non-smoking by-law in Cambridge, Massachusetts, conducted 3 months and 24 months after the law was implemented, identified several barriers to self-enforcement among small businesses (Rigotti, Stoto and Schelling, 1994). Some 65% of firms surveyed were small, with fewer than 10 employees, and about half were in the service sector. Many small businesses misunderstood the smoking by-law or were misinformed. Unlike large employers, small firms typically do not write formal policies for employee behaviour, so a law requiring them to do so went against established practice. Compliance with the by-law was associated with the proportion of smokers among employees. Also influential was whether the owner smoked, given his or her key role in setting policy.

Interestingly, there is one study showing that small firm employers are more likely to have smoking policies. Nelson, Sacks and Addiss (1993) documented smoking policies in licensed child-care centres in the United States. Some 55% of licensed centres in the study were smoke-free indoors and outdoors. These centres tended to be located in the west or south, were smaller and independently owned and had written smoking policies. The researchers offer no explanation of why size or ownership matters in this regard.

ETS Policy Implications

From a policy perspective, it is crucial to note regional and gender variations in workplaces with no smoking restrictions. The high smoking rates in the Quebec labour force, especially among females, underscore the need for more strenuous regulatory action in Quebec. In addition, rural-urban patterns of employment provide some clues to the higher prevalence of smoking among certain demographic groups. Our observations about firm size also have implications for ETS policy-making. Resistance to legislated restrictions is likely to be greater among small businesses. Given that large firms often voluntarily implement smoking restrictions, ETS legislation is more likely to affect small businesses. And, since knowledge of the law is a crucial factor in developing a workplace policy, concerted education, monitoring and worksite inspections must be a central component of any legislation aimed at small business.

Canadian Legislation

Provincial and Municipal Laws

Smoking is now banned in provincial government workplaces in British Columbia, Saskatchewan, Ontario, New Brunswick, Nova Scotia and Newfoundland and in government offices in the Northwest Territories. Quebec's 1987 legislation is certainly the weakest of the provincial laws (Quebec, 1987). Managers in the Quebec public sector are delegated to decide how, and to what extent, smoking is to be restricted.

Private workplaces are not directly regulated, with three important exceptions: 1) Newfoundland bans smoking in all workplaces and prescribes ventilation requirements for enclosed smoking rooms; 2) Ontario restricts smoking, permitting up to 25% of a workplace to be a designated smoking area without ventilation requirements; and 3) the 1988 federal Non-smokers' Health Act restricts smoking to designated areas or rooms in workplaces under federal jurisdiction. Federal prisons, which were exempted from the Non-smokers' Health Act, are planning to go smoke-free in April 1998 (Globe and Mail, 1995).

Ontario's 1990 Smoking in the Workplace Act contains many flaws, not the least of which are the inadequate protection afforded non-smokers by allowing up to 25% of a workplace to be designated a smoking area and the absence of ventilation requirements. Newfoundland's 1993 Smoke-Free Environment Act may go the furthest, with its requirement that designated smoking areas meet prescribed ventilation standards. Both the Ontario and Newfoundland laws protect employees from reprisals for seeking enforcement of the terms of the Act (National Clearinghouse on Tobacco and Health, 1995).

A growing number of municipal by-laws also limit smoking. These regulations apply mainly to public places, although workplaces are increasingly being covered. As of May 1995, 39% of municipalities (270 out of 698) with populations over 10,000 had smoking control by-laws; another 4% had a non-smoking policy (Health Canada, 1995b). Only 15% of the by-laws impose workplace smoking restrictions beyond municipal offices. Most by-laws require employers to have written policies, take responsibility for compliance and set up smoking areas that reduce smoke through ventilation, size or physical barriers.

Hospitality Industry Restrictions

Over half of municipalities (52%) restrict smoking in restaurants, with most requiring 50% of the seating to be non-smoking. As of June 1994, Canada had no municipalities with 100% smoke-free restaurant provisions, whereas the United States had 146 (Council for a Tobacco-Free Ontario, 1995). Few had restrictions applying to licensed establishments. By April 1995, many Ontario municipalities were considering tobacco control by-laws that would require total bans on smoking in restaurants and, in some cases, bars. These initiatives are being strongly resisted by the Ontario Restaurant Association and the Canadian Restaurant and Foodservice Association (Ontario Campaign for Action on Tobacco, 1995). Restaurant workers are already prohibited under the Health Protection and Promotion Act from smoking inside restaurants. Several large municipalities in British Columbia are investigating ways of achieving 100% smoke-free environments in restaurants, bars, bingo halls, casinos and other hospitality industry venues (City of Vancouver, 1994).

Voluntary bans in the private sector are most likely found in large corporations (Lowe and Neale, 1992). The most recent examples are in the fast-food industry: corporate-owned McDonald's, the Taco Bell chain and some Tim Horton's (Health Canada, 1995a; National Clearinghouse on Tobacco and Health, 1995). Some smaller establishments, too, have opted to go smoke-free. In British Columbia, for example, there are at least 550 eateries that are smoke-free (Ontario Campaign for Action on Tobacco, 1995:8).

American Progress Toward Smoke-free Workplaces

The Environmental Protection Agency Ruling on ETS

A watershed in the drive to ban ETS from workplaces is the EPA's 1992 report classifying ETS as a Group A (or known) human carcinogen (U.S. Environmental Protection Agency, 1993). That classification is based primarily on the causal association between ETS exposure and lung cancer. The EPA had earlier concluded that ETS is the most widespread and harmful indoor air pollutant and a major contributor to indoor air pollution (U.S. Environmental Protection Agency, 1993). Both the EPA and the National Institute for Occupational Safety and Health have recommended either eliminating smoking in buildings or creating separate, enclosed smoking areas that are separately ventilated and exhausted directly to the outside (Americans for Nonsmokers' Rights and Prospect Associates, 1994).

Top of Page

The EPA report sparked interest in ETS as a public health issue among legislators and policy-makers (Maskin, Connolly and Noonan, 1993). In addition, the EPA report has had a major impact on public perception, which in turn should prompt employers to reduce their legal liability by eliminating employees' exposure to ETS. The recent trend favouring total bans instead of separate smoking areas can be traced to the EPA ruling (Americans for Nonsmokers' Rights and Prospect Associates, 1994:10). This is buttressed by scientific evidence on the actual exposure of non-smoking workers to ETS, which suggests that total bans represent the only safe policy (Repace and Lowrey, 1993:470).

Other Developments

In May 1993, 16 U.S. attorneys general formed a working group to study tobacco-related issues, including ETS exposure in fast-food restaurants. The working group's major concern was with children's exposure to ETS. The study also included workers in the fast-food sector, given that 40% are estimated to be under the age of 18 (Council for a Tobacco-Free Ontario, 1995). After the Attorney General of Texas launched a lawsuit based on the working group's recommendation that fast-food restaurants go smoke-free, McDonald's and Taco Bell banned smoking in all corporate-owned outlets.

Also influential in encouraging employers to voluntarily adopt smoking bans was the 1990 Americans with Disabilities Act. It offered protection against ETS by requiring employers (with 15 or more employees) to ban smoking to protect non-smokers with documented sensitivities to tobacco smoke. However, there is legal opinion suggesting that smokers may be able to use the Americans with Disabilities Act as a basis for legal action if they are refused employment because they are "regarded as" having a disability in the form of their tobacco addiction. However, there is insufficient case law to predict how the Act will be applied to smokers (Sugarman, 1993:168).

California Legislation

The political issues surrounding the new California legislation banning smoking in enclosed places of employment presage the counter-pressures Canadian law-makers will face. The law's intent is to protect California workers from the "serious health effects" of ETS by banning smoking in all workplaces, with only hotel and motel guest rooms exempted (California State Senate, 1993). The bill was sponsored by a broad coalition that included the California Restaurant Association, California Medical Association, California Labour Federation AFL-CIO, American Heart Association, American Lung Association, California Hotel and Motel Association and Building Owners and Managers Association.

For its part, the California Restaurant Association was concerned about the increasing numbers of workers' compensation claims based on workplace exposure to ETS, especially in the wake of the EPA ruling noted above. It cited a case in which an otherwise healthy non-smoking waiter received over $80,000 in workers' compensation benefits for a heart attack he claimed was caused by ETS. Achieving uniform standards across the state, in the face of diverse local ordinances, was also an important consideration. Oddly enough, the Americans for Nonsmokers' Rights lined up with the Tobacco Institute in opposing the bill --for different reasons, of course. The opposition of the Americans for Nonsmokers' Rights stems from the lack of an anti-preemptive clause explicitly allowing stronger local ordinances and the lack of local enforcement provisions.

Implications for Canadian ETS Policy-making

New policy and legislative initiatives in the United States have pushed closer to the goal of smoke-free workplaces and public places (Maskin, Connolly and Noonan, 1993:65). The major new initiatives are in the hospitality sector. The states of Vermont, Utah and California have complete smoking bans in restaurants, and in 1997 California's ban extends to bars (Americans for Nonsmokers' Rights and Prospect Associates, 1994; Ontario Campaign for Action on Tobacco, 1995). While Canadian municipalities have yet to impose total smoking bans in restaurants, 136 U.S. municipalities had done so by June 1994 (Council for a Tobacco-Free Ontario, 1995). American experience suggests that an advantage of this local approach to ETS regulation is that the tobacco industry has had greater difficulty lobbying against hundreds of decentralized initiatives, compared with state or federal legislation (Kagan and Vogel, 1993:43). Current policy-making by the U.S. Occupational Safety and Health Administration (OSHA) may provide insights on how Canadian occupational health and safety (OHS) and workers' compensation legislation can be reformed to rid workplaces of ETS.

Ingredients of Effective Legislation

A review of clean indoor air legislation identifies the following minimum requirements for effective legislation:

  1. prohibit smoking in all workplaces; or
  2. prohibit smoking in all workplaces, except in designated smoking areas that are enclosed, separately ventilated and directly exhausted to the outside;
  3. if smoking is allowed in designated areas, prohibit smoking in all commonly used areas and specify that non-smoking employees' rights prevail when disputes arise over smoking in the workplace;
  4. protect non-smokers who assert their rights under the law from retaliation by an employer;
  5. require employers to develop a written workplace policy consistent with the law, to communicate this policy to employees and to post non-smoking signs;
  6. impose stiff penalties for non-compliance; and
  7. if the legislation is provincial, it must not preempt stronger municipal legislation.

In U.S. policy circles, it is also considered important to prohibit employers from discriminating against smokers in hiring, career decisions or firing (Americans for Nonsmokers' Rights and Prospect Associates, 1994:9).

As municipal by-laws become stricter and reach into the hospitality sector, it is crucial to launch any new legislative initiatives with public consultations (as Vancouver has done). This will help to galvanize community support.

A comprehensive information and education campaign, as in the 1993 Toronto by-laws upgrading, is equally important. Such campaigns need to be flexible and adapt their message to specific socio-demographic groups. If smoking restrictions are going to succeed in the small business sector, particularly in restaurants and bars, consultations and information campaigns must allay concerns about business loss. In this regard, it is important to recognize that support for smoking restrictions may be inhibited by the culture of these workplaces, fear of reprisals or a desire to avoid conflicts with smoking employees.

Regulating ETS Through Occupational Health and Safety Legislation

Recent actions by the EPA and the OSHA in the United States have focused the attention of Canadian policy-makers on OHS and workers' compensation legislation as a means of eliminating ETS from the workplace. Unlike the United States, where OSHA regulation of workplace smoking could preempt weaker local ordinances, in Canada there is no federal administrative body with national powers in this regard. However, some provinces, notably Saskatchewan, are beginning to move against ETS via the OHS route.

ETS

-- an Occupational Safety and Health Priority in the United States. The U.S. OSHA, which is responsible for ensuring a safe and healthy work environment, is taking the lead in addressing workplace ETS exposure. Prior to the EPA report, the OSHA did not consider ETS a priority. However, experts now argue that the Occupational Safety and Health Act may be the single most important weapon in the anti-smoking legislative arsenal (Short, 1992:68). The OSHA is coordinating its efforts with the Department of Health and Human Services on public education, and the Centres for Disease Control has issued guidelines for achieving smoke-free workplaces, proposing an outright ban on smoking in indoor workplaces. The Building Owners and Managers Association supports this, having requested such restrictions in 1993 on the grounds that smoking is the main cause of fires in office buildings (City of Vancouver, 1994).

Challenges Presented by an OHS Strategy

Devising appropriate strategies in the occupational health arena raises major questions for Canadian policy-makers. First, could existing provincial OHS laws and the Canada Labour Code be easily amended to explicitly incorporate ETS as an occupational health hazard? Second, what weaknesses in existing OHS legislation need to be rectified in order to improve its ability to deal with ETS? Third, how could OHS legislation bring about greater reductions in ETS exposure than those already achieved through existing non-smoking legislation and policies? Fourth, how will the variations now found in provincial OHS legislation on issues such as the right to refuse unsafe work and requirements for joint worksite health and safety committees affect the ability to achieve uniform ETS reduction goals and standards across Canada? Answers to these questions will help frame policy discussions about how to chart a course into this new regulatory terrain.

A Health Canada ETS workshop earlier this year recommended using existing OHS laws to regulate ETS (Health Canada, 1995a), because such laws require employers to provide a safe working environment. It may be possible for employees to use provisions of OHS laws to obtain enforcement of restrictions. In principle, they should have the right to refuse to work in environments with high levels of ETS. This would require legislating threshold levels of ETS exposure and empowering workers to monitor these ETS levels. As many workplaces still do not have complete bans, it is essential to establish an "enforceable air quality standard" for ETS to enable those responsible for monitoring and enforcement to measure the smoke in a worksite (Repace and Lowrey, 1993:464). Coupled with this objective, it may also be necessary to more accurately define the frequency of exposure to ETS among specific individuals in a worksite (Burns, Axelrad, Bal et al., 1992:S15). This may be one of the larger challenges legislators face in placing ETS under the OHS legislative umbrella.

Legal Precedents

There are some legal precedents from Ontario, as well as from Britain and the United States, which establish that an employer has a duty to maintain a safe workplace as grounds for requiring the restriction of ETS exposure (Council for a Tobacco-Free Ontario, 1995:24). Other aspects of Canadian OHS legislation that could apply indirectly to ETS include the regulation of substances found in tobacco smoke (possibly through the national Workplace Hazardous Materials Information System, or WHMIS) and the provision for joint worker--management health and safety committees. For example, the Canada Occupational Safety and Health Regulations under the Canada Labour Code set threshold limit values and biological exposure indices for carbon monoxide and carbon dioxide, as well as listing various human carcinogens. However, some legal experts consider these general provisions ineffective for dealing with ETS, as courts look for specific provisions (Grossman and Price, 1992:6-19).

Top of Page

In the United States, workers seeking a smoke-free workplace based on the provisions of the Occupational Safety and Health Act have not succeeded in legal action against employers. The Act does not give employees the right to sue employers, a right that would provide the basis for an employee complaint about ETS (Bowers, 1992:42). While the Act requires private employers to provide a work environment "free from recognized hazards that are causing or are likely to cause death or serious physical harm to [an] employee," as recently as 1990 courts have not ruled that this requires smoking bans (Vaughn, 1992:125). Yet, as suggested above, this regulatory framework could change as the OSHA moves to incorporate ETS.

Using the Workers' Compensation System Against ETS

British Columbia is adopting a related approach through the workers' compensation system. The B.C. Workers' Compensation Board (WCB) has drafted indoor air quality regulations that identify ETS as an occupational health hazard and has recommended designated smoking areas (Workers' Compensation Board of British Columbia, 1994:77-8). The WCB Occupational Hygiene Subcommittee report states that non-smokers should be provided a smoke-free work environment. It also recommends that employers be made responsible for bans or for establishing separately ventilated and enclosed designated smoking areas that conform to American Society of Heating, Refrigeration and Air-Conditioning Engineers (ASHRAE) standards. This report further recommends that the ASHRAE standards be applied to the fresh air supplied in restaurants, night clubs and games rooms -- as well as the use of smoke elimination equipment.

Using Ontario's Occupational Health and Safety Act Against ETS

Cunningham's (1995) analysis of the Ontario Occupational Health and Safety Act identified 11 ways in which the Act could be used to eliminate workplace ETS:

  1. the obligation on employers to acquaint workers with hazards and to "take every precaution reasonable in the circumstances for the protection of a worker";
  2. a similar obligation on supervisors;
  3. prohibiting employees from working in ways that endanger themselves or other workers;
  4. regulations governing industrial establishments prohibiting smoking in areas where there are poisonous substances;
  5. regulations prohibiting exposure to certain agents found in ETS above set levels; or
  6. regulations prohibiting any exposure;
  7. the right to refuse work when physical conditions of the workplace are likely to endanger the worker;
  8. an order by two certified members of the joint committee to stop dangerous circumstances;
  9. an order by the director to the employer to prohibit or control the use of agents likely to endanger workers' health;
  10. the ability of a joint committee to make recommendations to the employer; and
  11. using the authority under the Act for government to restrict or eliminate smoking in workplaces.

ETS Policy Implications

Before public health policy-makers pursue any of these strategies, they should be aware of the current limitations of OHS legislation. Most OHS legislation in Canada rests on the principles of the "internal responsibility system" (IRS). The IRS draws on European models, which assume that workers directly exposed to risks and hazards are in the best position to develop solutions suited to their particular workplace. The cornerstones of the IRS are a worker's rights to know the hazards of the workplace, supported by WHMIS; to be consulted on health and safety issues by management; and to refuse unsafe work (Krahn and Lowe, 1993:281-3). Yet workers are often unaware of their rights and usually lack the necessary knowledge to address complex safety and health problems (Sass, 1986; Walters and Haynes, 1988). Certainly, there is a much higher level of worker awareness of the health hazards associated with smoking; still, these basic concerns remain potential obstacles to eliminating ETS via OHS legislation.

Impact of Workplace Smoking Restriction

Compliance Issues

How effective are various legislated restrictions on workplace smoking, the vast majority of which are self-enforcing? Municipal by-laws typically place the onus on employers to notify employees of a non-smoking policy, implement the policy and ensure compliance. A 1994 survey of Canadian municipalities (Council for a Tobacco-Free Ontario, 1995:57) found the following compliance problems with workplace smoking by-laws:

  • employee and public complaints about non-compliance;
  • lack of employer enforcement;
  • difficulties of enforcement;
  • employees complaining that designated smoking areas are inadequate or not provided; and
  • complaints about smokers puffing away in doorways.
  • However, the extent of these specific complaints was not documented.

Evaluating City of Toronto By-laws

A comprehensive evaluation of the City of Toronto's 1993 by-laws restricting smoking in workplaces and public places found widespread compliance three months after the laws were implemented (Kendall, 1994). The Workplace Smoking By-law requires all workplaces to be completely smoke-free, unless a designated smoking area is provided (it must be fully enclosed and separately ventilated to the outside, less than 25% of the indoor space and not located in an essential part of the workplace). Health care facilities fall under the Public Place Smoking By-law and must be smoke-free, with exemptions only for patients or residents in chronic care facilities.

Nine out of 10 workplaces had smoke-free environments, with 83% banning smoking entirely and another 7% providing a fully enclosed and separately ventilated smoking area. The main factor associated with the provision of a smoke-free work environment was having 10% or fewer employees who smoked. Employers were concerned about possible reductions in productivity due to smoking breaks and having to force smokers outside in bad weather or when they had health problems. Interviews with smokers found general acceptance of restricted smoking areas in offices. Because these and other by-laws are self-enforcing, it is essential to inform the public and businesses about the need for, and requirements of, the law. Consequently, Toronto had a full-time information officer assigned to oversee the communications and education programs for its 1993 by-laws.

Restaurants

Restaurants had an 86% compliance level with the Public Place Smoking By-law. However, Kendall's (1994) report is critical of the 50% of restaurant seating set aside for smokers, because non-smokers are still exposed to high levels of ETS. A mere 4% of restaurants were voluntarily smoke-free. There were difficulties getting restaurants to comply with the by-law's duty-to-inform provisions. The proposed solution to non-compliance is to convince restaurant owners that they will not lose business and, furthermore, that there is indeed a "market force" for clean air in restaurants (Kendall, 1994:59). The report concludes that a 100% smoke-free strategy would be the best way of eliminating ETS in restaurants. However, it cautions that the timing and implementation of such a policy must be carefully orchestrated to maximize acceptance by the restaurant industry.

Vancouver's Proposed ETS Ban

The Vancouver Health Department's preparations for introducing a 100% smoke-free indoor environment by-law in January 1996 raises compliance issues (City of Vancouver, 1994). This law would be self-enforcing. To pave the way, the department held public consultations on how best to move toward eliminating ETS, plans a phased-in approach and aims to build compliance by allowing time for education, acceptance and dissipation of resistance.

Pockets of resistance and potential non-compliance were clearly evident from the public consultations. Associations representing bowling alleys, pubs and cabarets, hotels, casinos, and restaurant and food service establishments opposed the ban, mainly based on concerns about loss of business. Their ideological preference is to "let the marketplace decide" (Strategic Action Group, 1995). The Hotel Employees and Restaurant Employees Union (HERE) voiced concerns for the health of its members but opposed a total ban (Strategic Action Group, 1995). In the public consultations, a civic employee claimed that existing city policy was being violated in the engineering yards. He recommended better enforcement, noting that non-smoking employees are reluctant to complain, fearing backlash or believing nothing can be done. This complaint is unlikely to be an isolated incident. In offices, social norms are more likely to favour non-smokers. However, in other kinds of work environments, especially outdoors, the self-regulating framework breaks down, especially if employees' own knowledge, values and behaviours do not place a priority on health promotion.

Sociological Factors in Compliance

How can we explain the relatively high level of compliance to workplace smoking restrictions in North America? Kagan and Skolnick (1993) argue that compliance costs are low, violations are difficult to conceal and violators are not supported by a deviant subculture based on resistance to enforcement. Especially important, the values attached to smoking behaviour, what Kagan and Skolnick (1993:79) call "the rules governing the civility of smoking," have radically changed. "American society's traditions and health values turned carcinogenic' -- a taboo word -- into a trump card," they observe (Kagan and Skolnick, 1993:84), permitting non-smokers to assert their right to clean air.

Social values also underlie cross-national and even inter-regional differences in the tolerance of smoking. For example, when managers at a Hewlett-Packard branch in France tried to restrict smoking, angry staff threatened to barricade the firm's restaurant to keep overweight people out on the grounds that more food would be bad for their health (Kagan and Skolnick, 1993:85). Neither the availability of scientific evidence linking ETS with increased risk of morbidity and mortality nor legislation has altered the values underlying this reaction. Similar "cultural" differences also help to account for variations within Canada in smoking prevalence and receptivity to ETS regulations by region, industry, occupation and workplace.

Top of Page

Impact on Smoking Prevalence

Evidence from Population Studies

Considerable evidence suggests that restricting or banning smoking in workplaces reduces smoking prevalence and cigarette consumption among those who continue to smoke. A review of representative studies shows, however, that these gains are moderate, at best.

California's Tobacco Control Program has documented considerable success (Pierce, Evans, Farkas et al., 1994). The proportion of smoke-free workplaces almost doubled between 1990 and 1993, and non-smokers' exposure to ETS was cut by about 25%. Maintaining a smoke-free workplace was associated with a 14% overall decline in smoking prevalence. The introduction of a smoke-free work area policy resulted in a 10% decrease in per capita cigarette consumption. This figure rose to 26% when the policy was maintained. Similarly, maintaining such a policy contributed to increased quit rates over time, especially among light smokers. Prevalence and consumption increased among smokers who moved from a smoke-free to a less restricted worksite between 1990 and 1992.

Other researchers calculate from the 1990 California Tobacco Survey that cigarette consumption among indoor employees is 21% below what it would be without restrictions (Woodruff, Rosbrook, Pierce and Glantz, 1993). Moreover, multivariate analysis found that after controlling for the effects of age, education, ethnicity and gender, workers with only work area restrictions were 1.15 times more likely to be smokers than those in smoke-free workplaces. This ratio increased to 1.3 times when restrictions were weaker or absent (Woodruff, Rosbrook, Pierce and Glantz, 1993:1489). Similar results of legislated workplace smoking restrictions have also been found in Washington state (Kinne, Kristal, White and Hunt, 1993). While this suggests that stricter policies yield better quit rates, the cross-sectional design of these sample surveys prevents us from making causal inferences.

Evidence from Case Studies of Employers

Additional evidence of the positive impact of smoking bans can be found in case studies of specific employers. Millar's (1988a, 1988c) evaluation of the 1986 smoking restrictions at Health and Welfare Canada compared smoking behaviour 5 months before the policy was introduced and again 13 months into the new policy. Attempts at quitting were well above the national average, but only 3% had given up smoking for an entire year. Still, this was about 10 times the continuous one-year quit rate in the general population at the time. There was also a small reduction in the number of cigarettes smoked per day and a much larger decrease in the mean number of cigarettes smoked at work. A key feature of the policy was a self-help smoking cessation program. About 16% of smokers registered in such courses achieved continuous one-year quit rates, slightly above that for all smoking employees.

Several case studies augment these findings. For example, 18 months after a total smoking ban was introduced in Telecom Australia, smokers were consuming between three and four cigarettes less per workday, and the quit rate among smokers was about double the community rate (Hocking, Borland, Owen and Kemp, 1991). This study also examined how the ban affected work performance and discovered that 75% of staff surveyed reported positive effects. Negative effects were reported mainly by smokers and resulted from the additional time they spent going for a smoke.

A study of the smoking ban at Baltimore's Johns Hopkins Hospital surveyed employees six months prior to a total ban and then six months after the ban had been in place (Stillman, Becker, Swank et al., 1990). Quit rates among smokers ranged from 9% to 20%, depending on the estimation parameters used. Total daily cigarette consumption and consumption at work among employees who continued to smoke decreased by 25%. Equally interesting, multivariate analysis identified the number of cigarettes smoked at the baseline survey and education level as independent predictors of quitting for periods longer than three months. That is, controlling for other socio-demographic and employment characteristics, smokers with lower cigarette consumption and higher levels of education were the most likely to quit after the policy was implemented. The ban was extensively supported by educational, health promotion and smoking cessation activities. These factors, in addition to the prospective cohort design of the study, may partly account for the positive results.

A two-year randomized trial of different smoking restrictions in 32 worksites yielded mixed results (Jeffery, Kelder, Forster, French, Lando and Baxter, 1994). Longitudinally, there was little evidence that restrictive policies (compared with no restrictions) improved cessation rates. However, smokers in worksites that adopted restrictions during the course of the study had a 10% drop in daily cigarette consumption. The researchers question the health benefits of this. Another study of U.S. hospital employees found that five months after a total indoor smoking ban was introduced, workday cigarette consumption fell, but quit rates did not increase (Daughton, Andrews, Orona, Patil and Rennard, 1992).

Impact of Smoking Cessation Programs

Limited Benefits

Estimates suggest that a sizeable minority of employers subsidize smoking cessation programs (Benowitz and Leistikow, 1994). The Health and Welfare Canada study just noted suggests modest benefits in this regard. However, recent research on the impact of smoking cessation studies has had mixed results; we still do not fully understanding what constitutes an effective cessation program. A meta-analysis of 20 studies of worksite smoking cessation programs found a 13% average quit rate after 12 months (Fisher, Glasgow and Terborg, 1990). Successful interventions were more intensive, used a combination of worktime and employees' time and were conducted in smaller organizations. Quit rates seem to be higher for heavier smokers. However, these factors explained only part of the variation in cessation program success.

Evaluating Specific Programs

Some experts argue that the potential of worksites to generate high quit rates has yet to be realized (Hymowitz, Campbell and Feuerman, 1991:366). One attempt to improve quit rates is the Johnson & Johnson Live for Life Wellness Program. This offers quit-smoking clinics in an environment enriched by health education and promotion activities. A three-year evaluation of this program found that 31.6% of smokers who participated in the smoking clinic were not smoking and that the mean duration of quitting was 16 months. Researchers compared the effectiveness of group quit-smoking programs conducted in an "enriched milieu" (health education and promotion activities, plus worksite smoking policies) with cessation programs conducted in the absence of these enriched conditions. Across the six white-collar worksites in the study, 252 employees participated in 23 quit-smoking programs over three years. The quit-smoking programs offered eight successive two-hour sessions and a brief follow-up session. Company medical staff set up waiting rooms to facilitate smoking cessation and counselled smokers to quit. Extensive health education was also provided. Some of the worksites had comprehensive non-smoking policies.

Findings suggest that while sustained quit rates were impressive among all cessation program participants (20% were not smoking 12 months after the program), there were no significant differences between the sites with no smoking restrictions and those with restrictions. It is unclear whether organizational factors or variations in how the cessation programs may have been run (e.g., group leader skills) account for these findings.

Other evaluations of cessation programs find little positive effect on quitting behaviour (Sorensen, Rigotti, Rosen, Pinney and Prible, 1991). Some evidence suggests that reductions in smoking behaviour after the introduction of smoking restrictions are short-term (Hudzinski and Sirois, 1994). The design of the intervention is crucial, and this should be adapted to the particular organization (Lichtenstein and Glasgow, 1992:523). A study of three chemical plants concluded that a comprehensive program of cessation (called SmokeFree), discouragement and smoking control is more effective than just offering cessation programs (Dawley, Dawley, Glasgow, Rice and Correa, 1993). This study included blue-collar workers with high quit rates in the comprehensive program and suggested that even more concerted efforts are required to encourage this group to quit.

Impact on Cigarette Consumption and Uptake

Consumption During Non-work Hours

A related concern is the impact of smoking restrictions on cigarette consumption during non-work hours. One study recorded the number of cigarettes smoked on workdays and leisure days. It found that total or partial bans resulted in smokers consuming five fewer cigarettes per day, with no apparent compensation during leisure time (Wakefield, Wilson, Owen, Esterman and Roberts, 1992).

In another study, a small group of smokers (n=34) in a medical centre where smoking had been banned were compared with a control group of smokers in an unrestricted worksite. Physiological tests were used to measure smoking amount and toxic exposure (Brigham, Gross, Stitzer and Felch, 1994). The key finding was that smokers did not compensate for the pack-per-week reduction in smoking by either increasing their cigarette consumption outside work or smoking more intensively. However, the overall health benefits are limited, given that the decline in cotinine levels among smokers in the non-smoking worksite was not statistically significant.

Smoking Uptake

Far less is known about the role of smoking restrictions and bans on smoking uptake. This is a crucial question in those industries with high proportions of teenage employees -- such as fast-food outlets, retail stores and other sectors of the consumer service industry. As Burns, Axelrad, Bal et al. (1992:S16) state: "The large fraction of smokers who became regular or dependent smokers at ages when they would be entering the workforce raises questions about the role of socialization into the workforce and workplace smoking norms on the imitation of smoking and the development of nicotine dependence." This casts new light on the employer's responsibility to provide a healthy and safe work environment. While this concern should add support to smoking bans in areas of high teenage employment, scientific evidence is required to document a connection between workplace smoking bans and smoking uptake among teenagers.

Top of Page

Occupational Influences

Most of the major case studies of the impact of workplace smoking bans or restrictions on employee smoking behaviour have been limited to government offices, large corporations or the health care sector. Because the majority of workers in these settings are in white-collar jobs, less is known about the effects of worksite smoking restrictions on blue-collar workers (Wakefield, Wilson, Owen, Esterman and Roberts, 1992:693). Australian research on the impact of workplace restrictions on cigarette consumption with a representative sample of the adult population found that those facing no restrictions on workplace smoking tended to be smokers, male and in lower-status occupations (Wakefield, Wilson, Owen, Esterman and Roberts, 1992:695).

Occupational factors also influence the likelihood of a worker quitting smoking after the introduction of a workplace smoking ban. Borland, Owen, Hill and Schofield (1991) note that while a number of studies have documented a reduction in workday cigarette consumption associated with bans, there is less evidence about smoking cessation associated with bans. These researchers use a two-stage model (making the attempt to quit, maintaining the attempt) that combines cognitive, behavioural and environmental variables. Their findings suggest that desire to quit and self-efficacy are strongly related to attempting to quit, while strength of addiction was the best predictor of success, along with social support and a higher level of education. All of these factors are intertwined with occupational conditions. Hence, the researchers suggest that future cessation efforts and related research should focus on blue-collar workers. More generally, they recommend that future smoking cessation studies focus on maintenance of non-smoking among those who attempt to quit.

Public Support for Workplace Smoking Restrictions

The vast majority of Canadian adults (90%) believe that smoking is harmful to smokers' and non-smokers' health (Health Canada, 1994). Moreover, there is no doubt that the public supports legislated restrictions on smoking.

Support for Total Bans

A 1995 Angus Reid poll in the Greater Vancouver Area found that 66% of respondents support a municipal by-law that would ban smoking in all indoor places in their community. While support was 79% among non-smokers, only 30% of smokers were supportive. Individuals with higher levels of education and children at home were the most supportive (Angus Reid Group, 1995).

A 1991 survey of Ontarians documents majority support among non-smokers and smokers alike for the enactment and enforcement of smoking restrictions by local governments (Ashley, Bull and Pederson, 1994). However, only 35.8% of non-smokers and 14.2% of smokers agreed that smoking should be prohibited in the workplace. Interestingly, almost equivalent proportions supported total smoking bans in restaurants (35.9% and 9.9%, respectively). Support for workplace and restaurant smoking bans is well below the 85-90% levels of support for smoking bans in city buses, doctors' offices and day-care centres. A large majority (72.5%) of smokers said they would "go along with the rules"; only 12% would ignore prohibitions.

ETS Policy Implications

These findings led the researchers to conclude that enforcement should not be a major problem and that citizens place responsibility for enacting and enforcing restrictions on local governments. Most relevant for our purpose -- but keeping in mind that this study was conducted four years ago and that attitudes have been changing (e.g., Pederson, Bull, Ashley and Kozma, 1992) -- is the conclusion that there is a low level of support for workplace and restaurant restrictions.

Socio-demographic Variations in Attitudes and Knowledge

Another salient issue is how knowledge, attitudes and compliance to smoking regulations vary considerably by socio-economic status and smoking status. Smokers' greater opposition to smoking restrictions is influenced by their socio-economic status. Recall that smoking is most prevalent among the lowest socio-demographic groups (measured by occupation, education and income). Similarly, relevant public opinion research (e.g., Krahn and van Roosmalen, 1991; Pederson, Wanklin, Bull and Ashley, 1991; Makkai, McAllister and Goodin, 1994) identifies lower socio-economic groups as being the least knowledge-able about the health effects of smoking and less supportive of smoking restrictions.

An Ontario study found that the smokers who were least likely to support restrictions were younger, were less knowledgeable about the health effects of smoking, tended not to be married, were less likely to attend church regularly and tended to have a stronger smoking habit (Pederson, Wanklin, Bull and Ashley, 1991:106). There is reason to believe that different factors influence smokers' specific attitudes, knowledge and behaviour about smoking (Pederson, Bull, Ashley and Lefcoe, 1989a, 1989b). Nonetheless, current educational programs about ETS probably have minimal impact where they are most needed -- "the perennial problem of health education campaigns" (Makkai, McAllister and Goodin, 1994:425; see also Bull, Pederson and Ashley, 1994:104).

ETS Policy Implications

This research suggests a ground-up approach to legislation, whereby smokers' opinions about the acceptability of proposed laws are used in planning educational campaigns aimed at this group (Pederson, Wanklin, Bull and Ashley, 1991:109-10).

Union Perspectives on ETS

General Issues

Little has changed in the past five years regarding the basic stance of unions toward workplace smoking. This was confirmed in recent telephone discussions with six union officials. The central argument made in Lowe and Neale (1992) was this: Despite Canadian public sector unions having played a central role in shaping federal legislation, unions do not, on the whole, view ETS as a high priority. Many unions would actually prefer to see ETS dealt with through employer policies or government legislation, as long as the unions are fully consulted and the interests of their smoking members are not jeopardized. Industrial unions, in particular, object to employers' and governments' predisposition to single out workers' smoking behaviour, when exposure to other worksite hazards poses more immediate risks.

The Canadian Labour Congress, the national umbrella organization for unions, voiced two concerns regarding ETS. One concern is that the discrepancies found across the country in smoking restrictions reveal the need for national standards. The other concern is workers' ETS exposure in the hospitality industry. More generally, the Canadian Labour Congress would like to see the major public health issue of ETS become linked to a broader concern about pollutants in the work environment.

Major Industrial Unions

Two major industrial unions -- the Canadian Auto Workers (CAW) and the United Steel Workers (USW) -- increasingly represent workers in the service sector. The CAW is very concerned about the high levels of ETS exposure among its members in the hospitality industry. Indeed, this was a major issue in the 1994 strike at the Windsor Casino. Casino employees were prohibited from smoking in the workplace, while having to breathe excessive amounts of ETS from customers. The workers sought to negotiate improved ventilation, a designated smoking room and a direct say in designing the new casino so that state-of-the-art ventilation was included. The CAW has worked closely with public health organizations such as the Canadian Cancer Society to educate members and uses literature from the Non-smokers' Rights Association. However, it will not join publicly with these organizations for fear of alienating members who are smokers.

The USW has only recently begun to consider ETS as a health issue for its members. While the union has a smoking policy for its offices and meetings, it has not mounted an anti-smoking campaign among its members. Put simply, too many members are smokers, and too many of these smokers (notably miners and smelter workers) face equally serious workplace health hazards. A union official gave two examples. When smelter workers whose lead exposure levels were well above acceptable levels had smoking bans imposed by their employer, their typical response was a rather incredulous, "What, so they want me to quit smoking?" And in response to a recent Ontario Ministry of Labour report recommending that uranium miners be told to stop smoking, the union argued that the chronic problem is exposure to radon and uranium. At the same time, the USW has not resisted employers' attempts to introduce smoking policies.

Hospitality Sector Unions

Discussions with the two unions that traditionally have represented workers in hotels, restaurants and bars did not yield as promising an outlook. Officials of the HERE and the United Food and Commercial Workers (UFCW) took the position that ETS exposure was not a problem for their members. A UFCW official recognized the health risks of ETS, but the decentralized organization of the union gives considerable autonomy to locals, none of which had raised such concerns.

The HERE opposes Vancouver's proposed 100% smoke-free by-law (Strategic Action Group, 1995). Vancouver Health Department officials admitted that union involvement in the public consultation process was disappointing. In the only union brief submitted to the public consultations, the HERE based its opposition on concerns for its members' jobs, asserting that a ban would be bad for business (Strategic Action Group, 1995). It also noted that many customers are tourists whose cultural differences (i.e., preference for smoking) must be respected. The union claimed that its members could find themselves having to enforce smoking bans on behalf of management. The union's solution is to set up clearly marked smoking areas and use "technology" to minimize non-smokers' ETS exposure. Elsewhere, ETS seems to be a non-issue for the HERE. The president of the union's Alberta local responded to a question about ETS by saying, "We're all smokers here . . . the issue is freedom of choice: you can either smoke or not smoke."

Top of Page

ETS Policy Implications

Both of the large industrial unions, as they diversify further into the service sector, could play a role in bringing about reductions in smoking rates and ETS exposure in this sector of the labour force. The traditional hospitality sector unions, such as UFCW and HERE, will resist becoming involved in smoking control initiatives. Still, it is important for policy-makers to understand each union's position on ETS, recognizing that they must balance the interests of non-smokers and smokers among their membership. ETS will never be as salient an issue for unions as it is for public health policy-makers.

The Economics of Workplace Smoking Restrictions

Despite employers' proverbial focus on "the bottom line," most who have introduced smoking policies claim that they have been motivated by a concern for their employees' health (Douville, 1990:106). This may hold true more for those organizations that voluntarily impose smoking restrictions. Certainly, employer opposition to legislated restrictions is firmly grounded in economics.

Costs of Separately Ventilated Smoking Rooms

Critics of independent ventilation argue that it is expensive for small employers and that it will not help employees in restaurants and bars (Ontario Campaign for Action on Tobacco, 1995:9). More generally, some experts argue that the provision of designated smoking rooms is contrary to the objectives of public health policy, is costly, provides an amenity only for smokers and could be the basis for liability action because the employer has created an unsafe environment (Hocking, Borland, Owen and Kemp, 1991).

Increased Cancer Risk

A strong argument against separately ventilated smoking rooms is that they significantly increase lung cancer mortality risks among smokers (Siegel, Husten, Merritt, Giovino and Eriksen, 1995). Put simply, a total workplace ban will prevent many premature deaths among smokers. Siegel and his colleagues caution that there is limited research on the potential health effects of ETS on smokers and the actual level of exposure in smoking lounges. It is not clear, for example, whether the increased cancer risk they calculate is due to exposure to ETS in lounges or to a higher incidence of smoking. However, most employers are unlikely to be convinced by the economic implications of this argument.

Psychological Factors

Social psychologists identify possible benefits of designated smoking rooms, although their argument could be extended to support total bans.

A review of the psychological literature on smoker/non-smoker interaction concludes that these groups perceive each other negatively and that this affects the work performance of non-smokers: "Nonsmokers hold negative stereotypes of smokers, suffer from depressed mood states when near a smoker, perform worse when around smokers, are more aggressive toward smokers, help smokers less than nonsmokers, and require more interpersonal distance when interacting with smokers" (Gibson, 1994:1082).

Differential Treatment of Smokers

Issues in U.S. Employment Law

The American employment law literature has focused attention lately on two related issues: employer discrimination against smokers in hiring and firing decisions, and differential health and life insurance charges levied on smokers (Sculco, 1992; Short, 1992; Slade, 1993; Warner, 1994). Considerable publicity surrounds some of these policies, such as the Turner Broadcasting System's off-duty smoking ban and non-smokers-only hiring policy (Business Week, 1991). In response, close to half of the state legislatures have enacted smokers' rights laws that prohibit employers from refusing to hire smokers or from taking adverse actions against employees who smoke during non-work hours. Such measures are actively supported by the tobacco industry (Vaughn, 1992; Slade, 1993).

Whether these policies are a blatant infringement on smokers' rights and a form of discrimination is hotly debated. On one side, civil libertarians and unions view these attempts by employers to dictate individual behaviour as discriminatory and an invasion of individual privacy. They point out that smoking is not illegal and that smokers have an addiction. On the other side of the debate, some researchers conclude that this is only good social policy (Warner, 1994:137). Sidestepping for now the legal and ethical principles at stake, what is of interest for our purposes is the economic rationale underpinning such policies. Apparently, employers' economic analysis leads them to conclude that a range of higher costs is associated with smoking employees.

Lawsuits

Using the same logic, some U.S. private sector employers are voluntarily implementing smoking bans as a way of avoiding lawsuits from non-smoking employees claiming that their health has been impaired by exposure to ETS. While this is a theme in the employment law literature, it is difficult to say how pervasive the concern actually is (Bowers, 1992; cf. Vaughn, 1992).

Economic Benefits

Costs of Smoking

Here are some examples of the costs to employers of having smokers on the payroll. Dow Chemical Co. discovered that one of its divisions was losing about $600,000 annually from the absenteeism of ill smokers (Sculco, 1992:883). The Congressional Office of Technology Assessment estimates that each of the approximately 15 million smokers in the United States costs their respective employers between $2,000 and $5,000 annually in increased health care and fire insurance premiums, absenteeism, lost productivity and property damage (Warner, 1994:130). The OSHA's proposed ban on smoking in indoor workplaces has the support of the Building Owners and Managers Association, which views smoking as the major cause of fires in office buildings. National data for the United States document that male and female smokers have higher absenteeism rates than non-smokers, are sicker and require more medical care (Rice, Hodgson, Sinsheimer, Browner and Kopstein, 1986).

Cost--Benefit Analysis

Both the Canadian and American governments have conducted cost-benefit analyses of federal non-smoking legislation. Labour Canada's Regulatory Impact Analysis Statement prepared for the federal Non-smokers' Health Act estimated that $32.2 million (1989) could be saved from reduced smoke and related property damage, depreciation, maintenance and cleaning costs and savings to the health care system through reduced ill-health effects of ETS exposure (Canada Gazette, 1989:4540). Setting up separately ventilated smoking rooms was projected to cost $19.77 million during 1990, the first year of the Act.

The U.S. EPA assessed the impact of the proposed Smoke-Free Environment Act (U.S. Environmental Protection Agency, 1994). The bill would ban or restrict smoking in all non-residential indoor air environments. The main conclusion of the cost-benefit analysis was that the legislation would produce net benefits of between $39 and $72 billion. The report estimates that only 10-20% of buildings would construct separate smoking lounges, owing mainly to cost and feasibility. These smoking lounges would cost between $0.3 and $0.7 billion. While the study did not establish scientifically that ETS exposure reduces worker productivity, it did predict increased organizational efficiency due to reduced conflicts between smokers and non-smokers. Reduced absenteeism would also boost productivity, as, compared with non-smokers, smokers have about 50% more workdays lost, and former smokers about 30% more. Neither of these cost-benefit analyses assessed the enhanced quality of life accruing from reduced smoking or the reduced exposure of non-smokers to ETS.

Impact of Smoking Restrictions on Restaurant and Bar Revenues

American Studies

Opposition to non-smoking legislation in the restaurant industry rests on concerns about business loss. Perhaps the most comprehensive study of the economic impact of smoking restrictions on bar and restaurant sales is that by Glantz and Smith (1994). They examined 15 cities with smoke-free restaurant ordinances and 15 similar non-smoke-free control cities. Using sales tax data records reported to California and Colorado states, they calculated two ratios: restaurant sales as a percentage of total retail sales, and restaurant sales in smoke-free versus control cities. The study found no statistically significant effect of local non-smoking ordinances on either ratio.

Thus, smoke-free ordinances had no effect on revenues, refuting tobacco industry claims, in its campaign to rescind the Beverly Hills smoke-free restaurant ordinance, that business declined by 30%. The findings are replicated by Maroney, Sherwood and Stubblebine (1994), who examined sales tax data in 19 cities, 10 with partial restrictions and 9 that are 100% smoke-free.

Canadian Evidence

The only Canadian attempt to assess the impact of smoking restrictions on restaurants was the City of Toronto's evaluation of its 1993 by-law. It found that the setting aside of 50% of restaurant seating for non-smoking areas (which is a minimal restriction) had no negative impact on the restaurant business (Kendall, 1994). This assessment did not examine other economic costs or benefits.

The Vancouver Medical Officer of Health argues that the benefits of smoking bans in restaurants have been understated (City of Vancouver, 1994:6). Patron waiting time will be reduced with the abolition of smoking sections, maintenance costs will decrease and staff will not lose as much worktime as a result of respiratory illnesses. These points are merely asserted, with no supporting evidence. The report provides a brief discussion of the environmental and financial impacts of the proposed by-law for 100% smoke-free indoor environments, citing improved health and quality of life, reduced health costs, reduced employer liabilities and reduced building maintenance and fire risk.

Top of Page

An Angus Reid Group (1995) survey of B.C. Lower Mainland and Victoria residents addressed the issue of potential business loss if smoking were to be banned in restaurants and bars. It found that such a ban would actually increase patronage in restaurants. Among frequent and occasional patrons, 22% expected to go to restaurants more often, compared with 13% who would go less often. The by-law would not affect the patronage of drinking or gaming establishments. One issue is the effect of weaker regulations in neighbouring municipalities. The survey found that most smokers were not willing to travel more than 30 minutes to smoke in a bar.

Conclusion and Recommendations

Having reviewed the current state of research on workplace smoking restrictions and evaluated existing legislation, we now can suggest future research and policy directions for Canada. While Canada has made considerable progress in reducing workplace ETS exposure, the goals of the NSTRTU have been achieved in only a few industries and occupations. Blue-collar occupations and huge areas of the service sector have smoking rates well above average. Small firms, which employ the majority of Canadian workers, are less affected by smoking regulations. Moreover, total bans on ETS are not yet the standard for regulation. Generally, workers comply with smoking legislation or policies. However, achieving greater reductions in ETS exposure will require enforcement strategies aimed at non-conforming groups.

An integrated ETS reduction strategy must involve government, employers, unions and public health organizations. Uniform minimum standards are also essential and should be established at the provincial level. However, most progress in achieving smoke-free workplaces is now occurring at the municipal level. The use of enhanced OHS legislation may also be a useful strategy to reduce ETS in high-exposure worksites.

Recommendations for Further Research

  1. A more comprehensive analysis of industrial and occupational patterns of smoking is needed. This should attempt to link smoking prevalence with workers' exposure to other workplace health hazards.
  2. We require a better understanding of why members of specific blue-collar occupations and service occupations have high smoking rates. What are the socio-demographic correlates, and how do the organizational and "cultural" characteristics of these workplaces reinforce smoking behaviour? It may also be informative to examine why the majority of workers in these occupations do not smoke and if occupational conditions had any influence on the decision of former smokers to quit.
  3. We know little about how the characteristics of small workplaces contribute to a higher incidence of smoking among these firms' employees. In addition, research on the conditions under which small firms successfully go smoke-free would help in designing more effective intervention strategies.
  4. Why do female workers in Quebec have the highest smoking prevalence, compared with male or female workers in other provinces?
  5. Do provincial and rural-urban variations in workplace ETS policies simply reflect differences in regulatory environments, or are there more complex socio-economic factors that underlie this?
  6. As Canadian municipalities introduce 100% smoke-free policies for restaurants and bars, it is essential to systematically evaluate the economic impact of these policies. This requires designing evaluation research programs before the by-laws are in place to obtain pre-by-law data.
  7. What are the implications for patterns of ETS exposure as teleworking becomes more common? Will smokers smoke more at home?
  8. Research on the effects of worksite smoking cessation programs is inconclusive, showing modest benefits at best. Canadian workplaces with different types of cessation programs should be monitored to determine what ingredients are most effective in maintaining quit rates.

Policy Recommendations

  1. Minimum federal and provincial standards for ETS exposure are required.
  2. Amendments to existing laws and any new legislation should aim for a comprehensive and consistent definition of "workplace."
  3. Specifically, municipal by-laws enforcing 100% smoke-free bans in workplaces should be encouraged to include all workplaces. This would cover the hospitality sector.
  4. The self-enforcement provisions of most Canadian smoking restrictions need to be strengthened. Public opinion and government fiscal constraints militate against setting up state-enforced regulatory frameworks. It is therefore essential to consider how self-enforcement can be improved to achieve 100% compliance.
  5. The hospitality industry resists smoking regulations because of concerns about business loss. These concerns could be addressed with factual information campaigns and consultations (being cognizant, of course, of the fact that the tobacco industry is directly backing this opposition).
  6. Educational and information campaigns have helped to convince a large majority of Canadians that ETS is a major health hazard. Future campaigns must reach beyond this majority to the hard-core smoker -- that is, heavy users (Mecklenburg, 1994:11). These individuals are most likely to resist smoking restrictions in the workplace.
  7. Unions are potential allies, albeit playing a background role, in achieving NSTRTU objectives. However, they are unlikely to accept a limited focus on ETS. Broader strategies that address ETS as one of many occupational health hazards are more consistent with the union position.
  8. Occupational health and safety legislation has the potential to address ETS, but careful analysis is required of the limitations of existing legislation and the implications for ETS regulation. Workers' compensation is a related regulatory arena for addressing ETS. Again, current weaknesses and provincial variations could hamper policy actions on this front.

Acknowledgments

A number of individuals and organizations made valuable contributions to this paper. Margaret de Groh analyzed the data reported in the figures and table; without her assistance, there would be few or no "data" in the paper. Health Canada reviewers offered helpful suggestions on a detailed outline of the paper. Participants in the workshop provided insightful suggestions on an earlier draft of the paper during a very lively discussion period. Two anonymous reviewers also gave useful suggestions for improvements. The National Clearinghouse on Tobacco and Health greatly assisted me by providing research materials not usually found in a university library. Edna Djokoto helped with library research. Chris Mowat gave me timely editorial assistance. Representatives of various public health non-governmental organizations in Canada and the United States, municipal governments and unions gave freely of their time and publications. I take full responsibility for any errors or omissions.

References

  • Americans for Nonsmokers' Rights and Prospect Associates. 1994. Clean Indoor Air: A Guide to Developing Policy. June 1994.
  • Angus Reid Group. 1995. No Smoking Bylaw Survey. Submitted to the Metropolitan Board of Health of Greater Vancouver, B.C. Lung Association, Canadian Cancer Society, Heart and Stroke Foundation of B.C. and Yukon. May 1995.
  • Ashley MJ, Bull SB, Pederson LL. 1994. Restrictive Measures on Smoking in Ontario: Similarities and Differences Between Smokers and Nonsmokers in Knowledge, Attitudes and Predicted Behaviour and Implications for Tobacco Programs and Policies. Ontario Tobacco Research Unit, Toronto. Working Papers Series No. 1, 1994.
  • Benowitz NL, Leistikow B. 1994. Environmental tobacco smoke hazards and benefits of work-site nonsmoking programs. West J Med 1994;160:562-3.
  • Borland R, Owen N, Hill D, Schofield P. 1991. Predicted attempts and sustained cessation of smoking after the introduction of workplace smoking bans. Health Psychol 1991;10:336-42.
  • Bowers MH. 1992. What labour and management need to know about workplace smoking cases. Labor Law J 1992;43:40-9.
  • Brigham J, Gross J, Stitzer ML, Felch LJ. 1994. Effects of a restricted work-site smoking policy on employees who smoke. Am J Public Health 1994;84:773-8.
  • Bull SB, Pederson LL, Ashley MJ. 1994. Relationship of smoking status to changes in knowledge and attitudes concerning restrictions on smoking. Can J Public Health 1994;85:103-5.
  • Burns DM, Axelrad R, Bal D, Carol J, Davis RM, Myers MA et al. 1992. Report of the tobacco policy research study group on smoke-free indoor air policies. Tobacco Control 1992;1(suppl):S14-S18.
  • Business Week. 1991. "If you light up on Sunday, don't come in on Monday." Business Week, 26 August 1991:68-72.
  • California State Senate. 1993. Bill Analysis. T. Friedman et al., Bill 13. California State Senate Gopher Service, 1993.
  • Canada Gazette. 1989. Canada Gazette, Part I. Ottawa: Minister of Supply and Services, 1989.
  • City of Vancouver. 1994. Policy Report from Medical Officer of Health, 100% Smoke-Free Indoor Environments. 31 October 1994. Database: CC05, Document: 1994 1031:0002-00010.
  • Council for a Tobacco-Free Ontario. 1995. Local Tobacco Control: The Facts and Issues. March 1995.
  • Cunningham R. 1995. ETS and the Ontario Occupational Health and Safety Act. 30 May 1995. Unpublished manuscript.
  • Daughton DM, Andrews CE, Orona CP, Patil KD, Rennard SI. 1992. Total indoor smoking ban and smoker behaviour. Prev Med 1992;21:670-6.
    Top of Page
  • Dawley LT, Dawley HH, Glasgow RE, Rice J, Correa P. 1993. Worksite smoking control, discouragement, and cessation. Int J Addict 1993;28:719-33.
  • Douville JA. 1990. Active and Passive Smoking Hazards in the Workplace. New York: Van Nostrand Reinhold, 1990.
  • Fisher JK, Glasgow RE, Terborg JR. 1990. Work site smoking cessation: a meta-analysis of long-term quit rates from controlled studies. J Occup Med 1990;32:429-39.
  • Frank J. 1995. Preparing for the Information Highway: information technology in Canadian households. Can Soc Trends 1995;Autumn:2-7.
  • Gibson B. 1994. Psychological aspects of smoker-nonsmoker interaction: implications for public policy. Am Psychol 1994;49:1081-3.
  • Glantz SA, Smith LRA. 1994. The effect of ordinances requiring smoke-free restaurants on restaurant sales. Am J Public Health 1994;84:1081-5.
  • Globe and Mail. 1995. Globe and Mail, Toronto, 21 July 1995:A1, A5.
  • Grossman M, Price P. 1992. Tobacco Smoking and the Law in Canada. Toronto: Butterworths, 1992.
  • Hammond KS, Sorensen G, Youngstrom, R, Ockene JK. 1995. Occupational exposure to environmental tobacco smoke. J Am Med Assoc 1995;274:956-60.
  • Health Canada. 1993. Directions: The Directional Paper of the National Strategy to Reduce Tobacco Use. An Update, 1993.
  • Health Canada. 1994. Survey on Smoking in Canada, Cycle 2. Fact Sheet No. 7. Environmental Tobacco Smoke. November 1994.
  • Health Canada. 1995a. Health Canada's Workshops on Environmental Tobacco Smoke (ETS) in the Workplace. Summary of Proceedings. 17 and 23 March 1995.
  • Health Canada. 1995b. Smoking Bylaws in Canada 1995. Office of Tobacco Control, Health Protection Branch, 1995.
  • Health Promotion Survey. 1990. Microdata file. Statistics Canada, 1990.
  • Hocking B, Borland R, Owen N, Kemp G. 1991. A total smoking ban on workplace smoking is acceptable and effective. J Occup Med 1991;33:163-7.
  • Hudzinski LG, Sirois PA. 1994. Changes in smoking behaviour and body weight after implementation of a non-smoking policy in the workplace. South Med J 1994;87:322-7.
  • Hymowitz N, Campbell K, Feuerman M. 1991. Long-term smoking intervention at the worksite: effects of quit-smoking groups and an ?enriched milieu' on smoking cessation in adult white-collar employees. Health Psychol 1991;10:366-9.
  • Jarvis MJ, Foulds J, Feyerabend C. 1992. Exposure to passive smoking among bar staff. Br J Addict 1992;87:111-3.
  • Jeffery RW, Kelder SH, Forster JL, French SA, Lando HA, Baxter JE. 1994. Restrictive smoking policies in the workplace: effects on smoking prevalence and cigarette consumption. Prev Med 1994;23:78-82.
  • Kagan RA, Skolnick JH. 1993. Banning smoking: compliance without enforcement. In Rabin RL, Sugarman SD, editors. Smoking Policy: Law, Politics, and Culture. New York: Oxford University Press, 1993.
  • Kagan RA, Vogel D. 1993. The politics of smoking regulation: Canada, France, the United States. In Rabin RL, Sugarman SD, editors. Smoking Policy: Law, Politics, and Culture. New York: Oxford University Press, 1993.
  • Kendall PRW. 1994. Smoking By-laws in the City of Toronto: A Comprehensive Evaluation. City of Toronto, Department of Public Health, January 1994.
  • Kinne S, Kristal AR, White E, Hunt J. 1993. Work-site smoking policies: their population impact in Washington state. Am J Public Health 1993;83:1031-3.
  • Krahn H, Lowe G. 1993. Work, Industry and Canadian Society, 2nd edition. Scarborough: Nelson Canada, 1993.
  • Krahn H, van Roosmalen E. 1991. Smoking Policies and Public Reaction in Alberta: An Analysis of the 1990 Alberta Survey. Health and Welfare Canada, 1991.
  • Lando HA, Johnson KM, Graham-Tomasi RP, McGovern PG, Solberg L. 1992. Urban Indians' smoking patterns and interest in quitting. Public Health Rep 1992;107:340-4.
  • Letourneay G, Bujold M. 1990. Smoking in Groups at Risk: The Francophones of Québec. Health and Welfare Canada, June 1990.
  • Lichtenstein E, Glasgow RE. 1992. Smoking cessation: what have we learned over the past decade? J Consult Clin Psychol 1992;60:518-27.
  • Lindsay C, Devereaux MS, Bergob M. 1994. Youth in Canada, 2nd edition. Ottawa: Statistics Canada, 1994. Catalogue no. 89-511E.
  • Lowe GS, Neale DJ. 1992. Unions and Workplace Smoking Policy. Current Issues Series. Kingston, ON: Industrial Relations Centre, Queen's University, 1992.
  • Makkai T, McAllister I, Goodin M. 1994. Public knowledge about passive smoking: results from a survey in the Australian Capital Territory. Int J Addict 1994;29:415-27.
  • Maroney N, Sherwood D, Stubblebine WC. 1994. The Impact of Tobacco Control Ordinances on Restaurant Revenues in California. Claremont, CA: The Claremont Institute for Economic Policy Studies, January 1994.
  • Maskin A, Connolly A, Noonan EA. 1993. Environmental tobacco smoke: implications for the workplace. Occupational Safety & Health Reporter, Bureau of National Affairs, 24 February 1993:1663-9.
  • Mecklenburg RE. 1994. Managing hard-core smokers: oral health team challenges and opportunities. Health Values 1994;18:6-16.
  • Millar WJ. 1988a. Smoke in the Workplace: An Evaluation of Smoking Restrictions. Ottawa: Health and Welfare Canada, Health Promotion Directorate, Health Services and Promotion Branch, 1988.
  • Millar WJ. 1988b. The Smoking Behaviour of Canadians, 1986. Ottawa: Health and Welfare Canada, Health Promotion Directorate, Health Services and Promotion Branch, 1988.
  • Millar WJ. 1988c. Evaluation of the impact of smoking restrictions in a government work setting. Can J Public Health 1988;79:379-82.
  • Millar WJ. 1992. Place of birth and ethnic status: factors associated with smoking prevalence among Canadians. Health Rep 1992;4:7-24.
  • Millar WJ, Bisch L. 1989. Smoking in the workplace 1986: Labour Force Survey estimates. Can J Public Health 1989;80:261-5.
  • Millar WJ, Hunter L. 1990. The relationship between socioeconomic status and household smoking patterns in Canada. Am J Health Promot 1990;5:36-42.
  • Morissette R. 1991. Are jobs in large firms better jobs? Perspect Labour Income 1991;Autumn:40-50.
  • National Clearinghouse on Tobacco and Health. 1995. Federal and Provincial Legislation in Canada: An Overview. Ottawa: National Clearinghouse on Tobacco and Health, March 1995.
  • National Population Health Survey. 1994-95. Microdata file. Statistics Canada, 1995.
  • Nelson DE, Sacks JJ, Addiss DG. 1993. Smoking policies of licensed child day-care centres in the United States. Pediatrics 1993;91:460-3.
  • O'Connor AM, Harrison M. 1992. Survey of smoking prevalence among Canadian nursing students and registered nurses. Can J Public Health 1992;83:417-21.
  • Ontario Campaign for Action on Tobacco. 1995. In Support of Smoke-Free Restaurants: The Benefits of Restaurant Smoking Bans, and Responses to Arguments Against Restaurant and Bar Smoking Restrictions. 7 April 1995.
  • Orser B, Foster M. 1992. Home Enterprise: Canadians and Home-Based Work. Report Prepared for The Home-Based Business Project Committee. 1992.
  • Pederson LL, Bull SB, Ashley MJ, Lefcoe NM. 1989a. A population survey on legislative measures to restrict smoking in Ontario: 3. Variables related to attitudes of smokers and nonsmokers. Am J Prev Med 1989;5:313-22.
  • Pederson LL, Bull SB, Ashley MJ, Lefcoe NM. 1989b. A population survey on legislative measures to restrict smoking in Ontario: 4. Variables related to knowledge of active and passive smoking health effects and to predicted behaviour of smokers and nonsmokers. Am J Prev Med 1989;5:323-9.
  • Pederson LL, Wanklin JB, Bull SB, Ashley MJ. 1991. A conceptual framework for the roles of legislation and education in reducing exposure to environmental tobacco smoke. Am J Health Promot 1991;6:105-11.
  • Pederson LL, Bull SB, Ashley MJ, Kozma D. 1992. Restrictions on smoking: changes in knowledge, attitudes and predicted behaviour in Metropolitan Toronto from 1983 to 1988. Can J Public Health 1992;83:408-12.
  • Pierce JP, Evans N, Farkas AJ, Cavin SW, Berry C, Kramer M et al. 1994. Tobacco Use in California. An Evaluation of the Tobacco Control Program, 1989-1993. La Jolla, CA: University of California, San Diego, 1994.
  • Quebec. 1987. An Act Respecting the Protection of Non-smokers in Certain Public Places. Chapter P.38-01. 1 March 1987.
  • Repace JL, Lowrey AH. 1985. A quantitative estimate of nonsmokers' lung cancer risk from passive smoking. Environ Int 1985;11:3-22.
  • Repace JL, Lowrey AH. 1993. An enforceable indoor air quality standard for environmental tobacco smoke in the workplace. Risk Anal 1993;13:463-75.
  • Rice DP, Hodgson TA, Sinsheimer P, Browner W, Kopstein AN. 1986. The economic costs of the health effects of smoking. Milbank Q 1986;64:489-546.
  • Rigotti NA, Stoto MA, Schelling TC. 1994. Do businesses comply with a no-smoking law? Assessing the self-enforcement approach. Prev Med 1994;23:223-9.
  • Sass R. 1986. Workplace health and safety: report from Canada. Int J Health Serv 1986;16:565-82.
  • Sculco TW. 1992. Smokers' rights legislation: should the state ?butt out' of the workplace? Boston Coll Law Rev 1992;33:879-902.
  • Short LW. 1992. The legal implications of workplace smoking policies. Employee Responsibilities Rights J 1992;5:65-73.
  • Siegel M. 1993. Involuntary smoking in the restaurant workplace. A review of employee exposure and health effects. J Am Med Assoc 1993;270:490-3. [Also see Comment in J Am Med Assoc 1994;271:584-5.]
  • Siegel M, Husten C, Merritt RK, Giovino GA, Eriksen MP. 1995. Effects of separately ventilated smoking lounges on the health of smokers: is this an appropriate public health policy? Tobacco Control 1995;4:22-9.
  • Slade J. 1993. Protection from job bias for people who smoke. J Substance Abuse Treatment 1993;10:229-31.
  • Sorensen G, Rigotti N, Rosen A, Pinney J, Prible R. 1991. Effects of a worksite nonsmoking policy: evidence for increased cessation. Am J Public Health 1991;81:202-4.
  • Sterling T, Weinkam J. 1990. The confounding of occupation and smoking and its consequences. Soc Sci Med 1990;30:457-67.
  • Stillman FA, Becker DM, Swank RT, Hantula D, Moses H, Glantz S, Waranch HR. 1990. Ending smoking at the Johns Hopkins Medical Institutions. J Am Med Assoc 1990;264:1565-9.
  • Strategic Action Group. 1995. 100% Smoking By-law Initiative: Consultation Report. 3 August 1995.
  • Sugarman SD. 1993. Disparate treatment of smokers in employment and insurance. In Rabin RL, Sugarman SD, editors. Smoking Policy: Law, Politics, and Culture. New York: Oxford University Press, 1993.
  • Survey on Smoking in Canada. 1994. Cycle 2 Data. Statistics Canada, 1994.
  • U.S. Environmental Protection Agency. 1993. Respiratory Health Effects of Passive Smoking: Lung Cancer and Other Disorders (EPA/600/6-90/006F). Washington, DC, 1993.
  • U.S. Environmental Protection Agency. 1994. The Costs and Benefits of Smoking Restrictions. An Assessment of the Smoke-Free Environment Act of 1993. Office of Radiation and Indoor Air, Washington, DC, April 1994.
  • Vaughn DH. 1992. Smoking in the workplace: a management perspective. Employee Relations Law J 1992;18:123-39.
  • Wakefield MA, Wilson D, Owen N, Esterman A, Roberts L. 1992. Workplace smoking restrictions, occupational status, and reduced cigarette consumption. J Occup Med 1992;34:693-7.
  • Walters V, Haynes T. 1988. Workers' use and knowledge of the internal responsibility system': limits to participation in occupational health and safety. Can Public Policy 1988;14:411-23.
  • Warner D. 1994. "We do not hire smokers": may employers discriminate against smokers? Employee Responsibilities Rights J 1994;7:129-40.
  • Woodruff TJ, Rosbrook B, Pierce J, Glantz SA. 1993. Lower levels of cigarette consumption found in smoke-free workplaces in California. Arch Intern Med 1993;153:1485-93.
Date Modified: 2010-01-22

Top of Page
Important Notices